From 7682a4c624e0011b5f3e8dd3021dc54961260d97 Mon Sep 17 00:00:00 2001 From: Dean Nelson Date: Tue, 8 Aug 2006 15:03:29 -0500 Subject: [PATCH 001/115] [IA64-SGI] Silent data corruption caused by XPC V2. Jack Steiner identified a problem where XPC can cause a silent data corruption. On module load, the placement may cause the xpc_remote_copy_buffer to span two physical pages. DMA transfers are done to the start virtual address translated to physical. This patch changes the buffer from a statically allocated buffer to a kmalloc'd buffer. Dean Nelson reviewed this before posting. I have tested it in the configuration that was showing the memory corruption and verified it works. I also added a BUG_ON statement to help catch this if a similar situation is encountered. Signed-off-by: Robin Holt Signed-off-by: Dean Nelson Signed-off-by: Jack Steiner Signed-off-by: Tony Luck --- arch/ia64/sn/kernel/xpc_channel.c | 4 ++-- arch/ia64/sn/kernel/xpc_main.c | 28 ++++++++++++++++------------ arch/ia64/sn/kernel/xpc_partition.c | 24 ++++++++---------------- include/asm-ia64/sn/xp.h | 22 ++++++++++++++++++---- include/asm-ia64/sn/xpc.h | 4 +++- 5 files changed, 47 insertions(+), 35 deletions(-) diff --git a/arch/ia64/sn/kernel/xpc_channel.c b/arch/ia64/sn/kernel/xpc_channel.c index c2f69f7942a..1f3540826e6 100644 --- a/arch/ia64/sn/kernel/xpc_channel.c +++ b/arch/ia64/sn/kernel/xpc_channel.c @@ -279,8 +279,8 @@ xpc_pull_remote_cachelines(struct xpc_partition *part, void *dst, return part->reason; } - bte_ret = xp_bte_copy((u64) src, (u64) ia64_tpa((u64) dst), - (u64) cnt, (BTE_NORMAL | BTE_WACQUIRE), NULL); + bte_ret = xp_bte_copy((u64) src, (u64) dst, (u64) cnt, + (BTE_NORMAL | BTE_WACQUIRE), NULL); if (bte_ret == BTE_SUCCESS) { return xpcSuccess; } diff --git a/arch/ia64/sn/kernel/xpc_main.c b/arch/ia64/sn/kernel/xpc_main.c index 5e8e59efb34..4d026f9dd98 100644 --- a/arch/ia64/sn/kernel/xpc_main.c +++ b/arch/ia64/sn/kernel/xpc_main.c @@ -1052,6 +1052,8 @@ xpc_do_exit(enum xpc_retval reason) if (xpc_sysctl) { unregister_sysctl_table(xpc_sysctl); } + + kfree(xpc_remote_copy_buffer_base); } @@ -1212,24 +1214,20 @@ xpc_init(void) partid_t partid; struct xpc_partition *part; pid_t pid; + size_t buf_size; if (!ia64_platform_is("sn2")) { return -ENODEV; } - /* - * xpc_remote_copy_buffer is used as a temporary buffer for bte_copy'ng - * various portions of a partition's reserved page. Its size is based - * on the size of the reserved page header and part_nasids mask. So we - * need to ensure that the other items will fit as well. - */ - if (XPC_RP_VARS_SIZE > XPC_RP_HEADER_SIZE + XP_NASID_MASK_BYTES) { - dev_err(xpc_part, "xpc_remote_copy_buffer is not big enough\n"); - return -EPERM; - } - DBUG_ON((u64) xpc_remote_copy_buffer != - L1_CACHE_ALIGN((u64) xpc_remote_copy_buffer)); + + buf_size = max(XPC_RP_VARS_SIZE, + XPC_RP_HEADER_SIZE + XP_NASID_MASK_BYTES); + xpc_remote_copy_buffer = xpc_kmalloc_cacheline_aligned(buf_size, + GFP_KERNEL, &xpc_remote_copy_buffer_base); + if (xpc_remote_copy_buffer == NULL) + return -ENOMEM; snprintf(xpc_part->bus_id, BUS_ID_SIZE, "part"); snprintf(xpc_chan->bus_id, BUS_ID_SIZE, "chan"); @@ -1293,6 +1291,8 @@ xpc_init(void) if (xpc_sysctl) { unregister_sysctl_table(xpc_sysctl); } + + kfree(xpc_remote_copy_buffer_base); return -EBUSY; } @@ -1311,6 +1311,8 @@ xpc_init(void) if (xpc_sysctl) { unregister_sysctl_table(xpc_sysctl); } + + kfree(xpc_remote_copy_buffer_base); return -EBUSY; } @@ -1362,6 +1364,8 @@ xpc_init(void) if (xpc_sysctl) { unregister_sysctl_table(xpc_sysctl); } + + kfree(xpc_remote_copy_buffer_base); return -EBUSY; } diff --git a/arch/ia64/sn/kernel/xpc_partition.c b/arch/ia64/sn/kernel/xpc_partition.c index 2a89cfce495..57c723f5cba 100644 --- a/arch/ia64/sn/kernel/xpc_partition.c +++ b/arch/ia64/sn/kernel/xpc_partition.c @@ -71,19 +71,15 @@ struct xpc_partition xpc_partitions[XP_MAX_PARTITIONS + 1]; * Generic buffer used to store a local copy of portions of a remote * partition's reserved page (either its header and part_nasids mask, * or its vars). - * - * xpc_discovery runs only once and is a seperate thread that is - * very likely going to be processing in parallel with receiving - * interrupts. */ -char ____cacheline_aligned xpc_remote_copy_buffer[XPC_RP_HEADER_SIZE + - XP_NASID_MASK_BYTES]; +char *xpc_remote_copy_buffer; +void *xpc_remote_copy_buffer_base; /* * Guarantee that the kmalloc'd memory is cacheline aligned. */ -static void * +void * xpc_kmalloc_cacheline_aligned(size_t size, gfp_t flags, void **base) { /* see if kmalloc will give us cachline aligned memory by default */ @@ -148,7 +144,7 @@ xpc_get_rsvd_page_pa(int nasid) } } - bte_res = xp_bte_copy(rp_pa, ia64_tpa(buf), buf_len, + bte_res = xp_bte_copy(rp_pa, buf, buf_len, (BTE_NOTIFY | BTE_WACQUIRE), NULL); if (bte_res != BTE_SUCCESS) { dev_dbg(xpc_part, "xp_bte_copy failed %i\n", bte_res); @@ -447,7 +443,7 @@ xpc_check_remote_hb(void) /* pull the remote_hb cache line */ bres = xp_bte_copy(part->remote_vars_pa, - ia64_tpa((u64) remote_vars), + (u64) remote_vars, XPC_RP_VARS_SIZE, (BTE_NOTIFY | BTE_WACQUIRE), NULL); if (bres != BTE_SUCCESS) { @@ -498,8 +494,7 @@ xpc_get_remote_rp(int nasid, u64 *discovered_nasids, /* pull over the reserved page header and part_nasids mask */ - - bres = xp_bte_copy(*remote_rp_pa, ia64_tpa((u64) remote_rp), + bres = xp_bte_copy(*remote_rp_pa, (u64) remote_rp, XPC_RP_HEADER_SIZE + xp_nasid_mask_bytes, (BTE_NOTIFY | BTE_WACQUIRE), NULL); if (bres != BTE_SUCCESS) { @@ -554,11 +549,8 @@ xpc_get_remote_vars(u64 remote_vars_pa, struct xpc_vars *remote_vars) return xpcVarsNotSet; } - /* pull over the cross partition variables */ - - bres = xp_bte_copy(remote_vars_pa, ia64_tpa((u64) remote_vars), - XPC_RP_VARS_SIZE, + bres = xp_bte_copy(remote_vars_pa, (u64) remote_vars, XPC_RP_VARS_SIZE, (BTE_NOTIFY | BTE_WACQUIRE), NULL); if (bres != BTE_SUCCESS) { return xpc_map_bte_errors(bres); @@ -1239,7 +1231,7 @@ xpc_initiate_partid_to_nasids(partid_t partid, void *nasid_mask) part_nasid_pa = (u64) XPC_RP_PART_NASIDS(part->remote_rp_pa); - bte_res = xp_bte_copy(part_nasid_pa, ia64_tpa((u64) nasid_mask), + bte_res = xp_bte_copy(part_nasid_pa, (u64) nasid_mask, xp_nasid_mask_bytes, (BTE_NOTIFY | BTE_WACQUIRE), NULL); return xpc_map_bte_errors(bte_res); diff --git a/include/asm-ia64/sn/xp.h b/include/asm-ia64/sn/xp.h index 9bd2f9bf329..6f807e0193b 100644 --- a/include/asm-ia64/sn/xp.h +++ b/include/asm-ia64/sn/xp.h @@ -60,23 +60,37 @@ * the bte_copy() once in the hope that the failure was due to a temporary * aberration (i.e., the link going down temporarily). * - * See bte_copy for definition of the input parameters. + * src - physical address of the source of the transfer. + * vdst - virtual address of the destination of the transfer. + * len - number of bytes to transfer from source to destination. + * mode - see bte_copy() for definition. + * notification - see bte_copy() for definition. * * Note: xp_bte_copy() should never be called while holding a spinlock. */ static inline bte_result_t -xp_bte_copy(u64 src, u64 dest, u64 len, u64 mode, void *notification) +xp_bte_copy(u64 src, u64 vdst, u64 len, u64 mode, void *notification) { bte_result_t ret; + u64 pdst = ia64_tpa(vdst); - ret = bte_copy(src, dest, len, mode, notification); + /* + * Ensure that the physically mapped memory is contiguous. + * + * We do this by ensuring that the memory is from region 7 only. + * If the need should arise to use memory from one of the other + * regions, then modify the BUG_ON() statement to ensure that the + * memory from that region is always physically contiguous. + */ + BUG_ON(REGION_NUMBER(vdst) != RGN_KERNEL); + ret = bte_copy(src, pdst, len, mode, notification); if (ret != BTE_SUCCESS) { if (!in_interrupt()) { cond_resched(); } - ret = bte_copy(src, dest, len, mode, notification); + ret = bte_copy(src, pdst, len, mode, notification); } return ret; diff --git a/include/asm-ia64/sn/xpc.h b/include/asm-ia64/sn/xpc.h index b72af597878..35e1386f37a 100644 --- a/include/asm-ia64/sn/xpc.h +++ b/include/asm-ia64/sn/xpc.h @@ -683,7 +683,9 @@ extern struct xpc_vars *xpc_vars; extern struct xpc_rsvd_page *xpc_rsvd_page; extern struct xpc_vars_part *xpc_vars_part; extern struct xpc_partition xpc_partitions[XP_MAX_PARTITIONS + 1]; -extern char xpc_remote_copy_buffer[]; +extern char *xpc_remote_copy_buffer; +extern void *xpc_remote_copy_buffer_base; +extern void *xpc_kmalloc_cacheline_aligned(size_t, gfp_t, void **); extern struct xpc_rsvd_page *xpc_rsvd_page_init(void); extern void xpc_allow_IPI_ops(void); extern void xpc_restrict_IPI_ops(void); From 3a5ff61c18659443f76bad6cf06f60103046de5d Mon Sep 17 00:00:00 2001 From: Steve French Date: Fri, 14 Jul 2006 22:37:11 +0000 Subject: [PATCH 002/115] [CIFS] Do not time out posix brl requests when using new posix setfileinfo request and do not time out slow requests to a server that is still responding well to other threads Suggested by jra of Samba team Signed-off-by: Steve French (cherry picked from 89b57148115479eef074b8d3f86c4c86c96ac969 commit) --- fs/cifs/CHANGES | 6 ++++++ fs/cifs/cifsfs.c | 6 +----- fs/cifs/cifsfs.h | 2 +- fs/cifs/cifsglob.h | 3 ++- fs/cifs/cifssmb.c | 11 ++++++++--- fs/cifs/connect.c | 17 ++++++++++++++++- fs/cifs/file.c | 6 ++---- fs/cifs/transport.c | 16 +++++++++++++--- 8 files changed, 49 insertions(+), 18 deletions(-) diff --git a/fs/cifs/CHANGES b/fs/cifs/CHANGES index a61d17ed182..acb843b9bc3 100644 --- a/fs/cifs/CHANGES +++ b/fs/cifs/CHANGES @@ -1,3 +1,9 @@ +Version 1.45 +------------ +Do not time out lockw calls when using posix extensions. Do not +time out requests if server still responding reasonably fast +on requests on other threads + Version 1.44 ------------ Rewritten sessionsetup support, including support for legacy SMB diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index c28ede59994..3cd750029be 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -402,7 +402,6 @@ static struct quotactl_ops cifs_quotactl_ops = { }; #endif -#ifdef CONFIG_CIFS_EXPERIMENTAL static void cifs_umount_begin(struct vfsmount * vfsmnt, int flags) { struct cifs_sb_info *cifs_sb; @@ -422,7 +421,7 @@ static void cifs_umount_begin(struct vfsmount * vfsmnt, int flags) tcon->tidStatus = CifsExiting; up(&tcon->tconSem); - /* cancel_brl_requests(tcon); */ + /* cancel_brl_requests(tcon); */ /* BB mark all brl mids as exiting */ /* cancel_notify_requests(tcon); */ if(tcon->ses && tcon->ses->server) { @@ -438,7 +437,6 @@ static void cifs_umount_begin(struct vfsmount * vfsmnt, int flags) return; } -#endif static int cifs_remount(struct super_block *sb, int *flags, char *data) { @@ -457,9 +455,7 @@ struct super_operations cifs_super_ops = { unless later we add lazy close of inodes or unless the kernel forgets to call us with the same number of releases (closes) as opens */ .show_options = cifs_show_options, -#ifdef CONFIG_CIFS_EXPERIMENTAL .umount_begin = cifs_umount_begin, -#endif .remount_fs = cifs_remount, }; diff --git a/fs/cifs/cifsfs.h b/fs/cifs/cifsfs.h index 8f75c6f2470..39ee8ef3bde 100644 --- a/fs/cifs/cifsfs.h +++ b/fs/cifs/cifsfs.h @@ -100,5 +100,5 @@ extern ssize_t cifs_getxattr(struct dentry *, const char *, void *, size_t); extern ssize_t cifs_listxattr(struct dentry *, char *, size_t); extern int cifs_ioctl (struct inode * inode, struct file * filep, unsigned int command, unsigned long arg); -#define CIFS_VERSION "1.44" +#define CIFS_VERSION "1.45" #endif /* _CIFSFS_H */ diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h index 6d7cf5f3bc0..39b43e6a750 100644 --- a/fs/cifs/cifsglob.h +++ b/fs/cifs/cifsglob.h @@ -158,7 +158,8 @@ struct TCP_Server_Info { /* 16th byte of RFC1001 workstation name is always null */ char workstation_RFC1001_name[SERVER_NAME_LEN_WITH_NULL]; __u32 sequence_number; /* needed for CIFS PDU signature */ - char mac_signing_key[CIFS_SESS_KEY_SIZE + 16]; + char mac_signing_key[CIFS_SESS_KEY_SIZE + 16]; + unsigned long lstrp; /* when we got last response from this server */ }; /* diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index 19678c575df..c03c42ee9eb 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -1484,6 +1484,7 @@ CIFSSMBPosixLock(const int xid, struct cifsTconInfo *tcon, char *data_offset; struct cifs_posix_lock *parm_data; int rc = 0; + int timeout = 0; int bytes_returned = 0; __u16 params, param_offset, offset, byte_count, count; @@ -1503,7 +1504,6 @@ CIFSSMBPosixLock(const int xid, struct cifsTconInfo *tcon, pSMB->MaxSetupCount = 0; pSMB->Reserved = 0; pSMB->Flags = 0; - pSMB->Timeout = 0; pSMB->Reserved2 = 0; param_offset = offsetof(struct smb_com_transaction2_sfi_req, Fid) - 4; offset = param_offset + params; @@ -1529,8 +1529,13 @@ CIFSSMBPosixLock(const int xid, struct cifsTconInfo *tcon, (((char *) &pSMB->hdr.Protocol) + offset); parm_data->lock_type = cpu_to_le16(lock_type); - if(waitFlag) + if(waitFlag) { + timeout = 3; /* blocking operation, no timeout */ parm_data->lock_flags = cpu_to_le16(1); + pSMB->Timeout = cpu_to_le32(-1); + } else + pSMB->Timeout = 0; + parm_data->pid = cpu_to_le32(current->tgid); parm_data->start = cpu_to_le64(pLockData->fl_start); parm_data->length = cpu_to_le64(len); /* normalize negative numbers */ @@ -1542,7 +1547,7 @@ CIFSSMBPosixLock(const int xid, struct cifsTconInfo *tcon, pSMB->hdr.smb_buf_length += byte_count; pSMB->ByteCount = cpu_to_le16(byte_count); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, - (struct smb_hdr *) pSMBr, &bytes_returned, 0); + (struct smb_hdr *) pSMBr, &bytes_returned, timeout); if (rc) { cFYI(1, ("Send error in Posix Lock = %d", rc)); } else if (get_flag) { diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index 876eb9ef85f..c4aedcf5c92 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -612,6 +612,10 @@ multi_t2_fnd: #ifdef CONFIG_CIFS_STATS2 mid_entry->when_received = jiffies; #endif + /* so we do not time out requests to server + which is still responding (since server could + be busy but not dead) */ + server->lstrp = jiffies; break; } } @@ -1969,7 +1973,18 @@ cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, } cFYI(1,("Negotiate caps 0x%x",(int)cap)); - +#ifdef CONFIG_CIFS_DEBUG2 + if(cap & CIFS_UNIX_FCNTL_CAP) + cFYI(1,("FCNTL cap")); + if(cap & CIFS_UNIX_EXTATTR_CAP) + cFYI(1,("EXTATTR cap")); + if(cap & CIFS_UNIX_POSIX_PATHNAMES_CAP) + cFYI(1,("POSIX path cap")); + if(cap & CIFS_UNIX_XATTR_CAP) + cFYI(1,("XATTR cap")); + if(cap & CIFS_UNIX_POSIX_ACL_CAP) + cFYI(1,("POSIX ACL cap")); +#endif /* CIFS_DEBUG2 */ if (CIFSSMBSetFSUnixInfo(xid, tcon, cap)) { cFYI(1,("setting capabilities failed")); } diff --git a/fs/cifs/file.c b/fs/cifs/file.c index 944d2b9e092..52e2e4c8794 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -644,8 +644,7 @@ int cifs_lock(struct file *file, int cmd, struct file_lock *pfLock) account for negative length which we can not accept over the wire */ if (IS_GETLK(cmd)) { - if(experimEnabled && - (cifs_sb->tcon->ses->capabilities & CAP_UNIX) && + if((cifs_sb->tcon->ses->capabilities & CAP_UNIX) && (CIFS_UNIX_FCNTL_CAP & le64_to_cpu(cifs_sb->tcon->fsUnixInfo.Capability))) { int posix_lock_type; @@ -683,8 +682,7 @@ int cifs_lock(struct file *file, int cmd, struct file_lock *pfLock) FreeXid(xid); return rc; } - if (experimEnabled && - (cifs_sb->tcon->ses->capabilities & CAP_UNIX) && + if ((cifs_sb->tcon->ses->capabilities & CAP_UNIX) && (CIFS_UNIX_FCNTL_CAP & le64_to_cpu(cifs_sb->tcon->fsUnixInfo.Capability))) { int posix_lock_type; diff --git a/fs/cifs/transport.c b/fs/cifs/transport.c index 17ba329e2b3..95e23ca670a 100644 --- a/fs/cifs/transport.c +++ b/fs/cifs/transport.c @@ -444,8 +444,9 @@ SendReceive2(const unsigned int xid, struct cifsSesInfo *ses, if(timeout != MAX_SCHEDULE_TIMEOUT) { timeout += jiffies; wait_event(ses->server->response_q, - (!(midQ->midState & MID_REQUEST_SUBMITTED)) || - time_after(jiffies, timeout) || + (!(midQ->midState & MID_REQUEST_SUBMITTED)) || + (time_after(jiffies, timeout) && + time_after(jiffies, ses->server->lstrp + HZ)) || ((ses->server->tcpStatus != CifsGood) && (ses->server->tcpStatus != CifsNew))); } else { @@ -710,9 +711,18 @@ SendReceive(const unsigned int xid, struct cifsSesInfo *ses, /* No user interrupts in wait - wreaks havoc with performance */ if(timeout != MAX_SCHEDULE_TIMEOUT) { timeout += jiffies; + /* although we prefer not to time out if the server is still + responding - we will time out if the server takes + more than 15 (or 45 or 180) seconds to respond to this request + and has not responded to any request from other threads + on this client within a second (note that it is not worth + grabbing the GlobalMid_Lock and slowing things down in this + wait event to more accurately check the lstrsp field on some + arch since we are already in an error path that will retry */ wait_event(ses->server->response_q, (!(midQ->midState & MID_REQUEST_SUBMITTED)) || - time_after(jiffies, timeout) || + (time_after(jiffies, timeout) && + time_after(jiffies, ses->server->lstrp + HZ)) || ((ses->server->tcpStatus != CifsGood) && (ses->server->tcpStatus != CifsNew))); } else { From 14a441a2b4ee1dfc00ec822d91d9fb20f401c62f Mon Sep 17 00:00:00 2001 From: Steve French Date: Sun, 16 Jul 2006 04:32:51 +0000 Subject: [PATCH 003/115] [CIFS] spinlock protect read of last srv response time in timeout path Signed-off-by: Jeremy Allison Signed-off-by: Steve French (cherry picked from b33a3f55e54fd210fc043eafcf83728b03bc9e02 commit) --- fs/cifs/transport.c | 99 ++++++++++++++++++++++++++++++++++----------- 1 file changed, 76 insertions(+), 23 deletions(-) diff --git a/fs/cifs/transport.c b/fs/cifs/transport.c index 95e23ca670a..d16e6032d4c 100644 --- a/fs/cifs/transport.c +++ b/fs/cifs/transport.c @@ -3,7 +3,8 @@ * * Copyright (C) International Business Machines Corp., 2002,2005 * Author(s): Steve French (sfrench@us.ibm.com) - * + * Jeremy Allison (jra@samba.org) 2006. + * * This library is free software; you can redistribute it and/or modify * it under the terms of the GNU Lesser General Public License as published * by the Free Software Foundation; either version 2.1 of the License, or @@ -442,13 +443,46 @@ SendReceive2(const unsigned int xid, struct cifsSesInfo *ses, /* No user interrupts in wait - wreaks havoc with performance */ if(timeout != MAX_SCHEDULE_TIMEOUT) { - timeout += jiffies; - wait_event(ses->server->response_q, - (!(midQ->midState & MID_REQUEST_SUBMITTED)) || - (time_after(jiffies, timeout) && - time_after(jiffies, ses->server->lstrp + HZ)) || - ((ses->server->tcpStatus != CifsGood) && - (ses->server->tcpStatus != CifsNew))); + unsigned long curr_timeout; + + for (;;) { + curr_timeout = timeout + jiffies; + wait_event(ses->server->response_q, + (!(midQ->midState & MID_REQUEST_SUBMITTED)) || + time_after(jiffies, curr_timeout) || + ((ses->server->tcpStatus != CifsGood) && + (ses->server->tcpStatus != CifsNew))); + + if (time_after(jiffies, curr_timeout) && + (midQ->midState & MID_REQUEST_SUBMITTED) && + ((ses->server->tcpStatus == CifsGood) || + (ses->server->tcpStatus == CifsNew))) { + + unsigned long lrt; + + /* We timed out. Is the server still + sending replies ? */ + spin_lock(&GlobalMid_Lock); + lrt = ses->server->lstrp; + spin_unlock(&GlobalMid_Lock); + + /* Calculate 10 seconds past last receive time. + Although we prefer not to time out if the + server is still responding - we will time + out if the server takes more than 15 (or 45 + or 180) seconds to respond to this request + and has not responded to any request from + other threads on the client within 10 seconds */ + lrt += (10 * HZ); + if (time_after(jiffies, lrt)) { + /* No replies for 10 seconds. */ + cERROR(1,("server not responding")); + break; + } + } else { + break; + } + } } else { wait_event(ses->server->response_q, (!(midQ->midState & MID_REQUEST_SUBMITTED)) || @@ -710,21 +744,40 @@ SendReceive(const unsigned int xid, struct cifsSesInfo *ses, /* No user interrupts in wait - wreaks havoc with performance */ if(timeout != MAX_SCHEDULE_TIMEOUT) { - timeout += jiffies; - /* although we prefer not to time out if the server is still - responding - we will time out if the server takes - more than 15 (or 45 or 180) seconds to respond to this request - and has not responded to any request from other threads - on this client within a second (note that it is not worth - grabbing the GlobalMid_Lock and slowing things down in this - wait event to more accurately check the lstrsp field on some - arch since we are already in an error path that will retry */ - wait_event(ses->server->response_q, - (!(midQ->midState & MID_REQUEST_SUBMITTED)) || - (time_after(jiffies, timeout) && - time_after(jiffies, ses->server->lstrp + HZ)) || - ((ses->server->tcpStatus != CifsGood) && - (ses->server->tcpStatus != CifsNew))); + unsigned long curr_timeout; + + for (;;) { + curr_timeout = timeout + jiffies; + wait_event(ses->server->response_q, + (!(midQ->midState & MID_REQUEST_SUBMITTED)) || + time_after(jiffies, curr_timeout) || + ((ses->server->tcpStatus != CifsGood) && + (ses->server->tcpStatus != CifsNew))); + + if (time_after(jiffies, curr_timeout) && + (midQ->midState & MID_REQUEST_SUBMITTED) && + ((ses->server->tcpStatus == CifsGood) || + (ses->server->tcpStatus == CifsNew))) { + + unsigned long lrt; + + /* We timed out. Is the server still + sending replies ? */ + spin_lock(&GlobalMid_Lock); + lrt = ses->server->lstrp; + spin_unlock(&GlobalMid_Lock); + + /* Calculate 10 seconds past last receive time*/ + lrt += (10 * HZ); + if (time_after(jiffies, lrt)) { + /* Server sent no reply in 10 seconds */ + cERROR(1,("Server not responding")); + break; + } + } else { + break; + } + } } else { wait_event(ses->server->response_q, (!(midQ->midState & MID_REQUEST_SUBMITTED)) || From 5da07b0208066fd3544ecf2b521fc7a2e0228ad5 Mon Sep 17 00:00:00 2001 From: Steve French Date: Sun, 16 Jul 2006 21:33:15 +0000 Subject: [PATCH 004/115] [CIFS] Make midState usage more consistent Although harmless, we were sometimes treating midState like it contained flags but they are exclusive states, and this makes that more clear. Signed-off-by: Jeremy Allison Signed-off-by: Steve French (cherry picked from 586c057c3a68dd6ae0f3ba94fbf76798b1558074 commit) --- fs/cifs/transport.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/fs/cifs/transport.c b/fs/cifs/transport.c index d16e6032d4c..748b0df07a7 100644 --- a/fs/cifs/transport.c +++ b/fs/cifs/transport.c @@ -448,13 +448,13 @@ SendReceive2(const unsigned int xid, struct cifsSesInfo *ses, for (;;) { curr_timeout = timeout + jiffies; wait_event(ses->server->response_q, - (!(midQ->midState & MID_REQUEST_SUBMITTED)) || + (!(midQ->midState == MID_REQUEST_SUBMITTED)) || time_after(jiffies, curr_timeout) || ((ses->server->tcpStatus != CifsGood) && (ses->server->tcpStatus != CifsNew))); if (time_after(jiffies, curr_timeout) && - (midQ->midState & MID_REQUEST_SUBMITTED) && + (midQ->midState == MID_REQUEST_SUBMITTED) && ((ses->server->tcpStatus == CifsGood) || (ses->server->tcpStatus == CifsNew))) { @@ -485,7 +485,7 @@ SendReceive2(const unsigned int xid, struct cifsSesInfo *ses, } } else { wait_event(ses->server->response_q, - (!(midQ->midState & MID_REQUEST_SUBMITTED)) || + (!(midQ->midState == MID_REQUEST_SUBMITTED)) || ((ses->server->tcpStatus != CifsGood) && (ses->server->tcpStatus != CifsNew))); } @@ -749,13 +749,13 @@ SendReceive(const unsigned int xid, struct cifsSesInfo *ses, for (;;) { curr_timeout = timeout + jiffies; wait_event(ses->server->response_q, - (!(midQ->midState & MID_REQUEST_SUBMITTED)) || + (!(midQ->midState == MID_REQUEST_SUBMITTED)) || time_after(jiffies, curr_timeout) || ((ses->server->tcpStatus != CifsGood) && (ses->server->tcpStatus != CifsNew))); if (time_after(jiffies, curr_timeout) && - (midQ->midState & MID_REQUEST_SUBMITTED) && + (midQ->midState == MID_REQUEST_SUBMITTED) && ((ses->server->tcpStatus == CifsGood) || (ses->server->tcpStatus == CifsNew))) { @@ -780,7 +780,7 @@ SendReceive(const unsigned int xid, struct cifsSesInfo *ses, } } else { wait_event(ses->server->response_q, - (!(midQ->midState & MID_REQUEST_SUBMITTED)) || + (!(midQ->midState == MID_REQUEST_SUBMITTED)) || ((ses->server->tcpStatus != CifsGood) && (ses->server->tcpStatus != CifsNew))); } From 6c3d8909d85b2c18fd7a6e64f0ca757a257b40fa Mon Sep 17 00:00:00 2001 From: Steve French Date: Mon, 31 Jul 2006 22:46:20 +0000 Subject: [PATCH 005/115] [CIFS] Allow cifsd to suspend if connection is lost Make cifsd allow us to suspend if it has lost the connection with a server Ref: http://bugzilla.kernel.org/show_bug.cgi?id=6811 Signed-off-by: Rafael J. Wysocki Acked-by: Pavel Machek Signed-off-by: Steve French (cherry picked from 27bd6cd87b0ada66515ad49bc346d77d1e9d3e05 commit) --- fs/cifs/connect.c | 1 + 1 file changed, 1 insertion(+) diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index c4aedcf5c92..b706b4f48b1 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -182,6 +182,7 @@ cifs_reconnect(struct TCP_Server_Info *server) while ((server->tcpStatus != CifsExiting) && (server->tcpStatus != CifsGood)) { + try_to_freeze(); if(server->protocolType == IPV6) { rc = ipv6_connect(&server->addr.sockAddr6,&server->ssocket); } else { From 7ee1af765dfa3146aef958258003245e082284e5 Mon Sep 17 00:00:00 2001 From: Jeremy Allison Date: Wed, 2 Aug 2006 21:56:33 +0000 Subject: [PATCH 006/115] [CIFS] Allow Windows blocking locks to be cancelled via a CANCEL_LOCK call. TODO - restrict this to servers that support NT_STATUS codes (Win9x will probably not support this call). Signed-off-by: Jeremy Allison Signed-off-by: Steve French (cherry picked from 570d4d2d895569825d0d017d4e76b51138f68864 commit) --- fs/cifs/cifsglob.h | 15 +- fs/cifs/cifsproto.h | 4 + fs/cifs/cifssmb.c | 15 +- fs/cifs/file.c | 95 +++++- fs/cifs/netmisc.c | 1 + fs/cifs/smberr.h | 1 + fs/cifs/transport.c | 709 ++++++++++++++++++++++++++------------------ 7 files changed, 528 insertions(+), 312 deletions(-) diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h index 39b43e6a750..b24006c47df 100644 --- a/fs/cifs/cifsglob.h +++ b/fs/cifs/cifsglob.h @@ -3,6 +3,7 @@ * * Copyright (C) International Business Machines Corp., 2002,2006 * Author(s): Steve French (sfrench@us.ibm.com) + * Jeremy Allison (jra@samba.org) * * This library is free software; you can redistribute it and/or modify * it under the terms of the GNU Lesser General Public License as published @@ -267,14 +268,14 @@ struct cifsTconInfo { }; /* - * This info hangs off the cifsFileInfo structure. This is used to track - * byte stream locks on the file + * This info hangs off the cifsFileInfo structure, pointed to by llist. + * This is used to track byte stream locks on the file */ struct cifsLockInfo { - struct cifsLockInfo *next; - int start; - int length; - int type; + struct list_head llist; /* pointer to next cifsLockInfo */ + __u64 offset; + __u64 length; + __u8 type; }; /* @@ -305,6 +306,8 @@ struct cifsFileInfo { /* lock scope id (0 if none) */ struct file * pfile; /* needed for writepage */ struct inode * pInode; /* needed for oplock break */ + struct semaphore lock_sem; + struct list_head llist; /* list of byte range locks we have. */ unsigned closePend:1; /* file is marked to close */ unsigned invalidHandle:1; /* file closed via session abend */ atomic_t wrtPending; /* handle in use - defer close */ diff --git a/fs/cifs/cifsproto.h b/fs/cifs/cifsproto.h index a5ddc62d6fe..b35c55c3c8b 100644 --- a/fs/cifs/cifsproto.h +++ b/fs/cifs/cifsproto.h @@ -50,6 +50,10 @@ extern int SendReceive(const unsigned int /* xid */ , struct cifsSesInfo *, extern int SendReceive2(const unsigned int /* xid */ , struct cifsSesInfo *, struct kvec *, int /* nvec to send */, int * /* type of buf returned */ , const int long_op); +extern int SendReceiveBlockingLock(const unsigned int /* xid */ , struct cifsTconInfo *, + struct smb_hdr * /* input */ , + struct smb_hdr * /* out */ , + int * /* bytes returned */); extern int checkSMBhdr(struct smb_hdr *smb, __u16 mid); extern int checkSMB(struct smb_hdr *smb, __u16 mid, int length); extern int is_valid_oplock_break(struct smb_hdr *smb, struct TCP_Server_Info *); diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index c03c42ee9eb..dcbc3e075e8 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -1460,8 +1460,13 @@ CIFSSMBLock(const int xid, struct cifsTconInfo *tcon, pSMB->hdr.smb_buf_length += count; pSMB->ByteCount = cpu_to_le16(count); - rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, + if (waitFlag) { + rc = SendReceiveBlockingLock(xid, tcon, (struct smb_hdr *) pSMB, + (struct smb_hdr *) pSMBr, &bytes_returned); + } else { + rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, timeout); + } cifs_stats_inc(&tcon->num_locks); if (rc) { cFYI(1, ("Send error in Lock = %d", rc)); @@ -1546,8 +1551,14 @@ CIFSSMBPosixLock(const int xid, struct cifsTconInfo *tcon, pSMB->Reserved4 = 0; pSMB->hdr.smb_buf_length += byte_count; pSMB->ByteCount = cpu_to_le16(byte_count); - rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, + if (waitFlag) { + rc = SendReceiveBlockingLock(xid, tcon, (struct smb_hdr *) pSMB, + (struct smb_hdr *) pSMBr, &bytes_returned); + } else { + rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, timeout); + } + if (rc) { cFYI(1, ("Send error in Posix Lock = %d", rc)); } else if (get_flag) { diff --git a/fs/cifs/file.c b/fs/cifs/file.c index 52e2e4c8794..e9c5ba9084f 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -5,6 +5,7 @@ * * Copyright (C) International Business Machines Corp., 2002,2003 * Author(s): Steve French (sfrench@us.ibm.com) + * Jeremy Allison (jra@samba.org) * * This library is free software; you can redistribute it and/or modify * it under the terms of the GNU Lesser General Public License as published @@ -47,6 +48,8 @@ static inline struct cifsFileInfo *cifs_init_private( private_data->netfid = netfid; private_data->pid = current->tgid; init_MUTEX(&private_data->fh_sem); + init_MUTEX(&private_data->lock_sem); + INIT_LIST_HEAD(&private_data->llist); private_data->pfile = file; /* needed for writepage */ private_data->pInode = inode; private_data->invalidHandle = FALSE; @@ -473,6 +476,8 @@ int cifs_close(struct inode *inode, struct file *file) cifs_sb = CIFS_SB(inode->i_sb); pTcon = cifs_sb->tcon; if (pSMBFile) { + struct cifsLockInfo *li, *tmp; + pSMBFile->closePend = TRUE; if (pTcon) { /* no sense reconnecting to close a file that is @@ -496,6 +501,16 @@ int cifs_close(struct inode *inode, struct file *file) pSMBFile->netfid); } } + + /* Delete any outstanding lock records. + We'll lose them when the file is closed anyway. */ + down(&pSMBFile->lock_sem); + list_for_each_entry_safe(li, tmp, &pSMBFile->llist, llist) { + list_del(&li->llist); + kfree(li); + } + up(&pSMBFile->lock_sem); + write_lock(&GlobalSMBSeslock); list_del(&pSMBFile->flist); list_del(&pSMBFile->tlist); @@ -570,6 +585,21 @@ int cifs_closedir(struct inode *inode, struct file *file) return rc; } +static int store_file_lock(struct cifsFileInfo *fid, __u64 len, + __u64 offset, __u8 lockType) +{ + struct cifsLockInfo *li = kmalloc(sizeof(struct cifsLockInfo), GFP_KERNEL); + if (li == NULL) + return -ENOMEM; + li->offset = offset; + li->length = len; + li->type = lockType; + down(&fid->lock_sem); + list_add(&li->llist, &fid->llist); + up(&fid->lock_sem); + return 0; +} + int cifs_lock(struct file *file, int cmd, struct file_lock *pfLock) { int rc, xid; @@ -581,6 +611,7 @@ int cifs_lock(struct file *file, int cmd, struct file_lock *pfLock) struct cifsTconInfo *pTcon; __u16 netfid; __u8 lockType = LOCKING_ANDX_LARGE_FILES; + int posix_locking; length = 1 + pfLock->fl_end - pfLock->fl_start; rc = -EACCES; @@ -639,14 +670,14 @@ int cifs_lock(struct file *file, int cmd, struct file_lock *pfLock) } netfid = ((struct cifsFileInfo *)file->private_data)->netfid; + posix_locking = (cifs_sb->tcon->ses->capabilities & CAP_UNIX) && + (CIFS_UNIX_FCNTL_CAP & le64_to_cpu(cifs_sb->tcon->fsUnixInfo.Capability)); /* BB add code here to normalize offset and length to account for negative length which we can not accept over the wire */ if (IS_GETLK(cmd)) { - if((cifs_sb->tcon->ses->capabilities & CAP_UNIX) && - (CIFS_UNIX_FCNTL_CAP & - le64_to_cpu(cifs_sb->tcon->fsUnixInfo.Capability))) { + if(posix_locking) { int posix_lock_type; if(lockType & LOCKING_ANDX_SHARED_LOCK) posix_lock_type = CIFS_RDLCK; @@ -682,9 +713,15 @@ int cifs_lock(struct file *file, int cmd, struct file_lock *pfLock) FreeXid(xid); return rc; } - if ((cifs_sb->tcon->ses->capabilities & CAP_UNIX) && - (CIFS_UNIX_FCNTL_CAP & - le64_to_cpu(cifs_sb->tcon->fsUnixInfo.Capability))) { + + if (!numLock && !numUnlock) { + /* if no lock or unlock then nothing + to do since we do not know what it is */ + FreeXid(xid); + return -EOPNOTSUPP; + } + + if (posix_locking) { int posix_lock_type; if(lockType & LOCKING_ANDX_SHARED_LOCK) posix_lock_type = CIFS_RDLCK; @@ -693,18 +730,46 @@ int cifs_lock(struct file *file, int cmd, struct file_lock *pfLock) if(numUnlock == 1) posix_lock_type = CIFS_UNLCK; - else if(numLock == 0) { - /* if no lock or unlock then nothing - to do since we do not know what it is */ - FreeXid(xid); - return -EOPNOTSUPP; - } + rc = CIFSSMBPosixLock(xid, pTcon, netfid, 0 /* set */, length, pfLock, posix_lock_type, wait_flag); - } else - rc = CIFSSMBLock(xid, pTcon, netfid, length, pfLock->fl_start, - numUnlock, numLock, lockType, wait_flag); + } else { + struct cifsFileInfo *fid = (struct cifsFileInfo *)file->private_data; + + if (numLock) { + rc = CIFSSMBLock(xid, pTcon, netfid, length, pfLock->fl_start, + 0, numLock, lockType, wait_flag); + + if (rc == 0) { + /* For Windows locks we must store them. */ + rc = store_file_lock(fid, length, + pfLock->fl_start, lockType); + } + } else if (numUnlock) { + /* For each stored lock that this unlock overlaps + completely, unlock it. */ + int stored_rc = 0; + struct cifsLockInfo *li, *tmp; + + down(&fid->lock_sem); + list_for_each_entry_safe(li, tmp, &fid->llist, llist) { + if (pfLock->fl_start <= li->offset && + length >= li->length) { + stored_rc = CIFSSMBLock(xid, pTcon, netfid, + li->length, li->offset, + 1, 0, li->type, FALSE); + if (stored_rc) + rc = stored_rc; + + list_del(&li->llist); + kfree(li); + } + } + up(&fid->lock_sem); + } + } + if (pfLock->fl_flags & FL_POSIX) posix_lock_file_wait(file, pfLock); FreeXid(xid); diff --git a/fs/cifs/netmisc.c b/fs/cifs/netmisc.c index b66eff5dc62..ce87550e918 100644 --- a/fs/cifs/netmisc.c +++ b/fs/cifs/netmisc.c @@ -72,6 +72,7 @@ static const struct smb_to_posix_error mapping_table_ERRDOS[] = { {ERRinvlevel,-EOPNOTSUPP}, {ERRdirnotempty, -ENOTEMPTY}, {ERRnotlocked, -ENOLCK}, + {ERRcancelviolation, -ENOLCK}, {ERRalreadyexists, -EEXIST}, {ERRmoredata, -EOVERFLOW}, {ERReasnotsupported,-EOPNOTSUPP}, diff --git a/fs/cifs/smberr.h b/fs/cifs/smberr.h index cd41c67ff8d..212c3c29640 100644 --- a/fs/cifs/smberr.h +++ b/fs/cifs/smberr.h @@ -95,6 +95,7 @@ #define ERRinvlevel 124 #define ERRdirnotempty 145 #define ERRnotlocked 158 +#define ERRcancelviolation 173 #define ERRalreadyexists 183 #define ERRbadpipe 230 #define ERRpipebusy 231 diff --git a/fs/cifs/transport.c b/fs/cifs/transport.c index 748b0df07a7..48d47b46b1f 100644 --- a/fs/cifs/transport.c +++ b/fs/cifs/transport.c @@ -37,7 +37,7 @@ extern mempool_t *cifs_mid_poolp; extern kmem_cache_t *cifs_oplock_cachep; static struct mid_q_entry * -AllocMidQEntry(struct smb_hdr *smb_buffer, struct cifsSesInfo *ses) +AllocMidQEntry(const struct smb_hdr *smb_buffer, struct cifsSesInfo *ses) { struct mid_q_entry *temp; @@ -204,6 +204,10 @@ smb_send(struct socket *ssocket, struct smb_hdr *smb_buffer, rc = 0; } + /* Don't want to modify the buffer as a + side effect of this call. */ + smb_buffer->smb_buf_length = smb_buf_length; + return rc; } @@ -218,6 +222,7 @@ smb_send2(struct socket *ssocket, struct kvec *iov, int n_vec, unsigned int len = iov[0].iov_len; unsigned int total_len; int first_vec = 0; + unsigned int smb_buf_length = smb_buffer->smb_buf_length; if(ssocket == NULL) return -ENOTSOCK; /* BB eventually add reconnect code here */ @@ -294,9 +299,123 @@ smb_send2(struct socket *ssocket, struct kvec *iov, int n_vec, } else rc = 0; + /* Don't want to modify the buffer as a + side effect of this call. */ + smb_buffer->smb_buf_length = smb_buf_length; + return rc; } +static int wait_for_free_request(struct cifsSesInfo *ses, const int long_op) +{ + if(long_op == -1) { + /* oplock breaks must not be held up */ + atomic_inc(&ses->server->inFlight); + } else { + spin_lock(&GlobalMid_Lock); + while(1) { + if(atomic_read(&ses->server->inFlight) >= + cifs_max_pending){ + spin_unlock(&GlobalMid_Lock); +#ifdef CONFIG_CIFS_STATS2 + atomic_inc(&ses->server->num_waiters); +#endif + wait_event(ses->server->request_q, + atomic_read(&ses->server->inFlight) + < cifs_max_pending); +#ifdef CONFIG_CIFS_STATS2 + atomic_dec(&ses->server->num_waiters); +#endif + spin_lock(&GlobalMid_Lock); + } else { + if(ses->server->tcpStatus == CifsExiting) { + spin_unlock(&GlobalMid_Lock); + return -ENOENT; + } + + /* can not count locking commands against total since + they are allowed to block on server */ + + /* update # of requests on the wire to server */ + if (long_op < 3) + atomic_inc(&ses->server->inFlight); + spin_unlock(&GlobalMid_Lock); + break; + } + } + } + return 0; +} + +static int allocate_mid(struct cifsSesInfo *ses, struct smb_hdr *in_buf, + struct mid_q_entry **ppmidQ) +{ + if (ses->server->tcpStatus == CifsExiting) { + return -ENOENT; + } else if (ses->server->tcpStatus == CifsNeedReconnect) { + cFYI(1,("tcp session dead - return to caller to retry")); + return -EAGAIN; + } else if (ses->status != CifsGood) { + /* check if SMB session is bad because we are setting it up */ + if((in_buf->Command != SMB_COM_SESSION_SETUP_ANDX) && + (in_buf->Command != SMB_COM_NEGOTIATE)) { + return -EAGAIN; + } /* else ok - we are setting up session */ + } + *ppmidQ = AllocMidQEntry(in_buf, ses); + if (*ppmidQ == NULL) { + return -ENOMEM; + } + return 0; +} + +static int wait_for_response(struct cifsSesInfo *ses, + struct mid_q_entry *midQ, + unsigned long timeout, + unsigned long time_to_wait) +{ + unsigned long curr_timeout; + + for (;;) { + curr_timeout = timeout + jiffies; + wait_event(ses->server->response_q, + (!(midQ->midState == MID_REQUEST_SUBMITTED)) || + time_after(jiffies, curr_timeout) || + ((ses->server->tcpStatus != CifsGood) && + (ses->server->tcpStatus != CifsNew))); + + if (time_after(jiffies, curr_timeout) && + (midQ->midState == MID_REQUEST_SUBMITTED) && + ((ses->server->tcpStatus == CifsGood) || + (ses->server->tcpStatus == CifsNew))) { + + unsigned long lrt; + + /* We timed out. Is the server still + sending replies ? */ + spin_lock(&GlobalMid_Lock); + lrt = ses->server->lstrp; + spin_unlock(&GlobalMid_Lock); + + /* Calculate time_to_wait past last receive time. + Although we prefer not to time out if the + server is still responding - we will time + out if the server takes more than 15 (or 45 + or 180) seconds to respond to this request + and has not responded to any request from + other threads on the client within 10 seconds */ + lrt += time_to_wait; + if (time_after(jiffies, lrt)) { + /* No replies for time_to_wait. */ + cERROR(1,("server not responding")); + return -1; + } + } else { + return 0; + } + } +} + int SendReceive2(const unsigned int xid, struct cifsSesInfo *ses, struct kvec *iov, int n_vec, int * pRespBufType /* ret */, @@ -324,75 +443,27 @@ SendReceive2(const unsigned int xid, struct cifsSesInfo *ses, /* Ensure that we do not send more than 50 overlapping requests to the same server. We may make this configurable later or use ses->maxReq */ - if(long_op == -1) { - /* oplock breaks must not be held up */ - atomic_inc(&ses->server->inFlight); - } else { - spin_lock(&GlobalMid_Lock); - while(1) { - if(atomic_read(&ses->server->inFlight) >= - cifs_max_pending){ - spin_unlock(&GlobalMid_Lock); -#ifdef CONFIG_CIFS_STATS2 - atomic_inc(&ses->server->num_waiters); -#endif - wait_event(ses->server->request_q, - atomic_read(&ses->server->inFlight) - < cifs_max_pending); -#ifdef CONFIG_CIFS_STATS2 - atomic_dec(&ses->server->num_waiters); -#endif - spin_lock(&GlobalMid_Lock); - } else { - if(ses->server->tcpStatus == CifsExiting) { - spin_unlock(&GlobalMid_Lock); - cifs_small_buf_release(in_buf); - return -ENOENT; - } - /* can not count locking commands against total since - they are allowed to block on server */ - - if(long_op < 3) { - /* update # of requests on the wire to server */ - atomic_inc(&ses->server->inFlight); - } - spin_unlock(&GlobalMid_Lock); - break; - } - } + rc = wait_for_free_request(ses, long_op); + if (rc) { + cifs_small_buf_release(in_buf); + return rc; } + /* make sure that we sign in the same order that we send on this socket and avoid races inside tcp sendmsg code that could cause corruption of smb data */ down(&ses->server->tcpSem); - if (ses->server->tcpStatus == CifsExiting) { - rc = -ENOENT; - goto out_unlock2; - } else if (ses->server->tcpStatus == CifsNeedReconnect) { - cFYI(1,("tcp session dead - return to caller to retry")); - rc = -EAGAIN; - goto out_unlock2; - } else if (ses->status != CifsGood) { - /* check if SMB session is bad because we are setting it up */ - if((in_buf->Command != SMB_COM_SESSION_SETUP_ANDX) && - (in_buf->Command != SMB_COM_NEGOTIATE)) { - rc = -EAGAIN; - goto out_unlock2; - } /* else ok - we are setting up session */ - } - midQ = AllocMidQEntry(in_buf, ses); - if (midQ == NULL) { + rc = allocate_mid(ses, in_buf, &midQ); + if (rc) { up(&ses->server->tcpSem); cifs_small_buf_release(in_buf); - /* If not lock req, update # of requests on wire to server */ - if(long_op < 3) { - atomic_dec(&ses->server->inFlight); - wake_up(&ses->server->request_q); - } - return -ENOMEM; + /* Update # of requests on wire to server */ + atomic_dec(&ses->server->inFlight); + wake_up(&ses->server->request_q); + return rc; } rc = cifs_sign_smb2(iov, n_vec, ses->server, &midQ->sequence_number); @@ -407,32 +478,23 @@ SendReceive2(const unsigned int xid, struct cifsSesInfo *ses, atomic_dec(&ses->server->inSend); midQ->when_sent = jiffies; #endif - if(rc < 0) { - DeleteMidQEntry(midQ); - up(&ses->server->tcpSem); - cifs_small_buf_release(in_buf); - /* If not lock req, update # of requests on wire to server */ - if(long_op < 3) { - atomic_dec(&ses->server->inFlight); - wake_up(&ses->server->request_q); - } - return rc; - } else { - up(&ses->server->tcpSem); - cifs_small_buf_release(in_buf); - } + + up(&ses->server->tcpSem); + cifs_small_buf_release(in_buf); + + if(rc < 0) + goto out; if (long_op == -1) - goto cifs_no_response_exit2; + goto out; else if (long_op == 2) /* writes past end of file can take loong time */ timeout = 180 * HZ; else if (long_op == 1) timeout = 45 * HZ; /* should be greater than servers oplock break timeout (about 43 seconds) */ - else if (long_op > 2) { - timeout = MAX_SCHEDULE_TIMEOUT; - } else + else timeout = 15 * HZ; + /* wait for 15 seconds or until woken up due to response arriving or due to last connection to this server being unmounted */ if (signal_pending(current)) { @@ -442,53 +504,7 @@ SendReceive2(const unsigned int xid, struct cifsSesInfo *ses, } /* No user interrupts in wait - wreaks havoc with performance */ - if(timeout != MAX_SCHEDULE_TIMEOUT) { - unsigned long curr_timeout; - - for (;;) { - curr_timeout = timeout + jiffies; - wait_event(ses->server->response_q, - (!(midQ->midState == MID_REQUEST_SUBMITTED)) || - time_after(jiffies, curr_timeout) || - ((ses->server->tcpStatus != CifsGood) && - (ses->server->tcpStatus != CifsNew))); - - if (time_after(jiffies, curr_timeout) && - (midQ->midState == MID_REQUEST_SUBMITTED) && - ((ses->server->tcpStatus == CifsGood) || - (ses->server->tcpStatus == CifsNew))) { - - unsigned long lrt; - - /* We timed out. Is the server still - sending replies ? */ - spin_lock(&GlobalMid_Lock); - lrt = ses->server->lstrp; - spin_unlock(&GlobalMid_Lock); - - /* Calculate 10 seconds past last receive time. - Although we prefer not to time out if the - server is still responding - we will time - out if the server takes more than 15 (or 45 - or 180) seconds to respond to this request - and has not responded to any request from - other threads on the client within 10 seconds */ - lrt += (10 * HZ); - if (time_after(jiffies, lrt)) { - /* No replies for 10 seconds. */ - cERROR(1,("server not responding")); - break; - } - } else { - break; - } - } - } else { - wait_event(ses->server->response_q, - (!(midQ->midState == MID_REQUEST_SUBMITTED)) || - ((ses->server->tcpStatus != CifsGood) && - (ses->server->tcpStatus != CifsNew))); - } + wait_for_response(ses, midQ, timeout, 10 * HZ); spin_lock(&GlobalMid_Lock); if (midQ->resp_buf) { @@ -516,11 +532,9 @@ SendReceive2(const unsigned int xid, struct cifsSesInfo *ses, } spin_unlock(&GlobalMid_Lock); DeleteMidQEntry(midQ); - /* If not lock req, update # of requests on wire to server */ - if(long_op < 3) { - atomic_dec(&ses->server->inFlight); - wake_up(&ses->server->request_q); - } + /* Update # of requests on wire to server */ + atomic_dec(&ses->server->inFlight); + wake_up(&ses->server->request_q); return rc; } @@ -571,24 +585,12 @@ SendReceive2(const unsigned int xid, struct cifsSesInfo *ses, cFYI(1,("Bad MID state?")); } } -cifs_no_response_exit2: + +out: + DeleteMidQEntry(midQ); - - if(long_op < 3) { - atomic_dec(&ses->server->inFlight); - wake_up(&ses->server->request_q); - } - - return rc; - -out_unlock2: - up(&ses->server->tcpSem); - cifs_small_buf_release(in_buf); - /* If not lock req, update # of requests on wire to server */ - if(long_op < 3) { - atomic_dec(&ses->server->inFlight); - wake_up(&ses->server->request_q); - } + atomic_dec(&ses->server->inFlight); + wake_up(&ses->server->request_q); return rc; } @@ -618,85 +620,34 @@ SendReceive(const unsigned int xid, struct cifsSesInfo *ses, /* Ensure that we do not send more than 50 overlapping requests to the same server. We may make this configurable later or use ses->maxReq */ - if(long_op == -1) { - /* oplock breaks must not be held up */ - atomic_inc(&ses->server->inFlight); - } else { - spin_lock(&GlobalMid_Lock); - while(1) { - if(atomic_read(&ses->server->inFlight) >= - cifs_max_pending){ - spin_unlock(&GlobalMid_Lock); -#ifdef CONFIG_CIFS_STATS2 - atomic_inc(&ses->server->num_waiters); -#endif - wait_event(ses->server->request_q, - atomic_read(&ses->server->inFlight) - < cifs_max_pending); -#ifdef CONFIG_CIFS_STATS2 - atomic_dec(&ses->server->num_waiters); -#endif - spin_lock(&GlobalMid_Lock); - } else { - if(ses->server->tcpStatus == CifsExiting) { - spin_unlock(&GlobalMid_Lock); - return -ENOENT; - } - /* can not count locking commands against total since - they are allowed to block on server */ - - if(long_op < 3) { - /* update # of requests on the wire to server */ - atomic_inc(&ses->server->inFlight); - } - spin_unlock(&GlobalMid_Lock); - break; - } - } - } + rc = wait_for_free_request(ses, long_op); + if (rc) + return rc; + /* make sure that we sign in the same order that we send on this socket and avoid races inside tcp sendmsg code that could cause corruption of smb data */ down(&ses->server->tcpSem); - if (ses->server->tcpStatus == CifsExiting) { - rc = -ENOENT; - goto out_unlock; - } else if (ses->server->tcpStatus == CifsNeedReconnect) { - cFYI(1,("tcp session dead - return to caller to retry")); - rc = -EAGAIN; - goto out_unlock; - } else if (ses->status != CifsGood) { - /* check if SMB session is bad because we are setting it up */ - if((in_buf->Command != SMB_COM_SESSION_SETUP_ANDX) && - (in_buf->Command != SMB_COM_NEGOTIATE)) { - rc = -EAGAIN; - goto out_unlock; - } /* else ok - we are setting up session */ - } - midQ = AllocMidQEntry(in_buf, ses); - if (midQ == NULL) { + rc = allocate_mid(ses, in_buf, &midQ); + if (rc) { up(&ses->server->tcpSem); - /* If not lock req, update # of requests on wire to server */ - if(long_op < 3) { - atomic_dec(&ses->server->inFlight); - wake_up(&ses->server->request_q); - } - return -ENOMEM; + /* Update # of requests on wire to server */ + atomic_dec(&ses->server->inFlight); + wake_up(&ses->server->request_q); + return rc; } if (in_buf->smb_buf_length > CIFSMaxBufSize + MAX_CIFS_HDR_SIZE - 4) { - up(&ses->server->tcpSem); cERROR(1, ("Illegal length, greater than maximum frame, %d", in_buf->smb_buf_length)); DeleteMidQEntry(midQ); - /* If not lock req, update # of requests on wire to server */ - if(long_op < 3) { - atomic_dec(&ses->server->inFlight); - wake_up(&ses->server->request_q); - } + up(&ses->server->tcpSem); + /* Update # of requests on wire to server */ + atomic_dec(&ses->server->inFlight); + wake_up(&ses->server->request_q); return -EIO; } @@ -712,27 +663,19 @@ SendReceive(const unsigned int xid, struct cifsSesInfo *ses, atomic_dec(&ses->server->inSend); midQ->when_sent = jiffies; #endif - if(rc < 0) { - DeleteMidQEntry(midQ); - up(&ses->server->tcpSem); - /* If not lock req, update # of requests on wire to server */ - if(long_op < 3) { - atomic_dec(&ses->server->inFlight); - wake_up(&ses->server->request_q); - } - return rc; - } else - up(&ses->server->tcpSem); + up(&ses->server->tcpSem); + + if(rc < 0) + goto out; + if (long_op == -1) - goto cifs_no_response_exit; + goto out; else if (long_op == 2) /* writes past end of file can take loong time */ timeout = 180 * HZ; else if (long_op == 1) timeout = 45 * HZ; /* should be greater than servers oplock break timeout (about 43 seconds) */ - else if (long_op > 2) { - timeout = MAX_SCHEDULE_TIMEOUT; - } else + else timeout = 15 * HZ; /* wait for 15 seconds or until woken up due to response arriving or due to last connection to this server being unmounted */ @@ -743,47 +686,7 @@ SendReceive(const unsigned int xid, struct cifsSesInfo *ses, } /* No user interrupts in wait - wreaks havoc with performance */ - if(timeout != MAX_SCHEDULE_TIMEOUT) { - unsigned long curr_timeout; - - for (;;) { - curr_timeout = timeout + jiffies; - wait_event(ses->server->response_q, - (!(midQ->midState == MID_REQUEST_SUBMITTED)) || - time_after(jiffies, curr_timeout) || - ((ses->server->tcpStatus != CifsGood) && - (ses->server->tcpStatus != CifsNew))); - - if (time_after(jiffies, curr_timeout) && - (midQ->midState == MID_REQUEST_SUBMITTED) && - ((ses->server->tcpStatus == CifsGood) || - (ses->server->tcpStatus == CifsNew))) { - - unsigned long lrt; - - /* We timed out. Is the server still - sending replies ? */ - spin_lock(&GlobalMid_Lock); - lrt = ses->server->lstrp; - spin_unlock(&GlobalMid_Lock); - - /* Calculate 10 seconds past last receive time*/ - lrt += (10 * HZ); - if (time_after(jiffies, lrt)) { - /* Server sent no reply in 10 seconds */ - cERROR(1,("Server not responding")); - break; - } - } else { - break; - } - } - } else { - wait_event(ses->server->response_q, - (!(midQ->midState == MID_REQUEST_SUBMITTED)) || - ((ses->server->tcpStatus != CifsGood) && - (ses->server->tcpStatus != CifsNew))); - } + wait_for_response(ses, midQ, timeout, 10 * HZ); spin_lock(&GlobalMid_Lock); if (midQ->resp_buf) { @@ -811,11 +714,9 @@ SendReceive(const unsigned int xid, struct cifsSesInfo *ses, } spin_unlock(&GlobalMid_Lock); DeleteMidQEntry(midQ); - /* If not lock req, update # of requests on wire to server */ - if(long_op < 3) { - atomic_dec(&ses->server->inFlight); - wake_up(&ses->server->request_q); - } + /* Update # of requests on wire to server */ + atomic_dec(&ses->server->inFlight); + wake_up(&ses->server->request_q); return rc; } @@ -862,23 +763,253 @@ SendReceive(const unsigned int xid, struct cifsSesInfo *ses, cERROR(1,("Bad MID state?")); } } -cifs_no_response_exit: + +out: + DeleteMidQEntry(midQ); - - if(long_op < 3) { - atomic_dec(&ses->server->inFlight); - wake_up(&ses->server->request_q); - } - - return rc; - -out_unlock: - up(&ses->server->tcpSem); - /* If not lock req, update # of requests on wire to server */ - if(long_op < 3) { - atomic_dec(&ses->server->inFlight); - wake_up(&ses->server->request_q); - } + atomic_dec(&ses->server->inFlight); + wake_up(&ses->server->request_q); return rc; } + +/* Send an NT_CANCEL SMB to cause the POSIX blocking lock to return. */ + +static int +send_nt_cancel(struct cifsTconInfo *tcon, struct smb_hdr *in_buf, + struct mid_q_entry *midQ) +{ + int rc = 0; + struct cifsSesInfo *ses = tcon->ses; + __u16 mid = in_buf->Mid; + + header_assemble(in_buf, SMB_COM_NT_CANCEL, tcon, 0); + in_buf->Mid = mid; + down(&ses->server->tcpSem); + rc = cifs_sign_smb(in_buf, ses->server, &midQ->sequence_number); + if (rc) { + up(&ses->server->tcpSem); + return rc; + } + rc = smb_send(ses->server->ssocket, in_buf, in_buf->smb_buf_length, + (struct sockaddr *) &(ses->server->addr.sockAddr)); + up(&ses->server->tcpSem); + return rc; +} + +/* We send a LOCKINGX_CANCEL_LOCK to cause the Windows + blocking lock to return. */ + +static int +send_lock_cancel(const unsigned int xid, struct cifsTconInfo *tcon, + struct smb_hdr *in_buf, + struct smb_hdr *out_buf) +{ + int bytes_returned; + struct cifsSesInfo *ses = tcon->ses; + LOCK_REQ *pSMB = (LOCK_REQ *)in_buf; + + /* We just modify the current in_buf to change + the type of lock from LOCKING_ANDX_SHARED_LOCK + or LOCKING_ANDX_EXCLUSIVE_LOCK to + LOCKING_ANDX_CANCEL_LOCK. */ + + pSMB->LockType = LOCKING_ANDX_CANCEL_LOCK|LOCKING_ANDX_LARGE_FILES; + pSMB->Timeout = 0; + pSMB->hdr.Mid = GetNextMid(ses->server); + + return SendReceive(xid, ses, in_buf, out_buf, + &bytes_returned, 0); +} + +int +SendReceiveBlockingLock(const unsigned int xid, struct cifsTconInfo *tcon, + struct smb_hdr *in_buf, struct smb_hdr *out_buf, + int *pbytes_returned) +{ + int rc = 0; + int rstart = 0; + unsigned int receive_len; + struct mid_q_entry *midQ; + struct cifsSesInfo *ses; + + if (tcon == NULL || tcon->ses == NULL) { + cERROR(1,("Null smb session")); + return -EIO; + } + ses = tcon->ses; + + if(ses->server == NULL) { + cERROR(1,("Null tcp session")); + return -EIO; + } + + if(ses->server->tcpStatus == CifsExiting) + return -ENOENT; + + /* Ensure that we do not send more than 50 overlapping requests + to the same server. We may make this configurable later or + use ses->maxReq */ + + rc = wait_for_free_request(ses, 3); + if (rc) + return rc; + + /* make sure that we sign in the same order that we send on this socket + and avoid races inside tcp sendmsg code that could cause corruption + of smb data */ + + down(&ses->server->tcpSem); + + rc = allocate_mid(ses, in_buf, &midQ); + if (rc) { + up(&ses->server->tcpSem); + return rc; + } + + if (in_buf->smb_buf_length > CIFSMaxBufSize + MAX_CIFS_HDR_SIZE - 4) { + up(&ses->server->tcpSem); + cERROR(1, ("Illegal length, greater than maximum frame, %d", + in_buf->smb_buf_length)); + DeleteMidQEntry(midQ); + return -EIO; + } + + rc = cifs_sign_smb(in_buf, ses->server, &midQ->sequence_number); + + midQ->midState = MID_REQUEST_SUBMITTED; +#ifdef CONFIG_CIFS_STATS2 + atomic_inc(&ses->server->inSend); +#endif + rc = smb_send(ses->server->ssocket, in_buf, in_buf->smb_buf_length, + (struct sockaddr *) &(ses->server->addr.sockAddr)); +#ifdef CONFIG_CIFS_STATS2 + atomic_dec(&ses->server->inSend); + midQ->when_sent = jiffies; +#endif + up(&ses->server->tcpSem); + + if(rc < 0) { + DeleteMidQEntry(midQ); + return rc; + } + + /* Wait for a reply - allow signals to interrupt. */ + rc = wait_event_interruptible(ses->server->response_q, + (!(midQ->midState == MID_REQUEST_SUBMITTED)) || + ((ses->server->tcpStatus != CifsGood) && + (ses->server->tcpStatus != CifsNew))); + + /* Were we interrupted by a signal ? */ + if ((rc == -ERESTARTSYS) && + (midQ->midState == MID_REQUEST_SUBMITTED) && + ((ses->server->tcpStatus == CifsGood) || + (ses->server->tcpStatus == CifsNew))) { + + if (in_buf->Command == SMB_COM_TRANSACTION2) { + /* POSIX lock. We send a NT_CANCEL SMB to cause the + blocking lock to return. */ + + rc = send_nt_cancel(tcon, in_buf, midQ); + if (rc) { + DeleteMidQEntry(midQ); + return rc; + } + } else { + /* Windows lock. We send a LOCKINGX_CANCEL_LOCK + to cause the blocking lock to return. */ + + rc = send_lock_cancel(xid, tcon, in_buf, out_buf); + + /* If we get -ENOLCK back the lock may have + already been removed. Don't exit in this case. */ + if (rc && rc != -ENOLCK) { + DeleteMidQEntry(midQ); + return rc; + } + } + + /* Wait 5 seconds for the response. */ + if (wait_for_response(ses, midQ, 5 * HZ, 5 * HZ)==0) { + /* We got the response - restart system call. */ + rstart = 1; + } + } + + spin_lock(&GlobalMid_Lock); + if (midQ->resp_buf) { + spin_unlock(&GlobalMid_Lock); + receive_len = midQ->resp_buf->smb_buf_length; + } else { + cERROR(1,("No response for cmd %d mid %d", + midQ->command, midQ->mid)); + if(midQ->midState == MID_REQUEST_SUBMITTED) { + if(ses->server->tcpStatus == CifsExiting) + rc = -EHOSTDOWN; + else { + ses->server->tcpStatus = CifsNeedReconnect; + midQ->midState = MID_RETRY_NEEDED; + } + } + + if (rc != -EHOSTDOWN) { + if(midQ->midState == MID_RETRY_NEEDED) { + rc = -EAGAIN; + cFYI(1,("marking request for retry")); + } else { + rc = -EIO; + } + } + spin_unlock(&GlobalMid_Lock); + DeleteMidQEntry(midQ); + return rc; + } + + if (receive_len > CIFSMaxBufSize + MAX_CIFS_HDR_SIZE) { + cERROR(1, ("Frame too large received. Length: %d Xid: %d", + receive_len, xid)); + rc = -EIO; + } else { /* rcvd frame is ok */ + + if (midQ->resp_buf && out_buf + && (midQ->midState == MID_RESPONSE_RECEIVED)) { + out_buf->smb_buf_length = receive_len; + memcpy((char *)out_buf + 4, + (char *)midQ->resp_buf + 4, + receive_len); + + dump_smb(out_buf, 92); + /* convert the length into a more usable form */ + if((receive_len > 24) && + (ses->server->secMode & (SECMODE_SIGN_REQUIRED | + SECMODE_SIGN_ENABLED))) { + rc = cifs_verify_signature(out_buf, + ses->server->mac_signing_key, + midQ->sequence_number+1); + if(rc) { + cERROR(1,("Unexpected SMB signature")); + /* BB FIXME add code to kill session */ + } + } + + *pbytes_returned = out_buf->smb_buf_length; + + /* BB special case reconnect tid and uid here? */ + rc = map_smb_to_linux_error(out_buf); + + /* convert ByteCount if necessary */ + if (receive_len >= + sizeof (struct smb_hdr) - + 4 /* do not count RFC1001 header */ + + (2 * out_buf->WordCount) + 2 /* bcc */ ) + BCC(out_buf) = le16_to_cpu(BCC_LE(out_buf)); + } else { + rc = -EIO; + cERROR(1,("Bad MID state?")); + } + } + DeleteMidQEntry(midQ); + if (rstart && rc == -EACCES) + return -ERESTARTSYS; + return rc; +} From 66abda5e1fa48e12e06d0b68746b0e67202a97d2 Mon Sep 17 00:00:00 2001 From: Steve French Date: Fri, 11 Aug 2006 16:52:09 +0000 Subject: [PATCH 007/115] [CIFS] Fix oops when negotiating lanman and no password specified Pointed out by Guenter Kukkukk Signed-of-by: Steve French (cherry picked from bbf33d512da608c7221fec42b56b9ef89c25a5ee commit) --- fs/cifs/cifsencrypt.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/fs/cifs/cifsencrypt.c b/fs/cifs/cifsencrypt.c index a89efaf78a2..4bc250b2d9f 100644 --- a/fs/cifs/cifsencrypt.c +++ b/fs/cifs/cifsencrypt.c @@ -277,7 +277,8 @@ void calc_lanman_hash(struct cifsSesInfo * ses, char * lnm_session_key) return; memset(password_with_pad, 0, CIFS_ENCPWD_SIZE); - strncpy(password_with_pad, ses->password, CIFS_ENCPWD_SIZE); + if(ses->password) + strncpy(password_with_pad, ses->password, CIFS_ENCPWD_SIZE); if((ses->server->secMode & SECMODE_PW_ENCRYPT) == 0) if(extended_security & CIFSSEC_MAY_PLNTXT) { From e466e4876bf39474e15d0572f2204578137ae7f5 Mon Sep 17 00:00:00 2001 From: Steve French Date: Tue, 15 Aug 2006 13:07:18 +0000 Subject: [PATCH 008/115] [CIFS] Fix oops in cifs_close due to unitialized lock sem and list in new POSIX locking code Signed-off-by: Steve French --- fs/cifs/CHANGES | 5 ++++- fs/cifs/connect.c | 14 ++++++++------ fs/cifs/dir.c | 4 ++++ 3 files changed, 16 insertions(+), 7 deletions(-) diff --git a/fs/cifs/CHANGES b/fs/cifs/CHANGES index acb843b9bc3..bd37727526e 100644 --- a/fs/cifs/CHANGES +++ b/fs/cifs/CHANGES @@ -2,7 +2,10 @@ Version 1.45 ------------ Do not time out lockw calls when using posix extensions. Do not time out requests if server still responding reasonably fast -on requests on other threads +on requests on other threads. Improve POSIX locking emulation, +(lock cancel now works, and unlock of merged range works even +to Windows servers now). Fix oops on mount to lanman servers +(win9x, os/2 etc.) when null password. Version 1.44 ------------ diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index b706b4f48b1..5d394c72686 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -1271,33 +1271,35 @@ find_unc(__be32 new_target_ip_addr, char *uncName, char *userName) read_lock(&GlobalSMBSeslock); list_for_each(tmp, &GlobalTreeConnectionList) { - cFYI(1, ("Next tcon - ")); + cFYI(1, ("Next tcon")); tcon = list_entry(tmp, struct cifsTconInfo, cifsConnectionList); if (tcon->ses) { if (tcon->ses->server) { cFYI(1, - (" old ip addr: %x == new ip %x ?", + ("old ip addr: %x == new ip %x ?", tcon->ses->server->addr.sockAddr.sin_addr. s_addr, new_target_ip_addr)); if (tcon->ses->server->addr.sockAddr.sin_addr. s_addr == new_target_ip_addr) { - /* BB lock tcon and server and tcp session and increment use count here? */ + /* BB lock tcon, server and tcp session and increment use count here? */ /* found a match on the TCP session */ /* BB check if reconnection needed */ - cFYI(1,("Matched ip, old UNC: %s == new: %s ?", + cFYI(1,("IP match, old UNC: %s new: %s", tcon->treeName, uncName)); if (strncmp (tcon->treeName, uncName, MAX_TREE_SIZE) == 0) { cFYI(1, - ("Matched UNC, old user: %s == new: %s ?", + ("and old usr: %s new: %s", tcon->treeName, uncName)); if (strncmp (tcon->ses->userName, userName, MAX_USERNAME_SIZE) == 0) { read_unlock(&GlobalSMBSeslock); - return tcon;/* also matched user (smb session)*/ + /* matched smb session + (user name */ + return tcon; } } } diff --git a/fs/cifs/dir.c b/fs/cifs/dir.c index ba4cbe9b068..914239d5363 100644 --- a/fs/cifs/dir.c +++ b/fs/cifs/dir.c @@ -267,6 +267,10 @@ cifs_create(struct inode *inode, struct dentry *direntry, int mode, pCifsFile->invalidHandle = FALSE; pCifsFile->closePend = FALSE; init_MUTEX(&pCifsFile->fh_sem); + init_MUTEX(&pCifsFile->lock_sem); + INIT_LIST_HEAD(&pCifsFile->llist); + atomic_set(&pCifsFile->wrtPending,0); + /* set the following in open now pCifsFile->pfile = file; */ write_lock(&GlobalSMBSeslock); From 5ddaa683a513439081c9511b0d9ad490672c51c9 Mon Sep 17 00:00:00 2001 From: Steve French Date: Tue, 15 Aug 2006 13:35:48 +0000 Subject: [PATCH 009/115] [CIFS] endian errors in lanman protocol support le16 compared to host-endian constant u8 fed to le32_to_cpu() le16 compared to host-endian constant Signed-off-by: Al Viro Signed-off-by: Steve French --- fs/cifs/cifssmb.c | 2 +- fs/cifs/readdir.c | 2 +- fs/cifs/sess.c | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-) diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index dcbc3e075e8..075d8fb3d37 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -477,7 +477,7 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses) /* BB get server time for time conversions and add code to use it and timezone since this is not UTC */ - if (rsp->EncryptionKeyLength == CIFS_CRYPTO_KEY_SIZE) { + if (rsp->EncryptionKeyLength == cpu_to_le16(CIFS_CRYPTO_KEY_SIZE)) { memcpy(server->cryptKey, rsp->EncryptionKey, CIFS_CRYPTO_KEY_SIZE); } else if (server->secMode & SECMODE_PW_ENCRYPT) { diff --git a/fs/cifs/readdir.c b/fs/cifs/readdir.c index 03bbcb37791..105761e3ba0 100644 --- a/fs/cifs/readdir.c +++ b/fs/cifs/readdir.c @@ -556,7 +556,7 @@ static int cifs_entry_is_dot(char *current_entry, struct cifsFileInfo *cfile) FIND_FILE_STANDARD_INFO * pFindData = (FIND_FILE_STANDARD_INFO *)current_entry; filename = &pFindData->FileName[0]; - len = le32_to_cpu(pFindData->FileNameLength); + len = pFindData->FileNameLength; } else { cFYI(1,("Unknown findfirst level %d",cfile->srch_inf.info_level)); } diff --git a/fs/cifs/sess.c b/fs/cifs/sess.c index 7202d534ef0..d1705ab8136 100644 --- a/fs/cifs/sess.c +++ b/fs/cifs/sess.c @@ -372,7 +372,7 @@ CIFS_SessSetup(unsigned int xid, struct cifsSesInfo *ses, int first_time, /* no capabilities flags in old lanman negotiation */ - pSMB->old_req.PasswordLength = CIFS_SESS_KEY_SIZE; + pSMB->old_req.PasswordLength = cpu_to_le16(CIFS_SESS_KEY_SIZE); /* BB calculate hash with password */ /* and copy into bcc */ From ea4c07d780a6f7b7be2d984117bd3e0a2b772e3d Mon Sep 17 00:00:00 2001 From: Steve French Date: Wed, 16 Aug 2006 19:44:25 +0000 Subject: [PATCH 010/115] [CIFS] Do not send Query All EAs SMB when mount option nouser_xattr specified Pointed out by Bjoern Jacke Signed-off-by: Steve French --- fs/cifs/CHANGES | 3 ++- fs/cifs/README | 2 +- fs/cifs/xattr.c | 6 +++++- 3 files changed, 8 insertions(+), 3 deletions(-) diff --git a/fs/cifs/CHANGES b/fs/cifs/CHANGES index bd37727526e..0feb3bd49cb 100644 --- a/fs/cifs/CHANGES +++ b/fs/cifs/CHANGES @@ -5,7 +5,8 @@ time out requests if server still responding reasonably fast on requests on other threads. Improve POSIX locking emulation, (lock cancel now works, and unlock of merged range works even to Windows servers now). Fix oops on mount to lanman servers -(win9x, os/2 etc.) when null password. +(win9x, os/2 etc.) when null password. Do not send listxattr +(SMB to query all EAs) if nouser_xattr specified. Version 1.44 ------------ diff --git a/fs/cifs/README b/fs/cifs/README index 7986d0d97ac..5f0e1bd64fe 100644 --- a/fs/cifs/README +++ b/fs/cifs/README @@ -408,7 +408,7 @@ A partial list of the supported mount options follows: user_xattr Allow getting and setting user xattrs as OS/2 EAs (extended attributes) to the server (default) e.g. via setfattr and getfattr utilities. - nouser_xattr Do not allow getfattr/setfattr to get/set xattrs + nouser_xattr Do not allow getfattr/setfattr to get/set/list xattrs mapchars Translate six of the seven reserved characters (not backslash) *?<>|: to the remap range (above 0xF000), which also diff --git a/fs/cifs/xattr.c b/fs/cifs/xattr.c index 7754d641775..067648b7179 100644 --- a/fs/cifs/xattr.c +++ b/fs/cifs/xattr.c @@ -330,11 +330,15 @@ ssize_t cifs_listxattr(struct dentry * direntry, char * data, size_t buf_size) sb = direntry->d_inode->i_sb; if(sb == NULL) return -EIO; - xid = GetXid(); cifs_sb = CIFS_SB(sb); pTcon = cifs_sb->tcon; + if(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_XATTR) + return -EOPNOTSUPP; + + xid = GetXid(); + full_path = build_path_from_dentry(direntry); if(full_path == NULL) { FreeXid(xid); From a813213d73bb29d6986c3e93669a9cd5f7984364 Mon Sep 17 00:00:00 2001 From: Paul Jackson Date: Mon, 14 Aug 2006 22:45:49 -0700 Subject: [PATCH 011/115] [IA64] panic if topology_init kzalloc fails There really is no sense trying to continue if the kzalloc of sysfs_cpus[] fails in ia64 topology_init. The code calling into here doesn't check errors very well, and one ends up with a nonobvious boot failure that wastes peoples time debugging. See for example the lkml thread at: http://lkml.org/lkml/2006/3/2/215 Since the system is totally dead when this kzalloc fails, not having yet even booted, might as well announce one's death boldly and plainly. Signed-off-by: Paul Jackson Signed-off-by: Andrew Morton Signed-off-by: Tony Luck --- arch/ia64/kernel/topology.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/arch/ia64/kernel/topology.c b/arch/ia64/kernel/topology.c index d24fa393b18..f648c610b10 100644 --- a/arch/ia64/kernel/topology.c +++ b/arch/ia64/kernel/topology.c @@ -67,10 +67,8 @@ static int __init topology_init(void) #endif sysfs_cpus = kzalloc(sizeof(struct ia64_cpu) * NR_CPUS, GFP_KERNEL); - if (!sysfs_cpus) { - err = -ENOMEM; - goto out; - } + if (!sysfs_cpus) + panic("kzalloc in topology_init failed - NR_CPUS too big?"); for_each_present_cpu(i) { if((err = arch_register_cpu(i))) From af07ac276af6eec5111a6567cb7eaf4d222bd332 Mon Sep 17 00:00:00 2001 From: Matt Porter Date: Fri, 4 Aug 2006 11:41:51 -0500 Subject: [PATCH 012/115] [POWERPC] Remove flush_dcache_all export Removes the flush_dcache_all export for non coherent platforms. We removed the last in-kernel user of this years ago in arch/ppc so it no longer serves a purpose. Plus, it breaks the build at the moment. Signed-off-by: Matt Porter Signed-off-by: Paul Mackerras --- arch/powerpc/kernel/ppc_ksyms.c | 4 ---- 1 file changed, 4 deletions(-) diff --git a/arch/powerpc/kernel/ppc_ksyms.c b/arch/powerpc/kernel/ppc_ksyms.c index f6a05f090b2..39d3bfcabcd 100644 --- a/arch/powerpc/kernel/ppc_ksyms.c +++ b/arch/powerpc/kernel/ppc_ksyms.c @@ -126,10 +126,6 @@ EXPORT_SYMBOL(pci_bus_mem_base_phys); EXPORT_SYMBOL(pci_bus_to_hose); #endif /* CONFIG_PCI */ -#ifdef CONFIG_NOT_COHERENT_CACHE -EXPORT_SYMBOL(flush_dcache_all); -#endif - EXPORT_SYMBOL(start_thread); EXPORT_SYMBOL(kernel_thread); From 054389f114fc55d525926b55e82b473b783a7d77 Mon Sep 17 00:00:00 2001 From: Matt Porter Date: Fri, 4 Aug 2006 11:44:01 -0500 Subject: [PATCH 013/115] [POWERPC] Fix powerpc 44x_mmu build The PIN_SIZE definition name changed, update 44x_mmu.c accordingly. Signed-off-by: Matt Porter Signed-off-by: Paul Mackerras --- arch/powerpc/mm/44x_mmu.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/powerpc/mm/44x_mmu.c b/arch/powerpc/mm/44x_mmu.c index 376829ed221..0a0a0487b33 100644 --- a/arch/powerpc/mm/44x_mmu.c +++ b/arch/powerpc/mm/44x_mmu.c @@ -103,7 +103,7 @@ unsigned long __init mmu_mapin_ram(void) /* Determine number of entries necessary to cover lowmem */ pinned_tlbs = (unsigned int) - (_ALIGN(total_lowmem, PPC44x_PIN_SIZE) >> PPC44x_PIN_SHIFT); + (_ALIGN(total_lowmem, PPC_PIN_SIZE) >> PPC44x_PIN_SHIFT); /* Write upper watermark to save location */ tlb_44x_hwater = PPC44x_LOW_SLOT - pinned_tlbs; @@ -111,7 +111,7 @@ unsigned long __init mmu_mapin_ram(void) /* If necessary, set additional pinned TLBs */ if (pinned_tlbs > 1) for (i = (PPC44x_LOW_SLOT-(pinned_tlbs-1)); i < PPC44x_LOW_SLOT; i++) { - unsigned int phys_addr = (PPC44x_LOW_SLOT-i) * PPC44x_PIN_SIZE; + unsigned int phys_addr = (PPC44x_LOW_SLOT-i) * PPC_PIN_SIZE; ppc44x_pin_tlb(i, phys_addr+PAGE_OFFSET, phys_addr); } From 5dc599c206ad50e1b190edfbc98b7cf8ce361003 Mon Sep 17 00:00:00 2001 From: Jon Loeliger Date: Tue, 15 Aug 2006 16:19:02 -0500 Subject: [PATCH 014/115] [POWERPC] Allow MPC8641 HPCN to build with CONFIG_PCI disabled too. Signed-off-by: Jon Loeliger Signed-off-by: Paul Mackerras --- arch/powerpc/platforms/86xx/mpc86xx_hpcn.c | 9 +++++++-- arch/powerpc/platforms/86xx/pci.c | 3 ++- include/asm-powerpc/mpc86xx.h | 2 -- 3 files changed, 9 insertions(+), 5 deletions(-) diff --git a/arch/powerpc/platforms/86xx/mpc86xx_hpcn.c b/arch/powerpc/platforms/86xx/mpc86xx_hpcn.c index 146da3001c6..813eb5811eb 100644 --- a/arch/powerpc/platforms/86xx/mpc86xx_hpcn.c +++ b/arch/powerpc/platforms/86xx/mpc86xx_hpcn.c @@ -52,6 +52,7 @@ unsigned long pci_dram_offset = 0; #endif +#ifdef CONFIG_PCI static void mpc86xx_8259_cascade(unsigned int irq, struct irq_desc *desc, struct pt_regs *regs) { @@ -60,14 +61,18 @@ static void mpc86xx_8259_cascade(unsigned int irq, struct irq_desc *desc, generic_handle_irq(cascade_irq, regs); desc->chip->eoi(irq); } +#endif /* CONFIG_PCI */ void __init mpc86xx_hpcn_init_irq(void) { struct mpic *mpic1; - struct device_node *np, *cascade_node = NULL; - int cascade_irq; + struct device_node *np; phys_addr_t openpic_paddr; +#ifdef CONFIG_PCI + struct device_node *cascade_node = NULL; + int cascade_irq; +#endif np = of_find_node_by_type(NULL, "open-pic"); if (np == NULL) diff --git a/arch/powerpc/platforms/86xx/pci.c b/arch/powerpc/platforms/86xx/pci.c index bc513904311..a8c8f0a4405 100644 --- a/arch/powerpc/platforms/86xx/pci.c +++ b/arch/powerpc/platforms/86xx/pci.c @@ -188,7 +188,8 @@ int __init add_bridge(struct device_node *dev) printk(KERN_INFO "Found MPC86xx PCIE host bridge at 0x%08lx. " "Firmware bus number: %d->%d\n", - rsrc.start, hose->first_busno, hose->last_busno); + (unsigned long) rsrc.start, + hose->first_busno, hose->last_busno); DBG(" ->Hose at 0x%p, cfg_addr=0x%p,cfg_data=0x%p\n", hose, hose->cfg_addr, hose->cfg_data); diff --git a/include/asm-powerpc/mpc86xx.h b/include/asm-powerpc/mpc86xx.h index f260382739f..2d6ad859df7 100644 --- a/include/asm-powerpc/mpc86xx.h +++ b/include/asm-powerpc/mpc86xx.h @@ -23,8 +23,6 @@ #define _ISA_MEM_BASE isa_mem_base #ifdef CONFIG_PCI #define PCI_DRAM_OFFSET pci_dram_offset -#else -#define PCI_DRAM_OFFSET 0 #endif #define CPU0_BOOT_RELEASE 0x01000000 From c85c41ad73c6db4cf4cc98c595cc5e2fdbdb53d5 Mon Sep 17 00:00:00 2001 From: Jon Loeliger Date: Thu, 17 Aug 2006 14:27:57 -0500 Subject: [PATCH 015/115] [POWERPC] Use mpc8641hpcn PIC base address from dev tree. After going through the trouble of setting up the PIC base address in the pic@40000 device tree node, use it instead of the obsolete hard-coded value. Signed-off-by: Jon Loeliger Signed-off-by: Paul Mackerras --- arch/powerpc/platforms/86xx/mpc86xx_hpcn.c | 19 +++++++++---------- include/asm-powerpc/mpc86xx.h | 1 - 2 files changed, 9 insertions(+), 11 deletions(-) diff --git a/arch/powerpc/platforms/86xx/mpc86xx_hpcn.c b/arch/powerpc/platforms/86xx/mpc86xx_hpcn.c index 813eb5811eb..0b1b52168bb 100644 --- a/arch/powerpc/platforms/86xx/mpc86xx_hpcn.c +++ b/arch/powerpc/platforms/86xx/mpc86xx_hpcn.c @@ -68,37 +68,36 @@ mpc86xx_hpcn_init_irq(void) { struct mpic *mpic1; struct device_node *np; - phys_addr_t openpic_paddr; + struct resource res; #ifdef CONFIG_PCI struct device_node *cascade_node = NULL; int cascade_irq; #endif + /* Determine PIC address. */ np = of_find_node_by_type(NULL, "open-pic"); if (np == NULL) return; - - /* Determine the Physical Address of the OpenPIC regs */ - openpic_paddr = get_immrbase() + MPC86xx_OPENPIC_OFFSET; + of_address_to_resource(np, 0, &res); /* Alloc mpic structure and per isu has 16 INT entries. */ - mpic1 = mpic_alloc(np, openpic_paddr, + mpic1 = mpic_alloc(np, res.start, MPIC_PRIMARY | MPIC_WANTS_RESET | MPIC_BIG_ENDIAN, 16, NR_IRQS - 4, " MPIC "); BUG_ON(mpic1 == NULL); - mpic_assign_isu(mpic1, 0, openpic_paddr + 0x10000); + mpic_assign_isu(mpic1, 0, res.start + 0x10000); /* 48 Internal Interrupts */ - mpic_assign_isu(mpic1, 1, openpic_paddr + 0x10200); - mpic_assign_isu(mpic1, 2, openpic_paddr + 0x10400); - mpic_assign_isu(mpic1, 3, openpic_paddr + 0x10600); + mpic_assign_isu(mpic1, 1, res.start + 0x10200); + mpic_assign_isu(mpic1, 2, res.start + 0x10400); + mpic_assign_isu(mpic1, 3, res.start + 0x10600); /* 16 External interrupts * Moving them from [0 - 15] to [64 - 79] */ - mpic_assign_isu(mpic1, 4, openpic_paddr + 0x10000); + mpic_assign_isu(mpic1, 4, res.start + 0x10000); mpic_init(mpic1); diff --git a/include/asm-powerpc/mpc86xx.h b/include/asm-powerpc/mpc86xx.h index 2d6ad859df7..b85df45b1a8 100644 --- a/include/asm-powerpc/mpc86xx.h +++ b/include/asm-powerpc/mpc86xx.h @@ -31,7 +31,6 @@ #define MCM_PORT_CONFIG_OFFSET 0x1010 /* Offset from CCSRBAR */ -#define MPC86xx_OPENPIC_OFFSET (0x40000) #define MPC86xx_MCM_OFFSET (0x00000) #define MPC86xx_MCM_SIZE (0x02000) From 986e12fa74c837d7fe5bdfe80666e2e2d46711bd Mon Sep 17 00:00:00 2001 From: Russ Anderson Date: Thu, 24 Aug 2006 11:08:52 -0500 Subject: [PATCH 016/115] [IA64] remove redundant local_irq_save() calls from sn_sal.h sn_change_memprotect() does a local_irq_save() then calls ia64_sal_oemcall_nolock() which calls SAL_CALL_NOLOCK() which also does a local_irq_save(). This patch removes the redundant local_irq_save() and local_irq_restore() calls in sn_change_memprotect() and sn_inject_error(). Signed-off-by: Russ Anderson Signed-off-by: Tony Luck --- include/asm-ia64/sn/sn_sal.h | 6 ------ 1 file changed, 6 deletions(-) diff --git a/include/asm-ia64/sn/sn_sal.h b/include/asm-ia64/sn/sn_sal.h index bd4452bda35..ba826b3f75b 100644 --- a/include/asm-ia64/sn/sn_sal.h +++ b/include/asm-ia64/sn/sn_sal.h @@ -706,12 +706,9 @@ static inline int sn_change_memprotect(u64 paddr, u64 len, u64 perms, u64 *nasid_array) { struct ia64_sal_retval ret_stuff; - unsigned long irq_flags; - local_irq_save(irq_flags); ia64_sal_oemcall_nolock(&ret_stuff, SN_SAL_MEMPROTECT, paddr, len, (u64)nasid_array, perms, 0, 0, 0); - local_irq_restore(irq_flags); return ret_stuff.status; } #define SN_MEMPROT_ACCESS_CLASS_0 0x14a080 @@ -1143,12 +1140,9 @@ static inline int sn_inject_error(u64 paddr, u64 *data, u64 *ecc) { struct ia64_sal_retval ret_stuff; - unsigned long irq_flags; - local_irq_save(irq_flags); ia64_sal_oemcall_nolock(&ret_stuff, SN_SAL_INJECT_ERROR, paddr, (u64)data, (u64)ecc, 0, 0, 0, 0); - local_irq_restore(irq_flags); return ret_stuff.status; } From c57d68caeed7bc335e6d35c951a9abae733a580b Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Tue, 22 Aug 2006 19:43:27 -0700 Subject: [PATCH 017/115] [IA64] Increase default nodes shift to 10, nr_cpus to 1024 Change both the NODES_SHIFT and the NR_CPUS so that even big machines can boot all nodes and processors with a generic kernel. Signed-off-by: Christoph Lameter Signed-off-by: Tony Luck --- arch/ia64/Kconfig | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig index 47de9ee6bcd..674de894347 100644 --- a/arch/ia64/Kconfig +++ b/arch/ia64/Kconfig @@ -258,7 +258,7 @@ config NR_CPUS int "Maximum number of CPUs (2-1024)" range 2 1024 depends on SMP - default "64" + default "1024" help You should set this to the number of CPUs in your system, but keep in mind that a kernel compiled for, e.g., 2 CPUs will boot but @@ -354,7 +354,7 @@ config NUMA config NODES_SHIFT int "Max num nodes shift(3-10)" range 3 10 - default "8" + default "10" depends on NEED_MULTIPLE_NODES help This option specifies the maximum number of nodes in your SSI system. From bf4152dd7ccb6c060d786200a893dfe30193a07f Mon Sep 17 00:00:00 2001 From: Kim Phillips Date: Fri, 25 Aug 2006 11:58:53 -0500 Subject: [PATCH 018/115] [POWERPC] back up old school ipic.[hc] to arch/ppc Keep from breaking 83xx arch/ppc build. Back up old school arch/powerpc/sysdev/ipic.[hc] to arch/ppc/syslib. Signed-off-by: Kim Phillips Signed-off-by: Paul Mackerras --- arch/powerpc/sysdev/Makefile | 4 +- arch/ppc/syslib/Makefile | 2 +- arch/ppc/syslib/ipic.c | 646 +++++++++++++++++++++++++++++++++++ arch/ppc/syslib/ipic.h | 47 +++ include/asm-powerpc/ipic.h | 12 +- 5 files changed, 705 insertions(+), 6 deletions(-) create mode 100644 arch/ppc/syslib/ipic.c create mode 100644 arch/ppc/syslib/ipic.h diff --git a/arch/powerpc/sysdev/Makefile b/arch/powerpc/sysdev/Makefile index cebfae24260..e5e999ea891 100644 --- a/arch/powerpc/sysdev/Makefile +++ b/arch/powerpc/sysdev/Makefile @@ -9,11 +9,11 @@ obj-$(CONFIG_BOOKE) += dcr.o obj-$(CONFIG_40x) += dcr.o obj-$(CONFIG_U3_DART) += dart_iommu.o obj-$(CONFIG_MMIO_NVRAM) += mmio_nvram.o -obj-$(CONFIG_PPC_83xx) += ipic.o obj-$(CONFIG_FSL_SOC) += fsl_soc.o obj-$(CONFIG_PPC_TODC) += todc.o obj-$(CONFIG_TSI108_BRIDGE) += tsi108_pci.o tsi108_dev.o ifeq ($(CONFIG_PPC_MERGE),y) obj-$(CONFIG_PPC_I8259) += i8259.o - endif +obj-$(CONFIG_PPC_83xx) += ipic.o +endif diff --git a/arch/ppc/syslib/Makefile b/arch/ppc/syslib/Makefile index 2497bbc07e7..dca23f2ef85 100644 --- a/arch/ppc/syslib/Makefile +++ b/arch/ppc/syslib/Makefile @@ -93,7 +93,7 @@ obj-$(CONFIG_PCI) += pci_auto.o endif obj-$(CONFIG_RAPIDIO) += ppc85xx_rio.o obj-$(CONFIG_83xx) += ppc83xx_setup.o ppc_sys.o \ - mpc83xx_sys.o mpc83xx_devices.o + mpc83xx_sys.o mpc83xx_devices.o ipic.o ifeq ($(CONFIG_83xx),y) obj-$(CONFIG_PCI) += pci_auto.o endif diff --git a/arch/ppc/syslib/ipic.c b/arch/ppc/syslib/ipic.c new file mode 100644 index 00000000000..46801f5ec03 --- /dev/null +++ b/arch/ppc/syslib/ipic.c @@ -0,0 +1,646 @@ +/* + * include/asm-ppc/ipic.c + * + * IPIC routines implementations. + * + * Copyright 2005 Freescale Semiconductor, Inc. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation; either version 2 of the License, or (at your + * option) any later version. + */ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "ipic.h" + +static struct ipic p_ipic; +static struct ipic * primary_ipic; + +static struct ipic_info ipic_info[] = { + [9] = { + .pend = IPIC_SIPNR_H, + .mask = IPIC_SIMSR_H, + .prio = IPIC_SIPRR_D, + .force = IPIC_SIFCR_H, + .bit = 24, + .prio_mask = 0, + }, + [10] = { + .pend = IPIC_SIPNR_H, + .mask = IPIC_SIMSR_H, + .prio = IPIC_SIPRR_D, + .force = IPIC_SIFCR_H, + .bit = 25, + .prio_mask = 1, + }, + [11] = { + .pend = IPIC_SIPNR_H, + .mask = IPIC_SIMSR_H, + .prio = IPIC_SIPRR_D, + .force = IPIC_SIFCR_H, + .bit = 26, + .prio_mask = 2, + }, + [14] = { + .pend = IPIC_SIPNR_H, + .mask = IPIC_SIMSR_H, + .prio = IPIC_SIPRR_D, + .force = IPIC_SIFCR_H, + .bit = 29, + .prio_mask = 5, + }, + [15] = { + .pend = IPIC_SIPNR_H, + .mask = IPIC_SIMSR_H, + .prio = IPIC_SIPRR_D, + .force = IPIC_SIFCR_H, + .bit = 30, + .prio_mask = 6, + }, + [16] = { + .pend = IPIC_SIPNR_H, + .mask = IPIC_SIMSR_H, + .prio = IPIC_SIPRR_D, + .force = IPIC_SIFCR_H, + .bit = 31, + .prio_mask = 7, + }, + [17] = { + .pend = IPIC_SEPNR, + .mask = IPIC_SEMSR, + .prio = IPIC_SMPRR_A, + .force = IPIC_SEFCR, + .bit = 1, + .prio_mask = 5, + }, + [18] = { + .pend = IPIC_SEPNR, + .mask = IPIC_SEMSR, + .prio = IPIC_SMPRR_A, + .force = IPIC_SEFCR, + .bit = 2, + .prio_mask = 6, + }, + [19] = { + .pend = IPIC_SEPNR, + .mask = IPIC_SEMSR, + .prio = IPIC_SMPRR_A, + .force = IPIC_SEFCR, + .bit = 3, + .prio_mask = 7, + }, + [20] = { + .pend = IPIC_SEPNR, + .mask = IPIC_SEMSR, + .prio = IPIC_SMPRR_B, + .force = IPIC_SEFCR, + .bit = 4, + .prio_mask = 4, + }, + [21] = { + .pend = IPIC_SEPNR, + .mask = IPIC_SEMSR, + .prio = IPIC_SMPRR_B, + .force = IPIC_SEFCR, + .bit = 5, + .prio_mask = 5, + }, + [22] = { + .pend = IPIC_SEPNR, + .mask = IPIC_SEMSR, + .prio = IPIC_SMPRR_B, + .force = IPIC_SEFCR, + .bit = 6, + .prio_mask = 6, + }, + [23] = { + .pend = IPIC_SEPNR, + .mask = IPIC_SEMSR, + .prio = IPIC_SMPRR_B, + .force = IPIC_SEFCR, + .bit = 7, + .prio_mask = 7, + }, + [32] = { + .pend = IPIC_SIPNR_H, + .mask = IPIC_SIMSR_H, + .prio = IPIC_SIPRR_A, + .force = IPIC_SIFCR_H, + .bit = 0, + .prio_mask = 0, + }, + [33] = { + .pend = IPIC_SIPNR_H, + .mask = IPIC_SIMSR_H, + .prio = IPIC_SIPRR_A, + .force = IPIC_SIFCR_H, + .bit = 1, + .prio_mask = 1, + }, + [34] = { + .pend = IPIC_SIPNR_H, + .mask = IPIC_SIMSR_H, + .prio = IPIC_SIPRR_A, + .force = IPIC_SIFCR_H, + .bit = 2, + .prio_mask = 2, + }, + [35] = { + .pend = IPIC_SIPNR_H, + .mask = IPIC_SIMSR_H, + .prio = IPIC_SIPRR_A, + .force = IPIC_SIFCR_H, + .bit = 3, + .prio_mask = 3, + }, + [36] = { + .pend = IPIC_SIPNR_H, + .mask = IPIC_SIMSR_H, + .prio = IPIC_SIPRR_A, + .force = IPIC_SIFCR_H, + .bit = 4, + .prio_mask = 4, + }, + [37] = { + .pend = IPIC_SIPNR_H, + .mask = IPIC_SIMSR_H, + .prio = IPIC_SIPRR_A, + .force = IPIC_SIFCR_H, + .bit = 5, + .prio_mask = 5, + }, + [38] = { + .pend = IPIC_SIPNR_H, + .mask = IPIC_SIMSR_H, + .prio = IPIC_SIPRR_A, + .force = IPIC_SIFCR_H, + .bit = 6, + .prio_mask = 6, + }, + [39] = { + .pend = IPIC_SIPNR_H, + .mask = IPIC_SIMSR_H, + .prio = IPIC_SIPRR_A, + .force = IPIC_SIFCR_H, + .bit = 7, + .prio_mask = 7, + }, + [48] = { + .pend = IPIC_SEPNR, + .mask = IPIC_SEMSR, + .prio = IPIC_SMPRR_A, + .force = IPIC_SEFCR, + .bit = 0, + .prio_mask = 4, + }, + [64] = { + .pend = IPIC_SIPNR_H, + .mask = IPIC_SIMSR_L, + .prio = IPIC_SMPRR_A, + .force = IPIC_SIFCR_L, + .bit = 0, + .prio_mask = 0, + }, + [65] = { + .pend = IPIC_SIPNR_H, + .mask = IPIC_SIMSR_L, + .prio = IPIC_SMPRR_A, + .force = IPIC_SIFCR_L, + .bit = 1, + .prio_mask = 1, + }, + [66] = { + .pend = IPIC_SIPNR_H, + .mask = IPIC_SIMSR_L, + .prio = IPIC_SMPRR_A, + .force = IPIC_SIFCR_L, + .bit = 2, + .prio_mask = 2, + }, + [67] = { + .pend = IPIC_SIPNR_H, + .mask = IPIC_SIMSR_L, + .prio = IPIC_SMPRR_A, + .force = IPIC_SIFCR_L, + .bit = 3, + .prio_mask = 3, + }, + [68] = { + .pend = IPIC_SIPNR_H, + .mask = IPIC_SIMSR_L, + .prio = IPIC_SMPRR_B, + .force = IPIC_SIFCR_L, + .bit = 4, + .prio_mask = 0, + }, + [69] = { + .pend = IPIC_SIPNR_H, + .mask = IPIC_SIMSR_L, + .prio = IPIC_SMPRR_B, + .force = IPIC_SIFCR_L, + .bit = 5, + .prio_mask = 1, + }, + [70] = { + .pend = IPIC_SIPNR_H, + .mask = IPIC_SIMSR_L, + .prio = IPIC_SMPRR_B, + .force = IPIC_SIFCR_L, + .bit = 6, + .prio_mask = 2, + }, + [71] = { + .pend = IPIC_SIPNR_H, + .mask = IPIC_SIMSR_L, + .prio = IPIC_SMPRR_B, + .force = IPIC_SIFCR_L, + .bit = 7, + .prio_mask = 3, + }, + [72] = { + .pend = IPIC_SIPNR_H, + .mask = IPIC_SIMSR_L, + .prio = 0, + .force = IPIC_SIFCR_L, + .bit = 8, + }, + [73] = { + .pend = IPIC_SIPNR_H, + .mask = IPIC_SIMSR_L, + .prio = 0, + .force = IPIC_SIFCR_L, + .bit = 9, + }, + [74] = { + .pend = IPIC_SIPNR_H, + .mask = IPIC_SIMSR_L, + .prio = 0, + .force = IPIC_SIFCR_L, + .bit = 10, + }, + [75] = { + .pend = IPIC_SIPNR_H, + .mask = IPIC_SIMSR_L, + .prio = 0, + .force = IPIC_SIFCR_L, + .bit = 11, + }, + [76] = { + .pend = IPIC_SIPNR_H, + .mask = IPIC_SIMSR_L, + .prio = 0, + .force = IPIC_SIFCR_L, + .bit = 12, + }, + [77] = { + .pend = IPIC_SIPNR_H, + .mask = IPIC_SIMSR_L, + .prio = 0, + .force = IPIC_SIFCR_L, + .bit = 13, + }, + [78] = { + .pend = IPIC_SIPNR_H, + .mask = IPIC_SIMSR_L, + .prio = 0, + .force = IPIC_SIFCR_L, + .bit = 14, + }, + [79] = { + .pend = IPIC_SIPNR_H, + .mask = IPIC_SIMSR_L, + .prio = 0, + .force = IPIC_SIFCR_L, + .bit = 15, + }, + [80] = { + .pend = IPIC_SIPNR_H, + .mask = IPIC_SIMSR_L, + .prio = 0, + .force = IPIC_SIFCR_L, + .bit = 16, + }, + [84] = { + .pend = IPIC_SIPNR_H, + .mask = IPIC_SIMSR_L, + .prio = 0, + .force = IPIC_SIFCR_L, + .bit = 20, + }, + [85] = { + .pend = IPIC_SIPNR_H, + .mask = IPIC_SIMSR_L, + .prio = 0, + .force = IPIC_SIFCR_L, + .bit = 21, + }, + [90] = { + .pend = IPIC_SIPNR_H, + .mask = IPIC_SIMSR_L, + .prio = 0, + .force = IPIC_SIFCR_L, + .bit = 26, + }, + [91] = { + .pend = IPIC_SIPNR_H, + .mask = IPIC_SIMSR_L, + .prio = 0, + .force = IPIC_SIFCR_L, + .bit = 27, + }, +}; + +static inline u32 ipic_read(volatile u32 __iomem *base, unsigned int reg) +{ + return in_be32(base + (reg >> 2)); +} + +static inline void ipic_write(volatile u32 __iomem *base, unsigned int reg, u32 value) +{ + out_be32(base + (reg >> 2), value); +} + +static inline struct ipic * ipic_from_irq(unsigned int irq) +{ + return primary_ipic; +} + +static void ipic_enable_irq(unsigned int irq) +{ + struct ipic *ipic = ipic_from_irq(irq); + unsigned int src = irq - ipic->irq_offset; + u32 temp; + + temp = ipic_read(ipic->regs, ipic_info[src].mask); + temp |= (1 << (31 - ipic_info[src].bit)); + ipic_write(ipic->regs, ipic_info[src].mask, temp); +} + +static void ipic_disable_irq(unsigned int irq) +{ + struct ipic *ipic = ipic_from_irq(irq); + unsigned int src = irq - ipic->irq_offset; + u32 temp; + + temp = ipic_read(ipic->regs, ipic_info[src].mask); + temp &= ~(1 << (31 - ipic_info[src].bit)); + ipic_write(ipic->regs, ipic_info[src].mask, temp); +} + +static void ipic_disable_irq_and_ack(unsigned int irq) +{ + struct ipic *ipic = ipic_from_irq(irq); + unsigned int src = irq - ipic->irq_offset; + u32 temp; + + ipic_disable_irq(irq); + + temp = ipic_read(ipic->regs, ipic_info[src].pend); + temp |= (1 << (31 - ipic_info[src].bit)); + ipic_write(ipic->regs, ipic_info[src].pend, temp); +} + +static void ipic_end_irq(unsigned int irq) +{ + if (!(irq_desc[irq].status & (IRQ_DISABLED|IRQ_INPROGRESS))) + ipic_enable_irq(irq); +} + +struct hw_interrupt_type ipic = { + .typename = " IPIC ", + .enable = ipic_enable_irq, + .disable = ipic_disable_irq, + .ack = ipic_disable_irq_and_ack, + .end = ipic_end_irq, +}; + +void __init ipic_init(phys_addr_t phys_addr, + unsigned int flags, + unsigned int irq_offset, + unsigned char *senses, + unsigned int senses_count) +{ + u32 i, temp = 0; + + primary_ipic = &p_ipic; + primary_ipic->regs = ioremap(phys_addr, MPC83xx_IPIC_SIZE); + + primary_ipic->irq_offset = irq_offset; + + ipic_write(primary_ipic->regs, IPIC_SICNR, 0x0); + + /* default priority scheme is grouped. If spread mode is required + * configure SICFR accordingly */ + if (flags & IPIC_SPREADMODE_GRP_A) + temp |= SICFR_IPSA; + if (flags & IPIC_SPREADMODE_GRP_D) + temp |= SICFR_IPSD; + if (flags & IPIC_SPREADMODE_MIX_A) + temp |= SICFR_MPSA; + if (flags & IPIC_SPREADMODE_MIX_B) + temp |= SICFR_MPSB; + + ipic_write(primary_ipic->regs, IPIC_SICNR, temp); + + /* handle MCP route */ + temp = 0; + if (flags & IPIC_DISABLE_MCP_OUT) + temp = SERCR_MCPR; + ipic_write(primary_ipic->regs, IPIC_SERCR, temp); + + /* handle routing of IRQ0 to MCP */ + temp = ipic_read(primary_ipic->regs, IPIC_SEMSR); + + if (flags & IPIC_IRQ0_MCP) + temp |= SEMSR_SIRQ0; + else + temp &= ~SEMSR_SIRQ0; + + ipic_write(primary_ipic->regs, IPIC_SEMSR, temp); + + for (i = 0 ; i < NR_IPIC_INTS ; i++) { + irq_desc[i+irq_offset].chip = &ipic; + irq_desc[i+irq_offset].status = IRQ_LEVEL; + } + + temp = 0; + for (i = 0 ; i < senses_count ; i++) { + if ((senses[i] & IRQ_SENSE_MASK) == IRQ_SENSE_EDGE) { + temp |= 1 << (15 - i); + if (i != 0) + irq_desc[i + irq_offset + MPC83xx_IRQ_EXT1 - 1].status = 0; + else + irq_desc[irq_offset + MPC83xx_IRQ_EXT0].status = 0; + } + } + ipic_write(primary_ipic->regs, IPIC_SECNR, temp); + + printk ("IPIC (%d IRQ sources, %d External IRQs) at %p\n", NR_IPIC_INTS, + senses_count, primary_ipic->regs); +} + +int ipic_set_priority(unsigned int irq, unsigned int priority) +{ + struct ipic *ipic = ipic_from_irq(irq); + unsigned int src = irq - ipic->irq_offset; + u32 temp; + + if (priority > 7) + return -EINVAL; + if (src > 127) + return -EINVAL; + if (ipic_info[src].prio == 0) + return -EINVAL; + + temp = ipic_read(ipic->regs, ipic_info[src].prio); + + if (priority < 4) { + temp &= ~(0x7 << (20 + (3 - priority) * 3)); + temp |= ipic_info[src].prio_mask << (20 + (3 - priority) * 3); + } else { + temp &= ~(0x7 << (4 + (7 - priority) * 3)); + temp |= ipic_info[src].prio_mask << (4 + (7 - priority) * 3); + } + + ipic_write(ipic->regs, ipic_info[src].prio, temp); + + return 0; +} + +void ipic_set_highest_priority(unsigned int irq) +{ + struct ipic *ipic = ipic_from_irq(irq); + unsigned int src = irq - ipic->irq_offset; + u32 temp; + + temp = ipic_read(ipic->regs, IPIC_SICFR); + + /* clear and set HPI */ + temp &= 0x7f000000; + temp |= (src & 0x7f) << 24; + + ipic_write(ipic->regs, IPIC_SICFR, temp); +} + +void ipic_set_default_priority(void) +{ + ipic_set_priority(MPC83xx_IRQ_TSEC1_TX, 0); + ipic_set_priority(MPC83xx_IRQ_TSEC1_RX, 1); + ipic_set_priority(MPC83xx_IRQ_TSEC1_ERROR, 2); + ipic_set_priority(MPC83xx_IRQ_TSEC2_TX, 3); + ipic_set_priority(MPC83xx_IRQ_TSEC2_RX, 4); + ipic_set_priority(MPC83xx_IRQ_TSEC2_ERROR, 5); + ipic_set_priority(MPC83xx_IRQ_USB2_DR, 6); + ipic_set_priority(MPC83xx_IRQ_USB2_MPH, 7); + + ipic_set_priority(MPC83xx_IRQ_UART1, 0); + ipic_set_priority(MPC83xx_IRQ_UART2, 1); + ipic_set_priority(MPC83xx_IRQ_SEC2, 2); + ipic_set_priority(MPC83xx_IRQ_IIC1, 5); + ipic_set_priority(MPC83xx_IRQ_IIC2, 6); + ipic_set_priority(MPC83xx_IRQ_SPI, 7); + ipic_set_priority(MPC83xx_IRQ_RTC_SEC, 0); + ipic_set_priority(MPC83xx_IRQ_PIT, 1); + ipic_set_priority(MPC83xx_IRQ_PCI1, 2); + ipic_set_priority(MPC83xx_IRQ_PCI2, 3); + ipic_set_priority(MPC83xx_IRQ_EXT0, 4); + ipic_set_priority(MPC83xx_IRQ_EXT1, 5); + ipic_set_priority(MPC83xx_IRQ_EXT2, 6); + ipic_set_priority(MPC83xx_IRQ_EXT3, 7); + ipic_set_priority(MPC83xx_IRQ_RTC_ALR, 0); + ipic_set_priority(MPC83xx_IRQ_MU, 1); + ipic_set_priority(MPC83xx_IRQ_SBA, 2); + ipic_set_priority(MPC83xx_IRQ_DMA, 3); + ipic_set_priority(MPC83xx_IRQ_EXT4, 4); + ipic_set_priority(MPC83xx_IRQ_EXT5, 5); + ipic_set_priority(MPC83xx_IRQ_EXT6, 6); + ipic_set_priority(MPC83xx_IRQ_EXT7, 7); +} + +void ipic_enable_mcp(enum ipic_mcp_irq mcp_irq) +{ + struct ipic *ipic = primary_ipic; + u32 temp; + + temp = ipic_read(ipic->regs, IPIC_SERMR); + temp |= (1 << (31 - mcp_irq)); + ipic_write(ipic->regs, IPIC_SERMR, temp); +} + +void ipic_disable_mcp(enum ipic_mcp_irq mcp_irq) +{ + struct ipic *ipic = primary_ipic; + u32 temp; + + temp = ipic_read(ipic->regs, IPIC_SERMR); + temp &= (1 << (31 - mcp_irq)); + ipic_write(ipic->regs, IPIC_SERMR, temp); +} + +u32 ipic_get_mcp_status(void) +{ + return ipic_read(primary_ipic->regs, IPIC_SERMR); +} + +void ipic_clear_mcp_status(u32 mask) +{ + ipic_write(primary_ipic->regs, IPIC_SERMR, mask); +} + +/* Return an interrupt vector or -1 if no interrupt is pending. */ +int ipic_get_irq(struct pt_regs *regs) +{ + int irq; + + irq = ipic_read(primary_ipic->regs, IPIC_SIVCR) & 0x7f; + + if (irq == 0) /* 0 --> no irq is pending */ + irq = -1; + + return irq; +} + +static struct sysdev_class ipic_sysclass = { + set_kset_name("ipic"), +}; + +static struct sys_device device_ipic = { + .id = 0, + .cls = &ipic_sysclass, +}; + +static int __init init_ipic_sysfs(void) +{ + int rc; + + if (!primary_ipic->regs) + return -ENODEV; + printk(KERN_DEBUG "Registering ipic with sysfs...\n"); + + rc = sysdev_class_register(&ipic_sysclass); + if (rc) { + printk(KERN_ERR "Failed registering ipic sys class\n"); + return -ENODEV; + } + rc = sysdev_register(&device_ipic); + if (rc) { + printk(KERN_ERR "Failed registering ipic sys device\n"); + return -ENODEV; + } + return 0; +} + +subsys_initcall(init_ipic_sysfs); diff --git a/arch/ppc/syslib/ipic.h b/arch/ppc/syslib/ipic.h new file mode 100644 index 00000000000..a60c9d18bb7 --- /dev/null +++ b/arch/ppc/syslib/ipic.h @@ -0,0 +1,47 @@ +/* + * IPIC private definitions and structure. + * + * Maintainer: Kumar Gala + * + * Copyright 2005 Freescale Semiconductor, Inc + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation; either version 2 of the License, or (at your + * option) any later version. + */ +#ifndef __IPIC_H__ +#define __IPIC_H__ + +#include + +#define MPC83xx_IPIC_SIZE (0x00100) + +/* System Global Interrupt Configuration Register */ +#define SICFR_IPSA 0x00010000 +#define SICFR_IPSD 0x00080000 +#define SICFR_MPSA 0x00200000 +#define SICFR_MPSB 0x00400000 + +/* System External Interrupt Mask Register */ +#define SEMSR_SIRQ0 0x00008000 + +/* System Error Control Register */ +#define SERCR_MCPR 0x00000001 + +struct ipic { + volatile u32 __iomem *regs; + unsigned int irq_offset; +}; + +struct ipic_info { + u8 pend; /* pending register offset from base */ + u8 mask; /* mask register offset from base */ + u8 prio; /* priority register offset from base */ + u8 force; /* force register offset from base */ + u8 bit; /* register bit position (as per doc) + bit mask = 1 << (31 - bit) */ + u8 prio_mask; /* priority mask value */ +}; + +#endif /* __IPIC_H__ */ diff --git a/include/asm-powerpc/ipic.h b/include/asm-powerpc/ipic.h index 0fe396a2b66..53079ec3a51 100644 --- a/include/asm-powerpc/ipic.h +++ b/include/asm-powerpc/ipic.h @@ -69,9 +69,6 @@ enum ipic_mcp_irq { IPIC_MCP_MU = 7, }; -extern void ipic_init(phys_addr_t phys_addr, unsigned int flags, - unsigned int irq_offset, - unsigned char *senses, unsigned int senses_count); extern int ipic_set_priority(unsigned int irq, unsigned int priority); extern void ipic_set_highest_priority(unsigned int irq); extern void ipic_set_default_priority(void); @@ -79,7 +76,16 @@ extern void ipic_enable_mcp(enum ipic_mcp_irq mcp_irq); extern void ipic_disable_mcp(enum ipic_mcp_irq mcp_irq); extern u32 ipic_get_mcp_status(void); extern void ipic_clear_mcp_status(u32 mask); + +#ifdef CONFIG_PPC_MERGE +extern void ipic_init(struct device_node *node, unsigned int flags); +extern unsigned int ipic_get_irq(struct pt_regs *regs); +#else +extern void ipic_init(phys_addr_t phys_addr, unsigned int flags, + unsigned int irq_offset, + unsigned char *senses, unsigned int senses_count); extern int ipic_get_irq(struct pt_regs *regs); +#endif #endif /* __ASM_IPIC_H__ */ #endif /* __KERNEL__ */ From b9f0f1bb2bcaae96dd3267f6bd3ad1ca44a1f5ad Mon Sep 17 00:00:00 2001 From: Kim Phillips Date: Fri, 25 Aug 2006 11:59:07 -0500 Subject: [PATCH 019/115] [POWERPC] Adapt ipic driver to new host_ops interface, add set_irq_type to set IRQ sense This converts ipic code to Benh's IRQ mods. For the IPIC, IRQ sense values in the device tree equal those in include/linux/irq.h; that's 8 for low assertion (most internal IRQs on mpc83xx), and 2 for high-to-low change. spinlocks added to [un]mask, ack operations; default handler and type now set in host_map; and redundant condition check eliminated. Signed-off-by: Kim Phillips Signed-off-by: Li Yang Signed-off-by: Paul Mackerras --- arch/powerpc/sysdev/ipic.c | 303 +++++++++++++++++++++++++------------ arch/powerpc/sysdev/ipic.h | 23 ++- 2 files changed, 228 insertions(+), 98 deletions(-) diff --git a/arch/powerpc/sysdev/ipic.c b/arch/powerpc/sysdev/ipic.c index 46801f5ec03..70e707785d4 100644 --- a/arch/powerpc/sysdev/ipic.c +++ b/arch/powerpc/sysdev/ipic.c @@ -19,15 +19,18 @@ #include #include #include +#include +#include +#include #include #include +#include #include -#include #include "ipic.h" -static struct ipic p_ipic; static struct ipic * primary_ipic; +static DEFINE_SPINLOCK(ipic_lock); static struct ipic_info ipic_info[] = { [9] = { @@ -373,74 +376,220 @@ static inline void ipic_write(volatile u32 __iomem *base, unsigned int reg, u32 out_be32(base + (reg >> 2), value); } -static inline struct ipic * ipic_from_irq(unsigned int irq) +static inline struct ipic * ipic_from_irq(unsigned int virq) { return primary_ipic; } -static void ipic_enable_irq(unsigned int irq) +#define ipic_irq_to_hw(virq) ((unsigned int)irq_map[virq].hwirq) + +static void ipic_unmask_irq(unsigned int virq) { - struct ipic *ipic = ipic_from_irq(irq); - unsigned int src = irq - ipic->irq_offset; + struct ipic *ipic = ipic_from_irq(virq); + unsigned int src = ipic_irq_to_hw(virq); + unsigned long flags; u32 temp; + spin_lock_irqsave(&ipic_lock, flags); + temp = ipic_read(ipic->regs, ipic_info[src].mask); temp |= (1 << (31 - ipic_info[src].bit)); ipic_write(ipic->regs, ipic_info[src].mask, temp); + + spin_unlock_irqrestore(&ipic_lock, flags); } -static void ipic_disable_irq(unsigned int irq) +static void ipic_mask_irq(unsigned int virq) { - struct ipic *ipic = ipic_from_irq(irq); - unsigned int src = irq - ipic->irq_offset; + struct ipic *ipic = ipic_from_irq(virq); + unsigned int src = ipic_irq_to_hw(virq); + unsigned long flags; u32 temp; + spin_lock_irqsave(&ipic_lock, flags); + temp = ipic_read(ipic->regs, ipic_info[src].mask); temp &= ~(1 << (31 - ipic_info[src].bit)); ipic_write(ipic->regs, ipic_info[src].mask, temp); + + spin_unlock_irqrestore(&ipic_lock, flags); } -static void ipic_disable_irq_and_ack(unsigned int irq) +static void ipic_ack_irq(unsigned int virq) { - struct ipic *ipic = ipic_from_irq(irq); - unsigned int src = irq - ipic->irq_offset; + struct ipic *ipic = ipic_from_irq(virq); + unsigned int src = ipic_irq_to_hw(virq); + unsigned long flags; u32 temp; - ipic_disable_irq(irq); + spin_lock_irqsave(&ipic_lock, flags); temp = ipic_read(ipic->regs, ipic_info[src].pend); temp |= (1 << (31 - ipic_info[src].bit)); ipic_write(ipic->regs, ipic_info[src].pend, temp); + + spin_unlock_irqrestore(&ipic_lock, flags); } -static void ipic_end_irq(unsigned int irq) +static void ipic_mask_irq_and_ack(unsigned int virq) { - if (!(irq_desc[irq].status & (IRQ_DISABLED|IRQ_INPROGRESS))) - ipic_enable_irq(irq); + struct ipic *ipic = ipic_from_irq(virq); + unsigned int src = ipic_irq_to_hw(virq); + unsigned long flags; + u32 temp; + + spin_lock_irqsave(&ipic_lock, flags); + + temp = ipic_read(ipic->regs, ipic_info[src].mask); + temp &= ~(1 << (31 - ipic_info[src].bit)); + ipic_write(ipic->regs, ipic_info[src].mask, temp); + + temp = ipic_read(ipic->regs, ipic_info[src].pend); + temp |= (1 << (31 - ipic_info[src].bit)); + ipic_write(ipic->regs, ipic_info[src].pend, temp); + + spin_unlock_irqrestore(&ipic_lock, flags); } -struct hw_interrupt_type ipic = { - .typename = " IPIC ", - .enable = ipic_enable_irq, - .disable = ipic_disable_irq, - .ack = ipic_disable_irq_and_ack, - .end = ipic_end_irq, +static int ipic_set_irq_type(unsigned int virq, unsigned int flow_type) +{ + struct ipic *ipic = ipic_from_irq(virq); + unsigned int src = ipic_irq_to_hw(virq); + struct irq_desc *desc = get_irq_desc(virq); + unsigned int vold, vnew, edibit; + + if (flow_type == IRQ_TYPE_NONE) + flow_type = IRQ_TYPE_LEVEL_LOW; + + /* ipic supports only low assertion and high-to-low change senses + */ + if (!(flow_type & (IRQ_TYPE_LEVEL_LOW | IRQ_TYPE_EDGE_FALLING))) { + printk(KERN_ERR "ipic: sense type 0x%x not supported\n", + flow_type); + return -EINVAL; + } + + desc->status &= ~(IRQ_TYPE_SENSE_MASK | IRQ_LEVEL); + desc->status |= flow_type & IRQ_TYPE_SENSE_MASK; + if (flow_type & IRQ_TYPE_LEVEL_LOW) { + desc->status |= IRQ_LEVEL; + set_irq_handler(virq, handle_level_irq); + } else { + set_irq_handler(virq, handle_edge_irq); + } + + /* only EXT IRQ senses are programmable on ipic + * internal IRQ senses are LEVEL_LOW + */ + if (src == IPIC_IRQ_EXT0) + edibit = 15; + else + if (src >= IPIC_IRQ_EXT1 && src <= IPIC_IRQ_EXT7) + edibit = (14 - (src - IPIC_IRQ_EXT1)); + else + return (flow_type & IRQ_TYPE_LEVEL_LOW) ? 0 : -EINVAL; + + vold = ipic_read(ipic->regs, IPIC_SECNR); + if ((flow_type & IRQ_TYPE_SENSE_MASK) == IRQ_TYPE_EDGE_FALLING) { + vnew = vold | (1 << edibit); + } else { + vnew = vold & ~(1 << edibit); + } + if (vold != vnew) + ipic_write(ipic->regs, IPIC_SECNR, vnew); + return 0; +} + +static struct irq_chip ipic_irq_chip = { + .typename = " IPIC ", + .unmask = ipic_unmask_irq, + .mask = ipic_mask_irq, + .mask_ack = ipic_mask_irq_and_ack, + .ack = ipic_ack_irq, + .set_type = ipic_set_irq_type, }; -void __init ipic_init(phys_addr_t phys_addr, - unsigned int flags, - unsigned int irq_offset, - unsigned char *senses, - unsigned int senses_count) +static int ipic_host_match(struct irq_host *h, struct device_node *node) { - u32 i, temp = 0; + struct ipic *ipic = h->host_data; - primary_ipic = &p_ipic; - primary_ipic->regs = ioremap(phys_addr, MPC83xx_IPIC_SIZE); + /* Exact match, unless ipic node is NULL */ + return ipic->of_node == NULL || ipic->of_node == node; +} - primary_ipic->irq_offset = irq_offset; +static int ipic_host_map(struct irq_host *h, unsigned int virq, + irq_hw_number_t hw) +{ + struct ipic *ipic = h->host_data; + struct irq_chip *chip; - ipic_write(primary_ipic->regs, IPIC_SICNR, 0x0); + /* Default chip */ + chip = &ipic->hc_irq; + + set_irq_chip_data(virq, ipic); + set_irq_chip_and_handler(virq, chip, handle_level_irq); + + /* Set default irq type */ + set_irq_type(virq, IRQ_TYPE_NONE); + + return 0; +} + +static int ipic_host_xlate(struct irq_host *h, struct device_node *ct, + u32 *intspec, unsigned int intsize, + irq_hw_number_t *out_hwirq, unsigned int *out_flags) + +{ + /* interrupt sense values coming from the device tree equal either + * LEVEL_LOW (low assertion) or EDGE_FALLING (high-to-low change) + */ + *out_hwirq = intspec[0]; + if (intsize > 1) + *out_flags = intspec[1]; + else + *out_flags = IRQ_TYPE_NONE; + return 0; +} + +static struct irq_host_ops ipic_host_ops = { + .match = ipic_host_match, + .map = ipic_host_map, + .xlate = ipic_host_xlate, +}; + +void __init ipic_init(struct device_node *node, + unsigned int flags) +{ + struct ipic *ipic; + struct resource res; + u32 temp = 0, ret; + + ipic = alloc_bootmem(sizeof(struct ipic)); + if (ipic == NULL) + return; + + memset(ipic, 0, sizeof(struct ipic)); + ipic->of_node = node ? of_node_get(node) : NULL; + + ipic->irqhost = irq_alloc_host(IRQ_HOST_MAP_LINEAR, + NR_IPIC_INTS, + &ipic_host_ops, 0); + if (ipic->irqhost == NULL) { + of_node_put(node); + return; + } + + ret = of_address_to_resource(node, 0, &res); + if (ret) + return; + + ipic->regs = ioremap(res.start, res.end - res.start + 1); + + ipic->irqhost->host_data = ipic; + ipic->hc_irq = ipic_irq_chip; + + /* init hw */ + ipic_write(ipic->regs, IPIC_SICNR, 0x0); /* default priority scheme is grouped. If spread mode is required * configure SICFR accordingly */ @@ -453,49 +602,35 @@ void __init ipic_init(phys_addr_t phys_addr, if (flags & IPIC_SPREADMODE_MIX_B) temp |= SICFR_MPSB; - ipic_write(primary_ipic->regs, IPIC_SICNR, temp); + ipic_write(ipic->regs, IPIC_SICNR, temp); /* handle MCP route */ temp = 0; if (flags & IPIC_DISABLE_MCP_OUT) temp = SERCR_MCPR; - ipic_write(primary_ipic->regs, IPIC_SERCR, temp); + ipic_write(ipic->regs, IPIC_SERCR, temp); /* handle routing of IRQ0 to MCP */ - temp = ipic_read(primary_ipic->regs, IPIC_SEMSR); + temp = ipic_read(ipic->regs, IPIC_SEMSR); if (flags & IPIC_IRQ0_MCP) temp |= SEMSR_SIRQ0; else temp &= ~SEMSR_SIRQ0; - ipic_write(primary_ipic->regs, IPIC_SEMSR, temp); + ipic_write(ipic->regs, IPIC_SEMSR, temp); - for (i = 0 ; i < NR_IPIC_INTS ; i++) { - irq_desc[i+irq_offset].chip = &ipic; - irq_desc[i+irq_offset].status = IRQ_LEVEL; - } + primary_ipic = ipic; + irq_set_default_host(primary_ipic->irqhost); - temp = 0; - for (i = 0 ; i < senses_count ; i++) { - if ((senses[i] & IRQ_SENSE_MASK) == IRQ_SENSE_EDGE) { - temp |= 1 << (15 - i); - if (i != 0) - irq_desc[i + irq_offset + MPC83xx_IRQ_EXT1 - 1].status = 0; - else - irq_desc[irq_offset + MPC83xx_IRQ_EXT0].status = 0; - } - } - ipic_write(primary_ipic->regs, IPIC_SECNR, temp); - - printk ("IPIC (%d IRQ sources, %d External IRQs) at %p\n", NR_IPIC_INTS, - senses_count, primary_ipic->regs); + printk ("IPIC (%d IRQ sources) at %p\n", NR_IPIC_INTS, + primary_ipic->regs); } -int ipic_set_priority(unsigned int irq, unsigned int priority) +int ipic_set_priority(unsigned int virq, unsigned int priority) { - struct ipic *ipic = ipic_from_irq(irq); - unsigned int src = irq - ipic->irq_offset; + struct ipic *ipic = ipic_from_irq(virq); + unsigned int src = ipic_irq_to_hw(virq); u32 temp; if (priority > 7) @@ -520,10 +655,10 @@ int ipic_set_priority(unsigned int irq, unsigned int priority) return 0; } -void ipic_set_highest_priority(unsigned int irq) +void ipic_set_highest_priority(unsigned int virq) { - struct ipic *ipic = ipic_from_irq(irq); - unsigned int src = irq - ipic->irq_offset; + struct ipic *ipic = ipic_from_irq(virq); + unsigned int src = ipic_irq_to_hw(virq); u32 temp; temp = ipic_read(ipic->regs, IPIC_SICFR); @@ -537,37 +672,10 @@ void ipic_set_highest_priority(unsigned int irq) void ipic_set_default_priority(void) { - ipic_set_priority(MPC83xx_IRQ_TSEC1_TX, 0); - ipic_set_priority(MPC83xx_IRQ_TSEC1_RX, 1); - ipic_set_priority(MPC83xx_IRQ_TSEC1_ERROR, 2); - ipic_set_priority(MPC83xx_IRQ_TSEC2_TX, 3); - ipic_set_priority(MPC83xx_IRQ_TSEC2_RX, 4); - ipic_set_priority(MPC83xx_IRQ_TSEC2_ERROR, 5); - ipic_set_priority(MPC83xx_IRQ_USB2_DR, 6); - ipic_set_priority(MPC83xx_IRQ_USB2_MPH, 7); - - ipic_set_priority(MPC83xx_IRQ_UART1, 0); - ipic_set_priority(MPC83xx_IRQ_UART2, 1); - ipic_set_priority(MPC83xx_IRQ_SEC2, 2); - ipic_set_priority(MPC83xx_IRQ_IIC1, 5); - ipic_set_priority(MPC83xx_IRQ_IIC2, 6); - ipic_set_priority(MPC83xx_IRQ_SPI, 7); - ipic_set_priority(MPC83xx_IRQ_RTC_SEC, 0); - ipic_set_priority(MPC83xx_IRQ_PIT, 1); - ipic_set_priority(MPC83xx_IRQ_PCI1, 2); - ipic_set_priority(MPC83xx_IRQ_PCI2, 3); - ipic_set_priority(MPC83xx_IRQ_EXT0, 4); - ipic_set_priority(MPC83xx_IRQ_EXT1, 5); - ipic_set_priority(MPC83xx_IRQ_EXT2, 6); - ipic_set_priority(MPC83xx_IRQ_EXT3, 7); - ipic_set_priority(MPC83xx_IRQ_RTC_ALR, 0); - ipic_set_priority(MPC83xx_IRQ_MU, 1); - ipic_set_priority(MPC83xx_IRQ_SBA, 2); - ipic_set_priority(MPC83xx_IRQ_DMA, 3); - ipic_set_priority(MPC83xx_IRQ_EXT4, 4); - ipic_set_priority(MPC83xx_IRQ_EXT5, 5); - ipic_set_priority(MPC83xx_IRQ_EXT6, 6); - ipic_set_priority(MPC83xx_IRQ_EXT7, 7); + ipic_write(primary_ipic->regs, IPIC_SIPRR_A, IPIC_SIPRR_A_DEFAULT); + ipic_write(primary_ipic->regs, IPIC_SIPRR_D, IPIC_SIPRR_D_DEFAULT); + ipic_write(primary_ipic->regs, IPIC_SMPRR_A, IPIC_SMPRR_A_DEFAULT); + ipic_write(primary_ipic->regs, IPIC_SMPRR_B, IPIC_SMPRR_B_DEFAULT); } void ipic_enable_mcp(enum ipic_mcp_irq mcp_irq) @@ -600,17 +708,20 @@ void ipic_clear_mcp_status(u32 mask) ipic_write(primary_ipic->regs, IPIC_SERMR, mask); } -/* Return an interrupt vector or -1 if no interrupt is pending. */ -int ipic_get_irq(struct pt_regs *regs) +/* Return an interrupt vector or NO_IRQ if no interrupt is pending. */ +unsigned int ipic_get_irq(struct pt_regs *regs) { int irq; - irq = ipic_read(primary_ipic->regs, IPIC_SIVCR) & 0x7f; + BUG_ON(primary_ipic == NULL); + +#define IPIC_SIVCR_VECTOR_MASK 0x7f + irq = ipic_read(primary_ipic->regs, IPIC_SIVCR) & IPIC_SIVCR_VECTOR_MASK; if (irq == 0) /* 0 --> no irq is pending */ - irq = -1; + return NO_IRQ; - return irq; + return irq_linear_revmap(primary_ipic->irqhost, irq); } static struct sysdev_class ipic_sysclass = { diff --git a/arch/powerpc/sysdev/ipic.h b/arch/powerpc/sysdev/ipic.h index a60c9d18bb7..c28e589877e 100644 --- a/arch/powerpc/sysdev/ipic.h +++ b/arch/powerpc/sysdev/ipic.h @@ -15,7 +15,18 @@ #include -#define MPC83xx_IPIC_SIZE (0x00100) +#define NR_IPIC_INTS 128 + +/* External IRQS */ +#define IPIC_IRQ_EXT0 48 +#define IPIC_IRQ_EXT1 17 +#define IPIC_IRQ_EXT7 23 + +/* Default Priority Registers */ +#define IPIC_SIPRR_A_DEFAULT 0x05309770 +#define IPIC_SIPRR_D_DEFAULT 0x05309770 +#define IPIC_SMPRR_A_DEFAULT 0x05309770 +#define IPIC_SMPRR_B_DEFAULT 0x05309770 /* System Global Interrupt Configuration Register */ #define SICFR_IPSA 0x00010000 @@ -31,7 +42,15 @@ struct ipic { volatile u32 __iomem *regs; - unsigned int irq_offset; + + /* The remapper for this IPIC */ + struct irq_host *irqhost; + + /* The "linux" controller struct */ + struct irq_chip hc_irq; + + /* The device node of the interrupt controller */ + struct device_node *of_node; }; struct ipic_info { From f1f17716d13bfb709809a3f5c84bda105b646b9e Mon Sep 17 00:00:00 2001 From: Kim Phillips Date: Fri, 25 Aug 2006 11:59:22 -0500 Subject: [PATCH 020/115] [POWERPC] modify mpc83xx platforms to use new IRQ layer This fixes MPC834x MDS (formerly SYS) and ITX platform code to get IRQ data (including PCI) from the device tree, and to use the new IPIC code. renamed defconfig (sys -> mds), left one redundant NULL assignment in mpc83xx_pcibios_fixup to keep the compiler happy. Signed-off-by: Kim Phillips Signed-off-by: Li Yang Signed-off-by: Paul Mackerras --- ...4x_sys_defconfig => mpc834x_mds_defconfig} | 0 arch/powerpc/platforms/83xx/mpc834x_itx.c | 49 +++------------- arch/powerpc/platforms/83xx/mpc834x_sys.c | 56 +++---------------- arch/powerpc/platforms/83xx/mpc83xx.h | 1 + arch/powerpc/platforms/83xx/pci.c | 9 +++ 5 files changed, 28 insertions(+), 87 deletions(-) rename arch/powerpc/configs/{mpc834x_sys_defconfig => mpc834x_mds_defconfig} (100%) diff --git a/arch/powerpc/configs/mpc834x_sys_defconfig b/arch/powerpc/configs/mpc834x_mds_defconfig similarity index 100% rename from arch/powerpc/configs/mpc834x_sys_defconfig rename to arch/powerpc/configs/mpc834x_mds_defconfig diff --git a/arch/powerpc/platforms/83xx/mpc834x_itx.c b/arch/powerpc/platforms/83xx/mpc834x_itx.c index b46305645d3..cf3967a66fb 100644 --- a/arch/powerpc/platforms/83xx/mpc834x_itx.c +++ b/arch/powerpc/platforms/83xx/mpc834x_itx.c @@ -46,26 +46,6 @@ unsigned long isa_io_base = 0; unsigned long isa_mem_base = 0; #endif -#ifdef CONFIG_PCI -static int -mpc83xx_map_irq(struct pci_dev *dev, unsigned char idsel, unsigned char pin) -{ - static char pci_irq_table[][4] = - /* - * PCI IDSEL/INTPIN->INTLINE - * A B C D - */ - { - {PIRQB, PIRQC, PIRQD, PIRQA}, /* idsel 0x0e */ - {PIRQA, PIRQB, PIRQC, PIRQD}, /* idsel 0x0f */ - {PIRQC, PIRQD, PIRQA, PIRQB}, /* idsel 0x10 */ - }; - - const long min_idsel = 0x0e, max_idsel = 0x10, irqs_per_slot = 4; - return PCI_IRQ_TABLE_LOOKUP; -} -#endif /* CONFIG_PCI */ - /* ************************************************************************ * * Setup the architecture @@ -92,8 +72,6 @@ static void __init mpc834x_itx_setup_arch(void) for (np = NULL; (np = of_find_node_by_type(np, "pci")) != NULL;) add_bridge(np); - ppc_md.pci_swizzle = common_swizzle; - ppc_md.pci_map_irq = mpc83xx_map_irq; ppc_md.pci_exclude_device = mpc83xx_exclude_device; #endif @@ -106,25 +84,13 @@ static void __init mpc834x_itx_setup_arch(void) void __init mpc834x_itx_init_IRQ(void) { - u8 senses[8] = { - 0, /* EXT 0 */ - IRQ_SENSE_LEVEL, /* EXT 1 */ - IRQ_SENSE_LEVEL, /* EXT 2 */ - 0, /* EXT 3 */ -#ifdef CONFIG_PCI - IRQ_SENSE_LEVEL, /* EXT 4 */ - IRQ_SENSE_LEVEL, /* EXT 5 */ - IRQ_SENSE_LEVEL, /* EXT 6 */ - IRQ_SENSE_LEVEL, /* EXT 7 */ -#else - 0, /* EXT 4 */ - 0, /* EXT 5 */ - 0, /* EXT 6 */ - 0, /* EXT 7 */ -#endif - }; + struct device_node *np; - ipic_init(get_immrbase() + 0x00700, 0, 0, senses, 8); + np = of_find_node_by_type(NULL, "ipic"); + if (!np) + return; + + ipic_init(np, 0); /* Initialize the default interrupt mapping priorities, * in case the boot rom changed something on us. @@ -153,4 +119,7 @@ define_machine(mpc834x_itx) { .time_init = mpc83xx_time_init, .calibrate_decr = generic_calibrate_decr, .progress = udbg_progress, +#ifdef CONFIG_PCI + .pcibios_fixup = mpc83xx_pcibios_fixup, +#endif }; diff --git a/arch/powerpc/platforms/83xx/mpc834x_sys.c b/arch/powerpc/platforms/83xx/mpc834x_sys.c index 3e1c16eb4a6..32df239d1c4 100644 --- a/arch/powerpc/platforms/83xx/mpc834x_sys.c +++ b/arch/powerpc/platforms/83xx/mpc834x_sys.c @@ -43,33 +43,6 @@ unsigned long isa_io_base = 0; unsigned long isa_mem_base = 0; #endif -#ifdef CONFIG_PCI -static int -mpc83xx_map_irq(struct pci_dev *dev, unsigned char idsel, unsigned char pin) -{ - static char pci_irq_table[][4] = - /* - * PCI IDSEL/INTPIN->INTLINE - * A B C D - */ - { - {PIRQA, PIRQB, PIRQC, PIRQD}, /* idsel 0x11 */ - {PIRQC, PIRQD, PIRQA, PIRQB}, /* idsel 0x12 */ - {PIRQD, PIRQA, PIRQB, PIRQC}, /* idsel 0x13 */ - {0, 0, 0, 0}, - {PIRQA, PIRQB, PIRQC, PIRQD}, /* idsel 0x15 */ - {PIRQD, PIRQA, PIRQB, PIRQC}, /* idsel 0x16 */ - {PIRQC, PIRQD, PIRQA, PIRQB}, /* idsel 0x17 */ - {PIRQB, PIRQC, PIRQD, PIRQA}, /* idsel 0x18 */ - {0, 0, 0, 0}, /* idsel 0x19 */ - {0, 0, 0, 0}, /* idsel 0x20 */ - }; - - const long min_idsel = 0x11, max_idsel = 0x20, irqs_per_slot = 4; - return PCI_IRQ_TABLE_LOOKUP; -} -#endif /* CONFIG_PCI */ - /* ************************************************************************ * * Setup the architecture @@ -96,8 +69,6 @@ static void __init mpc834x_sys_setup_arch(void) for (np = NULL; (np = of_find_node_by_type(np, "pci")) != NULL;) add_bridge(np); - ppc_md.pci_swizzle = common_swizzle; - ppc_md.pci_map_irq = mpc83xx_map_irq; ppc_md.pci_exclude_device = mpc83xx_exclude_device; #endif @@ -110,25 +81,13 @@ static void __init mpc834x_sys_setup_arch(void) void __init mpc834x_sys_init_IRQ(void) { - u8 senses[8] = { - 0, /* EXT 0 */ - IRQ_SENSE_LEVEL, /* EXT 1 */ - IRQ_SENSE_LEVEL, /* EXT 2 */ - 0, /* EXT 3 */ -#ifdef CONFIG_PCI - IRQ_SENSE_LEVEL, /* EXT 4 */ - IRQ_SENSE_LEVEL, /* EXT 5 */ - IRQ_SENSE_LEVEL, /* EXT 6 */ - IRQ_SENSE_LEVEL, /* EXT 7 */ -#else - 0, /* EXT 4 */ - 0, /* EXT 5 */ - 0, /* EXT 6 */ - 0, /* EXT 7 */ -#endif - }; + struct device_node *np; - ipic_init(get_immrbase() + 0x00700, 0, 0, senses, 8); + np = of_find_node_by_type(NULL, "ipic"); + if (!np) + return; + + ipic_init(np, 0); /* Initialize the default interrupt mapping priorities, * in case the boot rom changed something on us. @@ -178,4 +137,7 @@ define_machine(mpc834x_sys) { .time_init = mpc83xx_time_init, .calibrate_decr = generic_calibrate_decr, .progress = udbg_progress, +#ifdef CONFIG_PCI + .pcibios_fixup = mpc83xx_pcibios_fixup, +#endif }; diff --git a/arch/powerpc/platforms/83xx/mpc83xx.h b/arch/powerpc/platforms/83xx/mpc83xx.h index 01cae106912..2c82bca9bfb 100644 --- a/arch/powerpc/platforms/83xx/mpc83xx.h +++ b/arch/powerpc/platforms/83xx/mpc83xx.h @@ -11,6 +11,7 @@ extern int add_bridge(struct device_node *dev); extern int mpc83xx_exclude_device(u_char bus, u_char devfn); +extern void mpc83xx_pcibios_fixup(void); extern void mpc83xx_restart(char *cmd); extern long mpc83xx_time_init(void); diff --git a/arch/powerpc/platforms/83xx/pci.c b/arch/powerpc/platforms/83xx/pci.c index 3b5e563c279..5d84a9ccd10 100644 --- a/arch/powerpc/platforms/83xx/pci.c +++ b/arch/powerpc/platforms/83xx/pci.c @@ -45,6 +45,15 @@ int mpc83xx_exclude_device(u_char bus, u_char devfn) return PCIBIOS_SUCCESSFUL; } +void __init mpc83xx_pcibios_fixup(void) +{ + struct pci_dev *dev = NULL; + + /* map all the PCI irqs */ + for_each_pci_dev(dev) + pci_read_irq_line(dev); +} + int __init add_bridge(struct device_node *dev) { int len; From 1b9a93eb4638dcde0c3af42fd6c05e3911baa7db Mon Sep 17 00:00:00 2001 From: Kim Phillips Date: Tue, 29 Aug 2006 18:13:31 -0500 Subject: [PATCH 021/115] [POWERPC] Add MPC8349E MDS device tree source file to arch/powerpc/boot/dts Add MPC8349E MDS device tree source file to arch/powerpc/boot/dts Signed-off-by: Kim Phillips Signed-off-by: Paul Mackerras --- arch/powerpc/boot/dts/mpc8349emds.dts | 328 ++++++++++++++++++++++++++ 1 file changed, 328 insertions(+) create mode 100644 arch/powerpc/boot/dts/mpc8349emds.dts diff --git a/arch/powerpc/boot/dts/mpc8349emds.dts b/arch/powerpc/boot/dts/mpc8349emds.dts new file mode 100644 index 00000000000..12f5dbf3055 --- /dev/null +++ b/arch/powerpc/boot/dts/mpc8349emds.dts @@ -0,0 +1,328 @@ +/* + * MPC8349E MDS Device Tree Source + * + * Copyright 2005, 2006 Freescale Semiconductor Inc. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation; either version 2 of the License, or (at your + * option) any later version. + */ + +/ { + model = "MPC8349EMDS"; + compatible = "MPC834xMDS"; + #address-cells = <1>; + #size-cells = <1>; + + cpus { + #cpus = <1>; + #address-cells = <1>; + #size-cells = <0>; + + PowerPC,8349@0 { + device_type = "cpu"; + reg = <0>; + d-cache-line-size = <20>; // 32 bytes + i-cache-line-size = <20>; // 32 bytes + d-cache-size = <8000>; // L1, 32K + i-cache-size = <8000>; // L1, 32K + timebase-frequency = <0>; // from bootloader + bus-frequency = <0>; // from bootloader + clock-frequency = <0>; // from bootloader + 32-bit; + }; + }; + + memory { + device_type = "memory"; + reg = <00000000 10000000>; // 256MB at 0 + }; + + soc8349@e0000000 { + #address-cells = <1>; + #size-cells = <1>; + #interrupt-cells = <2>; + device_type = "soc"; + ranges = <0 e0000000 00100000>; + reg = ; + bus-frequency = <0>; + + wdt@200 { + device_type = "watchdog"; + compatible = "mpc83xx_wdt"; + reg = <200 100>; + }; + + i2c@3000 { + device_type = "i2c"; + compatible = "fsl-i2c"; + reg = <3000 100>; + interrupts = ; + interrupt-parent = <700>; + dfsrr; + }; + + i2c@3100 { + device_type = "i2c"; + compatible = "fsl-i2c"; + reg = <3100 100>; + interrupts = ; + interrupt-parent = <700>; + dfsrr; + }; + + spi@7000 { + device_type = "spi"; + compatible = "mpc83xx_spi"; + reg = <7000 1000>; + interrupts = <10 8>; + interrupt-parent = <700>; + mode = <0>; + }; + + /* phy type (ULPI or SERIAL) are only types supportted for MPH */ + /* port = 0 or 1 */ + usb@22000 { + device_type = "usb"; + compatible = "fsl-usb2-mph"; + reg = <22000 1000>; + #address-cells = <1>; + #size-cells = <0>; + interrupt-parent = <700>; + interrupts = <27 2>; + phy_type = "ulpi"; + port1; + }; + /* phy type (ULPI, UTMI, UTMI_WIDE, SERIAL) */ + usb@23000 { + device_type = "usb"; + compatible = "fsl-usb2-dr"; + reg = <23000 1000>; + #address-cells = <1>; + #size-cells = <0>; + interrupt-parent = <700>; + interrupts = <26 2>; + phy_type = "ulpi"; + }; + + mdio@24520 { + device_type = "mdio"; + compatible = "gianfar"; + reg = <24520 20>; + #address-cells = <1>; + #size-cells = <0>; + linux,phandle = <24520>; + ethernet-phy@0 { + linux,phandle = <2452000>; + interrupt-parent = <700>; + interrupts = <11 2>; + reg = <0>; + device_type = "ethernet-phy"; + }; + ethernet-phy@1 { + linux,phandle = <2452001>; + interrupt-parent = <700>; + interrupts = <12 2>; + reg = <1>; + device_type = "ethernet-phy"; + }; + }; + + ethernet@24000 { + device_type = "network"; + model = "TSEC"; + compatible = "gianfar"; + reg = <24000 1000>; + address = [ 00 00 00 00 00 00 ]; + local-mac-address = [ 00 00 00 00 00 00 ]; + interrupts = <20 8 21 8 22 8>; + interrupt-parent = <700>; + phy-handle = <2452000>; + }; + + ethernet@25000 { + #address-cells = <1>; + #size-cells = <0>; + device_type = "network"; + model = "TSEC"; + compatible = "gianfar"; + reg = <25000 1000>; + address = [ 00 00 00 00 00 00 ]; + local-mac-address = [ 00 00 00 00 00 00 ]; + interrupts = <23 8 24 8 25 8>; + interrupt-parent = <700>; + phy-handle = <2452001>; + }; + + serial@4500 { + device_type = "serial"; + compatible = "ns16550"; + reg = <4500 100>; + clock-frequency = <0>; + interrupts = <9 8>; + interrupt-parent = <700>; + }; + + serial@4600 { + device_type = "serial"; + compatible = "ns16550"; + reg = <4600 100>; + clock-frequency = <0>; + interrupts = ; + interrupt-parent = <700>; + }; + + pci@8500 { + interrupt-map-mask = ; + interrupt-map = < + + /* IDSEL 0x11 */ + 8800 0 0 1 700 14 8 + 8800 0 0 2 700 15 8 + 8800 0 0 3 700 16 8 + 8800 0 0 4 700 17 8 + + /* IDSEL 0x12 */ + 9000 0 0 1 700 16 8 + 9000 0 0 2 700 17 8 + 9000 0 0 3 700 14 8 + 9000 0 0 4 700 15 8 + + /* IDSEL 0x13 */ + 9800 0 0 1 700 17 8 + 9800 0 0 2 700 14 8 + 9800 0 0 3 700 15 8 + 9800 0 0 4 700 16 8 + + /* IDSEL 0x15 */ + a800 0 0 1 700 14 8 + a800 0 0 2 700 15 8 + a800 0 0 3 700 16 8 + a800 0 0 4 700 17 8 + + /* IDSEL 0x16 */ + b000 0 0 1 700 17 8 + b000 0 0 2 700 14 8 + b000 0 0 3 700 15 8 + b000 0 0 4 700 16 8 + + /* IDSEL 0x17 */ + b800 0 0 1 700 16 8 + b800 0 0 2 700 17 8 + b800 0 0 3 700 14 8 + b800 0 0 4 700 15 8 + + /* IDSEL 0x18 */ + b000 0 0 1 700 15 8 + b000 0 0 2 700 16 8 + b000 0 0 3 700 17 8 + b000 0 0 4 700 14 8>; + interrupt-parent = <700>; + interrupts = <42 8>; + bus-range = <0 0>; + ranges = <02000000 0 a0000000 a0000000 0 10000000 + 42000000 0 80000000 80000000 0 10000000 + 01000000 0 00000000 e2000000 0 00100000>; + clock-frequency = <3f940aa>; + #interrupt-cells = <1>; + #size-cells = <2>; + #address-cells = <3>; + reg = <8500 100>; + compatible = "83xx"; + device_type = "pci"; + }; + + pci@8600 { + interrupt-map-mask = ; + interrupt-map = < + + /* IDSEL 0x11 */ + 8800 0 0 1 700 14 8 + 8800 0 0 2 700 15 8 + 8800 0 0 3 700 16 8 + 8800 0 0 4 700 17 8 + + /* IDSEL 0x12 */ + 9000 0 0 1 700 16 8 + 9000 0 0 2 700 17 8 + 9000 0 0 3 700 14 8 + 9000 0 0 4 700 15 8 + + /* IDSEL 0x13 */ + 9800 0 0 1 700 17 8 + 9800 0 0 2 700 14 8 + 9800 0 0 3 700 15 8 + 9800 0 0 4 700 16 8 + + /* IDSEL 0x15 */ + a800 0 0 1 700 14 8 + a800 0 0 2 700 15 8 + a800 0 0 3 700 16 8 + a800 0 0 4 700 17 8 + + /* IDSEL 0x16 */ + b000 0 0 1 700 17 8 + b000 0 0 2 700 14 8 + b000 0 0 3 700 15 8 + b000 0 0 4 700 16 8 + + /* IDSEL 0x17 */ + b800 0 0 1 700 16 8 + b800 0 0 2 700 17 8 + b800 0 0 3 700 14 8 + b800 0 0 4 700 15 8 + + /* IDSEL 0x18 */ + b000 0 0 1 700 15 8 + b000 0 0 2 700 16 8 + b000 0 0 3 700 17 8 + b000 0 0 4 700 14 8>; + interrupt-parent = <700>; + interrupts = <42 8>; + bus-range = <0 0>; + ranges = <02000000 0 b0000000 b0000000 0 10000000 + 42000000 0 90000000 90000000 0 10000000 + 01000000 0 00000000 e2100000 0 00100000>; + clock-frequency = <3f940aa>; + #interrupt-cells = <1>; + #size-cells = <2>; + #address-cells = <3>; + reg = <8600 100>; + compatible = "83xx"; + device_type = "pci"; + }; + + /* May need to remove if on a part without crypto engine */ + crypto@30000 { + device_type = "crypto"; + model = "SEC2"; + compatible = "talitos"; + reg = <30000 10000>; + interrupts = ; + interrupt-parent = <700>; + num-channels = <4>; + channel-fifo-len = <18>; + exec-units-mask = <0000007e>; + /* desc mask is for rev2.0, + * we need runtime fixup for >2.0 */ + descriptor-types-mask = <01010ebf>; + }; + + /* IPIC + * interrupts cell = + * sense values match linux IORESOURCE_IRQ_* defines: + * sense == 8: Level, low assertion + * sense == 2: Edge, high-to-low change + */ + pic@700 { + linux,phandle = <700>; + interrupt-controller; + #address-cells = <0>; + #interrupt-cells = <2>; + reg = <700 100>; + built-in; + device_type = "ipic"; + }; + }; +}; From 4b3afca9345f5beb9c607faeb2aef4f91dd91a13 Mon Sep 17 00:00:00 2001 From: Zang Roy-r61911 Date: Fri, 25 Aug 2006 16:43:25 +0800 Subject: [PATCH 022/115] [POWERPC] Add mpc7448hpc2 device tree source file This patch adds the mpc7448hpc2 device tree source file. Signed-off-by: Roy Zang Signed-off-by: Paul Mackerras --- arch/powerpc/boot/dts/mpc7448hpc2.dts | 190 ++++++++++++++++++++++++++ 1 file changed, 190 insertions(+) create mode 100644 arch/powerpc/boot/dts/mpc7448hpc2.dts diff --git a/arch/powerpc/boot/dts/mpc7448hpc2.dts b/arch/powerpc/boot/dts/mpc7448hpc2.dts new file mode 100644 index 00000000000..d7b985e6bd2 --- /dev/null +++ b/arch/powerpc/boot/dts/mpc7448hpc2.dts @@ -0,0 +1,190 @@ +/* + * MPC7448HPC2 (Taiga) board Device Tree Source + * + * Copyright 2006 Freescale Semiconductor Inc. + * 2006 Roy Zang . + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation; either version 2 of the License, or (at your + * option) any later version. + */ + + +/ { + model = "mpc7448hpc2"; + compatible = "mpc74xx"; + #address-cells = <1>; + #size-cells = <1>; + linux,phandle = <100>; + + cpus { + #cpus = <1>; + #address-cells = <1>; + #size-cells =<0>; + linux,phandle = <200>; + + PowerPC,7448@0 { + device_type = "cpu"; + reg = <0>; + d-cache-line-size = <20>; // 32 bytes + i-cache-line-size = <20>; // 32 bytes + d-cache-size = <8000>; // L1, 32K bytes + i-cache-size = <8000>; // L1, 32K bytes + timebase-frequency = <0>; // 33 MHz, from uboot + clock-frequency = <0>; // From U-Boot + bus-frequency = <0>; // From U-Boot + 32-bit; + linux,phandle = <201>; + linux,boot-cpu; + }; + }; + + memory { + device_type = "memory"; + linux,phandle = <300>; + reg = <00000000 20000000 // DDR2 512M at 0 + >; + }; + + tsi108@c0000000 { + #address-cells = <1>; + #size-cells = <1>; + #interrupt-cells = <2>; + device_type = "tsi-bridge"; + ranges = <00000000 c0000000 00010000>; + reg = ; + bus-frequency = <0>; + + i2c@7000 { + interrupt-parent = <7400>; + interrupts = ; + reg = <7000 400>; + device_type = "i2c"; + compatible = "tsi-i2c"; + }; + + mdio@6000 { + device_type = "mdio"; + compatible = "tsi-ethernet"; + + ethernet-phy@6000 { + linux,phandle = <6000>; + interrupt-parent = <7400>; + interrupts = <2 1>; + reg = <6000 50>; + phy-id = <8>; + device_type = "ethernet-phy"; + }; + + ethernet-phy@6400 { + linux,phandle = <6400>; + interrupt-parent = <7400>; + interrupts = <2 1>; + reg = <6000 50>; + phy-id = <9>; + device_type = "ethernet-phy"; + }; + + }; + + ethernet@6200 { + #size-cells = <0>; + device_type = "network"; + model = "TSI-ETH"; + compatible = "tsi-ethernet"; + reg = <6000 200>; + address = [ 00 06 D2 00 00 01 ]; + interrupts = <10 2>; + interrupt-parent = <7400>; + phy-handle = <6000>; + }; + + ethernet@6600 { + #address-cells = <1>; + #size-cells = <0>; + device_type = "network"; + model = "TSI-ETH"; + compatible = "tsi-ethernet"; + reg = <6400 200>; + address = [ 00 06 D2 00 00 02 ]; + interrupts = <11 2>; + interrupt-parent = <7400>; + phy-handle = <6400>; + }; + + serial@7808 { + device_type = "serial"; + compatible = "ns16550"; + reg = <7808 200>; + clock-frequency = <3f6b5a00>; + interrupts = ; + interrupt-parent = <7400>; + }; + + serial@7c08 { + device_type = "serial"; + compatible = "ns16550"; + reg = <7c08 200>; + clock-frequency = <3f6b5a00>; + interrupts = ; + interrupt-parent = <7400>; + }; + + pic@7400 { + linux,phandle = <7400>; + clock-frequency = <0>; + interrupt-controller; + #address-cells = <0>; + #interrupt-cells = <2>; + reg = <7400 400>; + built-in; + compatible = "chrp,open-pic"; + device_type = "open-pic"; + big-endian; + }; + pci@1000 { + compatible = "tsi10x"; + device_type = "pci"; + linux,phandle = <1000>; + #interrupt-cells = <1>; + #size-cells = <2>; + #address-cells = <3>; + reg = <1000 1000>; + bus-range = <0 0>; + ranges = <02000000 0 e0000000 e0000000 0 1A000000 + 01000000 0 00000000 fa000000 0 00010000>; + clock-frequency = <7f28154>; + interrupt-parent = <7400>; + interrupts = <17 2>; + interrupt-map-mask = ; + interrupt-map = < + + /* IDSEL 0x11 */ + 0800 0 0 1 7400 24 0 + 0800 0 0 2 7400 25 0 + 0800 0 0 3 7400 26 0 + 0800 0 0 4 7400 27 0 + + /* IDSEL 0x12 */ + 1000 0 0 1 7400 25 0 + 1000 0 0 2 7400 26 0 + 1000 0 0 3 7400 27 0 + 1000 0 0 4 7400 24 0 + + /* IDSEL 0x13 */ + 1800 0 0 1 7400 26 0 + 1800 0 0 2 7400 27 0 + 1800 0 0 3 7400 24 0 + 1800 0 0 4 7400 25 0 + + /* IDSEL 0x14 */ + 2000 0 0 1 7400 27 0 + 2000 0 0 2 7400 24 0 + 2000 0 0 3 7400 25 0 + 2000 0 0 4 7400 26 0 + >; + }; + }; + +}; From 8ec8f2e85c6b88b4a1641eb3902275bcf2c6d60a Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Mon, 28 Aug 2006 11:17:37 +1000 Subject: [PATCH 023/115] [POWERPC] Fix performance regression in IRQ radix tree locking When reworking the powerpc irq code, I figured out that we were using the radix tree in a racy way. As a temporary fix, I put a spinlock in there. However, this can have a significant impact on performances. This patch reworks that to use a smarter technique based on the fact that what we need is in fact a rwlock with extremely rare writers (thus optimized for the read path). Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Paul Mackerras --- arch/powerpc/kernel/irq.c | 84 ++++++++++++++++++++++++++++++--------- 1 file changed, 66 insertions(+), 18 deletions(-) diff --git a/arch/powerpc/kernel/irq.c b/arch/powerpc/kernel/irq.c index 7ee68543331..12c5971d656 100644 --- a/arch/powerpc/kernel/irq.c +++ b/arch/powerpc/kernel/irq.c @@ -322,7 +322,8 @@ EXPORT_SYMBOL(do_softirq); static LIST_HEAD(irq_hosts); static spinlock_t irq_big_lock = SPIN_LOCK_UNLOCKED; - +static DEFINE_PER_CPU(unsigned int, irq_radix_reader); +static unsigned int irq_radix_writer; struct irq_map_entry irq_map[NR_IRQS]; static unsigned int irq_virq_count = NR_IRQS; static struct irq_host *irq_default_host; @@ -455,6 +456,58 @@ void irq_set_virq_count(unsigned int count) irq_virq_count = count; } +/* radix tree not lockless safe ! we use a brlock-type mecanism + * for now, until we can use a lockless radix tree + */ +static void irq_radix_wrlock(unsigned long *flags) +{ + unsigned int cpu, ok; + + spin_lock_irqsave(&irq_big_lock, *flags); + irq_radix_writer = 1; + smp_mb(); + do { + barrier(); + ok = 1; + for_each_possible_cpu(cpu) { + if (per_cpu(irq_radix_reader, cpu)) { + ok = 0; + break; + } + } + if (!ok) + cpu_relax(); + } while(!ok); +} + +static void irq_radix_wrunlock(unsigned long flags) +{ + smp_wmb(); + irq_radix_writer = 0; + spin_unlock_irqrestore(&irq_big_lock, flags); +} + +static void irq_radix_rdlock(unsigned long *flags) +{ + local_irq_save(*flags); + __get_cpu_var(irq_radix_reader) = 1; + smp_mb(); + if (likely(irq_radix_writer == 0)) + return; + __get_cpu_var(irq_radix_reader) = 0; + smp_wmb(); + spin_lock(&irq_big_lock); + __get_cpu_var(irq_radix_reader) = 1; + spin_unlock(&irq_big_lock); +} + +static void irq_radix_rdunlock(unsigned long flags) +{ + __get_cpu_var(irq_radix_reader) = 0; + local_irq_restore(flags); +} + + unsigned int irq_create_mapping(struct irq_host *host, irq_hw_number_t hwirq) { @@ -604,13 +657,9 @@ void irq_dispose_mapping(unsigned int virq) /* Check if radix tree allocated yet */ if (host->revmap_data.tree.gfp_mask == 0) break; - /* XXX radix tree not safe ! remove lock whem it becomes safe - * and use some RCU sync to make sure everything is ok before we - * can re-use that map entry - */ - spin_lock_irqsave(&irq_big_lock, flags); + irq_radix_wrlock(&flags); radix_tree_delete(&host->revmap_data.tree, hwirq); - spin_unlock_irqrestore(&irq_big_lock, flags); + irq_radix_wrunlock(flags); break; } @@ -677,25 +726,24 @@ unsigned int irq_radix_revmap(struct irq_host *host, if (tree->gfp_mask == 0) return irq_find_mapping(host, hwirq); - /* XXX Current radix trees are NOT SMP safe !!! Remove that lock - * when that is fixed (when Nick's patch gets in - */ - spin_lock_irqsave(&irq_big_lock, flags); - /* Now try to resolve */ + irq_radix_rdlock(&flags); ptr = radix_tree_lookup(tree, hwirq); + irq_radix_rdunlock(flags); + /* Found it, return */ if (ptr) { virq = ptr - irq_map; - goto bail; + return virq; } /* If not there, try to insert it */ virq = irq_find_mapping(host, hwirq); - if (virq != NO_IRQ) + if (virq != NO_IRQ) { + irq_radix_wrlock(&flags); radix_tree_insert(tree, hwirq, &irq_map[virq]); - bail: - spin_unlock_irqrestore(&irq_big_lock, flags); + irq_radix_wrunlock(flags); + } return virq; } @@ -806,12 +854,12 @@ static int irq_late_init(void) struct irq_host *h; unsigned long flags; - spin_lock_irqsave(&irq_big_lock, flags); + irq_radix_wrlock(&flags); list_for_each_entry(h, &irq_hosts, link) { if (h->revmap_type == IRQ_HOST_MAP_TREE) INIT_RADIX_TREE(&h->revmap_data.tree, GFP_ATOMIC); } - spin_unlock_irqrestore(&irq_big_lock, flags); + irq_radix_wrunlock(flags); return 0; } From 3efbdd136e52ee4028b5bb5b848a6043cf61cd6e Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Wed, 30 Aug 2006 08:58:00 +1000 Subject: [PATCH 024/115] [POWERPC] Fix MPIC sense codes in documentation The booting-without-of.txt had incorrect definition for the sense codes for an OpenPIC controller Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Paul Mackerras --- Documentation/powerpc/booting-without-of.txt | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/Documentation/powerpc/booting-without-of.txt b/Documentation/powerpc/booting-without-of.txt index 8c48b8a27b9..5c0ba235f5a 100644 --- a/Documentation/powerpc/booting-without-of.txt +++ b/Documentation/powerpc/booting-without-of.txt @@ -1136,10 +1136,10 @@ Sense and level information should be encoded as follows: Devices connected to openPIC-compatible controllers should encode sense and polarity as follows: - 0 = high to low edge sensitive type enabled + 0 = low to high edge sensitive type enabled 1 = active low level sensitive type enabled - 2 = low to high edge sensitive type enabled - 3 = active high level sensitive type enabled + 2 = active high level sensitive type enabled + 3 = high to low edge sensitive type enabled ISA PIC interrupt controllers should adhere to the ISA PIC encodings listed below: From 1ad8f401b6e16e3ba8a70dcda466e7a0986f7e5e Mon Sep 17 00:00:00 2001 From: Nathan Scott Date: Wed, 30 Aug 2006 13:41:23 +1000 Subject: [PATCH 025/115] [XFS] Update the MAINTAINERS file entry for XFS. Signed-off-by: Nathan Scott --- MAINTAINERS | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index 3bab239e82f..25cd7073a20 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -3296,10 +3296,11 @@ S: Maintained XFS FILESYSTEM P: Silicon Graphics Inc +P: Tim Shimmin, David Chatterton M: xfs-masters@oss.sgi.com -M: nathans@sgi.com L: xfs@oss.sgi.com W: http://oss.sgi.com/projects/xfs +T: git git://oss.sgi.com:8090/xfs/xfs-2.6 S: Supported X86 3-LEVEL PAGING (PAE) SUPPORT From 7288026b8671061aff7663b1766037b3f2573627 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Wed, 30 Aug 2006 13:41:58 +1000 Subject: [PATCH 026/115] [XFS] Fix char size overflow in bmap_alloc call for unwritten extent conversion. Since bma.conv is a char and XFS_BMAPI_CONVERT is 0x1000, bma.conv was always assigned zero. Spotted by the GNU C compiler (SVN version). SGI-PV: 947312 SGI-Modid: xfs-linux-melb:xfs-kern:26887a Signed-off-by: Adrian Bunk Signed-off-by: Nathan Scott --- fs/xfs/xfs_bmap.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/xfs/xfs_bmap.c b/fs/xfs/xfs_bmap.c index 3a613753906..bf46fae303a 100644 --- a/fs/xfs/xfs_bmap.c +++ b/fs/xfs/xfs_bmap.c @@ -4993,7 +4993,7 @@ xfs_bmapi( bma.firstblock = *firstblock; bma.alen = alen; bma.off = aoff; - bma.conv = (flags & XFS_BMAPI_CONVERT); + bma.conv = !!(flags & XFS_BMAPI_CONVERT); bma.wasdel = wasdelay; bma.minlen = minlen; bma.low = flist->xbf_low; From b9ac86727fc02cc7117ef3fe518a4d51cd573c82 Mon Sep 17 00:00:00 2001 From: Sridhar Samudrala Date: Mon, 28 Aug 2006 13:53:01 -0700 Subject: [PATCH 027/115] [SCTP]: Fix sctp_primitive_ABORT() call in sctp_close(). With the recent fix, the callers of sctp_primitive_ABORT() need to create an ABORT chunk and pass it as an argument rather than msghdr that was passed earlier. Signed-off-by: Sridhar Samudrala Signed-off-by: David S. Miller --- net/sctp/socket.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/net/sctp/socket.c b/net/sctp/socket.c index fde3f55bfd4..dab15949958 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -1289,9 +1289,13 @@ SCTP_STATIC void sctp_close(struct sock *sk, long timeout) } } - if (sock_flag(sk, SOCK_LINGER) && !sk->sk_lingertime) - sctp_primitive_ABORT(asoc, NULL); - else + if (sock_flag(sk, SOCK_LINGER) && !sk->sk_lingertime) { + struct sctp_chunk *chunk; + + chunk = sctp_make_abort_user(asoc, NULL, 0); + if (chunk) + sctp_primitive_ABORT(asoc, chunk); + } else sctp_primitive_SHUTDOWN(asoc, NULL); } From 8fb6f732c389847dece403b7470d6d3d2778804a Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Mon, 28 Aug 2006 22:12:54 -0700 Subject: [PATCH 028/115] [E100]: Add module option to ignore bad EEPROM checksums. Several people run into the situation where the E100 EEPROM contents are fine, but the checksum hasn't been set properly. This renders the device useless for them even though it would function correctly. The default is off, which retains the current behavior. Signed-off-by: David S. Miller --- drivers/net/e100.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/drivers/net/e100.c b/drivers/net/e100.c index 91ef5f2fd76..ce850f1078b 100644 --- a/drivers/net/e100.c +++ b/drivers/net/e100.c @@ -173,8 +173,11 @@ MODULE_LICENSE("GPL"); MODULE_VERSION(DRV_VERSION); static int debug = 3; +static int eeprom_bad_csum_allow = 0; module_param(debug, int, 0); +module_param(eeprom_bad_csum_allow, int, 0); MODULE_PARM_DESC(debug, "Debug level (0=none,...,16=all)"); +MODULE_PARM_DESC(eeprom_bad_csum_allow, "Allow bad eeprom checksums"); #define DPRINTK(nlevel, klevel, fmt, args...) \ (void)((NETIF_MSG_##nlevel & nic->msg_enable) && \ printk(KERN_##klevel PFX "%s: %s: " fmt, nic->netdev->name, \ @@ -756,7 +759,8 @@ static int e100_eeprom_load(struct nic *nic) checksum = le16_to_cpu(0xBABA - checksum); if(checksum != nic->eeprom[nic->eeprom_wc - 1]) { DPRINTK(PROBE, ERR, "EEPROM corrupted\n"); - return -EAGAIN; + if (!eeprom_bad_csum_allow) + return -EAGAIN; } return 0; From 76d0cc1b64e1686b2b53e081c94142dd99f01ec5 Mon Sep 17 00:00:00 2001 From: Lv Liangying Date: Tue, 29 Aug 2006 00:00:47 -0700 Subject: [PATCH 029/115] [IPV6]: SNMPv2 "ipv6IfStatsInAddrErrors" counter error When I tested Linux kernel 2.6.17.7 about statistics "ipv6IfStatsInAddrErrors", found that this counter couldn't increase correctly. The criteria is RFC2465: ipv6IfStatsInAddrErrors OBJECT-TYPE SYNTAX Counter32 MAX-ACCESS read-only STATUS current DESCRIPTION "The number of input datagrams discarded because the IPv6 address in their IPv6 header's destination field was not a valid address to be received at this entity. This count includes invalid addresses (e.g., ::0) and unsupported addresses (e.g., addresses with unallocated prefixes). For entities which are not IPv6 routers and therefore do not forward datagrams, this counter includes datagrams discarded because the destination address was not a local address." ::= { ipv6IfStatsEntry 5 } When I send packet to host with destination that is ether invalid address(::0) or unsupported addresses(1::1), the Linux kernel just discard the packet, and the counter doesn't increase(in the function ip6_pkt_discard). Signed-off-by: Lv Liangying Signed-off-by: David S. Miller --- net/ipv6/route.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 4b163711f3a..d9baca062d2 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -1532,6 +1532,10 @@ int ipv6_route_ioctl(unsigned int cmd, void __user *arg) static int ip6_pkt_discard(struct sk_buff *skb) { + int type = ipv6_addr_type(&skb->nh.ipv6h->daddr); + if (type == IPV6_ADDR_ANY || type == IPV6_ADDR_RESERVED) + IP6_INC_STATS(IPSTATS_MIB_INADDRERRORS); + IP6_INC_STATS(IPSTATS_MIB_OUTNOROUTES); icmpv6_send(skb, ICMPV6_DEST_UNREACH, ICMPV6_NOROUTE, 0, skb->dev); kfree_skb(skb); From 3fdf3f0c99e90e167f0d0643fcc8739e27456697 Mon Sep 17 00:00:00 2001 From: Daikichi Osuga Date: Tue, 29 Aug 2006 02:01:44 -0700 Subject: [PATCH 030/115] [TCP]: Two RFC3465 Appropriate Byte Count fixes. 1) fix slow start after retransmit timeout 2) fix case of L=2*SMSS acked bytes comparison Signed-off-by: Daikichi Osuga Signed-off-by: David S. Miller --- net/ipv4/tcp_cong.c | 2 +- net/ipv4/tcp_input.c | 9 +++++++-- 2 files changed, 8 insertions(+), 3 deletions(-) diff --git a/net/ipv4/tcp_cong.c b/net/ipv4/tcp_cong.c index 5765f9d0317..7ff2e4273a7 100644 --- a/net/ipv4/tcp_cong.c +++ b/net/ipv4/tcp_cong.c @@ -189,7 +189,7 @@ void tcp_slow_start(struct tcp_sock *tp) return; /* We MAY increase by 2 if discovered delayed ack */ - if (sysctl_tcp_abc > 1 && tp->bytes_acked > 2*tp->mss_cache) { + if (sysctl_tcp_abc > 1 && tp->bytes_acked >= 2*tp->mss_cache) { if (tp->snd_cwnd < tp->snd_cwnd_clamp) tp->snd_cwnd++; } diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 104af5d5bcb..111ff39a08c 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -2505,8 +2505,13 @@ static int tcp_ack(struct sock *sk, struct sk_buff *skb, int flag) if (before(ack, prior_snd_una)) goto old_ack; - if (sysctl_tcp_abc && icsk->icsk_ca_state < TCP_CA_CWR) - tp->bytes_acked += ack - prior_snd_una; + if (sysctl_tcp_abc) { + if (icsk->icsk_ca_state < TCP_CA_CWR) + tp->bytes_acked += ack - prior_snd_una; + else if (icsk->icsk_ca_state == TCP_CA_Loss) + /* we assume just one segment left network */ + tp->bytes_acked += min(ack - prior_snd_una, tp->mss_cache); + } if (!(flag&FLAG_SLOWPATH) && after(ack, prior_snd_una)) { /* Window is constant, pure forward advance. From fab2caf62ed03d83bd3a3598b859c3c345a8e8b5 Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Tue, 29 Aug 2006 02:15:24 -0700 Subject: [PATCH 031/115] [NETLINK]: Call panic if nl_table allocation fails This patch makes crash happen if initialization of nl_table fails in initcalls. It is better than getting use after free crash later. Signed-off-by: Akinobu Mita Signed-off-by: David S. Miller --- net/netlink/af_netlink.c | 14 ++++++-------- 1 file changed, 6 insertions(+), 8 deletions(-) diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index b85c1f9f128..8b85036ba8e 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -1273,8 +1273,7 @@ netlink_kernel_create(int unit, unsigned int groups, struct netlink_sock *nlk; unsigned long *listeners = NULL; - if (!nl_table) - return NULL; + BUG_ON(!nl_table); if (unit<0 || unit>=MAX_LINKS) return NULL; @@ -1745,11 +1744,8 @@ static int __init netlink_proto_init(void) netlink_skb_parms_too_large(); nl_table = kcalloc(MAX_LINKS, sizeof(*nl_table), GFP_KERNEL); - if (!nl_table) { -enomem: - printk(KERN_CRIT "netlink_init: Cannot allocate nl_table\n"); - return -ENOMEM; - } + if (!nl_table) + goto panic; if (num_physpages >= (128 * 1024)) max = num_physpages >> (21 - PAGE_SHIFT); @@ -1769,7 +1765,7 @@ enomem: nl_pid_hash_free(nl_table[i].hash.table, 1 * sizeof(*hash->table)); kfree(nl_table); - goto enomem; + goto panic; } memset(hash->table, 0, 1 * sizeof(*hash->table)); hash->max_shift = order; @@ -1786,6 +1782,8 @@ enomem: rtnetlink_init(); out: return err; +panic: + panic("netlink_init: Cannot allocate nl_table\n"); } core_initcall(netlink_proto_init); From 57f5f544f58ffa1d6c38630d0522c9c0be67c533 Mon Sep 17 00:00:00 2001 From: Keir Fraser Date: Tue, 29 Aug 2006 02:43:49 -0700 Subject: [PATCH 032/115] [IPV6]: ipv6_add_addr should install dstentry earlier ipv6_add_addr allocates a struct inet6_ifaddr and a dstentry, but it doesn't install the dstentry in ifa->rt until after it releases the addrconf_hash_lock. This means other CPUs will be able to see the new address while it hasn't been initialized completely yet. One possible fix would be to grab the ifp->lock spinlock when creating the address struct; a simpler fix is to just move the assignment. Acked-by: jbeulich@novell.com Acked-by: okir@suse.de Acked-by: YOSHIFUJI Hideaki Signed-off-by: David S. Miller --- net/ipv6/addrconf.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 0c5042e7380..c7852b38e03 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -578,6 +578,8 @@ ipv6_add_addr(struct inet6_dev *idev, const struct in6_addr *addr, int pfxlen, ifa->flags = flags | IFA_F_TENTATIVE; ifa->cstamp = ifa->tstamp = jiffies; + ifa->rt = rt; + ifa->idev = idev; in6_dev_hold(idev); /* For caller */ @@ -603,8 +605,6 @@ ipv6_add_addr(struct inet6_dev *idev, const struct in6_addr *addr, int pfxlen, } #endif - ifa->rt = rt; - in6_ifa_hold(ifa); write_unlock(&idev->lock); out2: From ee1377c3eef4238d89b2f99fa4d0bbbad3078b64 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Tue, 29 Aug 2006 21:18:45 -0700 Subject: [PATCH 033/115] [STRIP]: Fix neighbour table refcount leak. Found by inspection. The STRIP driver does neigh_lookup() but never releases. This driver shouldn't being doing gratuitous arp anyway. Untested, obviously, because of lack of hardware. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- drivers/net/wireless/strip.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/net/wireless/strip.c b/drivers/net/wireless/strip.c index fd31885c684..ccaf28e8db0 100644 --- a/drivers/net/wireless/strip.c +++ b/drivers/net/wireless/strip.c @@ -467,6 +467,7 @@ static int arp_query(unsigned char *haddr, u32 paddr, struct net_device *dev) { struct neighbour *neighbor_entry; + int ret = 0; neighbor_entry = neigh_lookup(&arp_tbl, &paddr, dev); @@ -474,10 +475,11 @@ static int arp_query(unsigned char *haddr, u32 paddr, neighbor_entry->used = jiffies; if (neighbor_entry->nud_state & NUD_VALID) { memcpy(haddr, neighbor_entry->ha, dev->addr_len); - return 1; + ret = 1; } + neigh_release(neighbor_entry); } - return 0; + return ret; } static void DumpData(char *msg, struct strip *strip_info, __u8 * ptr, From 47f2c3604f47579ac5c173f8b402dc6cd8e2e8fa Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Mon, 28 Aug 2006 00:33:03 -0700 Subject: [PATCH 034/115] [SPARC64]: Fix X server hangs due to large pages. This problem was introduced by changeset 14778d9072e53d2171f66ffd9657daff41acfaed Unlike the hugetlb code paths, the normal fault code is not setup to propagate PTE changes for large page sizes correctly like the ones we make for I/O mappings in io_remap_pfn_range(). It is absolutely necessary to update all sub-ptes of a largepage mapping on a fault. Adding special handling for this would add considerably complexity to tlb_batch_add(). So let's just side-step the issue and forcefully dirty any writable PTEs created by io_remap_pfn_range(). The only other real option would be to disable to large PTE code of io_remap_pfn_range() and we really don't want to do that. Much thanks to Mikael Pettersson for tracking down this problem and testing debug patches. Signed-off-by: David S. Miller --- arch/sparc64/mm/generic.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/sparc64/mm/generic.c b/arch/sparc64/mm/generic.c index 8cb06205d26..af9d81db0b3 100644 --- a/arch/sparc64/mm/generic.c +++ b/arch/sparc64/mm/generic.c @@ -69,6 +69,8 @@ static inline void io_remap_pte_range(struct mm_struct *mm, pte_t * pte, } else offset += PAGE_SIZE; + if (pte_write(entry)) + entry = pte_mkdirty(entry); do { BUG_ON(!pte_none(*pte)); set_pte_at(mm, address, pte, entry); From 404dda854b8bb04df72405d5088fa3e6100aef36 Mon Sep 17 00:00:00 2001 From: Krzysztof Helt Date: Mon, 28 Aug 2006 23:28:16 -0700 Subject: [PATCH 035/115] [SUNLANCE]: Fix probing problem. The current probe table causes ledma and lebuffer "le" devices to get probed twice which is not what we want. Match just "le" and look directly at the parent to get the correct top-level node information. Signed-off-by: Krzysztof Helt Signed-off-by: David S. Miller --- drivers/net/sunlance.c | 29 ++++++++++++----------------- 1 file changed, 12 insertions(+), 17 deletions(-) diff --git a/drivers/net/sunlance.c b/drivers/net/sunlance.c index 0e3fdf7c6dd..ec0413609f3 100644 --- a/drivers/net/sunlance.c +++ b/drivers/net/sunlance.c @@ -1566,20 +1566,21 @@ static int __exit sunlance_sun4_remove(void) static int __devinit sunlance_sbus_probe(struct of_device *dev, const struct of_device_id *match) { struct sbus_dev *sdev = to_sbus_device(&dev->dev); - struct device_node *dp = dev->node; int err; - if (!strcmp(dp->name, "le")) { + if (sdev->parent) { + struct of_device *parent = &sdev->parent->ofdev; + + if (!strcmp(parent->node->name, "ledma")) { + struct sbus_dma *ledma = find_ledma(to_sbus_device(&parent->dev)); + + err = sparc_lance_probe_one(sdev, ledma, NULL); + } else if (!strcmp(parent->node->name, "lebuffer")) { + err = sparc_lance_probe_one(sdev, NULL, to_sbus_device(&parent->dev)); + } else + err = sparc_lance_probe_one(sdev, NULL, NULL); + } else err = sparc_lance_probe_one(sdev, NULL, NULL); - } else if (!strcmp(dp->name, "ledma")) { - struct sbus_dma *ledma = find_ledma(sdev); - - err = sparc_lance_probe_one(sdev->child, ledma, NULL); - } else { - BUG_ON(strcmp(dp->name, "lebuffer")); - - err = sparc_lance_probe_one(sdev->child, NULL, sdev); - } return err; } @@ -1604,12 +1605,6 @@ static struct of_device_id sunlance_sbus_match[] = { { .name = "le", }, - { - .name = "ledma", - }, - { - .name = "lebuffer", - }, {}, }; From 7233593b7844c2db930594ee9c0c872a6900bfcc Mon Sep 17 00:00:00 2001 From: Zang Roy-r61911 Date: Fri, 25 Aug 2006 14:16:30 +1000 Subject: [PATCH 036/115] [POWERPC] Support for "weird" MPICs and fixup mpc7448_hpc2 This adds a new hardware information table for mpic. This enables the mpic code to deal with mpic controllers with different register layouts and hardware behaviours. This introduces CONFIG_MPIC_WEIRD. For boards with non standard mpic controllers, select CONFIG_MPIC_WEIRD and add its hardware information in the mpic_infos[] array. TSI108/109 PIC takes the first index of weird hardware information table. :) The table can be extended. The Tsi108/109 PIC looks like standard OpenPIC but, in fact, is different in register mapping and behavior. The patch does not affect the behavior of standard mpic. If CONFIG_MPIC_WEIRD is not defined, the code is essentially identical to the current code. [benh@kernel.crashing.org: This patch is a slightly cleaned up version of Zang Roy's support for the TSI108 MPIC variant. It also fixes up MPC7448_hpc2 to use the new version of the type macros and changes the way MPIC is selected in Kconfig to better match what is done for other system devices. ] Signed-off-by: Roy Zang Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Paul Mackerras --- arch/powerpc/Kconfig | 20 +- arch/powerpc/platforms/embedded6xx/Kconfig | 1 + .../platforms/embedded6xx/mpc7448_hpc2.c | 2 +- arch/powerpc/sysdev/mpic.c | 223 +++++++++++++----- include/asm-powerpc/mpic.h | 125 ++++++++++ 5 files changed, 305 insertions(+), 66 deletions(-) diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig index abb325eb8f7..4d4b6fb156e 100644 --- a/arch/powerpc/Kconfig +++ b/arch/powerpc/Kconfig @@ -354,6 +354,7 @@ endchoice config PPC_PSERIES depends on PPC_MULTIPLATFORM && PPC64 bool "IBM pSeries & new (POWER5-based) iSeries" + select MPIC select PPC_I8259 select PPC_RTAS select RTAS_ERROR_LOGGING @@ -363,6 +364,7 @@ config PPC_PSERIES config PPC_CHRP bool "Common Hardware Reference Platform (CHRP) based machines" depends on PPC_MULTIPLATFORM && PPC32 + select MPIC select PPC_I8259 select PPC_INDIRECT_PCI select PPC_RTAS @@ -373,6 +375,7 @@ config PPC_CHRP config PPC_PMAC bool "Apple PowerMac based machines" depends on PPC_MULTIPLATFORM + select MPIC select PPC_INDIRECT_PCI if PPC32 select PPC_MPC106 if PPC32 default y @@ -380,6 +383,7 @@ config PPC_PMAC config PPC_PMAC64 bool depends on PPC_PMAC && POWER4 + select MPIC select U3_DART select MPIC_BROKEN_U3 select GENERIC_TBSYNC @@ -389,6 +393,7 @@ config PPC_PMAC64 config PPC_PREP bool "PowerPC Reference Platform (PReP) based machines" depends on PPC_MULTIPLATFORM && PPC32 && BROKEN + select MPIC select PPC_I8259 select PPC_INDIRECT_PCI select PPC_UDBG_16550 @@ -397,6 +402,7 @@ config PPC_PREP config PPC_MAPLE depends on PPC_MULTIPLATFORM && PPC64 bool "Maple 970FX Evaluation Board" + select MPIC select U3_DART select MPIC_BROKEN_U3 select GENERIC_TBSYNC @@ -439,12 +445,6 @@ config U3_DART depends on PPC_MULTIPLATFORM && PPC64 default n -config MPIC - depends on PPC_PSERIES || PPC_PMAC || PPC_MAPLE || PPC_CHRP \ - || MPC7448HPC2 - bool - default y - config PPC_RTAS bool default n @@ -812,6 +812,14 @@ config GENERIC_ISA_DMA depends on PPC64 || POWER4 || 6xx && !CPM2 default y +config MPIC + bool + default n + +config MPIC_WEIRD + bool + default n + config PPC_I8259 bool default n diff --git a/arch/powerpc/platforms/embedded6xx/Kconfig b/arch/powerpc/platforms/embedded6xx/Kconfig index ba07a9a7c03..234a861870a 100644 --- a/arch/powerpc/platforms/embedded6xx/Kconfig +++ b/arch/powerpc/platforms/embedded6xx/Kconfig @@ -80,6 +80,7 @@ config MPC7448HPC2 select DEFAULT_UIMAGE select PPC_UDBG_16550 select MPIC + select MPIC_WEIRD help Select MPC7448HPC2 if configuring for Freescale MPC7448HPC2 (Taiga) platform diff --git a/arch/powerpc/platforms/embedded6xx/mpc7448_hpc2.c b/arch/powerpc/platforms/embedded6xx/mpc7448_hpc2.c index ed00ed2455d..5d393eb9493 100644 --- a/arch/powerpc/platforms/embedded6xx/mpc7448_hpc2.c +++ b/arch/powerpc/platforms/embedded6xx/mpc7448_hpc2.c @@ -215,7 +215,7 @@ static void __init mpc7448_hpc2_init_IRQ(void) mpic = mpic_alloc(tsi_pic, mpic_paddr, MPIC_PRIMARY | MPIC_BIG_ENDIAN | MPIC_WANTS_RESET | - MPIC_SPV_EOI | MPIC_MOD_ID(MPIC_ID_TSI108), + MPIC_SPV_EOI | MPIC_NO_PTHROU_DIS | MPIC_REGSET_TSI108, 0, /* num_sources used */ 0, /* num_sources used */ "Tsi108_PIC"); diff --git a/arch/powerpc/sysdev/mpic.c b/arch/powerpc/sysdev/mpic.c index 6e0281afa6c..b604926401f 100644 --- a/arch/powerpc/sysdev/mpic.c +++ b/arch/powerpc/sysdev/mpic.c @@ -54,6 +54,94 @@ static DEFINE_SPINLOCK(mpic_lock); #endif #endif +#ifdef CONFIG_MPIC_WEIRD +static u32 mpic_infos[][MPIC_IDX_END] = { + [0] = { /* Original OpenPIC compatible MPIC */ + MPIC_GREG_BASE, + MPIC_GREG_FEATURE_0, + MPIC_GREG_GLOBAL_CONF_0, + MPIC_GREG_VENDOR_ID, + MPIC_GREG_IPI_VECTOR_PRI_0, + MPIC_GREG_IPI_STRIDE, + MPIC_GREG_SPURIOUS, + MPIC_GREG_TIMER_FREQ, + + MPIC_TIMER_BASE, + MPIC_TIMER_STRIDE, + MPIC_TIMER_CURRENT_CNT, + MPIC_TIMER_BASE_CNT, + MPIC_TIMER_VECTOR_PRI, + MPIC_TIMER_DESTINATION, + + MPIC_CPU_BASE, + MPIC_CPU_STRIDE, + MPIC_CPU_IPI_DISPATCH_0, + MPIC_CPU_IPI_DISPATCH_STRIDE, + MPIC_CPU_CURRENT_TASK_PRI, + MPIC_CPU_WHOAMI, + MPIC_CPU_INTACK, + MPIC_CPU_EOI, + + MPIC_IRQ_BASE, + MPIC_IRQ_STRIDE, + MPIC_IRQ_VECTOR_PRI, + MPIC_VECPRI_VECTOR_MASK, + MPIC_VECPRI_POLARITY_POSITIVE, + MPIC_VECPRI_POLARITY_NEGATIVE, + MPIC_VECPRI_SENSE_LEVEL, + MPIC_VECPRI_SENSE_EDGE, + MPIC_VECPRI_POLARITY_MASK, + MPIC_VECPRI_SENSE_MASK, + MPIC_IRQ_DESTINATION + }, + [1] = { /* Tsi108/109 PIC */ + TSI108_GREG_BASE, + TSI108_GREG_FEATURE_0, + TSI108_GREG_GLOBAL_CONF_0, + TSI108_GREG_VENDOR_ID, + TSI108_GREG_IPI_VECTOR_PRI_0, + TSI108_GREG_IPI_STRIDE, + TSI108_GREG_SPURIOUS, + TSI108_GREG_TIMER_FREQ, + + TSI108_TIMER_BASE, + TSI108_TIMER_STRIDE, + TSI108_TIMER_CURRENT_CNT, + TSI108_TIMER_BASE_CNT, + TSI108_TIMER_VECTOR_PRI, + TSI108_TIMER_DESTINATION, + + TSI108_CPU_BASE, + TSI108_CPU_STRIDE, + TSI108_CPU_IPI_DISPATCH_0, + TSI108_CPU_IPI_DISPATCH_STRIDE, + TSI108_CPU_CURRENT_TASK_PRI, + TSI108_CPU_WHOAMI, + TSI108_CPU_INTACK, + TSI108_CPU_EOI, + + TSI108_IRQ_BASE, + TSI108_IRQ_STRIDE, + TSI108_IRQ_VECTOR_PRI, + TSI108_VECPRI_VECTOR_MASK, + TSI108_VECPRI_POLARITY_POSITIVE, + TSI108_VECPRI_POLARITY_NEGATIVE, + TSI108_VECPRI_SENSE_LEVEL, + TSI108_VECPRI_SENSE_EDGE, + TSI108_VECPRI_POLARITY_MASK, + TSI108_VECPRI_SENSE_MASK, + TSI108_IRQ_DESTINATION + }, +}; + +#define MPIC_INFO(name) mpic->hw_set[MPIC_IDX_##name] + +#else /* CONFIG_MPIC_WEIRD */ + +#define MPIC_INFO(name) MPIC_##name + +#endif /* CONFIG_MPIC_WEIRD */ + /* * Register accessor functions */ @@ -80,7 +168,8 @@ static inline void _mpic_write(unsigned int be, volatile u32 __iomem *base, static inline u32 _mpic_ipi_read(struct mpic *mpic, unsigned int ipi) { unsigned int be = (mpic->flags & MPIC_BIG_ENDIAN) != 0; - unsigned int offset = MPIC_GREG_IPI_VECTOR_PRI_0 + (ipi * 0x10); + unsigned int offset = MPIC_INFO(GREG_IPI_VECTOR_PRI_0) + + (ipi * MPIC_INFO(GREG_IPI_STRIDE)); if (mpic->flags & MPIC_BROKEN_IPI) be = !be; @@ -89,7 +178,8 @@ static inline u32 _mpic_ipi_read(struct mpic *mpic, unsigned int ipi) static inline void _mpic_ipi_write(struct mpic *mpic, unsigned int ipi, u32 value) { - unsigned int offset = MPIC_GREG_IPI_VECTOR_PRI_0 + (ipi * 0x10); + unsigned int offset = MPIC_INFO(GREG_IPI_VECTOR_PRI_0) + + (ipi * MPIC_INFO(GREG_IPI_STRIDE)); _mpic_write(mpic->flags & MPIC_BIG_ENDIAN, mpic->gregs, offset, value); } @@ -120,7 +210,7 @@ static inline u32 _mpic_irq_read(struct mpic *mpic, unsigned int src_no, unsigne unsigned int idx = src_no & mpic->isu_mask; return _mpic_read(mpic->flags & MPIC_BIG_ENDIAN, mpic->isus[isu], - reg + (idx * MPIC_IRQ_STRIDE)); + reg + (idx * MPIC_INFO(IRQ_STRIDE))); } static inline void _mpic_irq_write(struct mpic *mpic, unsigned int src_no, @@ -130,7 +220,7 @@ static inline void _mpic_irq_write(struct mpic *mpic, unsigned int src_no, unsigned int idx = src_no & mpic->isu_mask; _mpic_write(mpic->flags & MPIC_BIG_ENDIAN, mpic->isus[isu], - reg + (idx * MPIC_IRQ_STRIDE), value); + reg + (idx * MPIC_INFO(IRQ_STRIDE)), value); } #define mpic_read(b,r) _mpic_read(mpic->flags & MPIC_BIG_ENDIAN,(b),(r)) @@ -156,8 +246,8 @@ static void __init mpic_test_broken_ipi(struct mpic *mpic) { u32 r; - mpic_write(mpic->gregs, MPIC_GREG_IPI_VECTOR_PRI_0, MPIC_VECPRI_MASK); - r = mpic_read(mpic->gregs, MPIC_GREG_IPI_VECTOR_PRI_0); + mpic_write(mpic->gregs, MPIC_INFO(GREG_IPI_VECTOR_PRI_0), MPIC_VECPRI_MASK); + r = mpic_read(mpic->gregs, MPIC_INFO(GREG_IPI_VECTOR_PRI_0)); if (r == le32_to_cpu(MPIC_VECPRI_MASK)) { printk(KERN_INFO "mpic: Detected reversed IPI registers\n"); @@ -394,8 +484,8 @@ static inline struct mpic * mpic_from_irq(unsigned int irq) /* Send an EOI */ static inline void mpic_eoi(struct mpic *mpic) { - mpic_cpu_write(MPIC_CPU_EOI, 0); - (void)mpic_cpu_read(MPIC_CPU_WHOAMI); + mpic_cpu_write(MPIC_INFO(CPU_EOI), 0); + (void)mpic_cpu_read(MPIC_INFO(CPU_WHOAMI)); } #ifdef CONFIG_SMP @@ -419,8 +509,8 @@ static void mpic_unmask_irq(unsigned int irq) DBG("%p: %s: enable_irq: %d (src %d)\n", mpic, mpic->name, irq, src); - mpic_irq_write(src, MPIC_IRQ_VECTOR_PRI, - mpic_irq_read(src, MPIC_IRQ_VECTOR_PRI) & + mpic_irq_write(src, MPIC_INFO(IRQ_VECTOR_PRI), + mpic_irq_read(src, MPIC_INFO(IRQ_VECTOR_PRI)) & ~MPIC_VECPRI_MASK); /* make sure mask gets to controller before we return to user */ do { @@ -428,7 +518,7 @@ static void mpic_unmask_irq(unsigned int irq) printk(KERN_ERR "mpic_enable_irq timeout\n"); break; } - } while(mpic_irq_read(src, MPIC_IRQ_VECTOR_PRI) & MPIC_VECPRI_MASK); + } while(mpic_irq_read(src, MPIC_INFO(IRQ_VECTOR_PRI)) & MPIC_VECPRI_MASK); } static void mpic_mask_irq(unsigned int irq) @@ -439,8 +529,8 @@ static void mpic_mask_irq(unsigned int irq) DBG("%s: disable_irq: %d (src %d)\n", mpic->name, irq, src); - mpic_irq_write(src, MPIC_IRQ_VECTOR_PRI, - mpic_irq_read(src, MPIC_IRQ_VECTOR_PRI) | + mpic_irq_write(src, MPIC_INFO(IRQ_VECTOR_PRI), + mpic_irq_read(src, MPIC_INFO(IRQ_VECTOR_PRI)) | MPIC_VECPRI_MASK); /* make sure mask gets to controller before we return to user */ @@ -449,7 +539,7 @@ static void mpic_mask_irq(unsigned int irq) printk(KERN_ERR "mpic_enable_irq timeout\n"); break; } - } while(!(mpic_irq_read(src, MPIC_IRQ_VECTOR_PRI) & MPIC_VECPRI_MASK)); + } while(!(mpic_irq_read(src, MPIC_INFO(IRQ_VECTOR_PRI)) & MPIC_VECPRI_MASK)); } static void mpic_end_irq(unsigned int irq) @@ -560,24 +650,28 @@ static void mpic_set_affinity(unsigned int irq, cpumask_t cpumask) cpus_and(tmp, cpumask, cpu_online_map); - mpic_irq_write(src, MPIC_IRQ_DESTINATION, + mpic_irq_write(src, MPIC_INFO(IRQ_DESTINATION), mpic_physmask(cpus_addr(tmp)[0])); } -static unsigned int mpic_type_to_vecpri(unsigned int type) +static unsigned int mpic_type_to_vecpri(struct mpic *mpic, unsigned int type) { /* Now convert sense value */ switch(type & IRQ_TYPE_SENSE_MASK) { case IRQ_TYPE_EDGE_RISING: - return MPIC_VECPRI_SENSE_EDGE | MPIC_VECPRI_POLARITY_POSITIVE; + return MPIC_INFO(VECPRI_SENSE_EDGE) | + MPIC_INFO(VECPRI_POLARITY_POSITIVE); case IRQ_TYPE_EDGE_FALLING: case IRQ_TYPE_EDGE_BOTH: - return MPIC_VECPRI_SENSE_EDGE | MPIC_VECPRI_POLARITY_NEGATIVE; + return MPIC_INFO(VECPRI_SENSE_EDGE) | + MPIC_INFO(VECPRI_POLARITY_NEGATIVE); case IRQ_TYPE_LEVEL_HIGH: - return MPIC_VECPRI_SENSE_LEVEL | MPIC_VECPRI_POLARITY_POSITIVE; + return MPIC_INFO(VECPRI_SENSE_LEVEL) | + MPIC_INFO(VECPRI_POLARITY_POSITIVE); case IRQ_TYPE_LEVEL_LOW: default: - return MPIC_VECPRI_SENSE_LEVEL | MPIC_VECPRI_POLARITY_NEGATIVE; + return MPIC_INFO(VECPRI_SENSE_LEVEL) | + MPIC_INFO(VECPRI_POLARITY_NEGATIVE); } } @@ -609,13 +703,14 @@ static int mpic_set_irq_type(unsigned int virq, unsigned int flow_type) vecpri = MPIC_VECPRI_POLARITY_POSITIVE | MPIC_VECPRI_SENSE_EDGE; else - vecpri = mpic_type_to_vecpri(flow_type); + vecpri = mpic_type_to_vecpri(mpic, flow_type); - vold = mpic_irq_read(src, MPIC_IRQ_VECTOR_PRI); - vnew = vold & ~(MPIC_VECPRI_POLARITY_MASK | MPIC_VECPRI_SENSE_MASK); + vold = mpic_irq_read(src, MPIC_INFO(IRQ_VECTOR_PRI)); + vnew = vold & ~(MPIC_INFO(VECPRI_POLARITY_MASK) | + MPIC_INFO(VECPRI_SENSE_MASK)); vnew |= vecpri; if (vold != vnew) - mpic_irq_write(src, MPIC_IRQ_VECTOR_PRI, vnew); + mpic_irq_write(src, MPIC_INFO(IRQ_VECTOR_PRI), vnew); return 0; } @@ -798,17 +893,22 @@ struct mpic * __init mpic_alloc(struct device_node *node, mpic->irq_count = irq_count; mpic->num_sources = 0; /* so far */ +#ifdef CONFIG_MPIC_WEIRD + mpic->hw_set = mpic_infos[MPIC_GET_REGSET(flags)]; +#endif + /* Map the global registers */ - mpic->gregs = ioremap(phys_addr + MPIC_GREG_BASE, 0x1000); - mpic->tmregs = mpic->gregs + ((MPIC_TIMER_BASE - MPIC_GREG_BASE) >> 2); + mpic->gregs = ioremap(phys_addr + MPIC_INFO(GREG_BASE), 0x1000); + mpic->tmregs = mpic->gregs + + ((MPIC_INFO(TIMER_BASE) - MPIC_INFO(GREG_BASE)) >> 2); BUG_ON(mpic->gregs == NULL); /* Reset */ if (flags & MPIC_WANTS_RESET) { - mpic_write(mpic->gregs, MPIC_GREG_GLOBAL_CONF_0, - mpic_read(mpic->gregs, MPIC_GREG_GLOBAL_CONF_0) + mpic_write(mpic->gregs, MPIC_INFO(GREG_GLOBAL_CONF_0), + mpic_read(mpic->gregs, MPIC_INFO(GREG_GLOBAL_CONF_0)) | MPIC_GREG_GCONF_RESET); - while( mpic_read(mpic->gregs, MPIC_GREG_GLOBAL_CONF_0) + while( mpic_read(mpic->gregs, MPIC_INFO(GREG_GLOBAL_CONF_0)) & MPIC_GREG_GCONF_RESET) mb(); } @@ -817,7 +917,7 @@ struct mpic * __init mpic_alloc(struct device_node *node, * MPICs, num sources as well. On ISU MPICs, sources are counted * as ISUs are added */ - reg = mpic_read(mpic->gregs, MPIC_GREG_FEATURE_0); + reg = mpic_read(mpic->gregs, MPIC_INFO(GREG_FEATURE_0)); mpic->num_cpus = ((reg & MPIC_GREG_FEATURE_LAST_CPU_MASK) >> MPIC_GREG_FEATURE_LAST_CPU_SHIFT) + 1; if (isu_size == 0) @@ -826,16 +926,16 @@ struct mpic * __init mpic_alloc(struct device_node *node, /* Map the per-CPU registers */ for (i = 0; i < mpic->num_cpus; i++) { - mpic->cpuregs[i] = ioremap(phys_addr + MPIC_CPU_BASE + - i * MPIC_CPU_STRIDE, 0x1000); + mpic->cpuregs[i] = ioremap(phys_addr + MPIC_INFO(CPU_BASE) + + i * MPIC_INFO(CPU_STRIDE), 0x1000); BUG_ON(mpic->cpuregs[i] == NULL); } /* Initialize main ISU if none provided */ if (mpic->isu_size == 0) { mpic->isu_size = mpic->num_sources; - mpic->isus[0] = ioremap(phys_addr + MPIC_IRQ_BASE, - MPIC_IRQ_STRIDE * mpic->isu_size); + mpic->isus[0] = ioremap(phys_addr + MPIC_INFO(IRQ_BASE), + MPIC_INFO(IRQ_STRIDE) * mpic->isu_size); BUG_ON(mpic->isus[0] == NULL); } mpic->isu_shift = 1 + __ilog2(mpic->isu_size - 1); @@ -879,7 +979,8 @@ void __init mpic_assign_isu(struct mpic *mpic, unsigned int isu_num, BUG_ON(isu_num >= MPIC_MAX_ISU); - mpic->isus[isu_num] = ioremap(phys_addr, MPIC_IRQ_STRIDE * mpic->isu_size); + mpic->isus[isu_num] = ioremap(phys_addr, + MPIC_INFO(IRQ_STRIDE) * mpic->isu_size); if ((isu_first + mpic->isu_size) > mpic->num_sources) mpic->num_sources = isu_first + mpic->isu_size; } @@ -904,14 +1005,16 @@ void __init mpic_init(struct mpic *mpic) printk(KERN_INFO "mpic: Initializing for %d sources\n", mpic->num_sources); /* Set current processor priority to max */ - mpic_cpu_write(MPIC_CPU_CURRENT_TASK_PRI, 0xf); + mpic_cpu_write(MPIC_INFO(CPU_CURRENT_TASK_PRI), 0xf); /* Initialize timers: just disable them all */ for (i = 0; i < 4; i++) { mpic_write(mpic->tmregs, - i * MPIC_TIMER_STRIDE + MPIC_TIMER_DESTINATION, 0); + i * MPIC_INFO(TIMER_STRIDE) + + MPIC_INFO(TIMER_DESTINATION), 0); mpic_write(mpic->tmregs, - i * MPIC_TIMER_STRIDE + MPIC_TIMER_VECTOR_PRI, + i * MPIC_INFO(TIMER_STRIDE) + + MPIC_INFO(TIMER_VECTOR_PRI), MPIC_VECPRI_MASK | (MPIC_VEC_TIMER_0 + i)); } @@ -940,21 +1043,22 @@ void __init mpic_init(struct mpic *mpic) (8 << MPIC_VECPRI_PRIORITY_SHIFT); /* init hw */ - mpic_irq_write(i, MPIC_IRQ_VECTOR_PRI, vecpri); - mpic_irq_write(i, MPIC_IRQ_DESTINATION, + mpic_irq_write(i, MPIC_INFO(IRQ_VECTOR_PRI), vecpri); + mpic_irq_write(i, MPIC_INFO(IRQ_DESTINATION), 1 << hard_smp_processor_id()); } /* Init spurrious vector */ - mpic_write(mpic->gregs, MPIC_GREG_SPURIOUS, MPIC_VEC_SPURRIOUS); + mpic_write(mpic->gregs, MPIC_INFO(GREG_SPURIOUS), MPIC_VEC_SPURRIOUS); - /* Disable 8259 passthrough */ - mpic_write(mpic->gregs, MPIC_GREG_GLOBAL_CONF_0, - mpic_read(mpic->gregs, MPIC_GREG_GLOBAL_CONF_0) - | MPIC_GREG_GCONF_8259_PTHROU_DIS); + /* Disable 8259 passthrough, if supported */ + if (!(mpic->flags & MPIC_NO_PTHROU_DIS)) + mpic_write(mpic->gregs, MPIC_INFO(GREG_GLOBAL_CONF_0), + mpic_read(mpic->gregs, MPIC_INFO(GREG_GLOBAL_CONF_0)) + | MPIC_GREG_GCONF_8259_PTHROU_DIS); /* Set current processor priority to 0 */ - mpic_cpu_write(MPIC_CPU_CURRENT_TASK_PRI, 0); + mpic_cpu_write(MPIC_INFO(CPU_CURRENT_TASK_PRI), 0); } void __init mpic_set_clk_ratio(struct mpic *mpic, u32 clock_ratio) @@ -997,9 +1101,9 @@ void mpic_irq_set_priority(unsigned int irq, unsigned int pri) mpic_ipi_write(src - MPIC_VEC_IPI_0, reg | (pri << MPIC_VECPRI_PRIORITY_SHIFT)); } else { - reg = mpic_irq_read(src, MPIC_IRQ_VECTOR_PRI) + reg = mpic_irq_read(src, MPIC_INFO(IRQ_VECTOR_PRI)) & ~MPIC_VECPRI_PRIORITY_MASK; - mpic_irq_write(src, MPIC_IRQ_VECTOR_PRI, + mpic_irq_write(src, MPIC_INFO(IRQ_VECTOR_PRI), reg | (pri << MPIC_VECPRI_PRIORITY_SHIFT)); } spin_unlock_irqrestore(&mpic_lock, flags); @@ -1017,7 +1121,7 @@ unsigned int mpic_irq_get_priority(unsigned int irq) if (is_ipi) reg = mpic_ipi_read(src = MPIC_VEC_IPI_0); else - reg = mpic_irq_read(src, MPIC_IRQ_VECTOR_PRI); + reg = mpic_irq_read(src, MPIC_INFO(IRQ_VECTOR_PRI)); spin_unlock_irqrestore(&mpic_lock, flags); return (reg & MPIC_VECPRI_PRIORITY_MASK) >> MPIC_VECPRI_PRIORITY_SHIFT; } @@ -1043,12 +1147,12 @@ void mpic_setup_this_cpu(void) */ if (distribute_irqs) { for (i = 0; i < mpic->num_sources ; i++) - mpic_irq_write(i, MPIC_IRQ_DESTINATION, - mpic_irq_read(i, MPIC_IRQ_DESTINATION) | msk); + mpic_irq_write(i, MPIC_INFO(IRQ_DESTINATION), + mpic_irq_read(i, MPIC_INFO(IRQ_DESTINATION)) | msk); } /* Set current processor priority to 0 */ - mpic_cpu_write(MPIC_CPU_CURRENT_TASK_PRI, 0); + mpic_cpu_write(MPIC_INFO(CPU_CURRENT_TASK_PRI), 0); spin_unlock_irqrestore(&mpic_lock, flags); #endif /* CONFIG_SMP */ @@ -1058,7 +1162,7 @@ int mpic_cpu_get_priority(void) { struct mpic *mpic = mpic_primary; - return mpic_cpu_read(MPIC_CPU_CURRENT_TASK_PRI); + return mpic_cpu_read(MPIC_INFO(CPU_CURRENT_TASK_PRI)); } void mpic_cpu_set_priority(int prio) @@ -1066,7 +1170,7 @@ void mpic_cpu_set_priority(int prio) struct mpic *mpic = mpic_primary; prio &= MPIC_CPU_TASKPRI_MASK; - mpic_cpu_write(MPIC_CPU_CURRENT_TASK_PRI, prio); + mpic_cpu_write(MPIC_INFO(CPU_CURRENT_TASK_PRI), prio); } /* @@ -1088,11 +1192,11 @@ void mpic_teardown_this_cpu(int secondary) /* let the mpic know we don't want intrs. */ for (i = 0; i < mpic->num_sources ; i++) - mpic_irq_write(i, MPIC_IRQ_DESTINATION, - mpic_irq_read(i, MPIC_IRQ_DESTINATION) & ~msk); + mpic_irq_write(i, MPIC_INFO(IRQ_DESTINATION), + mpic_irq_read(i, MPIC_INFO(IRQ_DESTINATION)) & ~msk); /* Set current processor priority to max */ - mpic_cpu_write(MPIC_CPU_CURRENT_TASK_PRI, 0xf); + mpic_cpu_write(MPIC_INFO(CPU_CURRENT_TASK_PRI), 0xf); spin_unlock_irqrestore(&mpic_lock, flags); } @@ -1108,7 +1212,8 @@ void mpic_send_ipi(unsigned int ipi_no, unsigned int cpu_mask) DBG("%s: send_ipi(ipi_no: %d)\n", mpic->name, ipi_no); #endif - mpic_cpu_write(MPIC_CPU_IPI_DISPATCH_0 + ipi_no * 0x10, + mpic_cpu_write(MPIC_INFO(CPU_IPI_DISPATCH_0) + + ipi_no * MPIC_INFO(CPU_IPI_DISPATCH_STRIDE), mpic_physmask(cpu_mask & cpus_addr(cpu_online_map)[0])); } @@ -1116,7 +1221,7 @@ unsigned int mpic_get_one_irq(struct mpic *mpic, struct pt_regs *regs) { u32 src; - src = mpic_cpu_read(MPIC_CPU_INTACK) & MPIC_VECPRI_VECTOR_MASK; + src = mpic_cpu_read(MPIC_INFO(CPU_INTACK)) & MPIC_INFO(VECPRI_VECTOR_MASK); #ifdef DEBUG_LOW DBG("%s: get_one_irq(): %d\n", mpic->name, src); #endif diff --git a/include/asm-powerpc/mpic.h b/include/asm-powerpc/mpic.h index eb241c99c45..a9f9604b9ef 100644 --- a/include/asm-powerpc/mpic.h +++ b/include/asm-powerpc/mpic.h @@ -41,6 +41,7 @@ #define MPIC_GREG_IPI_VECTOR_PRI_1 0x000b0 #define MPIC_GREG_IPI_VECTOR_PRI_2 0x000c0 #define MPIC_GREG_IPI_VECTOR_PRI_3 0x000d0 +#define MPIC_GREG_IPI_STRIDE 0x10 #define MPIC_GREG_SPURIOUS 0x000e0 #define MPIC_GREG_TIMER_FREQ 0x000f0 @@ -68,6 +69,7 @@ #define MPIC_CPU_IPI_DISPATCH_1 0x00050 #define MPIC_CPU_IPI_DISPATCH_2 0x00060 #define MPIC_CPU_IPI_DISPATCH_3 0x00070 +#define MPIC_CPU_IPI_DISPATCH_STRIDE 0x00010 #define MPIC_CPU_CURRENT_TASK_PRI 0x00080 #define MPIC_CPU_TASKPRI_MASK 0x0000000f #define MPIC_CPU_WHOAMI 0x00090 @@ -114,6 +116,103 @@ #define MPIC_VEC_TIMER_1 248 #define MPIC_VEC_TIMER_0 247 +/* + * Tsi108 implementation of MPIC has many differences from the original one + */ + +/* + * Global registers + */ + +#define TSI108_GREG_BASE 0x00000 +#define TSI108_GREG_FEATURE_0 0x00000 +#define TSI108_GREG_GLOBAL_CONF_0 0x00004 +#define TSI108_GREG_VENDOR_ID 0x0000c +#define TSI108_GREG_IPI_VECTOR_PRI_0 0x00204 /* Doorbell 0 */ +#define TSI108_GREG_IPI_STRIDE 0x0c +#define TSI108_GREG_SPURIOUS 0x00010 +#define TSI108_GREG_TIMER_FREQ 0x00014 + +/* + * Timer registers + */ +#define TSI108_TIMER_BASE 0x0030 +#define TSI108_TIMER_STRIDE 0x10 +#define TSI108_TIMER_CURRENT_CNT 0x00000 +#define TSI108_TIMER_BASE_CNT 0x00004 +#define TSI108_TIMER_VECTOR_PRI 0x00008 +#define TSI108_TIMER_DESTINATION 0x0000c + +/* + * Per-Processor registers + */ +#define TSI108_CPU_BASE 0x00300 +#define TSI108_CPU_STRIDE 0x00040 +#define TSI108_CPU_IPI_DISPATCH_0 0x00200 +#define TSI108_CPU_IPI_DISPATCH_STRIDE 0x00000 +#define TSI108_CPU_CURRENT_TASK_PRI 0x00000 +#define TSI108_CPU_WHOAMI 0xffffffff +#define TSI108_CPU_INTACK 0x00004 +#define TSI108_CPU_EOI 0x00008 + +/* + * Per-source registers + */ +#define TSI108_IRQ_BASE 0x00100 +#define TSI108_IRQ_STRIDE 0x00008 +#define TSI108_IRQ_VECTOR_PRI 0x00000 +#define TSI108_VECPRI_VECTOR_MASK 0x000000ff +#define TSI108_VECPRI_POLARITY_POSITIVE 0x01000000 +#define TSI108_VECPRI_POLARITY_NEGATIVE 0x00000000 +#define TSI108_VECPRI_SENSE_LEVEL 0x02000000 +#define TSI108_VECPRI_SENSE_EDGE 0x00000000 +#define TSI108_VECPRI_POLARITY_MASK 0x01000000 +#define TSI108_VECPRI_SENSE_MASK 0x02000000 +#define TSI108_IRQ_DESTINATION 0x00004 + +/* weird mpic register indices and mask bits in the HW info array */ +enum { + MPIC_IDX_GREG_BASE = 0, + MPIC_IDX_GREG_FEATURE_0, + MPIC_IDX_GREG_GLOBAL_CONF_0, + MPIC_IDX_GREG_VENDOR_ID, + MPIC_IDX_GREG_IPI_VECTOR_PRI_0, + MPIC_IDX_GREG_IPI_STRIDE, + MPIC_IDX_GREG_SPURIOUS, + MPIC_IDX_GREG_TIMER_FREQ, + + MPIC_IDX_TIMER_BASE, + MPIC_IDX_TIMER_STRIDE, + MPIC_IDX_TIMER_CURRENT_CNT, + MPIC_IDX_TIMER_BASE_CNT, + MPIC_IDX_TIMER_VECTOR_PRI, + MPIC_IDX_TIMER_DESTINATION, + + MPIC_IDX_CPU_BASE, + MPIC_IDX_CPU_STRIDE, + MPIC_IDX_CPU_IPI_DISPATCH_0, + MPIC_IDX_CPU_IPI_DISPATCH_STRIDE, + MPIC_IDX_CPU_CURRENT_TASK_PRI, + MPIC_IDX_CPU_WHOAMI, + MPIC_IDX_CPU_INTACK, + MPIC_IDX_CPU_EOI, + + MPIC_IDX_IRQ_BASE, + MPIC_IDX_IRQ_STRIDE, + MPIC_IDX_IRQ_VECTOR_PRI, + + MPIC_IDX_VECPRI_VECTOR_MASK, + MPIC_IDX_VECPRI_POLARITY_POSITIVE, + MPIC_IDX_VECPRI_POLARITY_NEGATIVE, + MPIC_IDX_VECPRI_SENSE_LEVEL, + MPIC_IDX_VECPRI_SENSE_EDGE, + MPIC_IDX_VECPRI_POLARITY_MASK, + MPIC_IDX_VECPRI_SENSE_MASK, + MPIC_IDX_IRQ_DESTINATION, + MPIC_IDX_END +}; + + #ifdef CONFIG_MPIC_BROKEN_U3 /* Fixup table entry */ struct mpic_irq_fixup @@ -171,15 +270,29 @@ struct mpic volatile u32 __iomem *cpuregs[MPIC_MAX_CPUS]; volatile u32 __iomem *isus[MPIC_MAX_ISU]; +#ifdef CONFIG_MPIC_WEIRD + /* Pointer to HW info array */ + u32 *hw_set; +#endif + /* link */ struct mpic *next; }; +/* + * MPIC flags (passed to mpic_alloc) + * + * The top 4 bits contain an MPIC bhw id that is used to index the + * register offsets and some masks when CONFIG_MPIC_WEIRD is set. + * Note setting any ID (leaving those bits to 0) means standard MPIC + */ + /* This is the primary controller, only that one has IPIs and * has afinity control. A non-primary MPIC always uses CPU0 * registers only */ #define MPIC_PRIMARY 0x00000001 + /* Set this for a big-endian MPIC */ #define MPIC_BIG_ENDIAN 0x00000002 /* Broken U3 MPIC */ @@ -188,6 +301,18 @@ struct mpic #define MPIC_BROKEN_IPI 0x00000008 /* MPIC wants a reset */ #define MPIC_WANTS_RESET 0x00000010 +/* Spurious vector requires EOI */ +#define MPIC_SPV_EOI 0x00000020 +/* No passthrough disable */ +#define MPIC_NO_PTHROU_DIS 0x00000040 + +/* MPIC HW modification ID */ +#define MPIC_REGSET_MASK 0xf0000000 +#define MPIC_REGSET(val) (((val) & 0xf ) << 28) +#define MPIC_GET_REGSET(flags) (((flags) >> 28) & 0xf) + +#define MPIC_REGSET_STANDARD MPIC_REGSET(0) /* Original MPIC */ +#define MPIC_REGSET_TSI108 MPIC_REGSET(1) /* Tsi108/109 PIC */ /* Allocate the controller structure and setup the linux irq descs * for the range if interrupts passed in. No HW initialization is From 006b64de607f895de2ba1e21d3179cddf059128f Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Fri, 25 Aug 2006 14:46:23 +1000 Subject: [PATCH 037/115] [POWERPC] Make OF irq map code detect more error cases Device-tree bugs on js20 with some versions of SLOF were causing the interrupt for IDE to not be parsed correctly and fail to boot. This patch adds a bit more sanity checking to the parser to detect some of those errors and fail instead of returning bogus information. The powerpc PCI code can then trigger a fallback that works on those machines. Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Paul Mackerras --- arch/powerpc/kernel/pci_64.c | 11 +++++++++-- arch/powerpc/kernel/prom_parse.c | 17 +++++++++++++---- include/asm-powerpc/prom.h | 4 +++- 3 files changed, 25 insertions(+), 7 deletions(-) diff --git a/arch/powerpc/kernel/pci_64.c b/arch/powerpc/kernel/pci_64.c index 2fce7738e9e..138134c8c17 100644 --- a/arch/powerpc/kernel/pci_64.c +++ b/arch/powerpc/kernel/pci_64.c @@ -1289,6 +1289,9 @@ int pci_read_irq_line(struct pci_dev *pci_dev) DBG("Try to map irq for %s...\n", pci_name(pci_dev)); +#ifdef DEBUG + memset(&oirq, 0xff, sizeof(oirq)); +#endif /* Try to get a mapping from the device-tree */ if (of_irq_map_pci(pci_dev, &oirq)) { u8 line, pin; @@ -1314,8 +1317,9 @@ int pci_read_irq_line(struct pci_dev *pci_dev) if (virq != NO_IRQ) set_irq_type(virq, IRQ_TYPE_LEVEL_LOW); } else { - DBG(" -> got one, spec %d cells (0x%08x...) on %s\n", - oirq.size, oirq.specifier[0], oirq.controller->full_name); + DBG(" -> got one, spec %d cells (0x%08x 0x%08x...) on %s\n", + oirq.size, oirq.specifier[0], oirq.specifier[1], + oirq.controller->full_name); virq = irq_create_of_mapping(oirq.controller, oirq.specifier, oirq.size); @@ -1324,6 +1328,9 @@ int pci_read_irq_line(struct pci_dev *pci_dev) DBG(" -> failed to map !\n"); return -1; } + + DBG(" -> mapped to linux irq %d\n", virq); + pci_dev->irq = virq; pci_write_config_byte(pci_dev, PCI_INTERRUPT_LINE, virq); diff --git a/arch/powerpc/kernel/prom_parse.c b/arch/powerpc/kernel/prom_parse.c index 11052c212ad..a10825a5dfe 100644 --- a/arch/powerpc/kernel/prom_parse.c +++ b/arch/powerpc/kernel/prom_parse.c @@ -639,14 +639,17 @@ void of_irq_map_init(unsigned int flags) } -int of_irq_map_raw(struct device_node *parent, u32 *intspec, u32 *addr, - struct of_irq *out_irq) +int of_irq_map_raw(struct device_node *parent, u32 *intspec, u32 ointsize, + u32 *addr, struct of_irq *out_irq) { struct device_node *ipar, *tnode, *old = NULL, *newpar = NULL; u32 *tmp, *imap, *imask; u32 intsize = 1, addrsize, newintsize = 0, newaddrsize = 0; int imaplen, match, i; + DBG("of_irq_map_raw: par=%s,intspec=[0x%08x 0x%08x...],ointsize=%d\n", + parent->full_name, intspec[0], intspec[1], ointsize); + ipar = of_node_get(parent); /* First get the #interrupt-cells property of the current cursor @@ -670,6 +673,9 @@ int of_irq_map_raw(struct device_node *parent, u32 *intspec, u32 *addr, DBG("of_irq_map_raw: ipar=%s, size=%d\n", ipar->full_name, intsize); + if (ointsize != intsize) + return -EINVAL; + /* Look for this #address-cells. We have to implement the old linux * trick of looking for the parent here as some device-trees rely on it */ @@ -875,12 +881,15 @@ int of_irq_map_one(struct device_node *device, int index, struct of_irq *out_irq } intsize = *tmp; + DBG(" intsize=%d intlen=%d\n", intsize, intlen); + /* Check index */ if ((index + 1) * intsize > intlen) return -EINVAL; /* Get new specifier and map it */ - res = of_irq_map_raw(p, intspec + index * intsize, addr, out_irq); + res = of_irq_map_raw(p, intspec + index * intsize, intsize, + addr, out_irq); of_node_put(p); return res; } @@ -965,7 +974,7 @@ int of_irq_map_pci(struct pci_dev *pdev, struct of_irq *out_irq) laddr[0] = (pdev->bus->number << 16) | (pdev->devfn << 8); laddr[1] = laddr[2] = 0; - return of_irq_map_raw(ppnode, &lspec, laddr, out_irq); + return of_irq_map_raw(ppnode, &lspec, 1, laddr, out_irq); } EXPORT_SYMBOL_GPL(of_irq_map_pci); #endif /* CONFIG_PCI */ diff --git a/include/asm-powerpc/prom.h b/include/asm-powerpc/prom.h index b095a285c84..d0fa1b9aed3 100644 --- a/include/asm-powerpc/prom.h +++ b/include/asm-powerpc/prom.h @@ -276,6 +276,7 @@ extern void of_irq_map_init(unsigned int flags); * of_irq_map_raw - Low level interrupt tree parsing * @parent: the device interrupt parent * @intspec: interrupt specifier ("interrupts" property of the device) + * @ointsize: size of the passed in interrupt specifier * @addr: address specifier (start of "reg" property of the device) * @out_irq: structure of_irq filled by this function * @@ -288,7 +289,8 @@ extern void of_irq_map_init(unsigned int flags); * */ -extern int of_irq_map_raw(struct device_node *parent, u32 *intspec, u32 *addr, +extern int of_irq_map_raw(struct device_node *parent, u32 *intspec, + u32 ointsize, u32 *addr, struct of_irq *out_irq); From 11e9ed43ca8a741c2858c33d12120cf8817d3bff Mon Sep 17 00:00:00 2001 From: Will Schmidt Date: Fri, 25 Aug 2006 15:46:59 -0500 Subject: [PATCH 038/115] [POWERPC] Fix up ibm_architecture_vec definition This problem was noticed by one of the Phyp firmware folks. Our ibm,client-architecture-support call was failing. This corrects the vector length parameters being passed in. Signed-off-by: Will Schmidt Signed-off-by: Paul Mackerras --- arch/powerpc/kernel/prom_init.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/arch/powerpc/kernel/prom_init.c b/arch/powerpc/kernel/prom_init.c index 462bced40c1..4394e545f9f 100644 --- a/arch/powerpc/kernel/prom_init.c +++ b/arch/powerpc/kernel/prom_init.c @@ -646,13 +646,13 @@ static unsigned char ibm_architecture_vec[] = { 5 - 1, /* 5 option vectors */ /* option vector 1: processor architectures supported */ - 3 - 1, /* length */ + 3 - 2, /* length */ 0, /* don't ignore, don't halt */ OV1_PPC_2_00 | OV1_PPC_2_01 | OV1_PPC_2_02 | OV1_PPC_2_03 | OV1_PPC_2_04 | OV1_PPC_2_05, /* option vector 2: Open Firmware options supported */ - 34 - 1, /* length */ + 34 - 2, /* length */ OV2_REAL_MODE, 0, 0, W(0xffffffff), /* real_base */ @@ -666,16 +666,16 @@ static unsigned char ibm_architecture_vec[] = { 48, /* max log_2(hash table size) */ /* option vector 3: processor options supported */ - 3 - 1, /* length */ + 3 - 2, /* length */ 0, /* don't ignore, don't halt */ OV3_FP | OV3_VMX, /* option vector 4: IBM PAPR implementation */ - 2 - 1, /* length */ + 2 - 2, /* length */ 0, /* don't halt */ /* option vector 5: PAPR/OF options */ - 3 - 1, /* length */ + 3 - 2, /* length */ 0, /* don't ignore, don't halt */ OV5_LPAR | OV5_SPLPAR | OV5_LARGE_PAGES, }; From fea23bfefb4e98efd3c36f00ccc0b60281dc99da Mon Sep 17 00:00:00 2001 From: Paul Mackerras Date: Wed, 30 Aug 2006 14:45:35 +1000 Subject: [PATCH 039/115] [POWERPC] Restore copyright notice in arch/powerpc/kernel/fpu.S This code got moved from head.S but the copyright notice on head.S didn't get transferred with it. Noticed by Cort Dougan . Signed-off-by: Paul Mackerras --- arch/powerpc/kernel/fpu.S | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/arch/powerpc/kernel/fpu.S b/arch/powerpc/kernel/fpu.S index 7e2c9fe44ac..821e152e093 100644 --- a/arch/powerpc/kernel/fpu.S +++ b/arch/powerpc/kernel/fpu.S @@ -2,6 +2,11 @@ * FPU support code, moved here from head.S so that it can be used * by chips which use other head-whatever.S files. * + * Copyright (C) 1995-1996 Gary Thomas (gdt@linuxppc.org) + * Copyright (C) 1996 Cort Dougan + * Copyright (C) 1996 Paul Mackerras. + * Copyright (C) 1997 Dan Malek (dmalek@jlc.net). + * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License * as published by the Free Software Foundation; either version From e0d872d536bb93335d5905b09fe374a163486d43 Mon Sep 17 00:00:00 2001 From: Paul Mackerras Date: Wed, 30 Aug 2006 15:55:32 +1000 Subject: [PATCH 040/115] [POWERPC] Fix problem with time not advancing on 32-bit platforms This fixes a problem introduced in 5db9fa9593e2ff69f2b95f9d59229dc4faaa564d. The last_jiffy per-cpu variable is only 32 bits on 32-bit machines, but it was being compared with a 64-bit quantity (tb_next_jiffy), which resulted in time not advancing. This fixes it by changing last_jiffy to be 64 bits on all platforms. With this, we no longer need tb_last_stamp as a 32-bit version of tb_last_jiffy, so this gets rid of tb_last_stamp and we just use tb_last_jiffy instead. This also fixes a bug when the boot cpu is not online, because using tb_last_stamp could have caused the wrong timebase origin value to be used when calculating the time of day. Signed-off-by: Paul Mackerras --- arch/powerpc/kernel/time.c | 25 ++++++++----------------- include/asm-powerpc/time.h | 4 ---- 2 files changed, 8 insertions(+), 21 deletions(-) diff --git a/arch/powerpc/kernel/time.c b/arch/powerpc/kernel/time.c index 18e59e43d2b..a124499e65d 100644 --- a/arch/powerpc/kernel/time.c +++ b/arch/powerpc/kernel/time.c @@ -125,15 +125,8 @@ static long timezone_offset; unsigned long ppc_proc_freq; unsigned long ppc_tb_freq; -u64 tb_last_jiffy __cacheline_aligned_in_smp; -unsigned long tb_last_stamp; - -/* - * Note that on ppc32 this only stores the bottom 32 bits of - * the timebase value, but that's enough to tell when a jiffy - * has passed. - */ -DEFINE_PER_CPU(unsigned long, last_jiffy); +static u64 tb_last_jiffy __cacheline_aligned_in_smp; +static DEFINE_PER_CPU(u64, last_jiffy); #ifdef CONFIG_VIRT_CPU_ACCOUNTING /* @@ -458,7 +451,7 @@ void do_gettimeofday(struct timeval *tv) do { seq = read_seqbegin_irqsave(&xtime_lock, flags); sec = xtime.tv_sec; - nsec = xtime.tv_nsec + tb_ticks_since(tb_last_stamp); + nsec = xtime.tv_nsec + tb_ticks_since(tb_last_jiffy); } while (read_seqretry_irqrestore(&xtime_lock, seq, flags)); usec = nsec / 1000; while (usec >= 1000000) { @@ -700,7 +693,6 @@ void timer_interrupt(struct pt_regs * regs) tb_next_jiffy = tb_last_jiffy + tb_ticks_per_jiffy; if (per_cpu(last_jiffy, cpu) >= tb_next_jiffy) { tb_last_jiffy = tb_next_jiffy; - tb_last_stamp = per_cpu(last_jiffy, cpu); do_timer(regs); timer_recalc_offset(tb_last_jiffy); timer_check_rtc(); @@ -749,7 +741,7 @@ void __init smp_space_timers(unsigned int max_cpus) int i; unsigned long half = tb_ticks_per_jiffy / 2; unsigned long offset = tb_ticks_per_jiffy / max_cpus; - unsigned long previous_tb = per_cpu(last_jiffy, boot_cpuid); + u64 previous_tb = per_cpu(last_jiffy, boot_cpuid); /* make sure tb > per_cpu(last_jiffy, cpu) for all cpus always */ previous_tb -= tb_ticks_per_jiffy; @@ -830,7 +822,7 @@ int do_settimeofday(struct timespec *tv) * and therefore the (jiffies - wall_jiffies) computation * has been removed. */ - tb_delta = tb_ticks_since(tb_last_stamp); + tb_delta = tb_ticks_since(tb_last_jiffy); tb_delta = mulhdu(tb_delta, do_gtod.varp->tb_to_xs); /* in xsec */ new_nsec -= SCALE_XSEC(tb_delta, 1000000000); @@ -950,8 +942,7 @@ void __init time_init(void) if (__USE_RTC()) { /* 601 processor: dec counts down by 128 every 128ns */ ppc_tb_freq = 1000000000; - tb_last_stamp = get_rtcl(); - tb_last_jiffy = tb_last_stamp; + tb_last_jiffy = get_rtcl(); } else { /* Normal PowerPC with timebase register */ ppc_md.calibrate_decr(); @@ -959,7 +950,7 @@ void __init time_init(void) ppc_tb_freq / 1000000, ppc_tb_freq % 1000000); printk(KERN_DEBUG "time_init: processor frequency = %lu.%.6lu MHz\n", ppc_proc_freq / 1000000, ppc_proc_freq % 1000000); - tb_last_stamp = tb_last_jiffy = get_tb(); + tb_last_jiffy = get_tb(); } tb_ticks_per_jiffy = ppc_tb_freq / HZ; @@ -1036,7 +1027,7 @@ void __init time_init(void) do_gtod.varp = &do_gtod.vars[0]; do_gtod.var_idx = 0; do_gtod.varp->tb_orig_stamp = tb_last_jiffy; - __get_cpu_var(last_jiffy) = tb_last_stamp; + __get_cpu_var(last_jiffy) = tb_last_jiffy; do_gtod.varp->stamp_xsec = (u64) xtime.tv_sec * XSEC_PER_SEC; do_gtod.tb_ticks_per_sec = tb_ticks_per_sec; do_gtod.varp->tb_to_xs = tb_to_xs; diff --git a/include/asm-powerpc/time.h b/include/asm-powerpc/time.h index dcde4410348..5785ac4737b 100644 --- a/include/asm-powerpc/time.h +++ b/include/asm-powerpc/time.h @@ -30,10 +30,6 @@ extern unsigned long tb_ticks_per_usec; extern unsigned long tb_ticks_per_sec; extern u64 tb_to_xs; extern unsigned tb_to_us; -extern unsigned long tb_last_stamp; -extern u64 tb_last_jiffy; - -DECLARE_PER_CPU(unsigned long, last_jiffy); struct rtc_time; extern void to_tm(int tim, struct rtc_time * tm); From 467c37801c453849a2fe243c3226476ee3985868 Mon Sep 17 00:00:00 2001 From: Paul Mackerras Date: Wed, 30 Aug 2006 16:10:47 +1000 Subject: [PATCH 041/115] [POWERPC] Fix irq enable/disable in smp_generic_take_timebase Eran Ben-Avi pointed out that the arch/ppc version of smp_generic_take_timebase disables interrupts on entry but exits without restoring them. However, both it and the arch/powerpc version have another problem, which is that they use local_irq_disable/enable rather than local_irq_save/restore, and they are called with interrupts disabled. This fixes both problems; it changes a return to a break in the arch/ppc version, and changes both versions to use local_irq_save/restore. Signed-off-by: Paul Mackerras --- arch/powerpc/kernel/smp-tbsync.c | 5 +++-- arch/ppc/kernel/smp-tbsync.c | 7 ++++--- 2 files changed, 7 insertions(+), 5 deletions(-) diff --git a/arch/powerpc/kernel/smp-tbsync.c b/arch/powerpc/kernel/smp-tbsync.c index f19e2e0e61e..de59c6c31a5 100644 --- a/arch/powerpc/kernel/smp-tbsync.c +++ b/arch/powerpc/kernel/smp-tbsync.c @@ -45,8 +45,9 @@ void __devinit smp_generic_take_timebase(void) { int cmd; u64 tb; + unsigned long flags; - local_irq_disable(); + local_irq_save(flags); while (!running) barrier(); rmb(); @@ -70,7 +71,7 @@ void __devinit smp_generic_take_timebase(void) set_tb(tb >> 32, tb & 0xfffffffful); enter_contest(tbsync->mark, -1); } - local_irq_enable(); + local_irq_restore(flags); } static int __devinit start_contest(int cmd, long offset, int num) diff --git a/arch/ppc/kernel/smp-tbsync.c b/arch/ppc/kernel/smp-tbsync.c index 1576758deba..d0cf3f86931 100644 --- a/arch/ppc/kernel/smp-tbsync.c +++ b/arch/ppc/kernel/smp-tbsync.c @@ -47,8 +47,9 @@ void __devinit smp_generic_take_timebase( void ) { int cmd, tbl, tbu; + unsigned long flags; - local_irq_disable(); + local_irq_save(flags); while( !running ) ; rmb(); @@ -64,7 +65,7 @@ smp_generic_take_timebase( void ) tbu = tbsync->tbu; tbsync->ack = 0; if( cmd == kExit ) - return; + break; if( cmd == kSetAndTest ) { while( tbsync->handshake ) @@ -77,7 +78,7 @@ smp_generic_take_timebase( void ) } enter_contest( tbsync->mark, -1 ); } - local_irq_enable(); + local_irq_restore(flags); } static int __devinit From d21b55d30b46c0b43821a2980e4998965fa37e33 Mon Sep 17 00:00:00 2001 From: Sergei Shtylyov Date: Mon, 28 Aug 2006 19:49:03 +0400 Subject: [PATCH 042/115] [SERIAL] Make uart_match_port() work with all memory mapped UARTs uart_match_port() always fails with UPIO_MEM32, UPIO_AU, and UPIO_TSI cases. Since they match to the memory mapped UARTs, they should be handled just like UPIO_MEM case. Signed-off-by: Sergei Shtylyov Signed-off-by: Russell King --- drivers/serial/serial_core.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/serial/serial_core.c b/drivers/serial/serial_core.c index 80ef7d48275..372e47f7d59 100644 --- a/drivers/serial/serial_core.c +++ b/drivers/serial/serial_core.c @@ -2377,6 +2377,9 @@ int uart_match_port(struct uart_port *port1, struct uart_port *port2) return (port1->iobase == port2->iobase) && (port1->hub6 == port2->hub6); case UPIO_MEM: + case UPIO_MEM32: + case UPIO_AU: + case UPIO_TSI: return (port1->mapbase == port2->mapbase); } return 0; From e9422e091531c5851da4ffb8ee6dcbc36dc5b7a9 Mon Sep 17 00:00:00 2001 From: Helge Deller Date: Tue, 29 Aug 2006 21:57:29 +0200 Subject: [PATCH 043/115] [SERIAL] 8250: constify some serial structs - some const- ification and usage of ARRAY_SIZE() in serial drivers Signed-off-by: Helge Deller Signed-off-by: Russell King --- drivers/serial/8250_pci.c | 17 ++++++++--------- 1 file changed, 8 insertions(+), 9 deletions(-) diff --git a/drivers/serial/8250_pci.c b/drivers/serial/8250_pci.c index cd1979daf2b..851e4839d6d 100644 --- a/drivers/serial/8250_pci.c +++ b/drivers/serial/8250_pci.c @@ -458,11 +458,11 @@ static int pci_siig_setup(struct serial_private *priv, * growing *huge*, we use this function to collapse some 70 entries * in the PCI table into one, for sanity's and compactness's sake. */ -static unsigned short timedia_single_port[] = { +static const unsigned short timedia_single_port[] = { 0x4025, 0x4027, 0x4028, 0x5025, 0x5027, 0 }; -static unsigned short timedia_dual_port[] = { +static const unsigned short timedia_dual_port[] = { 0x0002, 0x4036, 0x4037, 0x4038, 0x4078, 0x4079, 0x4085, 0x4088, 0x4089, 0x5037, 0x5078, 0x5079, 0x5085, 0x6079, 0x7079, 0x8079, 0x8137, 0x8138, 0x8237, 0x8238, 0x9079, @@ -470,35 +470,34 @@ static unsigned short timedia_dual_port[] = { 0xD079, 0 }; -static unsigned short timedia_quad_port[] = { +static const unsigned short timedia_quad_port[] = { 0x4055, 0x4056, 0x4095, 0x4096, 0x5056, 0x8156, 0x8157, 0x8256, 0x8257, 0x9056, 0x9156, 0x9157, 0x9158, 0x9159, 0x9256, 0x9257, 0xA056, 0xA157, 0xA158, 0xA159, 0xB056, 0xB157, 0 }; -static unsigned short timedia_eight_port[] = { +static const unsigned short timedia_eight_port[] = { 0x4065, 0x4066, 0x5065, 0x5066, 0x8166, 0x9066, 0x9166, 0x9167, 0x9168, 0xA066, 0xA167, 0xA168, 0 }; static const struct timedia_struct { int num; - unsigned short *ids; + const unsigned short *ids; } timedia_data[] = { { 1, timedia_single_port }, { 2, timedia_dual_port }, { 4, timedia_quad_port }, - { 8, timedia_eight_port }, - { 0, NULL } + { 8, timedia_eight_port } }; static int pci_timedia_init(struct pci_dev *dev) { - unsigned short *ids; + const unsigned short *ids; int i, j; - for (i = 0; timedia_data[i].num; i++) { + for (i = 0; i < ARRAY_SIZE(timedia_data); i++) { ids = timedia_data[i].ids; for (j = 0; ids[j]; j++) if (dev->subsystem_device == ids[j]) From e7498656b5e2e9e3806d263fecc90b2707d02093 Mon Sep 17 00:00:00 2001 From: Stephen Rothwell Date: Wed, 30 Aug 2006 17:11:34 +1000 Subject: [PATCH 044/115] [POWERPC] iseries: Define insw et al. so libata/ide will compile These are build fixes that enable (for example) libata and the ide code to actually build on iSeries. The associated hardware will never be supported on legacy iSeries, so the code paths don't actually need to work, but it is useful (especially for a combined kernel) if the code can build. Signed-off-by: Stephen Rothwell Signed-off-by: Paul Mackerras --- include/asm-powerpc/io.h | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/include/asm-powerpc/io.h b/include/asm-powerpc/io.h index a9496f34b04..36c4c34bf56 100644 --- a/include/asm-powerpc/io.h +++ b/include/asm-powerpc/io.h @@ -72,6 +72,9 @@ extern unsigned long pci_io_base; * Neither do the standard versions now, these are just here * for older code. */ +#define insb(port, buf, ns) _insb((u8 __iomem *)((port)+pci_io_base), (buf), (ns)) +#define insw(port, buf, ns) _insw_ns((u8 __iomem *)((port)+pci_io_base), (buf), (ns)) +#define insl(port, buf, nl) _insl_ns((u8 __iomem *)((port)+pci_io_base), (buf), (nl)) #define insw_ns(port, buf, ns) _insw_ns((u16 __iomem *)((port)+pci_io_base), (buf), (ns)) #define insl_ns(port, buf, nl) _insl_ns((u32 __iomem *)((port)+pci_io_base), (buf), (nl)) #else @@ -137,12 +140,12 @@ static inline void __raw_writeq(unsigned long v, volatile void __iomem *addr) #define insw_ns(port, buf, ns) eeh_insw_ns((port), (buf), (ns)) #define insl_ns(port, buf, nl) eeh_insl_ns((port), (buf), (nl)) +#endif + #define outsb(port, buf, ns) _outsb((u8 __iomem *)((port)+pci_io_base), (buf), (ns)) #define outsw(port, buf, ns) _outsw_ns((u16 __iomem *)((port)+pci_io_base), (buf), (ns)) #define outsl(port, buf, nl) _outsl_ns((u32 __iomem *)((port)+pci_io_base), (buf), (nl)) -#endif - #define readb_relaxed(addr) readb(addr) #define readw_relaxed(addr) readw(addr) #define readl_relaxed(addr) readl(addr) From af313e5a4f56b248767d667c3c2436dda767df77 Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Wed, 30 Aug 2006 14:33:30 +0200 Subject: [PATCH 045/115] [S390] broken copy_in_user function. The copy_in_user primitive does not work as advertised. If the source and target area are available copy_in_user copies one byte too much. If one of the memory areas is not available it does not copy as much data as it can, but up to 257 bytes less. Signed-off-by: Martin Schwidefsky --- arch/s390/lib/uaccess.S | 33 +++++++++++++++++---------------- arch/s390/lib/uaccess64.S | 37 +++++++++++++++++++------------------ 2 files changed, 36 insertions(+), 34 deletions(-) diff --git a/arch/s390/lib/uaccess.S b/arch/s390/lib/uaccess.S index 5d59e262504..837275284d9 100644 --- a/arch/s390/lib/uaccess.S +++ b/arch/s390/lib/uaccess.S @@ -88,30 +88,31 @@ __copy_to_user_asm: .globl __copy_in_user_asm # %r2 = from, %r3 = n, %r4 = to __copy_in_user_asm: + ahi %r3,-1 + jo 6f sacf 256 - bras 1,1f - mvc 0(1,%r4),0(%r2) -0: mvc 0(256,%r4),0(%r2) - la %r2,256(%r2) - la %r4,256(%r4) -1: ahi %r3,-256 - jnm 0b -2: ex %r3,0(%r1) - sacf 0 - slr %r2,%r2 - br 14 -3: mvc 0(1,%r4),0(%r2) + bras %r1,4f +0: ahi %r3,257 +1: mvc 0(1,%r4),0(%r2) la %r2,1(%r2) la %r4,1(%r4) ahi %r3,-1 + jnz 1b +2: lr %r2,%r3 + br %r14 +3: mvc 0(256,%r4),0(%r2) + la %r2,256(%r2) + la %r4,256(%r4) +4: ahi %r3,-256 jnm 3b -4: lr %r2,%r3 +5: ex %r3,4(%r1) sacf 0 +6: slr %r2,%r2 br %r14 .section __ex_table,"a" - .long 0b,3b - .long 2b,3b - .long 3b,4b + .long 1b,2b + .long 3b,0b + .long 5b,0b .previous .align 4 diff --git a/arch/s390/lib/uaccess64.S b/arch/s390/lib/uaccess64.S index 19b41a33c23..1f755be22f9 100644 --- a/arch/s390/lib/uaccess64.S +++ b/arch/s390/lib/uaccess64.S @@ -88,30 +88,31 @@ __copy_to_user_asm: .globl __copy_in_user_asm # %r2 = from, %r3 = n, %r4 = to __copy_in_user_asm: + aghi %r3,-1 + jo 6f sacf 256 - bras 1,1f - mvc 0(1,%r4),0(%r2) -0: mvc 0(256,%r4),0(%r2) - la %r2,256(%r2) - la %r4,256(%r4) -1: aghi %r3,-256 - jnm 0b -2: ex %r3,0(%r1) - sacf 0 - slgr %r2,%r2 - br 14 -3: mvc 0(1,%r4),0(%r2) + bras %r1,4f +0: aghi %r3,257 +1: mvc 0(1,%r4),0(%r2) la %r2,1(%r2) la %r4,1(%r4) aghi %r3,-1 - jnm 3b -4: lgr %r2,%r3 - sacf 0 + jnz 1b +2: lgr %r2,%r3 br %r14 +3: mvc 0(256,%r4),0(%r2) + la %r2,256(%r2) + la %r4,256(%r4) +4: aghi %r3,-256 + jnm 3b +5: ex %r3,4(%r1) + sacf 0 +6: slgr %r2,%r2 + br 14 .section __ex_table,"a" - .quad 0b,3b - .quad 2b,3b - .quad 3b,4b + .quad 1b,2b + .quad 3b,0b + .quad 5b,0b .previous .align 4 From 8f61701bdf536c7a80f0f614bac91c7883804c4c Mon Sep 17 00:00:00 2001 From: Horst Hummel Date: Wed, 30 Aug 2006 14:33:33 +0200 Subject: [PATCH 046/115] [S390] dasd: fix device shutdown process. Fix clear_IO handling (need to wait for interrupt) and introduced error-handling in shutdown processing. Signed-off-by: Horst Hummel Signed-off-by: Martin Schwidefsky --- drivers/s390/block/dasd.c | 192 ++++++++++++++++++++++---------- drivers/s390/block/dasd_genhd.c | 10 +- 2 files changed, 137 insertions(+), 65 deletions(-) diff --git a/drivers/s390/block/dasd.c b/drivers/s390/block/dasd.c index d8e9b95f0a1..25c1ef6dfd4 100644 --- a/drivers/s390/block/dasd.c +++ b/drivers/s390/block/dasd.c @@ -52,7 +52,7 @@ static void dasd_setup_queue(struct dasd_device * device); static void dasd_free_queue(struct dasd_device * device); static void dasd_flush_request_queue(struct dasd_device *); static void dasd_int_handler(struct ccw_device *, unsigned long, struct irb *); -static void dasd_flush_ccw_queue(struct dasd_device *, int); +static int dasd_flush_ccw_queue(struct dasd_device *, int); static void dasd_tasklet(struct dasd_device *); static void do_kick_device(void *data); @@ -60,6 +60,7 @@ static void do_kick_device(void *data); * SECTION: Operations on the device structure. */ static wait_queue_head_t dasd_init_waitq; +static wait_queue_head_t dasd_flush_wq; /* * Allocate memory for a new device structure. @@ -121,7 +122,7 @@ dasd_free_device(struct dasd_device *device) /* * Make a new device known to the system. */ -static inline int +static int dasd_state_new_to_known(struct dasd_device *device) { int rc; @@ -145,7 +146,7 @@ dasd_state_new_to_known(struct dasd_device *device) /* * Let the system forget about a device. */ -static inline void +static int dasd_state_known_to_new(struct dasd_device * device) { /* Disable extended error reporting for this device. */ @@ -163,12 +164,13 @@ dasd_state_known_to_new(struct dasd_device * device) /* Give up reference we took in dasd_state_new_to_known. */ dasd_put_device(device); + return 0; } /* * Request the irq line for the device. */ -static inline int +static int dasd_state_known_to_basic(struct dasd_device * device) { int rc; @@ -192,17 +194,23 @@ dasd_state_known_to_basic(struct dasd_device * device) /* * Release the irq line for the device. Terminate any running i/o. */ -static inline void +static int dasd_state_basic_to_known(struct dasd_device * device) { + int rc; + dasd_gendisk_free(device); - dasd_flush_ccw_queue(device, 1); + rc = dasd_flush_ccw_queue(device, 1); + if (rc) + return rc; + DBF_DEV_EVENT(DBF_EMERG, device, "%p debug area deleted", device); if (device->debug_area != NULL) { debug_unregister(device->debug_area); device->debug_area = NULL; } device->state = DASD_STATE_KNOWN; + return 0; } /* @@ -219,7 +227,7 @@ dasd_state_basic_to_known(struct dasd_device * device) * In case the analysis returns an error, the device setup is stopped * (a fake disk was already added to allow formatting). */ -static inline int +static int dasd_state_basic_to_ready(struct dasd_device * device) { int rc; @@ -247,25 +255,31 @@ dasd_state_basic_to_ready(struct dasd_device * device) * Forget format information. Check if the target level is basic * and if it is create fake disk for formatting. */ -static inline void +static int dasd_state_ready_to_basic(struct dasd_device * device) { - dasd_flush_ccw_queue(device, 0); + int rc; + + rc = dasd_flush_ccw_queue(device, 0); + if (rc) + return rc; dasd_destroy_partitions(device); dasd_flush_request_queue(device); device->blocks = 0; device->bp_block = 0; device->s2b_shift = 0; device->state = DASD_STATE_BASIC; + return 0; } /* * Back to basic. */ -static inline void +static int dasd_state_unfmt_to_basic(struct dasd_device * device) { device->state = DASD_STATE_BASIC; + return 0; } /* @@ -273,7 +287,7 @@ dasd_state_unfmt_to_basic(struct dasd_device * device) * the requeueing of requests from the linux request queue to the * ccw queue. */ -static inline int +static int dasd_state_ready_to_online(struct dasd_device * device) { device->state = DASD_STATE_ONLINE; @@ -284,16 +298,17 @@ dasd_state_ready_to_online(struct dasd_device * device) /* * Stop the requeueing of requests again. */ -static inline void +static int dasd_state_online_to_ready(struct dasd_device * device) { device->state = DASD_STATE_READY; + return 0; } /* * Device startup state changes. */ -static inline int +static int dasd_increase_state(struct dasd_device *device) { int rc; @@ -329,30 +344,37 @@ dasd_increase_state(struct dasd_device *device) /* * Device shutdown state changes. */ -static inline int +static int dasd_decrease_state(struct dasd_device *device) { + int rc; + + rc = 0; if (device->state == DASD_STATE_ONLINE && device->target <= DASD_STATE_READY) - dasd_state_online_to_ready(device); + rc = dasd_state_online_to_ready(device); - if (device->state == DASD_STATE_READY && + if (!rc && + device->state == DASD_STATE_READY && device->target <= DASD_STATE_BASIC) - dasd_state_ready_to_basic(device); + rc = dasd_state_ready_to_basic(device); - if (device->state == DASD_STATE_UNFMT && + if (!rc && + device->state == DASD_STATE_UNFMT && device->target <= DASD_STATE_BASIC) - dasd_state_unfmt_to_basic(device); + rc = dasd_state_unfmt_to_basic(device); - if (device->state == DASD_STATE_BASIC && + if (!rc && + device->state == DASD_STATE_BASIC && device->target <= DASD_STATE_KNOWN) - dasd_state_basic_to_known(device); + rc = dasd_state_basic_to_known(device); - if (device->state == DASD_STATE_KNOWN && + if (!rc && + device->state == DASD_STATE_KNOWN && device->target <= DASD_STATE_NEW) - dasd_state_known_to_new(device); + rc = dasd_state_known_to_new(device); - return 0; + return rc; } /* @@ -701,6 +723,7 @@ dasd_term_IO(struct dasd_ccw_req * cqr) cqr->retries--; cqr->status = DASD_CQR_CLEAR; cqr->stopclk = get_clock(); + cqr->starttime = 0; DBF_DEV_EVENT(DBF_DEBUG, device, "terminate cqr %p successful", cqr); @@ -978,6 +1001,7 @@ dasd_int_handler(struct ccw_device *cdev, unsigned long intparm, irb->scsw.fctl & SCSW_FCTL_CLEAR_FUNC) { cqr->status = DASD_CQR_QUEUED; dasd_clear_timer(device); + wake_up(&dasd_flush_wq); dasd_schedule_bh(device); return; } @@ -1241,6 +1265,10 @@ __dasd_check_expire(struct dasd_device * device) cqr = list_entry(device->ccw_queue.next, struct dasd_ccw_req, list); if (cqr->status == DASD_CQR_IN_IO && cqr->expires != 0) { if (time_after_eq(jiffies, cqr->expires + cqr->starttime)) { + DEV_MESSAGE(KERN_ERR, device, + "internal error - timeout (%is) expired " + "for cqr %p (%i retries left)", + (cqr->expires/HZ), cqr, cqr->retries); if (device->discipline->term_IO(cqr) != 0) /* Hmpf, try again in 1/10 sec */ dasd_set_timer(device, 10); @@ -1285,46 +1313,100 @@ __dasd_start_head(struct dasd_device * device) dasd_set_timer(device, 50); } +static inline int +_wait_for_clear(struct dasd_ccw_req *cqr) +{ + return (cqr->status == DASD_CQR_QUEUED); +} + /* - * Remove requests from the ccw queue. + * Remove all requests from the ccw queue (all = '1') or only block device + * requests in case all = '0'. + * Take care of the erp-chain (chained via cqr->refers) and remove either + * the whole erp-chain or none of the erp-requests. + * If a request is currently running, term_IO is called and the request + * is re-queued. Prior to removing the terminated request we need to wait + * for the clear-interrupt. + * In case termination is not possible we stop processing and just finishing + * the already moved requests. */ -static void +static int dasd_flush_ccw_queue(struct dasd_device * device, int all) { + struct dasd_ccw_req *cqr, *orig, *n; + int rc, i; + struct list_head flush_queue; - struct list_head *l, *n; - struct dasd_ccw_req *cqr; INIT_LIST_HEAD(&flush_queue); spin_lock_irq(get_ccwdev_lock(device->cdev)); - list_for_each_safe(l, n, &device->ccw_queue) { - cqr = list_entry(l, struct dasd_ccw_req, list); + rc = 0; +restart: + list_for_each_entry_safe(cqr, n, &device->ccw_queue, list) { + /* get original request of erp request-chain */ + for (orig = cqr; orig->refers != NULL; orig = orig->refers); + /* Flush all request or only block device requests? */ - if (all == 0 && cqr->callback == dasd_end_request_cb) + if (all == 0 && cqr->callback != dasd_end_request_cb && + orig->callback != dasd_end_request_cb) { continue; - if (cqr->status == DASD_CQR_IN_IO) - device->discipline->term_IO(cqr); - if (cqr->status != DASD_CQR_DONE || - cqr->status != DASD_CQR_FAILED) { - cqr->status = DASD_CQR_FAILED; + } + /* Check status and move request to flush_queue */ + switch (cqr->status) { + case DASD_CQR_IN_IO: + rc = device->discipline->term_IO(cqr); + if (rc) { + /* unable to terminate requeust */ + DEV_MESSAGE(KERN_ERR, device, + "dasd flush ccw_queue is unable " + " to terminate request %p", + cqr); + /* stop flush processing */ + goto finished; + } + break; + case DASD_CQR_QUEUED: + case DASD_CQR_ERROR: + /* set request to FAILED */ cqr->stopclk = get_clock(); + cqr->status = DASD_CQR_FAILED; + break; + default: /* do not touch the others */ + break; + } + /* Rechain request (including erp chain) */ + for (i = 0; cqr != NULL; cqr = cqr->refers, i++) { + cqr->endclk = get_clock(); + list_move_tail(&cqr->list, &flush_queue); + } + if (i > 1) + /* moved more than one request - need to restart */ + goto restart; + } + +finished: + spin_unlock_irq(get_ccwdev_lock(device->cdev)); + /* Now call the callback function of flushed requests */ +restart_cb: + list_for_each_entry_safe(cqr, n, &flush_queue, list) { + if (cqr->status == DASD_CQR_CLEAR) { + /* wait for clear interrupt! */ + wait_event(dasd_flush_wq, _wait_for_clear(cqr)); + cqr->status = DASD_CQR_FAILED; } /* Process finished ERP request. */ if (cqr->refers) { __dasd_process_erp(device, cqr); - continue; + /* restart list_for_xx loop since dasd_process_erp + * might remove multiple elements */ + goto restart_cb; } - /* Rechain request on device request queue */ + /* call the callback function */ cqr->endclk = get_clock(); - list_move_tail(&cqr->list, &flush_queue); - } - spin_unlock_irq(get_ccwdev_lock(device->cdev)); - /* Now call the callback function of flushed requests */ - list_for_each_safe(l, n, &flush_queue) { - cqr = list_entry(l, struct dasd_ccw_req, list); if (cqr->callback != NULL) (cqr->callback)(cqr, cqr->callback_data); } + return rc; } /* @@ -1510,10 +1592,8 @@ dasd_sleep_on_interruptible(struct dasd_ccw_req * cqr) if (device->discipline->term_IO) { cqr->retries = -1; device->discipline->term_IO(cqr); - /*nished = - * wait (non-interruptible) for final status - * because signal ist still pending - */ + /* wait (non-interruptible) for final status + * because signal ist still pending */ spin_unlock_irq(get_ccwdev_lock(device->cdev)); wait_event(wait_q, _wait_for_wakeup(cqr)); spin_lock_irq(get_ccwdev_lock(device->cdev)); @@ -1546,19 +1626,11 @@ static inline int _dasd_term_running_cqr(struct dasd_device *device) { struct dasd_ccw_req *cqr; - int rc; if (list_empty(&device->ccw_queue)) return 0; cqr = list_entry(device->ccw_queue.next, struct dasd_ccw_req, list); - rc = device->discipline->term_IO(cqr); - if (rc == 0) { - /* termination successful */ - cqr->status = DASD_CQR_QUEUED; - cqr->startclk = cqr->stopclk = 0; - cqr->starttime = 0; - } - return rc; + return device->discipline->term_IO(cqr); } int @@ -1726,10 +1798,7 @@ dasd_flush_request_queue(struct dasd_device * device) return; spin_lock_irq(&device->request_queue_lock); - while (!list_empty(&device->request_queue->queue_head)) { - req = elv_next_request(device->request_queue); - if (req == NULL) - break; + while ((req = elv_next_request(device->request_queue))) { blkdev_dequeue_request(req); dasd_end_request(req, 0); } @@ -2091,6 +2160,7 @@ dasd_init(void) int rc; init_waitqueue_head(&dasd_init_waitq); + init_waitqueue_head(&dasd_flush_wq); /* register 'common' DASD debug area, used for all DBF_XXX calls */ dasd_debug_area = debug_register("dasd", 1, 2, 8 * sizeof (long)); diff --git a/drivers/s390/block/dasd_genhd.c b/drivers/s390/block/dasd_genhd.c index 4c272b70f41..d163632101d 100644 --- a/drivers/s390/block/dasd_genhd.c +++ b/drivers/s390/block/dasd_genhd.c @@ -83,10 +83,12 @@ dasd_gendisk_alloc(struct dasd_device *device) void dasd_gendisk_free(struct dasd_device *device) { - del_gendisk(device->gdp); - device->gdp->queue = NULL; - put_disk(device->gdp); - device->gdp = NULL; + if (device->gdp) { + del_gendisk(device->gdp); + device->gdp->queue = NULL; + put_disk(device->gdp); + device->gdp = NULL; + } } /* From 292888c81e74115db5e5a4a838f730a7c3662982 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Wed, 30 Aug 2006 14:33:35 +0200 Subject: [PATCH 047/115] [S390] cio: kernel stack overflow. Use different kind of assignment to make sure gcc doesn't create code that creates temp variables on the stack, assigns values to it and copies the content of the whole temp variable to the destination. This reduces stack usage of e.g. ccwgroup_driver_register from 976 to 48 bytes instead. Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- drivers/s390/cio/ccwgroup.c | 14 +++++--------- drivers/s390/cio/chsc.c | 6 ++---- drivers/s390/cio/device.c | 19 +++++++------------ drivers/s390/cio/device_fsm.c | 20 ++++++++------------ 4 files changed, 22 insertions(+), 37 deletions(-) diff --git a/drivers/s390/cio/ccwgroup.c b/drivers/s390/cio/ccwgroup.c index 3cba6c9fab1..38954f5cd14 100644 --- a/drivers/s390/cio/ccwgroup.c +++ b/drivers/s390/cio/ccwgroup.c @@ -183,11 +183,9 @@ ccwgroup_create(struct device *root, gdev->creator_id = creator_id; gdev->count = argc; - gdev->dev = (struct device ) { - .bus = &ccwgroup_bus_type, - .parent = root, - .release = ccwgroup_release, - }; + gdev->dev.bus = &ccwgroup_bus_type; + gdev->dev.parent = root; + gdev->dev.release = ccwgroup_release; snprintf (gdev->dev.bus_id, BUS_ID_SIZE, "%s", gdev->cdev[0]->dev.bus_id); @@ -391,10 +389,8 @@ int ccwgroup_driver_register (struct ccwgroup_driver *cdriver) { /* register our new driver with the core */ - cdriver->driver = (struct device_driver) { - .bus = &ccwgroup_bus_type, - .name = cdriver->name, - }; + cdriver->driver.bus = &ccwgroup_bus_type; + cdriver->driver.name = cdriver->name; return driver_register(&cdriver->driver); } diff --git a/drivers/s390/cio/chsc.c b/drivers/s390/cio/chsc.c index 61ce3f1d522..6405e9bd6eb 100644 --- a/drivers/s390/cio/chsc.c +++ b/drivers/s390/cio/chsc.c @@ -1391,10 +1391,8 @@ new_channel_path(int chpid) /* fill in status, etc. */ chp->id = chpid; chp->state = 1; - chp->dev = (struct device) { - .parent = &css[0]->device, - .release = chp_release, - }; + chp->dev.parent = &css[0]->device; + chp->dev.release = chp_release; snprintf(chp->dev.bus_id, BUS_ID_SIZE, "chp0.%x", chpid); /* Obtain channel path description and fill it in. */ diff --git a/drivers/s390/cio/device.c b/drivers/s390/cio/device.c index 585fa04233c..646da564040 100644 --- a/drivers/s390/cio/device.c +++ b/drivers/s390/cio/device.c @@ -556,12 +556,11 @@ get_disc_ccwdev_by_devno(unsigned int devno, unsigned int ssid, struct ccw_device *sibling) { struct device *dev; - struct match_data data = { - .devno = devno, - .ssid = ssid, - .sibling = sibling, - }; + struct match_data data; + data.devno = devno; + data.ssid = ssid; + data.sibling = sibling; dev = bus_find_device(&ccw_bus_type, NULL, &data, match_devno); return dev ? to_ccwdev(dev) : NULL; @@ -835,10 +834,8 @@ io_subchannel_probe (struct subchannel *sch) return -ENOMEM; } atomic_set(&cdev->private->onoff, 0); - cdev->dev = (struct device) { - .parent = &sch->dev, - .release = ccw_device_release, - }; + cdev->dev.parent = &sch->dev; + cdev->dev.release = ccw_device_release; INIT_LIST_HEAD(&cdev->private->kick_work.entry); /* Do first half of device_register. */ device_initialize(&cdev->dev); @@ -977,9 +974,7 @@ ccw_device_console_enable (struct ccw_device *cdev, struct subchannel *sch) int rc; /* Initialize the ccw_device structure. */ - cdev->dev = (struct device) { - .parent = &sch->dev, - }; + cdev->dev.parent= &sch->dev; rc = io_subchannel_recog(cdev, sch); if (rc) return rc; diff --git a/drivers/s390/cio/device_fsm.c b/drivers/s390/cio/device_fsm.c index 6d91c2eb205..35e162ba6d5 100644 --- a/drivers/s390/cio/device_fsm.c +++ b/drivers/s390/cio/device_fsm.c @@ -267,12 +267,10 @@ ccw_device_recog_done(struct ccw_device *cdev, int state) notify = 1; } /* fill out sense information */ - cdev->id = (struct ccw_device_id) { - .cu_type = cdev->private->senseid.cu_type, - .cu_model = cdev->private->senseid.cu_model, - .dev_type = cdev->private->senseid.dev_type, - .dev_model = cdev->private->senseid.dev_model, - }; + cdev->id.cu_type = cdev->private->senseid.cu_type; + cdev->id.cu_model = cdev->private->senseid.cu_model; + cdev->id.dev_type = cdev->private->senseid.dev_type; + cdev->id.dev_model = cdev->private->senseid.dev_model; if (notify) { cdev->private->state = DEV_STATE_OFFLINE; if (same_dev) { @@ -566,12 +564,10 @@ ccw_device_verify_done(struct ccw_device *cdev, int err) /* Deliver fake irb to device driver, if needed. */ if (cdev->private->flags.fake_irb) { memset(&cdev->private->irb, 0, sizeof(struct irb)); - cdev->private->irb.scsw = (struct scsw) { - .cc = 1, - .fctl = SCSW_FCTL_START_FUNC, - .actl = SCSW_ACTL_START_PEND, - .stctl = SCSW_STCTL_STATUS_PEND, - }; + cdev->private->irb.scsw.cc = 1; + cdev->private->irb.scsw.fctl = SCSW_FCTL_START_FUNC; + cdev->private->irb.scsw.actl = SCSW_ACTL_START_PEND; + cdev->private->irb.scsw.stctl = SCSW_STCTL_STATUS_PEND; cdev->private->flags.fake_irb = 0; if (cdev->handler) cdev->handler(cdev, cdev->private->intparm, From 3b88508a31a77eb3487154922e1eff282dc1d863 Mon Sep 17 00:00:00 2001 From: Peter Oberparleiter Date: Wed, 30 Aug 2006 14:33:37 +0200 Subject: [PATCH 048/115] [S390] cio: no path after machine check. Devices enter no-path state after disabling a channel path via the SE even though another path has been reenabled at the SE. The devices are set into no-path state before triggering path verification even though other paths may have become available. To fix this trigger path verification before setting a device into no-path state. Signed-off-by: Peter Oberparleiter Signed-off-by: Martin Schwidefsky --- drivers/s390/cio/chsc.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/s390/cio/chsc.c b/drivers/s390/cio/chsc.c index 6405e9bd6eb..c28444af091 100644 --- a/drivers/s390/cio/chsc.c +++ b/drivers/s390/cio/chsc.c @@ -238,8 +238,6 @@ s390_subchannel_remove_chpid(struct device *dev, void *data) /* Check for single path devices. */ if (sch->schib.pmcw.pim == 0x80) goto out_unreg; - if (sch->vpm == mask) - goto out_unreg; if ((sch->schib.scsw.actl & SCSW_ACTL_DEVACT) && (sch->schib.scsw.actl & SCSW_ACTL_SCHACT) && @@ -258,6 +256,8 @@ s390_subchannel_remove_chpid(struct device *dev, void *data) /* trigger path verification. */ if (sch->driver && sch->driver->verify) sch->driver->verify(&sch->dev); + else if (sch->vpm == mask) + goto out_unreg; out_unlock: spin_unlock_irq(&sch->lock); return 0; From 7b7db1b59563aebe2f4d2ba850468afb2c87c82a Mon Sep 17 00:00:00 2001 From: Stefan Bader Date: Wed, 30 Aug 2006 14:33:39 +0200 Subject: [PATCH 049/115] [S390] cio: unsolicited interrupts during sense pgid. Calls to set a device online with path grouping may get stuck in some cases because certain device conditions where discarded after unsolicited interrupts. Check subchannel activity after unsolicited interrupts and retry the operation if the subchannel is idle. Signed-off-by: Stefan Bader Signed-off-by: Martin Schwidefsky --- drivers/s390/cio/device_pgid.c | 27 +++++++++++++++++++++------ 1 file changed, 21 insertions(+), 6 deletions(-) diff --git a/drivers/s390/cio/device_pgid.c b/drivers/s390/cio/device_pgid.c index 32610fd8868..1693a102dcf 100644 --- a/drivers/s390/cio/device_pgid.c +++ b/drivers/s390/cio/device_pgid.c @@ -23,6 +23,21 @@ #include "device.h" #include "ioasm.h" +/* + * Helper function called from interrupt context to decide whether an + * operation should be tried again. + */ +static int __ccw_device_should_retry(struct scsw *scsw) +{ + /* CC is only valid if start function bit is set. */ + if ((scsw->fctl & SCSW_FCTL_START_FUNC) && scsw->cc == 1) + return 1; + /* No more activity. For sense and set PGID we stubbornly try again. */ + if (!scsw->actl) + return 1; + return 0; +} + /* * Start Sense Path Group ID helper function. Used in ccw_device_recog * and ccw_device_sense_pgid. @@ -155,10 +170,10 @@ ccw_device_sense_pgid_irq(struct ccw_device *cdev, enum dev_event dev_event) int ret; irb = (struct irb *) __LC_IRB; - /* Retry sense pgid for cc=1. */ + if (irb->scsw.stctl == (SCSW_STCTL_STATUS_PEND | SCSW_STCTL_ALERT_STATUS)) { - if (irb->scsw.cc == 1) { + if (__ccw_device_should_retry(&irb->scsw)) { ret = __ccw_device_sense_pgid_start(cdev); if (ret && ret != -EBUSY) ccw_device_sense_pgid_done(cdev, ret); @@ -391,10 +406,10 @@ ccw_device_verify_irq(struct ccw_device *cdev, enum dev_event dev_event) int ret; irb = (struct irb *) __LC_IRB; - /* Retry set pgid for cc=1. */ + if (irb->scsw.stctl == (SCSW_STCTL_STATUS_PEND | SCSW_STCTL_ALERT_STATUS)) { - if (irb->scsw.cc == 1) + if (__ccw_device_should_retry(&irb->scsw)) __ccw_device_verify_start(cdev); return; } @@ -494,10 +509,10 @@ ccw_device_disband_irq(struct ccw_device *cdev, enum dev_event dev_event) int ret; irb = (struct irb *) __LC_IRB; - /* Retry set pgid for cc=1. */ + if (irb->scsw.stctl == (SCSW_STCTL_STATUS_PEND | SCSW_STCTL_ALERT_STATUS)) { - if (irb->scsw.cc == 1) + if (__ccw_device_should_retry(&irb->scsw)) __ccw_device_disband_start(cdev); return; } From 0355b3e039c621d15321fd0d5cf72d8bdb8f723d Mon Sep 17 00:00:00 2001 From: Daniel Jacobowitz Date: Wed, 30 Aug 2006 15:06:39 +0100 Subject: [PATCH 050/115] [ARM] 3750/3: Fix double VFP emulation for EABI kernels Patch from Daniel Jacobowitz vfp_put_double didn't work in a CONFIG_AEABI kernel. By swapping the arguments, we arrange for them to be in the same place regardless of ABI. I made the same change to vfp_put_float for consistency. Signed-off-by: Daniel Jacobowitz Signed-off-by: Russell King --- arch/arm/vfp/vfp.h | 10 ++-------- arch/arm/vfp/vfpdouble.c | 20 ++++++++++---------- arch/arm/vfp/vfphw.S | 10 +++++----- arch/arm/vfp/vfpsingle.c | 20 ++++++++++---------- 4 files changed, 27 insertions(+), 33 deletions(-) diff --git a/arch/arm/vfp/vfp.h b/arch/arm/vfp/vfp.h index 5fbdf81a8aa..96fdf30f6a3 100644 --- a/arch/arm/vfp/vfp.h +++ b/arch/arm/vfp/vfp.h @@ -156,7 +156,7 @@ struct vfp_single { }; extern s32 vfp_get_float(unsigned int reg); -extern void vfp_put_float(unsigned int reg, s32 val); +extern void vfp_put_float(s32 val, unsigned int reg); /* * VFP_SINGLE_MANTISSA_BITS - number of bits in the mantissa @@ -267,7 +267,7 @@ struct vfp_double { */ #define VFP_REG_ZERO 16 extern u64 vfp_get_double(unsigned int reg); -extern void vfp_put_double(unsigned int reg, u64 val); +extern void vfp_put_double(u64 val, unsigned int reg); #define VFP_DOUBLE_MANTISSA_BITS (52) #define VFP_DOUBLE_EXPONENT_BITS (11) @@ -341,12 +341,6 @@ static inline int vfp_double_type(struct vfp_double *s) u32 vfp_double_normaliseround(int dd, struct vfp_double *vd, u32 fpscr, u32 exceptions, const char *func); -/* - * System registers - */ -extern u32 vfp_get_sys(unsigned int reg); -extern void vfp_put_sys(unsigned int reg, u32 val); - u32 vfp_estimate_sqrt_significand(u32 exponent, u32 significand); /* diff --git a/arch/arm/vfp/vfpdouble.c b/arch/arm/vfp/vfpdouble.c index 04bd3425b29..add48e36c2d 100644 --- a/arch/arm/vfp/vfpdouble.c +++ b/arch/arm/vfp/vfpdouble.c @@ -195,7 +195,7 @@ u32 vfp_double_normaliseround(int dd, struct vfp_double *vd, u32 fpscr, u32 exce s64 d = vfp_double_pack(vd); pr_debug("VFP: %s: d(d%d)=%016llx exceptions=%08x\n", func, dd, d, exceptions); - vfp_put_double(dd, d); + vfp_put_double(d, dd); } return exceptions; } @@ -250,19 +250,19 @@ vfp_propagate_nan(struct vfp_double *vdd, struct vfp_double *vdn, */ static u32 vfp_double_fabs(int dd, int unused, int dm, u32 fpscr) { - vfp_put_double(dd, vfp_double_packed_abs(vfp_get_double(dm))); + vfp_put_double(vfp_double_packed_abs(vfp_get_double(dm)), dd); return 0; } static u32 vfp_double_fcpy(int dd, int unused, int dm, u32 fpscr) { - vfp_put_double(dd, vfp_get_double(dm)); + vfp_put_double(vfp_get_double(dm), dd); return 0; } static u32 vfp_double_fneg(int dd, int unused, int dm, u32 fpscr) { - vfp_put_double(dd, vfp_double_packed_negate(vfp_get_double(dm))); + vfp_put_double(vfp_double_packed_negate(vfp_get_double(dm)), dd); return 0; } @@ -287,7 +287,7 @@ static u32 vfp_double_fsqrt(int dd, int unused, int dm, u32 fpscr) vdp = &vfp_double_default_qnan; ret = FPSCR_IOC; } - vfp_put_double(dd, vfp_double_pack(vdp)); + vfp_put_double(vfp_double_pack(vdp), dd); return ret; } @@ -476,7 +476,7 @@ static u32 vfp_double_fcvts(int sd, int unused, int dm, u32 fpscr) return vfp_single_normaliseround(sd, &vsd, fpscr, exceptions, "fcvts"); pack_nan: - vfp_put_float(sd, vfp_single_pack(&vsd)); + vfp_put_float(vfp_single_pack(&vsd), sd); return exceptions; } @@ -573,7 +573,7 @@ static u32 vfp_double_ftoui(int sd, int unused, int dm, u32 fpscr) pr_debug("VFP: ftoui: d(s%d)=%08x exceptions=%08x\n", sd, d, exceptions); - vfp_put_float(sd, d); + vfp_put_float(d, sd); return exceptions; } @@ -648,7 +648,7 @@ static u32 vfp_double_ftosi(int sd, int unused, int dm, u32 fpscr) pr_debug("VFP: ftosi: d(s%d)=%08x exceptions=%08x\n", sd, d, exceptions); - vfp_put_float(sd, (s32)d); + vfp_put_float((s32)d, sd); return exceptions; } @@ -1084,7 +1084,7 @@ static u32 vfp_double_fdiv(int dd, int dn, int dm, u32 fpscr) vdn_nan: exceptions = vfp_propagate_nan(&vdd, &vdn, &vdm, fpscr); pack: - vfp_put_double(dd, vfp_double_pack(&vdd)); + vfp_put_double(vfp_double_pack(&vdd), dd); return exceptions; vdm_nan: @@ -1104,7 +1104,7 @@ static u32 vfp_double_fdiv(int dd, int dn, int dm, u32 fpscr) goto pack; invalid: - vfp_put_double(dd, vfp_double_pack(&vfp_double_default_qnan)); + vfp_put_double(vfp_double_pack(&vfp_double_default_qnan), dd); return FPSCR_IOC; } diff --git a/arch/arm/vfp/vfphw.S b/arch/arm/vfp/vfphw.S index eb683cd7716..e51e6679c40 100644 --- a/arch/arm/vfp/vfphw.S +++ b/arch/arm/vfp/vfphw.S @@ -178,12 +178,12 @@ vfp_get_float: .globl vfp_put_float vfp_put_float: - add pc, pc, r0, lsl #3 + add pc, pc, r1, lsl #3 mov r0, r0 .irp dr,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15 - mcr p10, 0, r1, c\dr, c0, 0 @ fmsr r0, s0 + mcr p10, 0, r0, c\dr, c0, 0 @ fmsr r0, s0 mov pc, lr - mcr p10, 0, r1, c\dr, c0, 4 @ fmsr r0, s1 + mcr p10, 0, r0, c\dr, c0, 4 @ fmsr r0, s1 mov pc, lr .endr @@ -203,9 +203,9 @@ vfp_get_double: .globl vfp_put_double vfp_put_double: - add pc, pc, r0, lsl #3 + add pc, pc, r2, lsl #3 mov r0, r0 .irp dr,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15 - fmdrr d\dr, r1, r2 + fmdrr d\dr, r0, r1 mov pc, lr .endr diff --git a/arch/arm/vfp/vfpsingle.c b/arch/arm/vfp/vfpsingle.c index 78d7cac5f36..8f6c179cafb 100644 --- a/arch/arm/vfp/vfpsingle.c +++ b/arch/arm/vfp/vfpsingle.c @@ -200,7 +200,7 @@ u32 vfp_single_normaliseround(int sd, struct vfp_single *vs, u32 fpscr, u32 exce s32 d = vfp_single_pack(vs); pr_debug("VFP: %s: d(s%d)=%08x exceptions=%08x\n", func, sd, d, exceptions); - vfp_put_float(sd, d); + vfp_put_float(d, sd); } return exceptions; @@ -257,19 +257,19 @@ vfp_propagate_nan(struct vfp_single *vsd, struct vfp_single *vsn, */ static u32 vfp_single_fabs(int sd, int unused, s32 m, u32 fpscr) { - vfp_put_float(sd, vfp_single_packed_abs(m)); + vfp_put_float(vfp_single_packed_abs(m), sd); return 0; } static u32 vfp_single_fcpy(int sd, int unused, s32 m, u32 fpscr) { - vfp_put_float(sd, m); + vfp_put_float(m, sd); return 0; } static u32 vfp_single_fneg(int sd, int unused, s32 m, u32 fpscr) { - vfp_put_float(sd, vfp_single_packed_negate(m)); + vfp_put_float(vfp_single_packed_negate(m), sd); return 0; } @@ -333,7 +333,7 @@ static u32 vfp_single_fsqrt(int sd, int unused, s32 m, u32 fpscr) vsp = &vfp_single_default_qnan; ret = FPSCR_IOC; } - vfp_put_float(sd, vfp_single_pack(vsp)); + vfp_put_float(vfp_single_pack(vsp), sd); return ret; } @@ -517,7 +517,7 @@ static u32 vfp_single_fcvtd(int dd, int unused, s32 m, u32 fpscr) return vfp_double_normaliseround(dd, &vdd, fpscr, exceptions, "fcvtd"); pack_nan: - vfp_put_double(dd, vfp_double_pack(&vdd)); + vfp_put_double(vfp_double_pack(&vdd), dd); return exceptions; } @@ -613,7 +613,7 @@ static u32 vfp_single_ftoui(int sd, int unused, s32 m, u32 fpscr) pr_debug("VFP: ftoui: d(s%d)=%08x exceptions=%08x\n", sd, d, exceptions); - vfp_put_float(sd, d); + vfp_put_float(d, sd); return exceptions; } @@ -692,7 +692,7 @@ static u32 vfp_single_ftosi(int sd, int unused, s32 m, u32 fpscr) pr_debug("VFP: ftosi: d(s%d)=%08x exceptions=%08x\n", sd, d, exceptions); - vfp_put_float(sd, (s32)d); + vfp_put_float((s32)d, sd); return exceptions; } @@ -1127,7 +1127,7 @@ static u32 vfp_single_fdiv(int sd, int sn, s32 m, u32 fpscr) vsn_nan: exceptions = vfp_propagate_nan(&vsd, &vsn, &vsm, fpscr); pack: - vfp_put_float(sd, vfp_single_pack(&vsd)); + vfp_put_float(vfp_single_pack(&vsd), sd); return exceptions; vsm_nan: @@ -1147,7 +1147,7 @@ static u32 vfp_single_fdiv(int sd, int sn, s32 m, u32 fpscr) goto pack; invalid: - vfp_put_float(sd, vfp_single_pack(&vfp_single_default_qnan)); + vfp_put_float(vfp_single_pack(&vfp_single_default_qnan), sd); return FPSCR_IOC; } From ddcf36511d213501a035f168334046ba30a61a11 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Wed, 30 Aug 2006 19:37:05 +0200 Subject: [PATCH 051/115] [PATCH] x86_64: Update defconfig Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds --- arch/x86_64/defconfig | 66 +++++++++++-------------------------------- 1 file changed, 16 insertions(+), 50 deletions(-) diff --git a/arch/x86_64/defconfig b/arch/x86_64/defconfig index 840d5d93d5c..5fb97071594 100644 --- a/arch/x86_64/defconfig +++ b/arch/x86_64/defconfig @@ -1,7 +1,7 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.18-rc2 -# Tue Jul 18 17:13:20 2006 +# Linux kernel version: 2.6.18-rc4 +# Thu Aug 24 21:05:55 2006 # CONFIG_X86_64=y CONFIG_64BIT=y @@ -201,7 +201,7 @@ CONFIG_ACPI_THERMAL=y CONFIG_ACPI_NUMA=y # CONFIG_ACPI_ASUS is not set # CONFIG_ACPI_IBM is not set -CONFIG_ACPI_TOSHIBA=y +# CONFIG_ACPI_TOSHIBA is not set CONFIG_ACPI_BLACKLIST_YEAR=0 # CONFIG_ACPI_DEBUG is not set CONFIG_ACPI_EC=y @@ -216,7 +216,7 @@ CONFIG_ACPI_CONTAINER=y # CONFIG_CPU_FREQ=y CONFIG_CPU_FREQ_TABLE=y -# CONFIG_CPU_FREQ_DEBUG is not set +CONFIG_CPU_FREQ_DEBUG=y CONFIG_CPU_FREQ_STAT=y # CONFIG_CPU_FREQ_STAT_DETAILS is not set CONFIG_CPU_FREQ_DEFAULT_GOV_PERFORMANCE=y @@ -495,8 +495,9 @@ CONFIG_SCSI=y CONFIG_BLK_DEV_SD=y # CONFIG_CHR_DEV_ST is not set # CONFIG_CHR_DEV_OSST is not set -# CONFIG_BLK_DEV_SR is not set -# CONFIG_CHR_DEV_SG is not set +CONFIG_BLK_DEV_SR=y +# CONFIG_BLK_DEV_SR_VENDOR is not set +CONFIG_CHR_DEV_SG=y # CONFIG_CHR_DEV_SCH is not set # @@ -512,7 +513,7 @@ CONFIG_SCSI_CONSTANTS=y CONFIG_SCSI_SPI_ATTRS=y CONFIG_SCSI_FC_ATTRS=y # CONFIG_SCSI_ISCSI_ATTRS is not set -# CONFIG_SCSI_SAS_ATTRS is not set +CONFIG_SCSI_SAS_ATTRS=y # # SCSI low-level drivers @@ -538,7 +539,7 @@ CONFIG_MEGARAID_MAILBOX=y CONFIG_MEGARAID_SAS=y CONFIG_SCSI_SATA=y CONFIG_SCSI_SATA_AHCI=y -# CONFIG_SCSI_SATA_SVW is not set +CONFIG_SCSI_SATA_SVW=y CONFIG_SCSI_ATA_PIIX=y # CONFIG_SCSI_SATA_MV is not set CONFIG_SCSI_SATA_NV=y @@ -589,7 +590,7 @@ CONFIG_BLK_DEV_DM=y CONFIG_FUSION=y CONFIG_FUSION_SPI=y # CONFIG_FUSION_FC is not set -# CONFIG_FUSION_SAS is not set +CONFIG_FUSION_SAS=y CONFIG_FUSION_MAX_SGE=128 # CONFIG_FUSION_CTL is not set @@ -675,7 +676,7 @@ CONFIG_NET_PCI=y # CONFIG_PCNET32 is not set # CONFIG_AMD8111_ETH is not set # CONFIG_ADAPTEC_STARFIRE is not set -# CONFIG_B44 is not set +CONFIG_B44=y CONFIG_FORCEDETH=y # CONFIG_DGRS is not set # CONFIG_EEPRO100 is not set @@ -712,7 +713,7 @@ CONFIG_E1000=y # CONFIG_SK98LIN is not set # CONFIG_VIA_VELOCITY is not set CONFIG_TIGON3=y -# CONFIG_BNX2 is not set +CONFIG_BNX2=y # # Ethernet (10000 Mbit) @@ -842,44 +843,7 @@ CONFIG_LEGACY_PTY_COUNT=256 # # Watchdog Cards # -CONFIG_WATCHDOG=y -# CONFIG_WATCHDOG_NOWAYOUT is not set - -# -# Watchdog Device Drivers -# -CONFIG_SOFT_WATCHDOG=y -# CONFIG_ACQUIRE_WDT is not set -# CONFIG_ADVANTECH_WDT is not set -# CONFIG_ALIM1535_WDT is not set -# CONFIG_ALIM7101_WDT is not set -# CONFIG_SC520_WDT is not set -# CONFIG_EUROTECH_WDT is not set -# CONFIG_IB700_WDT is not set -# CONFIG_IBMASR is not set -# CONFIG_WAFER_WDT is not set -# CONFIG_I6300ESB_WDT is not set -# CONFIG_I8XX_TCO is not set -# CONFIG_SC1200_WDT is not set -# CONFIG_60XX_WDT is not set -# CONFIG_SBC8360_WDT is not set -# CONFIG_CPU5_WDT is not set -# CONFIG_W83627HF_WDT is not set -# CONFIG_W83877F_WDT is not set -# CONFIG_W83977F_WDT is not set -# CONFIG_MACHZ_WDT is not set -# CONFIG_SBC_EPX_C3_WATCHDOG is not set - -# -# PCI-based Watchdog Cards -# -# CONFIG_PCIPCWATCHDOG is not set -# CONFIG_WDTPCI is not set - -# -# USB-based Watchdog Cards -# -# CONFIG_USBPCWATCHDOG is not set +# CONFIG_WATCHDOG is not set CONFIG_HW_RANDOM=y CONFIG_HW_RANDOM_INTEL=y CONFIG_HW_RANDOM_AMD=y @@ -1056,6 +1020,7 @@ CONFIG_VGACON_SOFT_SCROLLBACK=y CONFIG_VGACON_SOFT_SCROLLBACK_SIZE=256 CONFIG_VIDEO_SELECT=y CONFIG_DUMMY_CONSOLE=y +# CONFIG_BACKLIGHT_LCD_SUPPORT is not set # # Sound @@ -1301,7 +1266,7 @@ CONFIG_INOTIFY=y CONFIG_INOTIFY_USER=y # CONFIG_QUOTA is not set CONFIG_DNOTIFY=y -CONFIG_AUTOFS_FS=y +# CONFIG_AUTOFS_FS is not set CONFIG_AUTOFS4_FS=y # CONFIG_FUSE_FS is not set @@ -1494,4 +1459,5 @@ CONFIG_DEBUG_STACKOVERFLOW=y # CONFIG_CRC16 is not set CONFIG_CRC32=y # CONFIG_LIBCRC32C is not set +CONFIG_ZLIB_INFLATE=y CONFIG_PLIST=y From 11012d419cfc0e0f78ca356aca03674217910124 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Wed, 30 Aug 2006 19:37:06 +0200 Subject: [PATCH 052/115] [PATCH] x86: Revert e820 MCFG heuristics The check for the MCFG table being reserved in the e820 map was originally added to detect a broken BIOS in a preproduction Intel SDV. However it also breaks the Apple x86 Macs, which can't supply this properly, but need a working MCFG. With this patch they wouldn't use the MCFG and not work. After some discussion I think it's best to remove the heuristic again. It also failed on some other boxes (although it didn't cause much problems there because old style port access for PCI config space still works as fallback), but the preproduction SDVs can just use pci=nommcfg. Supporting production machines properly is more important. Edgar Hucek did all the debugging work. Cc: Arjan van de Ven Cc: Edgar Hucek Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds --- arch/i386/kernel/setup.c | 32 -------------------------------- arch/i386/pci/mmconfig.c | 9 --------- arch/x86_64/kernel/e820.c | 29 ----------------------------- arch/x86_64/pci/mmconfig.c | 9 --------- 4 files changed, 79 deletions(-) diff --git a/arch/i386/kernel/setup.c b/arch/i386/kernel/setup.c index f1682206d30..345ffb7d904 100644 --- a/arch/i386/kernel/setup.c +++ b/arch/i386/kernel/setup.c @@ -956,38 +956,6 @@ efi_memory_present_wrapper(unsigned long start, unsigned long end, void *arg) return 0; } - /* - * This function checks if the entire range is mapped with type. - * - * Note: this function only works correct if the e820 table is sorted and - * not-overlapping, which is the case - */ -int __init -e820_all_mapped(unsigned long s, unsigned long e, unsigned type) -{ - u64 start = s; - u64 end = e; - int i; - for (i = 0; i < e820.nr_map; i++) { - struct e820entry *ei = &e820.map[i]; - if (type && ei->type != type) - continue; - /* is the region (part) in overlap with the current region ?*/ - if (ei->addr >= end || ei->addr + ei->size <= start) - continue; - /* if the region is at the beginning of we move - * start to the end of the region since it's ok until there - */ - if (ei->addr <= start) - start = ei->addr + ei->size; - /* if start is now at or beyond end, we're done, full - * coverage */ - if (start >= end) - return 1; /* we're done */ - } - return 0; -} - /* * Find the highest page frame number we have available */ diff --git a/arch/i386/pci/mmconfig.c b/arch/i386/pci/mmconfig.c index 972180f738d..5effb2e663e 100644 --- a/arch/i386/pci/mmconfig.c +++ b/arch/i386/pci/mmconfig.c @@ -198,15 +198,6 @@ void __init pci_mmcfg_init(void) (pci_mmcfg_config[0].base_address == 0)) return; - if (!e820_all_mapped(pci_mmcfg_config[0].base_address, - pci_mmcfg_config[0].base_address + MMCONFIG_APER_MIN, - E820_RESERVED)) { - printk(KERN_ERR "PCI: BIOS Bug: MCFG area at %x is not E820-reserved\n", - pci_mmcfg_config[0].base_address); - printk(KERN_ERR "PCI: Not using MMCONFIG.\n"); - return; - } - printk(KERN_INFO "PCI: Using MMCONFIG\n"); raw_pci_ops = &pci_mmcfg; pci_probe = (pci_probe & ~PCI_PROBE_MASK) | PCI_PROBE_MMCONF; diff --git a/arch/x86_64/kernel/e820.c b/arch/x86_64/kernel/e820.c index e56c2adf57a..d81e3d6dd24 100644 --- a/arch/x86_64/kernel/e820.c +++ b/arch/x86_64/kernel/e820.c @@ -104,35 +104,6 @@ e820_any_mapped(unsigned long start, unsigned long end, unsigned type) return 0; } -/* - * This function checks if the entire range is mapped with type. - * - * Note: this function only works correct if the e820 table is sorted and - * not-overlapping, which is the case - */ -int __init e820_all_mapped(unsigned long start, unsigned long end, unsigned type) -{ - int i; - for (i = 0; i < e820.nr_map; i++) { - struct e820entry *ei = &e820.map[i]; - if (type && ei->type != type) - continue; - /* is the region (part) in overlap with the current region ?*/ - if (ei->addr >= end || ei->addr + ei->size <= start) - continue; - - /* if the region is at the beginning of we move - * start to the end of the region since it's ok until there - */ - if (ei->addr <= start) - start = ei->addr + ei->size; - /* if start is now at or beyond end, we're done, full coverage */ - if (start >= end) - return 1; /* we're done */ - } - return 0; -} - /* * Find a free area in a specific range. */ diff --git a/arch/x86_64/pci/mmconfig.c b/arch/x86_64/pci/mmconfig.c index 3c55c76c6fd..8a4a0f9263c 100644 --- a/arch/x86_64/pci/mmconfig.c +++ b/arch/x86_64/pci/mmconfig.c @@ -177,15 +177,6 @@ void __init pci_mmcfg_init(void) (pci_mmcfg_config[0].base_address == 0)) return; - if (!e820_all_mapped(pci_mmcfg_config[0].base_address, - pci_mmcfg_config[0].base_address + MMCONFIG_APER_MIN, - E820_RESERVED)) { - printk(KERN_ERR "PCI: BIOS Bug: MCFG area at %x is not E820-reserved\n", - pci_mmcfg_config[0].base_address); - printk(KERN_ERR "PCI: Not using MMCONFIG.\n"); - return; - } - /* RED-PEN i386 doesn't do _nocache right now */ pci_mmcfg_virt = kmalloc(sizeof(*pci_mmcfg_virt) * pci_mmcfg_config_num, GFP_KERNEL); if (pci_mmcfg_virt == NULL) { From c05991ed12fd71e539dd8de8f5663450cd0c934c Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Wed, 30 Aug 2006 19:37:08 +0200 Subject: [PATCH 053/115] [PATCH] x86_64: Add kernel thread stack frame termination for properly stopping stack unwinds. One open question: Should these added pushes perhaps be made conditional upon CONFIG_STACK_UNWIND or CONFIG_UNWIND_INFO? [AK: Not needed -- these are all very slow paths] Signed-off-by: Jan Beulich Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds --- arch/x86_64/kernel/entry.S | 3 +++ arch/x86_64/kernel/head.S | 1 + 2 files changed, 4 insertions(+) diff --git a/arch/x86_64/kernel/entry.S b/arch/x86_64/kernel/entry.S index 6f810424df4..aa8d8939abc 100644 --- a/arch/x86_64/kernel/entry.S +++ b/arch/x86_64/kernel/entry.S @@ -973,6 +973,8 @@ ENTRY(kernel_thread) ENDPROC(kernel_thread) child_rip: + pushq $0 # fake return address + CFI_STARTPROC /* * Here we are in the child and the registers are set as they were * at kernel_thread() invocation in the parent. @@ -983,6 +985,7 @@ child_rip: # exit xorl %edi, %edi call do_exit + CFI_ENDPROC ENDPROC(child_rip) /* diff --git a/arch/x86_64/kernel/head.S b/arch/x86_64/kernel/head.S index 6df05e6034f..c9739ca81d0 100644 --- a/arch/x86_64/kernel/head.S +++ b/arch/x86_64/kernel/head.S @@ -191,6 +191,7 @@ startup_64: * jump */ movq initial_code(%rip),%rax + pushq $0 # fake return address jmp *%rax /* SMP bootup changes these two */ From 29fe5f3bafb644c33269fb0f2a719d4809a07332 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Wed, 30 Aug 2006 19:37:09 +0200 Subject: [PATCH 054/115] [PATCH] i386: Add kernel thread stack frame termination for properly stopping stack unwinds. One open question: Should this added push perhaps be made conditional upon CONFIG_STACK_UNWIND or CONFIG_UNWIND_INFO? [AK: not needed, these are all very slow paths] Signed-off-by: Jan Beulich Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds --- arch/i386/kernel/head.S | 14 ++++---------- 1 file changed, 4 insertions(+), 10 deletions(-) diff --git a/arch/i386/kernel/head.S b/arch/i386/kernel/head.S index eb79aa2fa8b..a6b8bd89aa2 100644 --- a/arch/i386/kernel/head.S +++ b/arch/i386/kernel/head.S @@ -317,20 +317,14 @@ is386: movl $2,%ecx # set MP movl %eax,%gs lldt %ax cld # gcc2 wants the direction flag cleared at all times + pushl %eax # fake return address #ifdef CONFIG_SMP movb ready, %cl movb $1, ready - cmpb $0,%cl - je 1f # the first CPU calls start_kernel - # all other CPUs call initialize_secondary - call initialize_secondary - jmp L6 -1: + cmpb $0,%cl # the first CPU calls start_kernel + jne initialize_secondary # all other CPUs call initialize_secondary #endif /* CONFIG_SMP */ - call start_kernel -L6: - jmp L6 # main should never return here, but - # just in case, we know what happens. + jmp start_kernel /* * We depend on ET to be correct. This checks for 287/387. From 61171b8dbd36b0cc34d3813a59a8e4dc2984414d Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Wed, 30 Aug 2006 19:37:10 +0200 Subject: [PATCH 055/115] [PATCH] x86: fix x86 cpuid keys used in alternative_smp() By hard-coding the cpuid keys for alternative_smp() rather than using the symbolic constant it turned out that incorrect values were used on both i386 (0x68 instead of 0x69) and x86-64 (0x66 instead of 0x68). Signed-off-by: Jan Beulich Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds --- include/asm-i386/alternative.h | 2 +- include/asm-x86_64/alternative.h | 3 ++- 2 files changed, 3 insertions(+), 2 deletions(-) diff --git a/include/asm-i386/alternative.h b/include/asm-i386/alternative.h index 96adbabec74..bb3b6317c81 100644 --- a/include/asm-i386/alternative.h +++ b/include/asm-i386/alternative.h @@ -116,7 +116,7 @@ static inline void alternatives_smp_switch(int smp) {} " .align 4\n" \ " .long 661b\n" /* label */ \ " .long 663f\n" /* new instruction */ \ - " .byte 0x68\n" /* X86_FEATURE_UP */ \ + " .byte " __stringify(X86_FEATURE_UP) "\n" \ " .byte 662b-661b\n" /* sourcelen */ \ " .byte 664f-663f\n" /* replacementlen */ \ ".previous\n" \ diff --git a/include/asm-x86_64/alternative.h b/include/asm-x86_64/alternative.h index aa67bfd1b3c..709ad3f0d35 100644 --- a/include/asm-x86_64/alternative.h +++ b/include/asm-x86_64/alternative.h @@ -4,6 +4,7 @@ #ifdef __KERNEL__ #include +#include struct alt_instr { u8 *instr; /* original instruction */ @@ -130,7 +131,7 @@ static inline void alternatives_smp_switch(int smp) {} " .align 8\n" \ " .quad 661b\n" /* label */ \ " .quad 663f\n" /* new instruction */ \ - " .byte 0x66\n" /* X86_FEATURE_UP */ \ + " .byte " __stringify(X86_FEATURE_UP) "\n" \ " .byte 662b-661b\n" /* sourcelen */ \ " .byte 664f-663f\n" /* replacementlen */ \ ".previous\n" \ From ea424055b771a165c9abd3ae109255a3b825c745 Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Wed, 30 Aug 2006 19:37:11 +0200 Subject: [PATCH 056/115] [PATCH] x86: Make backtracer fallback logic more bullet-proof The unwinder fallback logic still had potential for falling through to the legacy stack trace code without printing an indication (at once serving as a separator) of this. Further, the stack pointer retrieval for the fallback should be as restrictive as possible (in order to avoid having the legacy stack tracer try to access invalid memory). The patch tightens that, but this could certainly be further improved. Also making the call_trace command line option now conditional upon CONFIG_STACK_UNWIND (as it's meaningless otherwise). Signed-off-by: Jan Beulich Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds --- arch/i386/kernel/traps.c | 27 ++++++++++++++++----------- arch/x86_64/kernel/traps.c | 28 +++++++++++++++++----------- include/asm-i386/unwind.h | 1 + include/asm-x86_64/unwind.h | 1 + 4 files changed, 35 insertions(+), 22 deletions(-) diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c index 82e0fd02af1..7e9edafffd8 100644 --- a/arch/i386/kernel/traps.c +++ b/arch/i386/kernel/traps.c @@ -92,7 +92,11 @@ asmlinkage void spurious_interrupt_bug(void); asmlinkage void machine_check(void); static int kstack_depth_to_print = 24; +#ifdef CONFIG_STACK_UNWIND static int call_trace = 1; +#else +#define call_trace (-1) +#endif ATOMIC_NOTIFIER_HEAD(i386die_chain); int register_die_notifier(struct notifier_block *nb) @@ -187,22 +191,21 @@ static void show_trace_log_lvl(struct task_struct *task, struct pt_regs *regs, if (unwind_init_blocked(&info, task) == 0) unw_ret = show_trace_unwind(&info, log_lvl); } - if (unw_ret > 0 && !arch_unw_user_mode(&info)) { -#ifdef CONFIG_STACK_UNWIND - print_symbol("DWARF2 unwinder stuck at %s\n", - UNW_PC(&info)); - if (call_trace == 1) { - printk("Leftover inexact backtrace:\n"); - if (UNW_SP(&info)) + if (unw_ret > 0) { + if (call_trace == 1 && !arch_unw_user_mode(&info)) { + print_symbol("DWARF2 unwinder stuck at %s\n", + UNW_PC(&info)); + if (UNW_SP(&info) >= PAGE_OFFSET) { + printk("Leftover inexact backtrace:\n"); stack = (void *)UNW_SP(&info); - } else if (call_trace > 1) + } else + printk("Full inexact backtrace again:\n"); + } else if (call_trace >= 1) return; else printk("Full inexact backtrace again:\n"); -#else + } else printk("Inexact backtrace:\n"); -#endif - } } if (task == current) { @@ -1241,6 +1244,7 @@ static int __init kstack_setup(char *s) } __setup("kstack=", kstack_setup); +#ifdef CONFIG_STACK_UNWIND static int __init call_trace_setup(char *s) { if (strcmp(s, "old") == 0) @@ -1254,3 +1258,4 @@ static int __init call_trace_setup(char *s) return 1; } __setup("call_trace=", call_trace_setup); +#endif diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c index 14052f08981..5e00af54af6 100644 --- a/arch/x86_64/kernel/traps.c +++ b/arch/x86_64/kernel/traps.c @@ -107,7 +107,11 @@ static inline void preempt_conditional_cli(struct pt_regs *regs) } static int kstack_depth_to_print = 12; +#ifdef CONFIG_STACK_UNWIND static int call_trace = 1; +#else +#define call_trace (-1) +#endif #ifdef CONFIG_KALLSYMS # include @@ -274,21 +278,21 @@ void show_trace(struct task_struct *tsk, struct pt_regs *regs, unsigned long * s if (unwind_init_blocked(&info, tsk) == 0) unw_ret = show_trace_unwind(&info, NULL); } - if (unw_ret > 0 && !arch_unw_user_mode(&info)) { -#ifdef CONFIG_STACK_UNWIND - unsigned long rip = info.regs.rip; - print_symbol("DWARF2 unwinder stuck at %s\n", rip); - if (call_trace == 1) { - printk("Leftover inexact backtrace:\n"); - stack = (unsigned long *)info.regs.rsp; - } else if (call_trace > 1) + if (unw_ret > 0) { + if (call_trace == 1 && !arch_unw_user_mode(&info)) { + print_symbol("DWARF2 unwinder stuck at %s\n", + UNW_PC(&info)); + if ((long)UNW_SP(&info) < 0) { + printk("Leftover inexact backtrace:\n"); + stack = (unsigned long *)UNW_SP(&info); + } else + printk("Full inexact backtrace again:\n"); + } else if (call_trace >= 1) return; else printk("Full inexact backtrace again:\n"); -#else + } else printk("Inexact backtrace:\n"); -#endif - } } /* @@ -1120,6 +1124,7 @@ static int __init kstack_setup(char *s) } __setup("kstack=", kstack_setup); +#ifdef CONFIG_STACK_UNWIND static int __init call_trace_setup(char *s) { if (strcmp(s, "old") == 0) @@ -1133,3 +1138,4 @@ static int __init call_trace_setup(char *s) return 1; } __setup("call_trace=", call_trace_setup); +#endif diff --git a/include/asm-i386/unwind.h b/include/asm-i386/unwind.h index 69f0f1df672..4c1a0b96856 100644 --- a/include/asm-i386/unwind.h +++ b/include/asm-i386/unwind.h @@ -87,6 +87,7 @@ static inline int arch_unw_user_mode(const struct unwind_frame_info *info) #else #define UNW_PC(frame) ((void)(frame), 0) +#define UNW_SP(frame) ((void)(frame), 0) static inline int arch_unw_user_mode(const void *info) { diff --git a/include/asm-x86_64/unwind.h b/include/asm-x86_64/unwind.h index f3e7124effe..1f6e9bfb569 100644 --- a/include/asm-x86_64/unwind.h +++ b/include/asm-x86_64/unwind.h @@ -95,6 +95,7 @@ static inline int arch_unw_user_mode(const struct unwind_frame_info *info) #else #define UNW_PC(frame) ((void)(frame), 0) +#define UNW_SP(frame) ((void)(frame), 0) static inline int arch_unw_user_mode(const void *info) { From ceee88223047749ad683d397b19904c3dfb6adeb Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Wed, 30 Aug 2006 19:37:12 +0200 Subject: [PATCH 057/115] [PATCH] x86_64: Recover 1MB of kernel memory Noticed by Jan Beulich. When the kernel was moved from 1MB to 2MB in 2.6.17 the kernel reservation code wasn't adjusted and it still reserved starting with 1MB. This means 1MB always were lost. This patch fixes this by reserving only starting with _text. Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds --- arch/x86_64/kernel/e820.c | 6 +++++- arch/x86_64/kernel/setup.c | 6 ++---- 2 files changed, 7 insertions(+), 5 deletions(-) diff --git a/arch/x86_64/kernel/e820.c b/arch/x86_64/kernel/e820.c index d81e3d6dd24..764bf23c710 100644 --- a/arch/x86_64/kernel/e820.c +++ b/arch/x86_64/kernel/e820.c @@ -71,7 +71,11 @@ static inline int bad_addr(unsigned long *addrp, unsigned long size) #endif /* kernel code + 640k memory hole (later should not be needed, but be paranoid for now) */ - if (last >= 640*1024 && addr < __pa_symbol(&_end)) { + if (last >= 640*1024 && addr < 1024*1024) { + *addrp = 1024*1024; + return 1; + } + if (last >= __pa_symbol(&_text) && last < __pa_symbol(&_end)) { *addrp = __pa_symbol(&_end); return 1; } diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c index 8a099ff1f8b..34afad70482 100644 --- a/arch/x86_64/kernel/setup.c +++ b/arch/x86_64/kernel/setup.c @@ -521,8 +521,6 @@ static void discover_ebda(void) void __init setup_arch(char **cmdline_p) { - unsigned long kernel_end; - ROOT_DEV = old_decode_dev(ORIG_ROOT_DEV); screen_info = SCREEN_INFO; edid_info = EDID_INFO; @@ -596,8 +594,8 @@ void __init setup_arch(char **cmdline_p) (table_end - table_start) << PAGE_SHIFT); /* reserve kernel */ - kernel_end = round_up(__pa_symbol(&_end),PAGE_SIZE); - reserve_bootmem_generic(HIGH_MEMORY, kernel_end - HIGH_MEMORY); + reserve_bootmem_generic(__pa_symbol(&_text), + __pa_symbol(&_end) - __pa_symbol(&_text)); /* * reserve physical page 0 - it's a special BIOS page on many boxes, From 841be8ddf92578e5b481ed9f9abb85649fc13238 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Wed, 30 Aug 2006 19:37:13 +0200 Subject: [PATCH 058/115] [PATCH] x86_64: Remove alternative_smp The .fill causes miscompilations with some binutils version. Instead just patch the lock prefix in the lock constructs. That is the majority of the cost and should be good enough. Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds --- include/asm-x86_64/alternative.h | 20 -------------------- include/asm-x86_64/spinlock.h | 11 ++++------- 2 files changed, 4 insertions(+), 27 deletions(-) diff --git a/include/asm-x86_64/alternative.h b/include/asm-x86_64/alternative.h index 709ad3f0d35..a584826cc57 100644 --- a/include/asm-x86_64/alternative.h +++ b/include/asm-x86_64/alternative.h @@ -103,9 +103,6 @@ static inline void alternatives_smp_switch(int smp) {} /* * Alternative inline assembly for SMP. * - * alternative_smp() takes two versions (SMP first, UP second) and is - * for more complex stuff such as spinlocks. - * * The LOCK_PREFIX macro defined here replaces the LOCK and * LOCK_PREFIX macros used everywhere in the source tree. * @@ -125,21 +122,6 @@ static inline void alternatives_smp_switch(int smp) {} */ #ifdef CONFIG_SMP -#define alternative_smp(smpinstr, upinstr, args...) \ - asm volatile ("661:\n\t" smpinstr "\n662:\n" \ - ".section .smp_altinstructions,\"a\"\n" \ - " .align 8\n" \ - " .quad 661b\n" /* label */ \ - " .quad 663f\n" /* new instruction */ \ - " .byte " __stringify(X86_FEATURE_UP) "\n" \ - " .byte 662b-661b\n" /* sourcelen */ \ - " .byte 664f-663f\n" /* replacementlen */ \ - ".previous\n" \ - ".section .smp_altinstr_replacement,\"awx\"\n" \ - "663:\n\t" upinstr "\n" /* replacement */ \ - "664:\n\t.fill 662b-661b,1,0x42\n" /* space for original */ \ - ".previous" : args) - #define LOCK_PREFIX \ ".section .smp_locks,\"a\"\n" \ " .align 8\n" \ @@ -148,8 +130,6 @@ static inline void alternatives_smp_switch(int smp) {} "661:\n\tlock; " #else /* ! CONFIG_SMP */ -#define alternative_smp(smpinstr, upinstr, args...) \ - asm volatile (upinstr : args) #define LOCK_PREFIX "" #endif diff --git a/include/asm-x86_64/spinlock.h b/include/asm-x86_64/spinlock.h index 8d3421996f9..248a79f0eaf 100644 --- a/include/asm-x86_64/spinlock.h +++ b/include/asm-x86_64/spinlock.h @@ -21,7 +21,7 @@ #define __raw_spin_lock_string \ "\n1:\t" \ - "lock ; decl %0\n\t" \ + LOCK_PREFIX " ; decl %0\n\t" \ "js 2f\n" \ LOCK_SECTION_START("") \ "2:\t" \ @@ -40,10 +40,7 @@ static inline void __raw_spin_lock(raw_spinlock_t *lock) { - alternative_smp( - __raw_spin_lock_string, - __raw_spin_lock_string_up, - "=m" (lock->slock) : : "memory"); + asm volatile(__raw_spin_lock_string : "=m" (lock->slock) : : "memory"); } #define __raw_spin_lock_flags(lock, flags) __raw_spin_lock(lock) @@ -125,12 +122,12 @@ static inline int __raw_write_trylock(raw_rwlock_t *lock) static inline void __raw_read_unlock(raw_rwlock_t *rw) { - asm volatile("lock ; incl %0" :"=m" (rw->lock) : : "memory"); + asm volatile(LOCK_PREFIX " ; incl %0" :"=m" (rw->lock) : : "memory"); } static inline void __raw_write_unlock(raw_rwlock_t *rw) { - asm volatile("lock ; addl $" RW_LOCK_BIAS_STR ",%0" + asm volatile(LOCK_PREFIX " ; addl $" RW_LOCK_BIAS_STR ",%0" : "=m" (rw->lock) : : "memory"); } From 8c74932779fc6f61b4c30145863a17125c1a296c Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Wed, 30 Aug 2006 19:37:14 +0200 Subject: [PATCH 059/115] [PATCH] i386: Remove alternative_smp The .fill causes miscompilations with some binutils version. Instead just patch the lock prefix in the lock constructs. That is the majority of the cost and should be good enough. Cc: Gerd Hoffmann Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds --- include/asm-i386/alternative.h | 20 -------------------- include/asm-i386/rwlock.h | 14 ++++++-------- include/asm-i386/spinlock.h | 17 ++++------------- 3 files changed, 10 insertions(+), 41 deletions(-) diff --git a/include/asm-i386/alternative.h b/include/asm-i386/alternative.h index bb3b6317c81..b01a7ec409c 100644 --- a/include/asm-i386/alternative.h +++ b/include/asm-i386/alternative.h @@ -88,9 +88,6 @@ static inline void alternatives_smp_switch(int smp) {} /* * Alternative inline assembly for SMP. * - * alternative_smp() takes two versions (SMP first, UP second) and is - * for more complex stuff such as spinlocks. - * * The LOCK_PREFIX macro defined here replaces the LOCK and * LOCK_PREFIX macros used everywhere in the source tree. * @@ -110,21 +107,6 @@ static inline void alternatives_smp_switch(int smp) {} */ #ifdef CONFIG_SMP -#define alternative_smp(smpinstr, upinstr, args...) \ - asm volatile ("661:\n\t" smpinstr "\n662:\n" \ - ".section .smp_altinstructions,\"a\"\n" \ - " .align 4\n" \ - " .long 661b\n" /* label */ \ - " .long 663f\n" /* new instruction */ \ - " .byte " __stringify(X86_FEATURE_UP) "\n" \ - " .byte 662b-661b\n" /* sourcelen */ \ - " .byte 664f-663f\n" /* replacementlen */ \ - ".previous\n" \ - ".section .smp_altinstr_replacement,\"awx\"\n" \ - "663:\n\t" upinstr "\n" /* replacement */ \ - "664:\n\t.fill 662b-661b,1,0x42\n" /* space for original */ \ - ".previous" : args) - #define LOCK_PREFIX \ ".section .smp_locks,\"a\"\n" \ " .align 4\n" \ @@ -133,8 +115,6 @@ static inline void alternatives_smp_switch(int smp) {} "661:\n\tlock; " #else /* ! CONFIG_SMP */ -#define alternative_smp(smpinstr, upinstr, args...) \ - asm volatile (upinstr : args) #define LOCK_PREFIX "" #endif diff --git a/include/asm-i386/rwlock.h b/include/asm-i386/rwlock.h index 96b0bef2ea5..3ac1ba98b1b 100644 --- a/include/asm-i386/rwlock.h +++ b/include/asm-i386/rwlock.h @@ -21,22 +21,20 @@ #define RW_LOCK_BIAS_STR "0x01000000" #define __build_read_lock_ptr(rw, helper) \ - alternative_smp("lock; subl $1,(%0)\n\t" \ + asm volatile(LOCK_PREFIX " ; subl $1,(%0)\n\t" \ "jns 1f\n" \ "call " helper "\n\t" \ - "1:\n", \ - "subl $1,(%0)\n\t", \ + "1:\n" \ :"a" (rw) : "memory") #define __build_read_lock_const(rw, helper) \ - alternative_smp("lock; subl $1,%0\n\t" \ + asm volatile(LOCK_PREFIX " ; subl $1,%0\n\t" \ "jns 1f\n" \ "pushl %%eax\n\t" \ "leal %0,%%eax\n\t" \ "call " helper "\n\t" \ "popl %%eax\n\t" \ - "1:\n", \ - "subl $1,%0\n\t", \ + "1:\n" : \ "+m" (*(volatile int *)rw) : : "memory") #define __build_read_lock(rw, helper) do { \ @@ -47,7 +45,7 @@ } while (0) #define __build_write_lock_ptr(rw, helper) \ - alternative_smp("lock; subl $" RW_LOCK_BIAS_STR ",(%0)\n\t" \ + asm volatile(LOCK_PREFIX " ; subl $" RW_LOCK_BIAS_STR ",(%0)\n\t" \ "jz 1f\n" \ "call " helper "\n\t" \ "1:\n", \ @@ -55,7 +53,7 @@ :"a" (rw) : "memory") #define __build_write_lock_const(rw, helper) \ - alternative_smp("lock; subl $" RW_LOCK_BIAS_STR ",%0\n\t" \ + asm volatile(LOCK_PREFIX " ; subl $" RW_LOCK_BIAS_STR ",%0\n\t" \ "jz 1f\n" \ "pushl %%eax\n\t" \ "leal %0,%%eax\n\t" \ diff --git a/include/asm-i386/spinlock.h b/include/asm-i386/spinlock.h index d816c62a7a1..d1020363c41 100644 --- a/include/asm-i386/spinlock.h +++ b/include/asm-i386/spinlock.h @@ -22,7 +22,7 @@ #define __raw_spin_lock_string \ "\n1:\t" \ - "lock ; decb %0\n\t" \ + LOCK_PREFIX " ; decb %0\n\t" \ "jns 3f\n" \ "2:\t" \ "rep;nop\n\t" \ @@ -38,7 +38,7 @@ */ #define __raw_spin_lock_string_flags \ "\n1:\t" \ - "lock ; decb %0\n\t" \ + LOCK_PREFIX " ; decb %0\n\t" \ "jns 5f\n" \ "2:\t" \ "testl $0x200, %1\n\t" \ @@ -57,15 +57,9 @@ "jmp 4b\n" \ "5:\n\t" -#define __raw_spin_lock_string_up \ - "\n\tdecb %0" - static inline void __raw_spin_lock(raw_spinlock_t *lock) { - alternative_smp( - __raw_spin_lock_string, - __raw_spin_lock_string_up, - "+m" (lock->slock) : : "memory"); + asm(__raw_spin_lock_string : "+m" (lock->slock) : : "memory"); } /* @@ -76,10 +70,7 @@ static inline void __raw_spin_lock(raw_spinlock_t *lock) #ifndef CONFIG_PROVE_LOCKING static inline void __raw_spin_lock_flags(raw_spinlock_t *lock, unsigned long flags) { - alternative_smp( - __raw_spin_lock_string_flags, - __raw_spin_lock_string_up, - "+m" (lock->slock) : "r" (flags) : "memory"); + asm(__raw_spin_lock_string_flags : "+m" (lock->slock) : "r" (flags) : "memory"); } #endif From 40dd2d20f220eda1cd0da8ea3f0f9db8971ba237 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Wed, 30 Aug 2006 19:37:15 +0200 Subject: [PATCH 060/115] [PATCH] x86: Disable MMCONFIG on Intel SDV using DMI blacklist As a replacement for the earlier removal of the e820 MCFG check we blacklist the Intel SDV with the original BIOS bug that motivated that check. On those machines don't use MMCONFIG. This also adds a new pci=mmconf parameter to override the blacklist. Cc: Greg KH Cc: Arjan van de Ven Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds --- Documentation/kernel-parameters.txt | 2 ++ arch/i386/pci/common.c | 5 +++++ arch/i386/pci/mmconfig.c | 25 ++++++++++++++++++++++++- arch/i386/pci/pci.h | 3 ++- arch/x86_64/pci/mmconfig.c | 25 ++++++++++++++++++++++++- 5 files changed, 57 insertions(+), 3 deletions(-) diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index b50595a0550..7947cede871 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -1183,6 +1183,8 @@ running once the system is up. Mechanism 2. nommconf [IA-32,X86_64] Disable use of MMCONFIG for PCI Configuration + mmconf [IA-32,X86_64] Force MMCONFIG. This is useful + to override the builtin blacklist. nomsi [MSI] If the PCI_MSI kernel config parameter is enabled, this kernel boot option can be used to disable the use of MSI interrupts system-wide. diff --git a/arch/i386/pci/common.c b/arch/i386/pci/common.c index 0a362e3aeac..1220dd828ce 100644 --- a/arch/i386/pci/common.c +++ b/arch/i386/pci/common.c @@ -237,6 +237,11 @@ char * __devinit pcibios_setup(char *str) pci_probe &= ~PCI_PROBE_MMCONF; return NULL; } + /* override DMI blacklist */ + else if (!strcmp(str, "mmconf")) { + pci_probe |= PCI_PROBE_MMCONF_FORCE; + return NULL; + } #endif else if (!strcmp(str, "noacpi")) { acpi_noirq_set(); diff --git a/arch/i386/pci/mmconfig.c b/arch/i386/pci/mmconfig.c index 5effb2e663e..ef5a2faa7d8 100644 --- a/arch/i386/pci/mmconfig.c +++ b/arch/i386/pci/mmconfig.c @@ -12,6 +12,7 @@ #include #include #include +#include #include #include "pci.h" @@ -187,9 +188,31 @@ static __init void unreachable_devices(void) } } +static int disable_mcfg(struct dmi_system_id *d) +{ + printk("PCI: %s detected. Disabling MCFG.\n", d->ident); + pci_probe &= ~PCI_PROBE_MMCONF; + return 0; +} + +static struct dmi_system_id __initdata dmi_bad_mcfg[] = { + /* Has broken MCFG table that makes the system hang when used */ + { + .callback = disable_mcfg, + .ident = "Intel D3C5105 SDV", + .matches = { + DMI_MATCH(DMI_BIOS_VENDOR, "Intel"), + DMI_MATCH(DMI_BOARD_NAME, "D26928"), + }, + }, + {} +}; + void __init pci_mmcfg_init(void) { - if ((pci_probe & PCI_PROBE_MMCONF) == 0) + dmi_check_system(dmi_bad_mcfg); + + if ((pci_probe & (PCI_PROBE_MMCONF_FORCE|PCI_PROBE_MMCONF)) == 0) return; acpi_table_parse(ACPI_MCFG, acpi_parse_mcfg); diff --git a/arch/i386/pci/pci.h b/arch/i386/pci/pci.h index bf4e7933538..49a849b3a24 100644 --- a/arch/i386/pci/pci.h +++ b/arch/i386/pci/pci.h @@ -16,7 +16,8 @@ #define PCI_PROBE_CONF1 0x0002 #define PCI_PROBE_CONF2 0x0004 #define PCI_PROBE_MMCONF 0x0008 -#define PCI_PROBE_MASK 0x000f +#define PCI_PROBE_MMCONF_FORCE 0x0010 +#define PCI_PROBE_MASK 0x00ff #define PCI_NO_SORT 0x0100 #define PCI_BIOS_SORT 0x0200 diff --git a/arch/x86_64/pci/mmconfig.c b/arch/x86_64/pci/mmconfig.c index 8a4a0f9263c..2d48a7941d4 100644 --- a/arch/x86_64/pci/mmconfig.c +++ b/arch/x86_64/pci/mmconfig.c @@ -9,6 +9,7 @@ #include #include #include +#include #include #include "pci.h" @@ -164,11 +165,33 @@ static __init void unreachable_devices(void) } } +static int disable_mcfg(struct dmi_system_id *d) +{ + printk("PCI: %s detected. Disabling MCFG.\n", d->ident); + pci_probe &= ~PCI_PROBE_MMCONF; + return 0; +} + +static struct dmi_system_id __initdata dmi_bad_mcfg[] = { + /* Has broken MCFG table that makes the system hang when used */ + { + .callback = disable_mcfg, + .ident = "Intel D3C5105 SDV", + .matches = { + DMI_MATCH(DMI_BIOS_VENDOR, "Intel"), + DMI_MATCH(DMI_BOARD_NAME, "D26928"), + }, + }, + {} +}; + void __init pci_mmcfg_init(void) { int i; - if ((pci_probe & PCI_PROBE_MMCONF) == 0) + dmi_check_system(dmi_bad_mcfg); + + if ((pci_probe & (PCI_PROBE_MMCONF|PCI_PROBE_MMCONF_FORCE)) == 0) return; acpi_table_parse(ACPI_MCFG, acpi_parse_mcfg); From bbad0b669d33f3023cfe70ec79ba5ea487afca59 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Wed, 30 Aug 2006 19:37:16 +0200 Subject: [PATCH 061/115] [PATCH] x86_64: Remove __KERNEL__ ifdef around _syscall*() After all their only point is having them in user space. On x86-64 they don't even work in kernel space. Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds --- include/asm-x86_64/unistd.h | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) diff --git a/include/asm-x86_64/unistd.h b/include/asm-x86_64/unistd.h index 94387c915e5..2d89d309a2a 100644 --- a/include/asm-x86_64/unistd.h +++ b/include/asm-x86_64/unistd.h @@ -620,8 +620,6 @@ __SYSCALL(__NR_vmsplice, sys_vmsplice) #define __NR_move_pages 279 __SYSCALL(__NR_move_pages, sys_move_pages) -#ifdef __KERNEL__ - #define __NR_syscall_max __NR_move_pages #ifndef __NO_STUBS @@ -746,6 +744,8 @@ __syscall_return(type,__res); \ #else /* __KERNEL_SYSCALLS__ */ +#ifdef __KERNEL__ + #include #include @@ -838,9 +838,9 @@ asmlinkage long sys_rt_sigaction(int sig, struct sigaction __user *oact, size_t sigsetsize); -#endif /* __ASSEMBLY__ */ +#endif -#endif /* __NO_STUBS */ +#endif /* * "Conditional" syscalls @@ -850,5 +850,6 @@ asmlinkage long sys_rt_sigaction(int sig, */ #define cond_syscall(x) asm(".weak\t" #x "\n\t.set\t" #x ",sys_ni_syscall") -#endif /* __KERNEL__ */ +#endif + #endif From 266f0566761cf88906d634727b3d9fc2556f5cbd Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Wed, 30 Aug 2006 19:37:17 +0200 Subject: [PATCH 062/115] [PATCH] i386: Fix stack switching in do_IRQ There was a bogus hunk from the genirq merge that essentially broke stack switching for hard interrupts. Remove it since it isn't needed. Cc: Thomas Gleixner Cc: Ingo Molnar Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds --- arch/i386/kernel/irq.c | 5 ----- 1 file changed, 5 deletions(-) diff --git a/arch/i386/kernel/irq.c b/arch/i386/kernel/irq.c index 6cb529f60dc..5fe547cd8f9 100644 --- a/arch/i386/kernel/irq.c +++ b/arch/i386/kernel/irq.c @@ -82,10 +82,6 @@ fastcall unsigned int do_IRQ(struct pt_regs *regs) } #endif - if (!irq_desc[irq].handle_irq) { - __do_IRQ(irq, regs); - goto out_exit; - } #ifdef CONFIG_4KSTACKS curctx = (union irq_ctx *) current_thread_info(); @@ -125,7 +121,6 @@ fastcall unsigned int do_IRQ(struct pt_regs *regs) #endif __do_IRQ(irq, regs); -out_exit: irq_exit(); return 1; From 386dcafaacd212ef4a8aeed67a7db3ffbb44c7b2 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Wed, 30 Aug 2006 19:37:18 +0200 Subject: [PATCH 063/115] [PATCH] i386: Remove __KERNEL__ ifdef around _syscall*() After all their only point is having them in user space. Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds --- include/asm-i386/unistd.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/include/asm-i386/unistd.h b/include/asm-i386/unistd.h index fc1c8ddae14..d983b74e4d9 100644 --- a/include/asm-i386/unistd.h +++ b/include/asm-i386/unistd.h @@ -324,8 +324,6 @@ #define __NR_vmsplice 316 #define __NR_move_pages 317 -#ifdef __KERNEL__ - #define NR_syscalls 318 /* @@ -425,6 +423,8 @@ __asm__ volatile ("push %%ebp ; push %%ebx ; movl 4(%2),%%ebp ; " \ __syscall_return(type,__res); \ } +#ifdef __KERNEL__ + #define __ARCH_WANT_IPC_PARSE_VERSION #define __ARCH_WANT_OLD_READDIR #define __ARCH_WANT_OLD_STAT From 01ebb77b31149d847726a8847ad0d37631d7f049 Mon Sep 17 00:00:00 2001 From: Keith Owens Date: Wed, 30 Aug 2006 19:37:19 +0200 Subject: [PATCH 064/115] [PATCH] x86_64: Save original IST values for checking stack addresses The values in init_tss.ist[] can change when an IST event occurs. Save the original IST values for checking stack addresses when debugging or doing stack traces. Signed-off-by: Keith Owens Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds --- arch/x86_64/kernel/init_task.c | 5 +++++ arch/x86_64/kernel/setup64.c | 3 ++- arch/x86_64/kernel/traps.c | 2 +- include/asm-x86_64/processor.h | 6 ++++++ 4 files changed, 14 insertions(+), 2 deletions(-) diff --git a/arch/x86_64/kernel/init_task.c b/arch/x86_64/kernel/init_task.c index ce31d904d60..3dc5854ba21 100644 --- a/arch/x86_64/kernel/init_task.c +++ b/arch/x86_64/kernel/init_task.c @@ -46,4 +46,9 @@ EXPORT_SYMBOL(init_task); */ DEFINE_PER_CPU(struct tss_struct, init_tss) ____cacheline_internodealigned_in_smp = INIT_TSS; +/* Copies of the original ist values from the tss are only accessed during + * debugging, no special alignment required. + */ +DEFINE_PER_CPU(struct orig_ist, orig_ist); + #define ALIGN_TO_4K __attribute__((section(".data.init_task"))) diff --git a/arch/x86_64/kernel/setup64.c b/arch/x86_64/kernel/setup64.c index 6fe58a634b5..417de564456 100644 --- a/arch/x86_64/kernel/setup64.c +++ b/arch/x86_64/kernel/setup64.c @@ -189,6 +189,7 @@ void __cpuinit cpu_init (void) { int cpu = stack_smp_processor_id(); struct tss_struct *t = &per_cpu(init_tss, cpu); + struct orig_ist *orig_ist = &per_cpu(orig_ist, cpu); unsigned long v; char *estacks = NULL; struct task_struct *me; @@ -256,7 +257,7 @@ void __cpuinit cpu_init (void) estacks += EXCEPTION_STKSZ; break; } - t->ist[v] = (unsigned long)estacks; + orig_ist->ist[v] = t->ist[v] = (unsigned long)estacks; } t->io_bitmap_base = offsetof(struct tss_struct, io_bitmap); diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c index 5e00af54af6..b1249774d1e 100644 --- a/arch/x86_64/kernel/traps.c +++ b/arch/x86_64/kernel/traps.c @@ -178,7 +178,7 @@ static unsigned long *in_exception_stack(unsigned cpu, unsigned long stack, break; #endif default: - end = per_cpu(init_tss, cpu).ist[k]; + end = per_cpu(orig_ist, cpu).ist[k]; break; } /* diff --git a/include/asm-x86_64/processor.h b/include/asm-x86_64/processor.h index 3b3c1217fe6..de9c3147ee4 100644 --- a/include/asm-x86_64/processor.h +++ b/include/asm-x86_64/processor.h @@ -232,8 +232,14 @@ struct tss_struct { unsigned long io_bitmap[IO_BITMAP_LONGS + 1]; } __attribute__((packed)) ____cacheline_aligned; + extern struct cpuinfo_x86 boot_cpu_data; DECLARE_PER_CPU(struct tss_struct,init_tss); +/* Save the original ist values for checking stack pointers during debugging */ +struct orig_ist { + unsigned long ist[7]; +}; +DECLARE_PER_CPU(struct orig_ist, orig_ist); #ifdef CONFIG_X86_VSMP #define ARCH_MIN_TASKALIGN (1 << INTERNODE_CACHE_SHIFT) From ec0063b40a2291e095faf07fbeb94be92e525c54 Mon Sep 17 00:00:00 2001 From: Suleiman Souhlal Date: Wed, 30 Aug 2006 19:37:20 +0200 Subject: [PATCH 065/115] [PATCH] x86_64: Don't write out segments from vsyscall32 DSO if it is not mapped It's possible to get an invalid page fault in kernel mode when we try to write out segments from vsyscall32 when dumping core for a 32bit process if the vsyscall32 DSO is not mapped in its address space (which can happen if, for example, ulimit -v 100 is run). Signed-off-by: Suleiman Souhlal Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds --- arch/x86_64/ia32/ia32_binfmt.c | 57 ++++++++++++++++++---------------- 1 file changed, 31 insertions(+), 26 deletions(-) diff --git a/arch/x86_64/ia32/ia32_binfmt.c b/arch/x86_64/ia32/ia32_binfmt.c index a9dc0f3b5b5..2fd5a67fd43 100644 --- a/arch/x86_64/ia32/ia32_binfmt.c +++ b/arch/x86_64/ia32/ia32_binfmt.c @@ -73,39 +73,44 @@ typedef elf_greg_t elf_gregset_t[ELF_NGREG]; * Dumping its extra ELF program headers includes all the other information * a debugger needs to easily find how the vsyscall DSO was being used. */ -#define ELF_CORE_EXTRA_PHDRS (VSYSCALL32_EHDR->e_phnum) +#define ELF_CORE_EXTRA_PHDRS (find_vma(current->mm, VSYSCALL32_BASE) ? \ + (VSYSCALL32_EHDR->e_phnum) : 0) #define ELF_CORE_WRITE_EXTRA_PHDRS \ do { \ - const struct elf32_phdr *const vsyscall_phdrs = \ - (const struct elf32_phdr *) (VSYSCALL32_BASE \ - + VSYSCALL32_EHDR->e_phoff); \ - int i; \ - Elf32_Off ofs = 0; \ - for (i = 0; i < VSYSCALL32_EHDR->e_phnum; ++i) { \ - struct elf32_phdr phdr = vsyscall_phdrs[i]; \ - if (phdr.p_type == PT_LOAD) { \ - BUG_ON(ofs != 0); \ - ofs = phdr.p_offset = offset; \ - phdr.p_memsz = PAGE_ALIGN(phdr.p_memsz); \ - phdr.p_filesz = phdr.p_memsz; \ - offset += phdr.p_filesz; \ + if (find_vma(current->mm, VSYSCALL32_BASE)) { \ + const struct elf32_phdr *const vsyscall_phdrs = \ + (const struct elf32_phdr *) (VSYSCALL32_BASE \ + + VSYSCALL32_EHDR->e_phoff);\ + int i; \ + Elf32_Off ofs = 0; \ + for (i = 0; i < VSYSCALL32_EHDR->e_phnum; ++i) { \ + struct elf32_phdr phdr = vsyscall_phdrs[i]; \ + if (phdr.p_type == PT_LOAD) { \ + BUG_ON(ofs != 0); \ + ofs = phdr.p_offset = offset; \ + phdr.p_memsz = PAGE_ALIGN(phdr.p_memsz); \ + phdr.p_filesz = phdr.p_memsz; \ + offset += phdr.p_filesz; \ + } \ + else \ + phdr.p_offset += ofs; \ + phdr.p_paddr = 0; /* match other core phdrs */ \ + DUMP_WRITE(&phdr, sizeof(phdr)); \ } \ - else \ - phdr.p_offset += ofs; \ - phdr.p_paddr = 0; /* match other core phdrs */ \ - DUMP_WRITE(&phdr, sizeof(phdr)); \ } \ } while (0) #define ELF_CORE_WRITE_EXTRA_DATA \ do { \ - const struct elf32_phdr *const vsyscall_phdrs = \ - (const struct elf32_phdr *) (VSYSCALL32_BASE \ - + VSYSCALL32_EHDR->e_phoff); \ - int i; \ - for (i = 0; i < VSYSCALL32_EHDR->e_phnum; ++i) { \ - if (vsyscall_phdrs[i].p_type == PT_LOAD) \ - DUMP_WRITE((void *) (u64) vsyscall_phdrs[i].p_vaddr, \ - PAGE_ALIGN(vsyscall_phdrs[i].p_memsz)); \ + if (find_vma(current->mm, VSYSCALL32_BASE)) { \ + const struct elf32_phdr *const vsyscall_phdrs = \ + (const struct elf32_phdr *) (VSYSCALL32_BASE \ + + VSYSCALL32_EHDR->e_phoff); \ + int i; \ + for (i = 0; i < VSYSCALL32_EHDR->e_phnum; ++i) { \ + if (vsyscall_phdrs[i].p_type == PT_LOAD) \ + DUMP_WRITE((void *) (u64) vsyscall_phdrs[i].p_vaddr,\ + PAGE_ALIGN(vsyscall_phdrs[i].p_memsz)); \ + } \ } \ } while (0) From a7dec1e0dbb9e8e032b56a62d07ab6ac009109d3 Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Wed, 30 Aug 2006 19:14:25 +0100 Subject: [PATCH 066/115] [PATCH] Missing PCI id update for VIA IDE The following change from -mm is important to 2.6.18 (actually to 2.6.17 but its too late for that). This was contributed over three months ago by VIA to Bartlomiej and nothing happened. As a result the new chipset is now out and Linux won't run on it. By the time 2.6.18 is finalised this will be the defacto standard VIA chipset so support would be a good plan. Tested in -mm for a while, its essentially a PCI ident update but for the bridge chip because VIA do things in weird ways. Signed-off-by: Linus Torvalds --- drivers/ide/pci/via82cxxx.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/ide/pci/via82cxxx.c b/drivers/ide/pci/via82cxxx.c index afdaee3c15c..9b7589e8e93 100644 --- a/drivers/ide/pci/via82cxxx.c +++ b/drivers/ide/pci/via82cxxx.c @@ -6,7 +6,7 @@ * * vt82c576, vt82c586, vt82c586a, vt82c586b, vt82c596a, vt82c596b, * vt82c686, vt82c686a, vt82c686b, vt8231, vt8233, vt8233c, vt8233a, - * vt8235, vt8237 + * vt8235, vt8237, vt8237a * * Copyright (c) 2000-2002 Vojtech Pavlik * @@ -81,6 +81,7 @@ static struct via_isa_bridge { { "vt6410", PCI_DEVICE_ID_VIA_6410, 0x00, 0x2f, VIA_UDMA_133 | VIA_BAD_AST }, { "vt8251", PCI_DEVICE_ID_VIA_8251, 0x00, 0x2f, VIA_UDMA_133 | VIA_BAD_AST }, { "vt8237", PCI_DEVICE_ID_VIA_8237, 0x00, 0x2f, VIA_UDMA_133 | VIA_BAD_AST }, + { "vt8237a", PCI_DEVICE_ID_VIA_8237A, 0x00, 0x2f, VIA_UDMA_133 | VIA_BAD_AST }, { "vt8235", PCI_DEVICE_ID_VIA_8235, 0x00, 0x2f, VIA_UDMA_133 | VIA_BAD_AST }, { "vt8233a", PCI_DEVICE_ID_VIA_8233A, 0x00, 0x2f, VIA_UDMA_133 | VIA_BAD_AST }, { "vt8233c", PCI_DEVICE_ID_VIA_8233C_0, 0x00, 0x2f, VIA_UDMA_100 }, From 18f2905fcec3e06deafd25a02e37eabaaaaef744 Mon Sep 17 00:00:00 2001 From: Roland Scheidegger Date: Wed, 30 Aug 2006 23:17:55 +0100 Subject: [PATCH 067/115] [PATCH] drm: radeon flush TCL VAP for vertex program enable/disable The radeon requires a VAP state flush when enabling/disabling vertex programs on the r200 cards. Signed-off-by: Dave Airlie Signed-off-by: Linus Torvalds --- drivers/char/drm/radeon_state.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/drivers/char/drm/radeon_state.c b/drivers/char/drm/radeon_state.c index 5bb2234a909..39a7f685e3f 100644 --- a/drivers/char/drm/radeon_state.c +++ b/drivers/char/drm/radeon_state.c @@ -175,6 +175,14 @@ static __inline__ int radeon_check_and_fixup_packets(drm_radeon_private_t * } break; + case R200_EMIT_VAP_CTL:{ + RING_LOCALS; + BEGIN_RING(2); + OUT_RING_REG(RADEON_SE_TCL_STATE_FLUSH, 0); + ADVANCE_RING(); + } + break; + case RADEON_EMIT_RB3D_COLORPITCH: case RADEON_EMIT_RE_LINE_PATTERN: case RADEON_EMIT_SE_LINE_WIDTH: @@ -202,7 +210,6 @@ static __inline__ int radeon_check_and_fixup_packets(drm_radeon_private_t * case R200_EMIT_TCL_LIGHT_MODEL_CTL_0: case R200_EMIT_TFACTOR_0: case R200_EMIT_VTX_FMT_0: - case R200_EMIT_VAP_CTL: case R200_EMIT_MATRIX_SELECT_0: case R200_EMIT_TEX_PROC_CTL_2: case R200_EMIT_TCL_UCP_VERT_BLEND_CTL: From d0027bf09f09d95a23b8f476ba8cea28f2576781 Mon Sep 17 00:00:00 2001 From: Paul Mackerras Date: Thu, 31 Aug 2006 13:22:58 +1000 Subject: [PATCH 068/115] [POWERPC] Fix return value from memcpy As pointed out by Herbert Xu , our memcpy implementation didn't return the destination pointer as its return value, and there is code in the kernel that expects that. This fixes it. Signed-off-by: Paul Mackerras --- arch/powerpc/lib/memcpy_64.S | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) diff --git a/arch/powerpc/lib/memcpy_64.S b/arch/powerpc/lib/memcpy_64.S index fd66acfd3e3..7173ba98f42 100644 --- a/arch/powerpc/lib/memcpy_64.S +++ b/arch/powerpc/lib/memcpy_64.S @@ -11,6 +11,7 @@ .align 7 _GLOBAL(memcpy) + std r3,48(r1) /* save destination pointer for return value */ mtcrf 0x01,r5 cmpldi cr1,r5,16 neg r6,r3 # LS 3 bits = # bytes to 8-byte dest bdry @@ -38,7 +39,7 @@ _GLOBAL(memcpy) stdu r9,16(r3) bdnz 1b 3: std r8,8(r3) - beqlr + beq 3f addi r3,r3,16 ld r9,8(r4) .Ldo_tail: @@ -53,7 +54,8 @@ _GLOBAL(memcpy) 2: bf cr7*4+3,3f rotldi r9,r9,8 stb r9,0(r3) -3: blr +3: ld r3,48(r1) /* return dest pointer */ + blr .Lsrc_unaligned: srdi r6,r5,3 @@ -115,7 +117,7 @@ _GLOBAL(memcpy) 5: srd r12,r9,r11 or r12,r8,r12 std r12,24(r3) - beqlr + beq 4f cmpwi cr1,r5,8 addi r3,r3,32 sld r9,r9,r10 @@ -167,4 +169,5 @@ _GLOBAL(memcpy) 3: bf cr7*4+3,4f lbz r0,0(r4) stb r0,0(r3) -4: blr +4: ld r3,48(r1) /* return dest pointer */ + blr From 98a3c781057fa43494e7e8b39d639e93fca0ecbf Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Thu, 31 Aug 2006 14:04:34 +1000 Subject: [PATCH 069/115] [PATCH] fbdev: Fix crashes in various fbdev's blank routines The backlight changes that went in had a bug where they could cause the kernel to access an unitialized pointer when blanking if there is no backlight control on a machine. The bug affects atyfb, aty128fb, nvidiafb and rivafb. radeonfb seems to be ok. This fixes it. Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Linus Torvalds --- drivers/video/aty/aty128fb.c | 5 ++++- drivers/video/aty/atyfb_base.c | 5 ++++- drivers/video/nvidia/nv_backlight.c | 4 +++- drivers/video/riva/fbdev.c | 5 ++++- 4 files changed, 15 insertions(+), 4 deletions(-) diff --git a/drivers/video/aty/aty128fb.c b/drivers/video/aty/aty128fb.c index 3e827e04a2a..106d428b72c 100644 --- a/drivers/video/aty/aty128fb.c +++ b/drivers/video/aty/aty128fb.c @@ -1800,6 +1800,9 @@ static struct backlight_properties aty128_bl_data = { static void aty128_bl_set_power(struct fb_info *info, int power) { + if (info->bl_dev == NULL) + return; + mutex_lock(&info->bl_mutex); up(&info->bl_dev->sem); info->bl_dev->props->power = power; @@ -1828,7 +1831,7 @@ static void aty128_bl_init(struct aty128fb_par *par) bd = backlight_device_register(name, par, &aty128_bl_data); if (IS_ERR(bd)) { info->bl_dev = NULL; - printk("aty128: Backlight registration failed\n"); + printk(KERN_WARNING "aty128: Backlight registration failed\n"); goto error; } diff --git a/drivers/video/aty/atyfb_base.c b/drivers/video/aty/atyfb_base.c index 053ff63365b..510e4ea296e 100644 --- a/drivers/video/aty/atyfb_base.c +++ b/drivers/video/aty/atyfb_base.c @@ -2199,6 +2199,9 @@ static struct backlight_properties aty_bl_data = { static void aty_bl_set_power(struct fb_info *info, int power) { + if (info->bl_dev == NULL) + return; + mutex_lock(&info->bl_mutex); up(&info->bl_dev->sem); info->bl_dev->props->power = power; @@ -2223,7 +2226,7 @@ static void aty_bl_init(struct atyfb_par *par) bd = backlight_device_register(name, par, &aty_bl_data); if (IS_ERR(bd)) { info->bl_dev = NULL; - printk("aty: Backlight registration failed\n"); + printk(KERN_WARNING "aty: Backlight registration failed\n"); goto error; } diff --git a/drivers/video/nvidia/nv_backlight.c b/drivers/video/nvidia/nv_backlight.c index b45f577094a..14c37c42191 100644 --- a/drivers/video/nvidia/nv_backlight.c +++ b/drivers/video/nvidia/nv_backlight.c @@ -112,6 +112,8 @@ static struct backlight_properties nvidia_bl_data = { void nvidia_bl_set_power(struct fb_info *info, int power) { + if (info->bl_dev == NULL) + return; mutex_lock(&info->bl_mutex); up(&info->bl_dev->sem); info->bl_dev->props->power = power; @@ -140,7 +142,7 @@ void nvidia_bl_init(struct nvidia_par *par) bd = backlight_device_register(name, par, &nvidia_bl_data); if (IS_ERR(bd)) { info->bl_dev = NULL; - printk("nvidia: Backlight registration failed\n"); + printk(KERN_WARNING "nvidia: Backlight registration failed\n"); goto error; } diff --git a/drivers/video/riva/fbdev.c b/drivers/video/riva/fbdev.c index 76fc9d355eb..90363943bd5 100644 --- a/drivers/video/riva/fbdev.c +++ b/drivers/video/riva/fbdev.c @@ -354,6 +354,9 @@ static struct backlight_properties riva_bl_data = { static void riva_bl_set_power(struct fb_info *info, int power) { + if (info->bl_dev == NULL) + return; + mutex_lock(&info->bl_mutex); up(&info->bl_dev->sem); info->bl_dev->props->power = power; @@ -382,7 +385,7 @@ static void riva_bl_init(struct riva_par *par) bd = backlight_device_register(name, par, &riva_bl_data); if (IS_ERR(bd)) { info->bl_dev = NULL; - printk("riva: Backlight registration failed\n"); + printk(KERN_WARNING "riva: Backlight registration failed\n"); goto error; } From eb36c2884a1a2190791afe65fd833b2d3cd4b999 Mon Sep 17 00:00:00 2001 From: Paul Mackerras Date: Wed, 30 Aug 2006 16:13:16 +1000 Subject: [PATCH 070/115] [PATCH] ppc32: fix last_jiffy time comparison This fixes a hang on ppc32. The problem was that I was comparing a 32-bit quantity with a 64-bit quantity, and consequently time wasn't advancing. This makes us use a 64-bit quantity on all platforms, which ends up simplifying the code since we can now get rid of the tb_last_stamp variable (which actually fixes another bug that Ben H and I noticed while going carefully through the code). This works fine on my G4 tibook. Let me know how it goes on your machines. Acked-by: Olaf Hering Acked-by: Mikael Pettersson Signed-off-by: Linus Torvalds --- arch/powerpc/kernel/time.c | 25 ++++++++----------------- include/asm-powerpc/time.h | 4 ---- 2 files changed, 8 insertions(+), 21 deletions(-) diff --git a/arch/powerpc/kernel/time.c b/arch/powerpc/kernel/time.c index 18e59e43d2b..a124499e65d 100644 --- a/arch/powerpc/kernel/time.c +++ b/arch/powerpc/kernel/time.c @@ -125,15 +125,8 @@ static long timezone_offset; unsigned long ppc_proc_freq; unsigned long ppc_tb_freq; -u64 tb_last_jiffy __cacheline_aligned_in_smp; -unsigned long tb_last_stamp; - -/* - * Note that on ppc32 this only stores the bottom 32 bits of - * the timebase value, but that's enough to tell when a jiffy - * has passed. - */ -DEFINE_PER_CPU(unsigned long, last_jiffy); +static u64 tb_last_jiffy __cacheline_aligned_in_smp; +static DEFINE_PER_CPU(u64, last_jiffy); #ifdef CONFIG_VIRT_CPU_ACCOUNTING /* @@ -458,7 +451,7 @@ void do_gettimeofday(struct timeval *tv) do { seq = read_seqbegin_irqsave(&xtime_lock, flags); sec = xtime.tv_sec; - nsec = xtime.tv_nsec + tb_ticks_since(tb_last_stamp); + nsec = xtime.tv_nsec + tb_ticks_since(tb_last_jiffy); } while (read_seqretry_irqrestore(&xtime_lock, seq, flags)); usec = nsec / 1000; while (usec >= 1000000) { @@ -700,7 +693,6 @@ void timer_interrupt(struct pt_regs * regs) tb_next_jiffy = tb_last_jiffy + tb_ticks_per_jiffy; if (per_cpu(last_jiffy, cpu) >= tb_next_jiffy) { tb_last_jiffy = tb_next_jiffy; - tb_last_stamp = per_cpu(last_jiffy, cpu); do_timer(regs); timer_recalc_offset(tb_last_jiffy); timer_check_rtc(); @@ -749,7 +741,7 @@ void __init smp_space_timers(unsigned int max_cpus) int i; unsigned long half = tb_ticks_per_jiffy / 2; unsigned long offset = tb_ticks_per_jiffy / max_cpus; - unsigned long previous_tb = per_cpu(last_jiffy, boot_cpuid); + u64 previous_tb = per_cpu(last_jiffy, boot_cpuid); /* make sure tb > per_cpu(last_jiffy, cpu) for all cpus always */ previous_tb -= tb_ticks_per_jiffy; @@ -830,7 +822,7 @@ int do_settimeofday(struct timespec *tv) * and therefore the (jiffies - wall_jiffies) computation * has been removed. */ - tb_delta = tb_ticks_since(tb_last_stamp); + tb_delta = tb_ticks_since(tb_last_jiffy); tb_delta = mulhdu(tb_delta, do_gtod.varp->tb_to_xs); /* in xsec */ new_nsec -= SCALE_XSEC(tb_delta, 1000000000); @@ -950,8 +942,7 @@ void __init time_init(void) if (__USE_RTC()) { /* 601 processor: dec counts down by 128 every 128ns */ ppc_tb_freq = 1000000000; - tb_last_stamp = get_rtcl(); - tb_last_jiffy = tb_last_stamp; + tb_last_jiffy = get_rtcl(); } else { /* Normal PowerPC with timebase register */ ppc_md.calibrate_decr(); @@ -959,7 +950,7 @@ void __init time_init(void) ppc_tb_freq / 1000000, ppc_tb_freq % 1000000); printk(KERN_DEBUG "time_init: processor frequency = %lu.%.6lu MHz\n", ppc_proc_freq / 1000000, ppc_proc_freq % 1000000); - tb_last_stamp = tb_last_jiffy = get_tb(); + tb_last_jiffy = get_tb(); } tb_ticks_per_jiffy = ppc_tb_freq / HZ; @@ -1036,7 +1027,7 @@ void __init time_init(void) do_gtod.varp = &do_gtod.vars[0]; do_gtod.var_idx = 0; do_gtod.varp->tb_orig_stamp = tb_last_jiffy; - __get_cpu_var(last_jiffy) = tb_last_stamp; + __get_cpu_var(last_jiffy) = tb_last_jiffy; do_gtod.varp->stamp_xsec = (u64) xtime.tv_sec * XSEC_PER_SEC; do_gtod.tb_ticks_per_sec = tb_ticks_per_sec; do_gtod.varp->tb_to_xs = tb_to_xs; diff --git a/include/asm-powerpc/time.h b/include/asm-powerpc/time.h index dcde4410348..5785ac4737b 100644 --- a/include/asm-powerpc/time.h +++ b/include/asm-powerpc/time.h @@ -30,10 +30,6 @@ extern unsigned long tb_ticks_per_usec; extern unsigned long tb_ticks_per_sec; extern u64 tb_to_xs; extern unsigned tb_to_us; -extern unsigned long tb_last_stamp; -extern u64 tb_last_jiffy; - -DECLARE_PER_CPU(unsigned long, last_jiffy); struct rtc_time; extern void to_tm(int tim, struct rtc_time * tm); From 8e34703b9315688305306d26148088b0a8292563 Mon Sep 17 00:00:00 2001 From: Russell King Date: Thu, 31 Aug 2006 15:09:30 +0100 Subject: [PATCH 071/115] [ARM] Fix ARM __raw_read_trylock() implementation Matthew Wilcox pointed out that the generic implementation of this is unfit for use. Here's an ARM optimised version instead. Signed-off-by: Russell King --- include/asm-arm/spinlock.h | 16 +++++++++++++++- 1 file changed, 15 insertions(+), 1 deletion(-) diff --git a/include/asm-arm/spinlock.h b/include/asm-arm/spinlock.h index 406ca97a8ab..e2f1d75171d 100644 --- a/include/asm-arm/spinlock.h +++ b/include/asm-arm/spinlock.h @@ -199,7 +199,21 @@ static inline void __raw_read_unlock(raw_rwlock_t *rw) : "cc"); } -#define __raw_read_trylock(lock) generic__raw_read_trylock(lock) +static inline int __raw_read_trylock(raw_rwlock_t *rw) +{ + unsigned long tmp tmp2 = 1; + + __asm__ __volatile__( +"1: ldrex %0, [%2]\n" +" adds %0, %0, #1\n" +" strexpl %1, %0, [%2]\n" + : "=&r" (tmp), "+r" (tmp2) + : "r" (&rw->lock) + : "cc"); + + smp_mb(); + return tmp2 == 0; +} /* read_can_lock - would read_trylock() succeed? */ #define __raw_read_can_lock(x) ((x)->lock < 0x80000000) From 851fb304b510fc1e5dc5852cda25361219c1c4b1 Mon Sep 17 00:00:00 2001 From: David Brownell Date: Thu, 31 Aug 2006 15:26:35 +0100 Subject: [PATCH 072/115] [ARM] 3763/1: add both rtcs to csb337 defconfig Patch from David Brownell This adds RTC support to the csb337 default config. Both the AT91 and the ds1307 RTCs are enabled (rtc0 and rtc1 respectively). The ds1307 is used to initialize the system time, since it's battery-backed. From then on the AT91 RTC is used, since it's more capable (with both alarm and update irqs, and system wakeup capability) even though it needs manual initialization (symlink /dev/rtc to /dev/rtc0 for older versions of hwclock, then "hwclock --systohc") in an rc script or from inittab. Signed-off-by: David Brownell Signed-off-by: Russell King --- arch/arm/configs/csb337_defconfig | 37 ++++++++++++++++++++++++++++--- 1 file changed, 34 insertions(+), 3 deletions(-) diff --git a/arch/arm/configs/csb337_defconfig b/arch/arm/configs/csb337_defconfig index 3594155a813..cf3fa5cb26e 100644 --- a/arch/arm/configs/csb337_defconfig +++ b/arch/arm/configs/csb337_defconfig @@ -621,9 +621,8 @@ CONFIG_AT91_WATCHDOG=y # USB-based Watchdog Cards # # CONFIG_USBPCWATCHDOG is not set +# CONFIG_HW_RANDOM is not set # CONFIG_NVRAM is not set -CONFIG_RTC=y -# CONFIG_AT91_RTC is not set # CONFIG_DTLK is not set # CONFIG_R3964 is not set @@ -956,9 +955,41 @@ CONFIG_USB_AT91=y CONFIG_MMC=y # CONFIG_MMC_DEBUG is not set CONFIG_MMC_BLOCK=y -# CONFIG_MMC_WBSD is not set CONFIG_MMC_AT91RM9200=y +# +# Real Time Clock +# +CONFIG_RTC_LIB=y +CONFIG_RTC_CLASS=y +CONFIG_RTC_HCTOSYS=y +CONFIG_RTC_HCTOSYS_DEVICE="rtc1" + +# +# RTC interfaces +# +# CONFIG_RTC_INTF_SYSFS is not set +CONFIG_RTC_INTF_PROC=y +CONFIG_RTC_INTF_DEV=y +# CONFIG_RTC_INTF_DEV_UIE_EMUL is not set + +# +# RTC drivers +# +# CONFIG_RTC_DRV_X1205 is not set +CONFIG_RTC_DRV_DS1307=y +# CONFIG_RTC_DRV_DS1553 is not set +# CONFIG_RTC_DRV_ISL1208 is not set +# CONFIG_RTC_DRV_DS1672 is not set +# CONFIG_RTC_DRV_DS1742 is not set +# CONFIG_RTC_DRV_PCF8563 is not set +# CONFIG_RTC_DRV_PCF8583 is not set +# CONFIG_RTC_DRV_RS5C372 is not set +# CONFIG_RTC_DRV_M48T86 is not set +CONFIG_RTC_DRV_AT91=y +# CONFIG_RTC_DRV_TEST is not set +# CONFIG_RTC_DRV_V3020 is not set + # # File systems # From f105a7dfc5e81c28dd23ebf2328e42972e2cf240 Mon Sep 17 00:00:00 2001 From: Ben Dooks Date: Thu, 31 Aug 2006 15:26:37 +0100 Subject: [PATCH 073/115] [ARM] 3764/1: S3C24XX: change type naming to kernel style Patch from Ben Dooks The type naming in the s3c24xx dma code is riddled with typedefs creating _t types, from the code import from 2.4 which is contrary to the current Kernel coding style. This patch cleans this up, removing the typedefs and and fixing up the resultant code changes. Signed-off-by: Ben Dooks Signed-off-by: Russell King --- arch/arm/mach-s3c2410/dma.c | 88 +++++++++++++++--------------- include/asm-arm/arch-s3c2410/dma.h | 80 ++++++++++++--------------- 2 files changed, 80 insertions(+), 88 deletions(-) diff --git a/arch/arm/mach-s3c2410/dma.c b/arch/arm/mach-s3c2410/dma.c index 25855452fe8..cc92a7b2db8 100644 --- a/arch/arm/mach-s3c2410/dma.c +++ b/arch/arm/mach-s3c2410/dma.c @@ -60,7 +60,7 @@ static void __iomem *dma_base; static kmem_cache_t *dma_kmem; /* dma channel state information */ -s3c2410_dma_chan_t s3c2410_chans[S3C2410_DMA_CHANNELS]; +struct s3c2410_dma_chan s3c2410_chans[S3C2410_DMA_CHANNELS]; /* debugging functions */ @@ -74,7 +74,7 @@ s3c2410_dma_chan_t s3c2410_chans[S3C2410_DMA_CHANNELS]; #define dma_wrreg(chan, reg, val) writel((val), (chan)->regs + (reg)) #else static inline void -dma_wrreg(s3c2410_dma_chan_t *chan, int reg, unsigned long val) +dma_wrreg(struct s3c2410_dma_chan *chan, int reg, unsigned long val) { pr_debug("writing %08x to register %08x\n",(unsigned int)val,reg); writel(val, dma_regaddr(chan, reg)); @@ -102,7 +102,7 @@ struct s3c2410_dma_regstate { */ static void -dmadbg_capture(s3c2410_dma_chan_t *chan, struct s3c2410_dma_regstate *regs) +dmadbg_capture(struct s3c2410_dma_chan *chan, struct s3c2410_dma_regstate *regs) { regs->dcsrc = dma_rdreg(chan, S3C2410_DMA_DCSRC); regs->disrc = dma_rdreg(chan, S3C2410_DMA_DISRC); @@ -112,7 +112,7 @@ dmadbg_capture(s3c2410_dma_chan_t *chan, struct s3c2410_dma_regstate *regs) } static void -dmadbg_dumpregs(const char *fname, int line, s3c2410_dma_chan_t *chan, +dmadbg_dumpregs(const char *fname, int line, struct s3c2410_dma_chan *chan, struct s3c2410_dma_regstate *regs) { printk(KERN_DEBUG "dma%d: %s:%d: DCSRC=%08lx, DISRC=%08lx, DSTAT=%08lx DMT=%02lx, DCON=%08lx\n", @@ -122,7 +122,7 @@ dmadbg_dumpregs(const char *fname, int line, s3c2410_dma_chan_t *chan, } static void -dmadbg_showchan(const char *fname, int line, s3c2410_dma_chan_t *chan) +dmadbg_showchan(const char *fname, int line, struct s3c2410_dma_chan *chan) { struct s3c2410_dma_regstate state; @@ -136,7 +136,7 @@ dmadbg_showchan(const char *fname, int line, s3c2410_dma_chan_t *chan) } static void -dmadbg_showregs(const char *fname, int line, s3c2410_dma_chan_t *chan) +dmadbg_showregs(const char *fname, int line, struct s3c2410_dma_chan *chan) { struct s3c2410_dma_regstate state; @@ -164,7 +164,7 @@ dmadbg_showregs(const char *fname, int line, s3c2410_dma_chan_t *chan) */ static void -s3c2410_dma_stats_timeout(s3c2410_dma_stats_t *stats, int val) +s3c2410_dma_stats_timeout(struct s3c2410_dma_stats *stats, int val) { if (stats == NULL) return; @@ -183,7 +183,7 @@ s3c2410_dma_stats_timeout(s3c2410_dma_stats_t *stats, int val) */ static int -s3c2410_dma_waitforload(s3c2410_dma_chan_t *chan, int line) +s3c2410_dma_waitforload(struct s3c2410_dma_chan *chan, int line) { int timeout = chan->load_timeout; int took; @@ -230,8 +230,8 @@ s3c2410_dma_waitforload(s3c2410_dma_chan_t *chan, int line) */ static inline int -s3c2410_dma_loadbuffer(s3c2410_dma_chan_t *chan, - s3c2410_dma_buf_t *buf) +s3c2410_dma_loadbuffer(struct s3c2410_dma_chan *chan, + struct s3c2410_dma_buf *buf) { unsigned long reload; @@ -304,7 +304,7 @@ s3c2410_dma_loadbuffer(s3c2410_dma_chan_t *chan, */ static void -s3c2410_dma_call_op(s3c2410_dma_chan_t *chan, s3c2410_chan_op_t op) +s3c2410_dma_call_op(struct s3c2410_dma_chan *chan, enum s3c2410_chan_op op) { if (chan->op_fn != NULL) { (chan->op_fn)(chan, op); @@ -318,8 +318,8 @@ s3c2410_dma_call_op(s3c2410_dma_chan_t *chan, s3c2410_chan_op_t op) */ static inline void -s3c2410_dma_buffdone(s3c2410_dma_chan_t *chan, s3c2410_dma_buf_t *buf, - s3c2410_dma_buffresult_t result) +s3c2410_dma_buffdone(struct s3c2410_dma_chan *chan, struct s3c2410_dma_buf *buf, + enum s3c2410_dma_buffresult result) { pr_debug("callback_fn=%p, buf=%p, id=%p, size=%d, result=%d\n", chan->callback_fn, buf, buf->id, buf->size, result); @@ -334,7 +334,7 @@ s3c2410_dma_buffdone(s3c2410_dma_chan_t *chan, s3c2410_dma_buf_t *buf, * start a dma channel going */ -static int s3c2410_dma_start(s3c2410_dma_chan_t *chan) +static int s3c2410_dma_start(struct s3c2410_dma_chan *chan) { unsigned long tmp; unsigned long flags; @@ -430,7 +430,7 @@ static int s3c2410_dma_start(s3c2410_dma_chan_t *chan) */ static int -s3c2410_dma_canload(s3c2410_dma_chan_t *chan) +s3c2410_dma_canload(struct s3c2410_dma_chan *chan) { if (chan->load_state == S3C2410_DMALOAD_NONE || chan->load_state == S3C2410_DMALOAD_1RUNNING) @@ -460,8 +460,8 @@ s3c2410_dma_canload(s3c2410_dma_chan_t *chan) int s3c2410_dma_enqueue(unsigned int channel, void *id, dma_addr_t data, int size) { - s3c2410_dma_chan_t *chan = &s3c2410_chans[channel]; - s3c2410_dma_buf_t *buf; + struct s3c2410_dma_chan *chan = &s3c2410_chans[channel]; + struct s3c2410_dma_buf *buf; unsigned long flags; check_channel(channel); @@ -540,7 +540,7 @@ int s3c2410_dma_enqueue(unsigned int channel, void *id, EXPORT_SYMBOL(s3c2410_dma_enqueue); static inline void -s3c2410_dma_freebuf(s3c2410_dma_buf_t *buf) +s3c2410_dma_freebuf(struct s3c2410_dma_buf *buf) { int magicok = (buf->magic == BUF_MAGIC); @@ -560,7 +560,7 @@ s3c2410_dma_freebuf(s3c2410_dma_buf_t *buf) */ static inline void -s3c2410_dma_lastxfer(s3c2410_dma_chan_t *chan) +s3c2410_dma_lastxfer(struct s3c2410_dma_chan *chan) { pr_debug("dma%d: s3c2410_dma_lastxfer: load_state %d\n", chan->number, chan->load_state); @@ -601,8 +601,8 @@ s3c2410_dma_lastxfer(s3c2410_dma_chan_t *chan) static irqreturn_t s3c2410_dma_irq(int irq, void *devpw, struct pt_regs *regs) { - s3c2410_dma_chan_t *chan = (s3c2410_dma_chan_t *)devpw; - s3c2410_dma_buf_t *buf; + struct s3c2410_dma_chan *chan = (struct s3c2410_dma_chan *)devpw; + struct s3c2410_dma_buf *buf; buf = chan->curr; @@ -731,10 +731,10 @@ s3c2410_dma_irq(int irq, void *devpw, struct pt_regs *regs) * get control of an dma channel */ -int s3c2410_dma_request(unsigned int channel, s3c2410_dma_client_t *client, +int s3c2410_dma_request(unsigned int channel, struct s3c2410_dma_client *client, void *dev) { - s3c2410_dma_chan_t *chan = &s3c2410_chans[channel]; + struct s3c2410_dma_chan *chan = &s3c2410_chans[channel]; unsigned long flags; int err; @@ -807,9 +807,9 @@ EXPORT_SYMBOL(s3c2410_dma_request); * allowed to go through. */ -int s3c2410_dma_free(dmach_t channel, s3c2410_dma_client_t *client) +int s3c2410_dma_free(dmach_t channel, struct s3c2410_dma_client *client) { - s3c2410_dma_chan_t *chan = &s3c2410_chans[channel]; + struct s3c2410_dma_chan *chan = &s3c2410_chans[channel]; unsigned long flags; check_channel(channel); @@ -846,7 +846,7 @@ int s3c2410_dma_free(dmach_t channel, s3c2410_dma_client_t *client) EXPORT_SYMBOL(s3c2410_dma_free); -static int s3c2410_dma_dostop(s3c2410_dma_chan_t *chan) +static int s3c2410_dma_dostop(struct s3c2410_dma_chan *chan) { unsigned long tmp; unsigned long flags; @@ -880,7 +880,7 @@ static int s3c2410_dma_dostop(s3c2410_dma_chan_t *chan) return 0; } -void s3c2410_dma_waitforstop(s3c2410_dma_chan_t *chan) +void s3c2410_dma_waitforstop(struct s3c2410_dma_chan *chan) { unsigned long tmp; unsigned int timeout = 0x10000; @@ -901,9 +901,9 @@ void s3c2410_dma_waitforstop(s3c2410_dma_chan_t *chan) * stop the channel, and remove all current and pending transfers */ -static int s3c2410_dma_flush(s3c2410_dma_chan_t *chan) +static int s3c2410_dma_flush(struct s3c2410_dma_chan *chan) { - s3c2410_dma_buf_t *buf, *next; + struct s3c2410_dma_buf *buf, *next; unsigned long flags; pr_debug("%s: chan %p (%d)\n", __FUNCTION__, chan, chan->number); @@ -958,7 +958,7 @@ static int s3c2410_dma_flush(s3c2410_dma_chan_t *chan) } int -s3c2410_dma_started(s3c2410_dma_chan_t *chan) +s3c2410_dma_started(struct s3c2410_dma_chan *chan) { unsigned long flags; @@ -995,9 +995,9 @@ s3c2410_dma_started(s3c2410_dma_chan_t *chan) } int -s3c2410_dma_ctrl(dmach_t channel, s3c2410_chan_op_t op) +s3c2410_dma_ctrl(dmach_t channel, enum s3c2410_chan_op op) { - s3c2410_dma_chan_t *chan = &s3c2410_chans[channel]; + struct s3c2410_dma_chan *chan = &s3c2410_chans[channel]; check_channel(channel); @@ -1046,7 +1046,7 @@ int s3c2410_dma_config(dmach_t channel, int xferunit, int dcon) { - s3c2410_dma_chan_t *chan = &s3c2410_chans[channel]; + struct s3c2410_dma_chan *chan = &s3c2410_chans[channel]; pr_debug("%s: chan=%d, xfer_unit=%d, dcon=%08x\n", __FUNCTION__, channel, xferunit, dcon); @@ -1086,7 +1086,7 @@ EXPORT_SYMBOL(s3c2410_dma_config); int s3c2410_dma_setflags(dmach_t channel, unsigned int flags) { - s3c2410_dma_chan_t *chan = &s3c2410_chans[channel]; + struct s3c2410_dma_chan *chan = &s3c2410_chans[channel]; check_channel(channel); @@ -1106,7 +1106,7 @@ EXPORT_SYMBOL(s3c2410_dma_setflags); int s3c2410_dma_set_opfn(dmach_t channel, s3c2410_dma_opfn_t rtn) { - s3c2410_dma_chan_t *chan = &s3c2410_chans[channel]; + struct s3c2410_dma_chan *chan = &s3c2410_chans[channel]; check_channel(channel); @@ -1121,7 +1121,7 @@ EXPORT_SYMBOL(s3c2410_dma_set_opfn); int s3c2410_dma_set_buffdone_fn(dmach_t channel, s3c2410_dma_cbfn_t rtn) { - s3c2410_dma_chan_t *chan = &s3c2410_chans[channel]; + struct s3c2410_dma_chan *chan = &s3c2410_chans[channel]; check_channel(channel); @@ -1149,11 +1149,11 @@ EXPORT_SYMBOL(s3c2410_dma_set_buffdone_fn); */ int s3c2410_dma_devconfig(int channel, - s3c2410_dmasrc_t source, + enum s3c2410_dmasrc source, int hwcfg, unsigned long devaddr) { - s3c2410_dma_chan_t *chan = &s3c2410_chans[channel]; + struct s3c2410_dma_chan *chan = &s3c2410_chans[channel]; check_channel(channel); @@ -1200,7 +1200,7 @@ EXPORT_SYMBOL(s3c2410_dma_devconfig); int s3c2410_dma_getposition(dmach_t channel, dma_addr_t *src, dma_addr_t *dst) { - s3c2410_dma_chan_t *chan = &s3c2410_chans[channel]; + struct s3c2410_dma_chan *chan = &s3c2410_chans[channel]; check_channel(channel); @@ -1222,7 +1222,7 @@ EXPORT_SYMBOL(s3c2410_dma_getposition); static int s3c2410_dma_suspend(struct sys_device *dev, pm_message_t state) { - s3c2410_dma_chan_t *cp = container_of(dev, s3c2410_dma_chan_t, dev); + struct s3c2410_dma_chan *cp = container_of(dev, struct s3c2410_dma_chan, dev); printk(KERN_DEBUG "suspending dma channel %d\n", cp->number); @@ -1262,7 +1262,7 @@ static struct sysdev_class dma_sysclass = { static void s3c2410_dma_cache_ctor(void *p, kmem_cache_t *c, unsigned long f) { - memset(p, 0, sizeof(s3c2410_dma_buf_t)); + memset(p, 0, sizeof(struct s3c2410_dma_buf)); } @@ -1270,7 +1270,7 @@ static void s3c2410_dma_cache_ctor(void *p, kmem_cache_t *c, unsigned long f) static int __init s3c2410_init_dma(void) { - s3c2410_dma_chan_t *cp; + struct s3c2410_dma_chan *cp; int channel; int ret; @@ -1288,7 +1288,7 @@ static int __init s3c2410_init_dma(void) goto err; } - dma_kmem = kmem_cache_create("dma_desc", sizeof(s3c2410_dma_buf_t), 0, + dma_kmem = kmem_cache_create("dma_desc", sizeof(struct s3c2410_dma_buf), 0, SLAB_HWCACHE_ALIGN, s3c2410_dma_cache_ctor, NULL); @@ -1301,7 +1301,7 @@ static int __init s3c2410_init_dma(void) for (channel = 0; channel < S3C2410_DMA_CHANNELS; channel++) { cp = &s3c2410_chans[channel]; - memset(cp, 0, sizeof(s3c2410_dma_chan_t)); + memset(cp, 0, sizeof(struct s3c2410_dma_chan)); /* dma channel irqs are in order.. */ cp->number = channel; diff --git a/include/asm-arm/arch-s3c2410/dma.h b/include/asm-arm/arch-s3c2410/dma.h index 7463fd5252c..46e65409fcc 100644 --- a/include/asm-arm/arch-s3c2410/dma.h +++ b/include/asm-arm/arch-s3c2410/dma.h @@ -35,14 +35,14 @@ /* types */ -typedef enum { +enum s3c2410_dma_state { S3C2410_DMA_IDLE, S3C2410_DMA_RUNNING, S3C2410_DMA_PAUSED -} s3c2410_dma_state_t; +}; -/* s3c2410_dma_loadst_t +/* enum s3c2410_dma_loadst * * This represents the state of the DMA engine, wrt to the loaded / running * transfers. Since we don't have any way of knowing exactly the state of @@ -70,34 +70,32 @@ typedef enum { * currently running. */ -typedef enum { +enum s3c2410_dma_loadst { S3C2410_DMALOAD_NONE, S3C2410_DMALOAD_1LOADED, S3C2410_DMALOAD_1RUNNING, S3C2410_DMALOAD_1LOADED_1RUNNING, -} s3c2410_dma_loadst_t; +}; -typedef enum { +enum s3c2410_dma_buffresult { S3C2410_RES_OK, S3C2410_RES_ERR, S3C2410_RES_ABORT -} s3c2410_dma_buffresult_t; +}; -typedef enum s3c2410_dmasrc_e s3c2410_dmasrc_t; - -enum s3c2410_dmasrc_e { +enum s3c2410_dmasrc { S3C2410_DMASRC_HW, /* source is memory */ S3C2410_DMASRC_MEM /* source is hardware */ }; -/* enum s3c2410_chan_op_e +/* enum s3c2410_chan_op * * operation codes passed to the DMA code by the user, and also used * to inform the current channel owner of any changes to the system state */ -enum s3c2410_chan_op_e { +enum s3c2410_chan_op { S3C2410_DMAOP_START, S3C2410_DMAOP_STOP, S3C2410_DMAOP_PAUSE, @@ -107,8 +105,6 @@ enum s3c2410_chan_op_e { S3C2410_DMAOP_STARTED, /* indicate channel started */ }; -typedef enum s3c2410_chan_op_e s3c2410_chan_op_t; - /* flags */ #define S3C2410_DMAF_SLOW (1<<0) /* slow, so don't worry about @@ -117,22 +113,19 @@ typedef enum s3c2410_chan_op_e s3c2410_chan_op_t; /* dma buffer */ -typedef struct s3c2410_dma_buf_s s3c2410_dma_buf_t; - struct s3c2410_dma_client { char *name; }; -typedef struct s3c2410_dma_client s3c2410_dma_client_t; - /* s3c2410_dma_buf_s * * internally used buffer structure to describe a queued or running * buffer. */ -struct s3c2410_dma_buf_s { - s3c2410_dma_buf_t *next; +struct s3c2410_dma_buf; +struct s3c2410_dma_buf { + struct s3c2410_dma_buf *next; int magic; /* magic */ int size; /* buffer size in bytes */ dma_addr_t data; /* start of DMA data */ @@ -142,20 +135,21 @@ struct s3c2410_dma_buf_s { /* [1] is this updated for both recv/send modes? */ -typedef struct s3c2410_dma_chan_s s3c2410_dma_chan_t; +struct s3c2410_dma_chan; /* s3c2410_dma_cbfn_t * * buffer callback routine type */ -typedef void (*s3c2410_dma_cbfn_t)(s3c2410_dma_chan_t *, void *buf, int size, - s3c2410_dma_buffresult_t result); +typedef void (*s3c2410_dma_cbfn_t)(struct s3c2410_dma_chan *, + void *buf, int size, + enum s3c2410_dma_buffresult result); -typedef int (*s3c2410_dma_opfn_t)(s3c2410_dma_chan_t *, - s3c2410_chan_op_t ); +typedef int (*s3c2410_dma_opfn_t)(struct s3c2410_dma_chan *, + enum s3c2410_chan_op ); -struct s3c2410_dma_stats_s { +struct s3c2410_dma_stats { unsigned long loads; unsigned long timeout_longest; unsigned long timeout_shortest; @@ -163,14 +157,12 @@ struct s3c2410_dma_stats_s { unsigned long timeout_failed; }; -typedef struct s3c2410_dma_stats_s s3c2410_dma_stats_t; - -/* struct s3c2410_dma_chan_s +/* struct s3c2410_dma_chan * * full state information for each DMA channel */ -struct s3c2410_dma_chan_s { +struct s3c2410_dma_chan { /* channel state flags and information */ unsigned char number; /* number of this dma channel */ unsigned char in_use; /* channel allocated */ @@ -180,12 +172,12 @@ struct s3c2410_dma_chan_s { /* channel state */ - s3c2410_dma_state_t state; - s3c2410_dma_loadst_t load_state; - s3c2410_dma_client_t *client; + enum s3c2410_dma_state state; + enum s3c2410_dma_loadst load_state; + struct s3c2410_dma_client *client; /* channel configuration */ - s3c2410_dmasrc_t source; + enum s3c2410_dmasrc source; unsigned long dev_addr; unsigned long load_timeout; unsigned int flags; /* channel flags */ @@ -201,20 +193,20 @@ struct s3c2410_dma_chan_s { s3c2410_dma_opfn_t op_fn; /* channel operation callback */ /* stats gathering */ - s3c2410_dma_stats_t *stats; - s3c2410_dma_stats_t stats_store; + struct s3c2410_dma_stats *stats; + struct s3c2410_dma_stats stats_store; /* buffer list and information */ - s3c2410_dma_buf_t *curr; /* current dma buffer */ - s3c2410_dma_buf_t *next; /* next buffer to load */ - s3c2410_dma_buf_t *end; /* end of queue */ + struct s3c2410_dma_buf *curr; /* current dma buffer */ + struct s3c2410_dma_buf *next; /* next buffer to load */ + struct s3c2410_dma_buf *end; /* end of queue */ /* system device */ struct sys_device dev; }; /* the currently allocated channel information */ -extern s3c2410_dma_chan_t s3c2410_chans[]; +extern struct s3c2410_dma_chan s3c2410_chans[]; /* note, we don't really use dma_device_t at the moment */ typedef unsigned long dma_device_t; @@ -227,7 +219,7 @@ typedef unsigned long dma_device_t; */ extern int s3c2410_dma_request(dmach_t channel, - s3c2410_dma_client_t *, void *dev); + struct s3c2410_dma_client *, void *dev); /* s3c2410_dma_ctrl @@ -235,7 +227,7 @@ extern int s3c2410_dma_request(dmach_t channel, * change the state of the dma channel */ -extern int s3c2410_dma_ctrl(dmach_t channel, s3c2410_chan_op_t op); +extern int s3c2410_dma_ctrl(dmach_t channel, enum s3c2410_chan_op op); /* s3c2410_dma_setflags * @@ -250,7 +242,7 @@ extern int s3c2410_dma_setflags(dmach_t channel, * free the dma channel (will also abort any outstanding operations) */ -extern int s3c2410_dma_free(dmach_t channel, s3c2410_dma_client_t *); +extern int s3c2410_dma_free(dmach_t channel, struct s3c2410_dma_client *); /* s3c2410_dma_enqueue * @@ -274,7 +266,7 @@ extern int s3c2410_dma_config(dmach_t channel, int xferunit, int dcon); * configure the device we're talking to */ -extern int s3c2410_dma_devconfig(int channel, s3c2410_dmasrc_t source, +extern int s3c2410_dma_devconfig(int channel, enum s3c2410_dmasrc source, int hwcfg, unsigned long devaddr); /* s3c2410_dma_getposition From 57bcdafcb1e0782e7ae13471d9223c69e3a6cba2 Mon Sep 17 00:00:00 2001 From: Ben Dooks Date: Thu, 31 Aug 2006 15:26:41 +0100 Subject: [PATCH 074/115] [ARM] 3765/1: S3C24XX: cleanup include/asm-arm/arch-s3c2410/dma.h Patch from Ben Dooks Cleanup for include/asm-arma/arch-s3c2410/dma.h, by using tab characters to indent items, remove the now un-necessary changelog, and update the copyright information. Signed-off-by: Ben Dooks Signed-off-by: Russell King --- include/asm-arm/arch-s3c2410/dma.h | 90 ++++++++++++++---------------- 1 file changed, 41 insertions(+), 49 deletions(-) diff --git a/include/asm-arm/arch-s3c2410/dma.h b/include/asm-arm/arch-s3c2410/dma.h index 46e65409fcc..3661e465b0a 100644 --- a/include/asm-arm/arch-s3c2410/dma.h +++ b/include/asm-arm/arch-s3c2410/dma.h @@ -1,18 +1,13 @@ -/* linux/include/asm-arm/arch-bast/dma.h +/* linux/include/asm-arm/arch-s3c2410/dma.h * - * Copyright (C) 2003,2004 Simtec Electronics + * Copyright (C) 2003,2004,2006 Simtec Electronics * Ben Dooks * - * Samsung S3C2410X DMA support + * Samsung S3C241XX DMA support * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 as * published by the Free Software Foundation. - * - * Changelog: - * ??-May-2003 BJD Created file - * ??-Jun-2003 BJD Added more dma functionality to go with arch - * 10-Nov-2004 BJD Added sys_device support */ #ifndef __ASM_ARCH_DMA_H @@ -21,15 +16,13 @@ #include #include "hardware.h" - /* * This is the maximum DMA address(physical address) that can be DMAd to. * */ -#define MAX_DMA_ADDRESS 0x20000000 +#define MAX_DMA_ADDRESS 0x40000000 #define MAX_DMA_TRANSFER_SIZE 0x100000 /* Data Unit is half word */ - /* we have 4 dma channels */ #define S3C2410_DMA_CHANNELS (4) @@ -83,10 +76,9 @@ enum s3c2410_dma_buffresult { S3C2410_RES_ABORT }; - enum s3c2410_dmasrc { - S3C2410_DMASRC_HW, /* source is memory */ - S3C2410_DMASRC_MEM /* source is hardware */ + S3C2410_DMASRC_HW, /* source is memory */ + S3C2410_DMASRC_MEM /* source is hardware */ }; /* enum s3c2410_chan_op @@ -101,7 +93,7 @@ enum s3c2410_chan_op { S3C2410_DMAOP_PAUSE, S3C2410_DMAOP_RESUME, S3C2410_DMAOP_FLUSH, - S3C2410_DMAOP_TIMEOUT, /* internal signal to handler */ + S3C2410_DMAOP_TIMEOUT, /* internal signal to handler */ S3C2410_DMAOP_STARTED, /* indicate channel started */ }; @@ -125,12 +117,12 @@ struct s3c2410_dma_client { struct s3c2410_dma_buf; struct s3c2410_dma_buf { - struct s3c2410_dma_buf *next; - int magic; /* magic */ - int size; /* buffer size in bytes */ - dma_addr_t data; /* start of DMA data */ - dma_addr_t ptr; /* where the DMA got to [1] */ - void *id; /* client's id */ + struct s3c2410_dma_buf *next; + int magic; /* magic */ + int size; /* buffer size in bytes */ + dma_addr_t data; /* start of DMA data */ + dma_addr_t ptr; /* where the DMA got to [1] */ + void *id; /* client's id */ }; /* [1] is this updated for both recv/send modes? */ @@ -150,11 +142,11 @@ typedef int (*s3c2410_dma_opfn_t)(struct s3c2410_dma_chan *, enum s3c2410_chan_op ); struct s3c2410_dma_stats { - unsigned long loads; - unsigned long timeout_longest; - unsigned long timeout_shortest; - unsigned long timeout_avg; - unsigned long timeout_failed; + unsigned long loads; + unsigned long timeout_longest; + unsigned long timeout_shortest; + unsigned long timeout_avg; + unsigned long timeout_failed; }; /* struct s3c2410_dma_chan @@ -164,42 +156,42 @@ struct s3c2410_dma_stats { struct s3c2410_dma_chan { /* channel state flags and information */ - unsigned char number; /* number of this dma channel */ - unsigned char in_use; /* channel allocated */ - unsigned char irq_claimed; /* irq claimed for channel */ - unsigned char irq_enabled; /* irq enabled for channel */ - unsigned char xfer_unit; /* size of an transfer */ + unsigned char number; /* number of this dma channel */ + unsigned char in_use; /* channel allocated */ + unsigned char irq_claimed; /* irq claimed for channel */ + unsigned char irq_enabled; /* irq enabled for channel */ + unsigned char xfer_unit; /* size of an transfer */ /* channel state */ - enum s3c2410_dma_state state; - enum s3c2410_dma_loadst load_state; - struct s3c2410_dma_client *client; + enum s3c2410_dma_state state; + enum s3c2410_dma_loadst load_state; + struct s3c2410_dma_client *client; /* channel configuration */ - enum s3c2410_dmasrc source; - unsigned long dev_addr; - unsigned long load_timeout; - unsigned int flags; /* channel flags */ + enum s3c2410_dmasrc source; + unsigned long dev_addr; + unsigned long load_timeout; + unsigned int flags; /* channel flags */ /* channel's hardware position and configuration */ - void __iomem *regs; /* channels registers */ - void __iomem *addr_reg; /* data address register */ - unsigned int irq; /* channel irq */ - unsigned long dcon; /* default value of DCON */ + void __iomem *regs; /* channels registers */ + void __iomem *addr_reg; /* data address register */ + unsigned int irq; /* channel irq */ + unsigned long dcon; /* default value of DCON */ /* driver handles */ - s3c2410_dma_cbfn_t callback_fn; /* buffer done callback */ - s3c2410_dma_opfn_t op_fn; /* channel operation callback */ + s3c2410_dma_cbfn_t callback_fn; /* buffer done callback */ + s3c2410_dma_opfn_t op_fn; /* channel op callback */ /* stats gathering */ - struct s3c2410_dma_stats *stats; - struct s3c2410_dma_stats stats_store; + struct s3c2410_dma_stats *stats; + struct s3c2410_dma_stats stats_store; /* buffer list and information */ - struct s3c2410_dma_buf *curr; /* current dma buffer */ - struct s3c2410_dma_buf *next; /* next buffer to load */ - struct s3c2410_dma_buf *end; /* end of queue */ + struct s3c2410_dma_buf *curr; /* current dma buffer */ + struct s3c2410_dma_buf *next; /* next buffer to load */ + struct s3c2410_dma_buf *end; /* end of queue */ /* system device */ struct sys_device dev; From 22db37ec5fd51b0c77b1dd5751b1cdc2672c08d6 Mon Sep 17 00:00:00 2001 From: Chris Wright Date: Thu, 31 Aug 2006 00:53:22 -0700 Subject: [PATCH 075/115] [PATCH] i386: rwlock.h fix smp alternatives fix Commit 8c74932779fc6f61b4c30145863a17125c1a296c ("i386: Remove alternative_smp") did not actually compile on x86 with CONFIG_SMP. This fixes the __build_read/write_lock helpers. I've boot tested on SMP. [ Andi: "Oops, I think that was a quilt unrefreshed patch. Sorry. I fixed those before testing, but then still send out the old patch." ] Signed-off-by: Chris Wright Cc: Gerd Hoffmann Acked-by: Andi Kleen Signed-off-by: Linus Torvalds --- include/asm-i386/rwlock.h | 24 +++++++++++------------- 1 file changed, 11 insertions(+), 13 deletions(-) diff --git a/include/asm-i386/rwlock.h b/include/asm-i386/rwlock.h index 3ac1ba98b1b..87c069ccba0 100644 --- a/include/asm-i386/rwlock.h +++ b/include/asm-i386/rwlock.h @@ -21,21 +21,21 @@ #define RW_LOCK_BIAS_STR "0x01000000" #define __build_read_lock_ptr(rw, helper) \ - asm volatile(LOCK_PREFIX " ; subl $1,(%0)\n\t" \ + asm volatile(LOCK_PREFIX " subl $1,(%0)\n\t" \ "jns 1f\n" \ "call " helper "\n\t" \ "1:\n" \ - :"a" (rw) : "memory") + ::"a" (rw) : "memory") #define __build_read_lock_const(rw, helper) \ - asm volatile(LOCK_PREFIX " ; subl $1,%0\n\t" \ + asm volatile(LOCK_PREFIX " subl $1,%0\n\t" \ "jns 1f\n" \ "pushl %%eax\n\t" \ "leal %0,%%eax\n\t" \ "call " helper "\n\t" \ "popl %%eax\n\t" \ - "1:\n" : \ - "+m" (*(volatile int *)rw) : : "memory") + "1:\n" \ + :"+m" (*(volatile int *)rw) : : "memory") #define __build_read_lock(rw, helper) do { \ if (__builtin_constant_p(rw)) \ @@ -45,23 +45,21 @@ } while (0) #define __build_write_lock_ptr(rw, helper) \ - asm volatile(LOCK_PREFIX " ; subl $" RW_LOCK_BIAS_STR ",(%0)\n\t" \ + asm volatile(LOCK_PREFIX " subl $" RW_LOCK_BIAS_STR ",(%0)\n\t" \ "jz 1f\n" \ "call " helper "\n\t" \ - "1:\n", \ - "subl $" RW_LOCK_BIAS_STR ",(%0)\n\t", \ - :"a" (rw) : "memory") + "1:\n" \ + ::"a" (rw) : "memory") #define __build_write_lock_const(rw, helper) \ - asm volatile(LOCK_PREFIX " ; subl $" RW_LOCK_BIAS_STR ",%0\n\t" \ + asm volatile(LOCK_PREFIX " subl $" RW_LOCK_BIAS_STR ",%0\n\t" \ "jz 1f\n" \ "pushl %%eax\n\t" \ "leal %0,%%eax\n\t" \ "call " helper "\n\t" \ "popl %%eax\n\t" \ - "1:\n", \ - "subl $" RW_LOCK_BIAS_STR ",%0\n\t", \ - "+m" (*(volatile int *)rw) : : "memory") + "1:\n" \ + :"+m" (*(volatile int *)rw) : : "memory") #define __build_write_lock(rw, helper) do { \ if (__builtin_constant_p(rw)) \ From 99c7bc0133b875280fdd2bf78e4ffbd58cc609e3 Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Thu, 31 Aug 2006 14:52:17 -0700 Subject: [PATCH 076/115] [IPV6]: Fix kernel OOPs when setting sticky socket options. Bug noticed by Remi Denis-Courmont . Signed-off-by: YOSHIFUJI Hideaki Signed-off-by: David S. Miller --- net/ipv6/exthdrs.c | 29 ++++++++++++++++------------- 1 file changed, 16 insertions(+), 13 deletions(-) diff --git a/net/ipv6/exthdrs.c b/net/ipv6/exthdrs.c index 9d0ee7f0eeb..86dac106873 100644 --- a/net/ipv6/exthdrs.c +++ b/net/ipv6/exthdrs.c @@ -635,14 +635,17 @@ ipv6_renew_options(struct sock *sk, struct ipv6_txoptions *opt, struct ipv6_txoptions *opt2; int err; - if (newtype != IPV6_HOPOPTS && opt->hopopt) - tot_len += CMSG_ALIGN(ipv6_optlen(opt->hopopt)); - if (newtype != IPV6_RTHDRDSTOPTS && opt->dst0opt) - tot_len += CMSG_ALIGN(ipv6_optlen(opt->dst0opt)); - if (newtype != IPV6_RTHDR && opt->srcrt) - tot_len += CMSG_ALIGN(ipv6_optlen(opt->srcrt)); - if (newtype != IPV6_DSTOPTS && opt->dst1opt) - tot_len += CMSG_ALIGN(ipv6_optlen(opt->dst1opt)); + if (opt) { + if (newtype != IPV6_HOPOPTS && opt->hopopt) + tot_len += CMSG_ALIGN(ipv6_optlen(opt->hopopt)); + if (newtype != IPV6_RTHDRDSTOPTS && opt->dst0opt) + tot_len += CMSG_ALIGN(ipv6_optlen(opt->dst0opt)); + if (newtype != IPV6_RTHDR && opt->srcrt) + tot_len += CMSG_ALIGN(ipv6_optlen(opt->srcrt)); + if (newtype != IPV6_DSTOPTS && opt->dst1opt) + tot_len += CMSG_ALIGN(ipv6_optlen(opt->dst1opt)); + } + if (newopt && newoptlen) tot_len += CMSG_ALIGN(newoptlen); @@ -659,25 +662,25 @@ ipv6_renew_options(struct sock *sk, struct ipv6_txoptions *opt, opt2->tot_len = tot_len; p = (char *)(opt2 + 1); - err = ipv6_renew_option(opt->hopopt, newopt, newoptlen, + err = ipv6_renew_option(opt ? opt->hopopt : NULL, newopt, newoptlen, newtype != IPV6_HOPOPTS, &opt2->hopopt, &p); if (err) goto out; - err = ipv6_renew_option(opt->dst0opt, newopt, newoptlen, + err = ipv6_renew_option(opt ? opt->dst0opt : NULL, newopt, newoptlen, newtype != IPV6_RTHDRDSTOPTS, &opt2->dst0opt, &p); if (err) goto out; - err = ipv6_renew_option(opt->srcrt, newopt, newoptlen, + err = ipv6_renew_option(opt ? opt->srcrt : NULL, newopt, newoptlen, newtype != IPV6_RTHDR, - (struct ipv6_opt_hdr **)opt2->srcrt, &p); + (struct ipv6_opt_hdr **)&opt2->srcrt, &p); if (err) goto out; - err = ipv6_renew_option(opt->dst1opt, newopt, newoptlen, + err = ipv6_renew_option(opt ? opt->dst1opt : NULL, newopt, newoptlen, newtype != IPV6_DSTOPTS, &opt2->dst1opt, &p); if (err) From 5991c84421f3ea1e281954610a39e4993e5c39e4 Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Thu, 31 Aug 2006 15:21:50 -0700 Subject: [PATCH 077/115] [NET]: Rate limiting for socket allocation failure messages. This patch limits the warning messages when socket allocation failures happen. It happens under memory pressure. Signed-off-by: Akinobu Mita Signed-off-by: David S. Miller --- net/socket.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/net/socket.c b/net/socket.c index b4848ce0d6a..6d261bf206f 100644 --- a/net/socket.c +++ b/net/socket.c @@ -1178,7 +1178,8 @@ static int __sock_create(int family, int type, int protocol, struct socket **res */ if (!(sock = sock_alloc())) { - printk(KERN_WARNING "socket: no more sockets\n"); + if (net_ratelimit()) + printk(KERN_WARNING "socket: no more sockets\n"); err = -ENFILE; /* Not exactly a match, but its the closest posix thing */ goto out; From 0668b47205e42c04e9c1b594573be5a822ac7f09 Mon Sep 17 00:00:00 2001 From: Wei Dong Date: Thu, 31 Aug 2006 15:24:48 -0700 Subject: [PATCH 078/115] [IPV4]: Fix SNMPv2 "ipFragFails" counter error When I tested Linux kernel 2.6.17.7 about statistics "ipFragFails",found that this counter couldn't increase correctly. The criteria is RFC2011: RFC2011 ipFragFails OBJECT-TYPE SYNTAX Counter32 MAX-ACCESS read-only STATUS current DESCRIPTION "The number of IP datagrams that have been discarded because they needed to be fragmented at this entity but could not be, e.g., because their Don't Fragment flag was set." ::= { ip 18 } When I send big IP packet to a router with DF bit set to 1 which need to be fragmented, and router just sends an ICMP error message ICMP_FRAG_NEEDED but no increments for this counter(in the function ip_fragment). Signed-off-by: Wei Dong Signed-off-by: David S. Miller --- net/ipv4/ip_output.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index 4c20f554689..a2ede167e04 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -440,6 +440,7 @@ int ip_fragment(struct sk_buff *skb, int (*output)(struct sk_buff*)) iph = skb->nh.iph; if (unlikely((iph->frag_off & htons(IP_DF)) && !skb->local_df)) { + IP_INC_STATS(IPSTATS_MIB_FRAGFAILS); icmp_send(skb, ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED, htonl(dst_mtu(&rt->u.dst))); kfree_skb(skb); From 5a4e6dccbc0cd1b726820b782daebf887dcb95e9 Mon Sep 17 00:00:00 2001 From: Roland Dreier Date: Thu, 31 Aug 2006 16:43:06 -0700 Subject: [PATCH 079/115] IB/mthca: Use IRQ safe locks to protect allocation bitmaps It is supposed to be OK to call mthca_create_ah() and mthca_destroy_ah() from any context. However, for mem-full HCAs, these functions use the mthca_alloc() and mthca_free() bitmap helpers, and those helpers use non-IRQ-safe spin_lock() internally. Lockdep correctly warns that this could lead to a deadlock. Fix this by changing mthca_alloc() and mthca_free() to use spin_lock_irqsave(). Signed-off-by: Roland Dreier --- drivers/infiniband/hw/mthca/mthca_allocator.c | 15 +++++++++++---- 1 file changed, 11 insertions(+), 4 deletions(-) diff --git a/drivers/infiniband/hw/mthca/mthca_allocator.c b/drivers/infiniband/hw/mthca/mthca_allocator.c index 25157f57a6d..f930e55b58f 100644 --- a/drivers/infiniband/hw/mthca/mthca_allocator.c +++ b/drivers/infiniband/hw/mthca/mthca_allocator.c @@ -41,9 +41,11 @@ /* Trivial bitmap-based allocator */ u32 mthca_alloc(struct mthca_alloc *alloc) { + unsigned long flags; u32 obj; - spin_lock(&alloc->lock); + spin_lock_irqsave(&alloc->lock, flags); + obj = find_next_zero_bit(alloc->table, alloc->max, alloc->last); if (obj >= alloc->max) { alloc->top = (alloc->top + alloc->max) & alloc->mask; @@ -56,19 +58,24 @@ u32 mthca_alloc(struct mthca_alloc *alloc) } else obj = -1; - spin_unlock(&alloc->lock); + spin_unlock_irqrestore(&alloc->lock, flags); return obj; } void mthca_free(struct mthca_alloc *alloc, u32 obj) { + unsigned long flags; + obj &= alloc->max - 1; - spin_lock(&alloc->lock); + + spin_lock_irqsave(&alloc->lock, flags); + clear_bit(obj, alloc->table); alloc->last = min(alloc->last, obj); alloc->top = (alloc->top + alloc->max) & alloc->mask; - spin_unlock(&alloc->lock); + + spin_unlock_irqrestore(&alloc->lock, flags); } int mthca_alloc_init(struct mthca_alloc *alloc, u32 num, u32 mask, From 082fdd12b15c28ab74e5f6559fb3ba15bf9fd393 Mon Sep 17 00:00:00 2001 From: "juergen.mell@t-online.de" Date: Mon, 28 Aug 2006 13:53:53 -0700 Subject: [PATCH 080/115] USB floppy drive SAMSUNG SFD-321U/EP detected 8 times USB floppy drive SAMSUNG SFD-321U/EP detected 8 times Acked-by: mantel@suse.de Signed-off-by: Greg Kroah-Hartman --- drivers/usb/storage/unusual_devs.h | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/drivers/usb/storage/unusual_devs.h b/drivers/usb/storage/unusual_devs.h index 4a803d69fa3..43ed6b19535 100644 --- a/drivers/usb/storage/unusual_devs.h +++ b/drivers/usb/storage/unusual_devs.h @@ -599,6 +599,13 @@ UNUSUAL_DEV( 0x054c, 0x0099, 0x0000, 0x9999, US_SC_DEVICE, US_PR_DEVICE, NULL, US_FL_FIX_INQUIRY ), +/* floppy reports multiple luns */ +UNUSUAL_DEV( 0x055d, 0x2020, 0x0000, 0x0210, + "SAMSUNG", + "SFD-321U [FW 0C]", + US_SC_DEVICE, US_PR_DEVICE, NULL, + US_FL_SINGLE_LUN ), + UNUSUAL_DEV( 0x057b, 0x0000, 0x0000, 0x0299, "Y-E Data", From 6f8d9e26e7deecb1296c221aa979542bc5d63f20 Mon Sep 17 00:00:00 2001 From: Jeremy Roberson Date: Mon, 28 Aug 2006 19:58:28 -0700 Subject: [PATCH 081/115] hid-core.c: Adds all GTCO CalComp Digitizers and InterWrite School Products to blacklist Adds all GTCO CalComp Digitizers and InterWrite School Products to hid-core.c blacklist. Signed-off-by: Jeremy A. Roberson Signed-off-by: Greg Kroah-Hartman --- drivers/usb/input/hid-core.c | 92 ++++++++++++++++++++++++++++++++++++ 1 file changed, 92 insertions(+) diff --git a/drivers/usb/input/hid-core.c b/drivers/usb/input/hid-core.c index 8ea9c915fbf..59a92149a0d 100644 --- a/drivers/usb/input/hid-core.c +++ b/drivers/usb/input/hid-core.c @@ -1411,6 +1411,53 @@ void hid_init_reports(struct hid_device *hid) warn("timeout initializing reports"); } +#define USB_VENDOR_ID_GTCO 0x078c +#define USB_DEVICE_ID_GTCO_90 0x0090 +#define USB_DEVICE_ID_GTCO_100 0x0100 +#define USB_DEVICE_ID_GTCO_101 0x0101 +#define USB_DEVICE_ID_GTCO_103 0x0103 +#define USB_DEVICE_ID_GTCO_104 0x0104 +#define USB_DEVICE_ID_GTCO_105 0x0105 +#define USB_DEVICE_ID_GTCO_106 0x0106 +#define USB_DEVICE_ID_GTCO_107 0x0107 +#define USB_DEVICE_ID_GTCO_108 0x0108 +#define USB_DEVICE_ID_GTCO_200 0x0200 +#define USB_DEVICE_ID_GTCO_201 0x0201 +#define USB_DEVICE_ID_GTCO_202 0x0202 +#define USB_DEVICE_ID_GTCO_203 0x0203 +#define USB_DEVICE_ID_GTCO_204 0x0204 +#define USB_DEVICE_ID_GTCO_205 0x0205 +#define USB_DEVICE_ID_GTCO_206 0x0206 +#define USB_DEVICE_ID_GTCO_207 0x0207 +#define USB_DEVICE_ID_GTCO_300 0x0300 +#define USB_DEVICE_ID_GTCO_301 0x0301 +#define USB_DEVICE_ID_GTCO_302 0x0302 +#define USB_DEVICE_ID_GTCO_303 0x0303 +#define USB_DEVICE_ID_GTCO_304 0x0304 +#define USB_DEVICE_ID_GTCO_305 0x0305 +#define USB_DEVICE_ID_GTCO_306 0x0306 +#define USB_DEVICE_ID_GTCO_307 0x0307 +#define USB_DEVICE_ID_GTCO_308 0x0308 +#define USB_DEVICE_ID_GTCO_309 0x0309 +#define USB_DEVICE_ID_GTCO_400 0x0400 +#define USB_DEVICE_ID_GTCO_401 0x0401 +#define USB_DEVICE_ID_GTCO_402 0x0402 +#define USB_DEVICE_ID_GTCO_403 0x0403 +#define USB_DEVICE_ID_GTCO_404 0x0404 +#define USB_DEVICE_ID_GTCO_404 0x0405 +#define USB_DEVICE_ID_GTCO_500 0x0500 +#define USB_DEVICE_ID_GTCO_501 0x0501 +#define USB_DEVICE_ID_GTCO_502 0x0502 +#define USB_DEVICE_ID_GTCO_503 0x0503 +#define USB_DEVICE_ID_GTCO_504 0x0504 +#define USB_DEVICE_ID_GTCO_1000 0x1000 +#define USB_DEVICE_ID_GTCO_1001 0x1001 +#define USB_DEVICE_ID_GTCO_1002 0x1002 +#define USB_DEVICE_ID_GTCO_1003 0x1003 +#define USB_DEVICE_ID_GTCO_1004 0x1004 +#define USB_DEVICE_ID_GTCO_1005 0x1005 +#define USB_DEVICE_ID_GTCO_1006 0x1006 + #define USB_VENDOR_ID_WACOM 0x056a #define USB_DEVICE_ID_WACOM_PENPARTNER 0x0000 #define USB_DEVICE_ID_WACOM_GRAPHIRE 0x0010 @@ -1588,6 +1635,51 @@ static const struct hid_blacklist { { USB_VENDOR_ID_GLAB, USB_DEVICE_ID_0_8_8_IF_KIT, HID_QUIRK_IGNORE }, { USB_VENDOR_ID_GRIFFIN, USB_DEVICE_ID_POWERMATE, HID_QUIRK_IGNORE }, { USB_VENDOR_ID_GRIFFIN, USB_DEVICE_ID_SOUNDKNOB, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_90, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_100, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_101, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_103, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_104, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_105, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_106, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_107, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_108, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_200, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_201, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_202, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_203, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_204, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_205, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_206, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_207, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_300, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_301, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_302, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_303, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_304, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_305, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_306, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_307, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_308, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_309, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_400, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_401, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_402, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_403, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_404, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_404, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_500, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_501, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_502, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_503, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_504, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_1000, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_1001, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_1002, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_1003, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_1004, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_1005, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_1006, HID_QUIRK_IGNORE }, { USB_VENDOR_ID_KBGEAR, USB_DEVICE_ID_KBGEAR_JAMSTUDIO, HID_QUIRK_IGNORE }, { USB_VENDOR_ID_LD, USB_DEVICE_ID_LD_CASSY, HID_QUIRK_IGNORE }, { USB_VENDOR_ID_LD, USB_DEVICE_ID_LD_POCKETCASSY, HID_QUIRK_IGNORE }, From ea186651d5a7b822c855de5de505c5c19812bf0b Mon Sep 17 00:00:00 2001 From: Ping Cheng Date: Fri, 14 Jul 2006 16:50:04 -0700 Subject: [PATCH 082/115] USB: add all wacom device to hid-core.c blacklist Add all Wacom devices to hid-core.c blacklist Signed-off-by: Ping Cheng Signed-off-by: Greg Kroah-Hartman --- drivers/usb/input/hid-core.c | 57 +++--------------------------------- 1 file changed, 4 insertions(+), 53 deletions(-) diff --git a/drivers/usb/input/hid-core.c b/drivers/usb/input/hid-core.c index 59a92149a0d..acb24c6219d 100644 --- a/drivers/usb/input/hid-core.c +++ b/drivers/usb/input/hid-core.c @@ -1459,16 +1459,6 @@ void hid_init_reports(struct hid_device *hid) #define USB_DEVICE_ID_GTCO_1006 0x1006 #define USB_VENDOR_ID_WACOM 0x056a -#define USB_DEVICE_ID_WACOM_PENPARTNER 0x0000 -#define USB_DEVICE_ID_WACOM_GRAPHIRE 0x0010 -#define USB_DEVICE_ID_WACOM_INTUOS 0x0020 -#define USB_DEVICE_ID_WACOM_PL 0x0030 -#define USB_DEVICE_ID_WACOM_INTUOS2 0x0040 -#define USB_DEVICE_ID_WACOM_VOLITO 0x0060 -#define USB_DEVICE_ID_WACOM_PTU 0x0003 -#define USB_DEVICE_ID_WACOM_INTUOS3 0x00B0 -#define USB_DEVICE_ID_WACOM_CINTIQ 0x003F -#define USB_DEVICE_ID_WACOM_DTF 0x00C0 #define USB_VENDOR_ID_ACECAD 0x0460 #define USB_DEVICE_ID_ACECAD_FLAIR 0x0004 @@ -1709,49 +1699,6 @@ static const struct hid_blacklist { { USB_VENDOR_ID_VERNIER, USB_DEVICE_ID_VERNIER_GOTEMP, HID_QUIRK_IGNORE }, { USB_VENDOR_ID_VERNIER, USB_DEVICE_ID_VERNIER_SKIP, HID_QUIRK_IGNORE }, { USB_VENDOR_ID_VERNIER, USB_DEVICE_ID_VERNIER_CYCLOPS, HID_QUIRK_IGNORE }, - { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_PENPARTNER, HID_QUIRK_IGNORE }, - { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_GRAPHIRE, HID_QUIRK_IGNORE }, - { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_GRAPHIRE + 1, HID_QUIRK_IGNORE }, - { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_GRAPHIRE + 2, HID_QUIRK_IGNORE }, - { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_GRAPHIRE + 3, HID_QUIRK_IGNORE }, - { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_GRAPHIRE + 4, HID_QUIRK_IGNORE }, - { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_INTUOS, HID_QUIRK_IGNORE }, - { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_INTUOS + 1, HID_QUIRK_IGNORE }, - { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_INTUOS + 2, HID_QUIRK_IGNORE }, - { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_INTUOS + 3, HID_QUIRK_IGNORE }, - { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_INTUOS + 4, HID_QUIRK_IGNORE }, - { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_PL, HID_QUIRK_IGNORE }, - { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_PL + 1, HID_QUIRK_IGNORE }, - { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_PL + 2, HID_QUIRK_IGNORE }, - { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_PL + 3, HID_QUIRK_IGNORE }, - { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_PL + 4, HID_QUIRK_IGNORE }, - { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_PL + 5, HID_QUIRK_IGNORE }, - { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_PL + 7, HID_QUIRK_IGNORE }, - { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_PL + 8, HID_QUIRK_IGNORE }, - { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_PL + 9, HID_QUIRK_IGNORE }, - { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_INTUOS2 + 1, HID_QUIRK_IGNORE }, - { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_INTUOS2 + 2, HID_QUIRK_IGNORE }, - { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_INTUOS2 + 3, HID_QUIRK_IGNORE }, - { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_INTUOS2 + 4, HID_QUIRK_IGNORE }, - { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_INTUOS2 + 5, HID_QUIRK_IGNORE }, - { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_INTUOS2 + 7, HID_QUIRK_IGNORE }, - { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_VOLITO, HID_QUIRK_IGNORE }, - { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_VOLITO + 1, HID_QUIRK_IGNORE }, - { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_VOLITO + 2, HID_QUIRK_IGNORE }, - { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_VOLITO + 3, HID_QUIRK_IGNORE }, - { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_VOLITO + 4, HID_QUIRK_IGNORE }, - { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_GRAPHIRE + 5, HID_QUIRK_IGNORE }, - { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_GRAPHIRE + 6, HID_QUIRK_IGNORE }, - { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_PTU, HID_QUIRK_IGNORE }, - { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_INTUOS3, HID_QUIRK_IGNORE }, - { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_INTUOS3 + 1, HID_QUIRK_IGNORE }, - { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_INTUOS3 + 2, HID_QUIRK_IGNORE }, - { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_INTUOS3 + 3, HID_QUIRK_IGNORE }, - { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_INTUOS3 + 4, HID_QUIRK_IGNORE }, - { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_INTUOS3 + 5, HID_QUIRK_IGNORE }, - { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_CINTIQ, HID_QUIRK_IGNORE }, - { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_DTF, HID_QUIRK_IGNORE }, - { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_DTF + 3, HID_QUIRK_IGNORE }, { USB_VENDOR_ID_WISEGROUP, USB_DEVICE_ID_4_PHIDGETSERVO_20, HID_QUIRK_IGNORE }, { USB_VENDOR_ID_WISEGROUP, USB_DEVICE_ID_1_PHIDGETSERVO_20, HID_QUIRK_IGNORE }, { USB_VENDOR_ID_YEALINK, USB_DEVICE_ID_YEALINK_P1K_P4K_B2K, HID_QUIRK_IGNORE }, @@ -1870,6 +1817,10 @@ static struct hid_device *usb_hid_configure(struct usb_interface *intf) char *rdesc; int n, len, insize = 0; + /* Ignore all Wacom devices */ + if (dev->descriptor.idVendor == USB_VENDOR_ID_WACOM) + return NULL; + for (n = 0; hid_blacklist[n].idVendor; n++) if ((hid_blacklist[n].idVendor == le16_to_cpu(dev->descriptor.idVendor)) && (hid_blacklist[n].idProduct == le16_to_cpu(dev->descriptor.idProduct))) From 789851cf0005b946557340c9bbfc7728906cdbfc Mon Sep 17 00:00:00 2001 From: David Brownell Date: Mon, 21 Aug 2006 15:26:38 -0700 Subject: [PATCH 083/115] usb gadget: g_ether spinlock recursion fix The new spinlock debug code turned up a spinlock recursion bug in the Ethernet gadget driver on a disconnect path; it would show up with any UDC driver where the cancellation of active requests was synchronous, rather than e.g. delayed until a controller's completion IRQ. That recursion is fixed here by creating and using a new spinlock to protect the relevant lists. Signed-off-by: David Brownell Signed-off-by: Greg Kroah-Hartman --- drivers/usb/gadget/ether.c | 45 ++++++++++++++++++++++++++------------ 1 file changed, 31 insertions(+), 14 deletions(-) diff --git a/drivers/usb/gadget/ether.c b/drivers/usb/gadget/ether.c index 4fe1bec1c25..30299c620d9 100644 --- a/drivers/usb/gadget/ether.c +++ b/drivers/usb/gadget/ether.c @@ -117,6 +117,8 @@ struct eth_dev { struct usb_ep *in_ep, *out_ep, *status_ep; const struct usb_endpoint_descriptor *in, *out, *status; + + spinlock_t req_lock; struct list_head tx_reqs, rx_reqs; struct net_device *net; @@ -1066,21 +1068,31 @@ static void eth_reset_config (struct eth_dev *dev) */ if (dev->in) { usb_ep_disable (dev->in_ep); + spin_lock(&dev->req_lock); while (likely (!list_empty (&dev->tx_reqs))) { req = container_of (dev->tx_reqs.next, struct usb_request, list); list_del (&req->list); + + spin_unlock(&dev->req_lock); usb_ep_free_request (dev->in_ep, req); + spin_lock(&dev->req_lock); } + spin_unlock(&dev->req_lock); } if (dev->out) { usb_ep_disable (dev->out_ep); + spin_lock(&dev->req_lock); while (likely (!list_empty (&dev->rx_reqs))) { req = container_of (dev->rx_reqs.next, struct usb_request, list); list_del (&req->list); + + spin_unlock(&dev->req_lock); usb_ep_free_request (dev->out_ep, req); + spin_lock(&dev->req_lock); } + spin_unlock(&dev->req_lock); } if (dev->status) { @@ -1659,9 +1671,9 @@ enomem: if (retval) { DEBUG (dev, "rx submit --> %d\n", retval); dev_kfree_skb_any (skb); - spin_lock (&dev->lock); + spin_lock(&dev->req_lock); list_add (&req->list, &dev->rx_reqs); - spin_unlock (&dev->lock); + spin_unlock(&dev->req_lock); } return retval; } @@ -1730,8 +1742,9 @@ quiesce: dev_kfree_skb_any (skb); if (!netif_running (dev->net)) { clean: - /* nobody reading rx_reqs, so no dev->lock */ + spin_lock(&dev->req_lock); list_add (&req->list, &dev->rx_reqs); + spin_unlock(&dev->req_lock); req = NULL; } if (req) @@ -1782,15 +1795,18 @@ static int alloc_requests (struct eth_dev *dev, unsigned n, gfp_t gfp_flags) { int status; + spin_lock(&dev->req_lock); status = prealloc (&dev->tx_reqs, dev->in_ep, n, gfp_flags); if (status < 0) goto fail; status = prealloc (&dev->rx_reqs, dev->out_ep, n, gfp_flags); if (status < 0) goto fail; - return 0; + goto done; fail: DEBUG (dev, "can't alloc requests\n"); +done: + spin_unlock(&dev->req_lock); return status; } @@ -1800,21 +1816,21 @@ static void rx_fill (struct eth_dev *dev, gfp_t gfp_flags) unsigned long flags; /* fill unused rxq slots with some skb */ - spin_lock_irqsave (&dev->lock, flags); + spin_lock_irqsave(&dev->req_lock, flags); while (!list_empty (&dev->rx_reqs)) { req = container_of (dev->rx_reqs.next, struct usb_request, list); list_del_init (&req->list); - spin_unlock_irqrestore (&dev->lock, flags); + spin_unlock_irqrestore(&dev->req_lock, flags); if (rx_submit (dev, req, gfp_flags) < 0) { defer_kevent (dev, WORK_RX_MEMORY); return; } - spin_lock_irqsave (&dev->lock, flags); + spin_lock_irqsave(&dev->req_lock, flags); } - spin_unlock_irqrestore (&dev->lock, flags); + spin_unlock_irqrestore(&dev->req_lock, flags); } static void eth_work (void *_dev) @@ -1848,9 +1864,9 @@ static void tx_complete (struct usb_ep *ep, struct usb_request *req) } dev->stats.tx_packets++; - spin_lock (&dev->lock); + spin_lock(&dev->req_lock); list_add (&req->list, &dev->tx_reqs); - spin_unlock (&dev->lock); + spin_unlock(&dev->req_lock); dev_kfree_skb_any (skb); atomic_dec (&dev->tx_qlen); @@ -1896,12 +1912,12 @@ static int eth_start_xmit (struct sk_buff *skb, struct net_device *net) /* ignores USB_CDC_PACKET_TYPE_DIRECTED */ } - spin_lock_irqsave (&dev->lock, flags); + spin_lock_irqsave(&dev->req_lock, flags); req = container_of (dev->tx_reqs.next, struct usb_request, list); list_del (&req->list); if (list_empty (&dev->tx_reqs)) netif_stop_queue (net); - spin_unlock_irqrestore (&dev->lock, flags); + spin_unlock_irqrestore(&dev->req_lock, flags); /* no buffer copies needed, unless the network stack did it * or the hardware can't use skb buffers. @@ -1955,11 +1971,11 @@ static int eth_start_xmit (struct sk_buff *skb, struct net_device *net) drop: dev->stats.tx_dropped++; dev_kfree_skb_any (skb); - spin_lock_irqsave (&dev->lock, flags); + spin_lock_irqsave(&dev->req_lock, flags); if (list_empty (&dev->tx_reqs)) netif_start_queue (net); list_add (&req->list, &dev->tx_reqs); - spin_unlock_irqrestore (&dev->lock, flags); + spin_unlock_irqrestore(&dev->req_lock, flags); } return 0; } @@ -2378,6 +2394,7 @@ autoconf_fail: return status; dev = netdev_priv(net); spin_lock_init (&dev->lock); + spin_lock_init (&dev->req_lock); INIT_WORK (&dev->work, eth_work, dev); INIT_LIST_HEAD (&dev->tx_reqs); INIT_LIST_HEAD (&dev->rx_reqs); From 7ceb932f489e86b555258e5f7d7f061f9c1863eb Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Mon, 21 Aug 2006 11:58:50 -0400 Subject: [PATCH 084/115] UHCI: don't stop at an Iso error Unlike other sorts of endpoint queues, Isochronous queues don't stop when an error is encountered. This patch (as772) fixes the scanning routine in uhci-hcd, to make it keep on going when it finds an Iso error. Signed-off-by: Alan Stern Signed-off-by: Greg Kroah-Hartman --- drivers/usb/host/uhci-q.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/usb/host/uhci-q.c b/drivers/usb/host/uhci-q.c index 66c3f61bc9d..026dad06f82 100644 --- a/drivers/usb/host/uhci-q.c +++ b/drivers/usb/host/uhci-q.c @@ -1348,7 +1348,7 @@ static void uhci_scan_qh(struct uhci_hcd *uhci, struct uhci_qh *qh, } uhci_giveback_urb(uhci, qh, urb, regs); - if (status < 0) + if (status < 0 && qh->type != USB_ENDPOINT_XFER_ISOC) break; } From f84c749f1ce6a54b84b098f5fa53a01686401fac Mon Sep 17 00:00:00 2001 From: Phil Dibowitz Date: Wed, 23 Aug 2006 21:11:01 -0700 Subject: [PATCH 085/115] USB Storage: Remove the finecam3 unusual_devs entry This patch removes the Kyocera Finecam L3 entry in unusual devices originally submitted by Michael Krauth and Alessandro Fracchetti given that Gerriet finds he doesn't need it and Alessandro confirms it isn't needed anymore as well. Signed-off-by: Phil Dibowitz Signed-off-by: Greg Kroah-Hartman --- drivers/usb/storage/unusual_devs.h | 10 ---------- 1 file changed, 10 deletions(-) diff --git a/drivers/usb/storage/unusual_devs.h b/drivers/usb/storage/unusual_devs.h index 43ed6b19535..793045167af 100644 --- a/drivers/usb/storage/unusual_devs.h +++ b/drivers/usb/storage/unusual_devs.h @@ -241,16 +241,6 @@ UNUSUAL_DEV( 0x0482, 0x0103, 0x0100, 0x0100, "Finecam S5", US_SC_DEVICE, US_PR_DEVICE, NULL, US_FL_FIX_INQUIRY), -/* Patch for Kyocera Finecam L3 - * Submitted by Michael Krauth - * and Alessandro Fracchetti - */ -UNUSUAL_DEV( 0x0482, 0x0105, 0x0100, 0x0100, - "Kyocera", - "Finecam L3", - US_SC_SCSI, US_PR_BULK, NULL, - US_FL_FIX_INQUIRY), - /* Reported by Paul Stewart * This entry is needed because the device reports Sub=ff */ UNUSUAL_DEV( 0x04a4, 0x0004, 0x0001, 0x0001, From c9c770ed1fe2698c9a3b18e5fc12d1262e099c0c Mon Sep 17 00:00:00 2001 From: Phil Dibowitz Date: Sun, 27 Aug 2006 23:54:37 -0700 Subject: [PATCH 086/115] USB Storage: unusual_devs.h for Sony Ericsson M600i This entry was sent in by Emmanuel Vasilakis , turned into a patch by yours truly. Signed-off-by: Phil Dibowitz Signed-off-by: Greg Kroah-Hartman --- drivers/usb/storage/unusual_devs.h | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/drivers/usb/storage/unusual_devs.h b/drivers/usb/storage/unusual_devs.h index 793045167af..b130e170b4a 100644 --- a/drivers/usb/storage/unusual_devs.h +++ b/drivers/usb/storage/unusual_devs.h @@ -1254,6 +1254,13 @@ UNUSUAL_DEV( 0x0fce, 0xd008, 0x0000, 0x0000, US_SC_DEVICE, US_PR_DEVICE, NULL, US_FL_NO_WP_DETECT ), +/* Reported by Emmanuel Vasilakis */ +UNUSUAL_DEV( 0x0fce, 0xe031, 0x0000, 0x0000, + "Sony Ericsson", + "M600i", + US_SC_DEVICE, US_PR_DEVICE, NULL, + US_FL_FIX_CAPACITY ), + /* Reported by Kevin Cernekee * Tested on hardware version 1.10. * Entry is needed only for the initializer function override. From eff674a514bd3f59e0cae9b843e0665b576a5ed8 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Mon, 14 Aug 2006 23:11:09 -0700 Subject: [PATCH 087/115] USB: rtl8150_disconnect() needs tasklet_kill() We need to wait until any currently-running handler has completed. Fixes an unplug-time oops reported by "Miles Lane" . Cc: "Petko Manolov" Signed-off-by: Andrew Morton Signed-off-by: Greg Kroah-Hartman --- drivers/usb/net/rtl8150.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/usb/net/rtl8150.c b/drivers/usb/net/rtl8150.c index bd09232ce13..a72685b9606 100644 --- a/drivers/usb/net/rtl8150.c +++ b/drivers/usb/net/rtl8150.c @@ -972,6 +972,7 @@ static void rtl8150_disconnect(struct usb_interface *intf) if (dev) { set_bit(RTL8150_UNPLUG, &dev->flags); tasklet_disable(&dev->tl); + tasklet_kill(&dev->tl); unregister_netdev(dev->netdev); unlink_all_urbs(dev); free_all_urbs(dev); From 1ae4f9ba84b94b85d995a6ae0064b869ff15b080 Mon Sep 17 00:00:00 2001 From: Mark Hindley Date: Mon, 28 Aug 2006 20:43:25 +0100 Subject: [PATCH 088/115] USB: Add VIA quirk fixup for VT8235 usb2 Patch to add VIA PCI quirk for Enhanced/Extended USB on VT8235 southbridge. It is needed in order to use EHCI/USB 2.0 with ACPI. Without it IRQs are not routed correctly, you get an "Unlink after no-IRQ?" error and the device is unusable. I belive this could also be a fix for Bugzilla Bug 5835. Signed-off-by: Mark Hindley Signed-off-by: Greg Kroah-Hartman --- drivers/pci/quirks.c | 1 + include/linux/pci_ids.h | 1 + 2 files changed, 2 insertions(+) diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c index 73177429fe7..17e709e7d72 100644 --- a/drivers/pci/quirks.c +++ b/drivers/pci/quirks.c @@ -667,6 +667,7 @@ DECLARE_PCI_FIXUP_ENABLE(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_82C586_0, quirk_vi DECLARE_PCI_FIXUP_ENABLE(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_82C586_1, quirk_via_irq); DECLARE_PCI_FIXUP_ENABLE(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_82C586_2, quirk_via_irq); DECLARE_PCI_FIXUP_ENABLE(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_82C586_3, quirk_via_irq); +DECLARE_PCI_FIXUP_ENABLE(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_8235_USB_2, quirk_via_irq); DECLARE_PCI_FIXUP_ENABLE(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_82C686, quirk_via_irq); DECLARE_PCI_FIXUP_ENABLE(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_82C686_4, quirk_via_irq); DECLARE_PCI_FIXUP_ENABLE(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_82C686_5, quirk_via_irq); diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index 4c2839eab7f..c91164ea3de 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -1292,6 +1292,7 @@ #define PCI_DEVICE_ID_VIA_8367_0 0x3099 #define PCI_DEVICE_ID_VIA_8653_0 0x3101 #define PCI_DEVICE_ID_VIA_8622 0x3102 +#define PCI_DEVICE_ID_VIA_8235_USB_2 0x3104 #define PCI_DEVICE_ID_VIA_8233C_0 0x3109 #define PCI_DEVICE_ID_VIA_8361 0x3112 #define PCI_DEVICE_ID_VIA_XM266 0x3116 From 5df3d8b53f436a26fad8077b1ceb39a7708e95ec Mon Sep 17 00:00:00 2001 From: Nobuhiro Iwamatsu Date: Tue, 29 Aug 2006 10:47:41 +0300 Subject: [PATCH 089/115] USB: Support for ELECOM LD-USB20 in pegasus This patch is support LD-USB20 of the USB LAN device. http://www2.elecom.co.jp/products/LD-USB20.html ( Japanese only ) I am using this device. And, I confirmed work by using this patch. Signed-off-by: Nobuhiro Iwamatsu Acked-by: Petko Manolov Signed-off-by: Greg Kroah-Hartman --- drivers/usb/net/pegasus.h | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/usb/net/pegasus.h b/drivers/usb/net/pegasus.h index a54752ce149..006438069b6 100644 --- a/drivers/usb/net/pegasus.h +++ b/drivers/usb/net/pegasus.h @@ -131,6 +131,7 @@ struct usb_eth_dev { #define VENDOR_COREGA 0x07aa #define VENDOR_DLINK 0x2001 #define VENDOR_ELCON 0x0db7 +#define VENDOR_ELECOM 0x056e #define VENDOR_ELSA 0x05cc #define VENDOR_GIGABYTE 0x1044 #define VENDOR_HAWKING 0x0e66 @@ -233,6 +234,8 @@ PEGASUS_DEV( "D-Link DSB-650", VENDOR_DLINK, 0xabc1, DEFAULT_GPIO_RESET ) PEGASUS_DEV( "GOLDPFEIL USB Adapter", VENDOR_ELCON, 0x0002, DEFAULT_GPIO_RESET | PEGASUS_II | HAS_HOME_PNA ) +PEGASUS_DEV( "ELECOM USB Ethernet LD-USB20", VENDOR_ELECOM, 0x4010, + DEFAULT_GPIO_RESET | PEGASUS_II ) PEGASUS_DEV( "EasiDock Ethernet", VENDOR_MOBILITY, 0x0304, DEFAULT_GPIO_RESET ) PEGASUS_DEV( "Elsa Micolink USB2Ethernet", VENDOR_ELSA, 0x3000, From db59b464f8708cdba857f16b183cff0b7466d6b5 Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Thu, 31 Aug 2006 14:18:39 -0400 Subject: [PATCH 090/115] uhci-hcd: fix list access bug When skipping to the last TD of an URB, go to the _last_ entry in the list instead of the _first_ entry (as780). This fixes Bugzilla #6747 and possibly others. Signed-off-by: Alan Stern Signed-off-by: Greg Kroah-Hartman --- drivers/usb/host/uhci-q.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/usb/host/uhci-q.c b/drivers/usb/host/uhci-q.c index 026dad06f82..431e8f31f1a 100644 --- a/drivers/usb/host/uhci-q.c +++ b/drivers/usb/host/uhci-q.c @@ -372,7 +372,7 @@ static void uhci_fixup_toggles(struct uhci_qh *qh, int skip_first) * need to change any toggles in this URB */ td = list_entry(urbp->td_list.next, struct uhci_td, list); if (toggle > 1 || uhci_toggle(td_token(td)) == toggle) { - td = list_entry(urbp->td_list.next, struct uhci_td, + td = list_entry(urbp->td_list.prev, struct uhci_td, list); toggle = uhci_toggle(td_token(td)) ^ 1; From a302eb4e4602d6444ae75a0e516fb2f2c62d6642 Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Thu, 31 Aug 2006 21:27:34 -0700 Subject: [PATCH 091/115] [PATCH] ZVC: Overstep counters Increments and decrements are usually grouped rather than mixed. We can optimize the inc and dec functions for that case. Increment and decrement the counters by 50% more than the threshold in those cases and set the differential accordingly. This decreases the need to update the atomic counters. The idea came originally from Andrew Morton. The overstepping alone was sufficient to address the contention issue found when updating the global and the per zone counters from 160 processors. Also remove some code in dec_zone_page_state. Signed-off-by: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/vmstat.c | 20 +++++--------------- 1 file changed, 5 insertions(+), 15 deletions(-) diff --git a/mm/vmstat.c b/mm/vmstat.c index dfdf2413390..3799a0f7543 100644 --- a/mm/vmstat.c +++ b/mm/vmstat.c @@ -190,8 +190,8 @@ static void __inc_zone_state(struct zone *zone, enum zone_stat_item item) (*p)++; if (unlikely(*p > STAT_THRESHOLD)) { - zone_page_state_add(*p, zone, item); - *p = 0; + zone_page_state_add(*p + STAT_THRESHOLD / 2, zone, item); + *p = -STAT_THRESHOLD / 2; } } @@ -209,8 +209,8 @@ void __dec_zone_page_state(struct page *page, enum zone_stat_item item) (*p)--; if (unlikely(*p < -STAT_THRESHOLD)) { - zone_page_state_add(*p, zone, item); - *p = 0; + zone_page_state_add(*p - STAT_THRESHOLD / 2, zone, item); + *p = STAT_THRESHOLD /2; } } EXPORT_SYMBOL(__dec_zone_page_state); @@ -239,19 +239,9 @@ EXPORT_SYMBOL(inc_zone_page_state); void dec_zone_page_state(struct page *page, enum zone_stat_item item) { unsigned long flags; - struct zone *zone; - s8 *p; - zone = page_zone(page); local_irq_save(flags); - p = diff_pointer(zone, item); - - (*p)--; - - if (unlikely(*p < -STAT_THRESHOLD)) { - zone_page_state_add(*p, zone, item); - *p = 0; - } + __dec_zone_page_state(page, item); local_irq_restore(flags); } EXPORT_SYMBOL(dec_zone_page_state); From df9ecaba3f152d1ea79f2a5e0b87505e03f47590 Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Thu, 31 Aug 2006 21:27:35 -0700 Subject: [PATCH 092/115] [PATCH] ZVC: Scale thresholds depending on the size of the system The ZVC counter update threshold is currently set to a fixed value of 32. This patch sets up the threshold depending on the number of processors and the sizes of the zones in the system. With the current threshold of 32, I was able to observe slight contention when more than 130-140 processors concurrently updated the counters. The contention vanished when I either increased the threshold to 64 or used Andrew's idea of overstepping the interval (see ZVC overstep patch). However, we saw contention again at 220-230 processors. So we need higher values for larger systems. But the current default is already a bit of an overkill for smaller systems. Some systems have tiny zones where precision matters. For example i386 and x86_64 have 16M DMA zones and either 900M ZONE_NORMAL or ZONE_DMA32. These are even present on SMP and NUMA systems. The patch here sets up a threshold based on the number of processors in the system and the size of the zone that these counters are used for. The threshold should grow logarithmically, so we use fls() as an easy approximation. Results of tests on a system with 1024 processors (4TB RAM) The following output is from a test allocating 1GB of memory concurrently on each processor (Forking the process. So contention on mmap_sem and the pte locks is not a factor): X MIN TYPE: CPUS WALL WALL SYS USER TOTCPU fork 1 0.552 0.552 0.540 0.012 0.552 fork 4 0.552 0.548 2.164 0.036 2.200 fork 16 0.564 0.548 8.812 0.164 8.976 fork 128 0.580 0.572 72.204 1.208 73.412 fork 256 1.300 0.660 310.400 2.160 312.560 fork 512 3.512 0.696 1526.836 4.816 1531.652 fork 1020 20.024 0.700 17243.176 6.688 17249.863 So a threshold of 32 is fine up to 128 processors. At 256 processors contention becomes a factor. Overstepping the counter (earlier patch) improves the numbers a bit: fork 4 0.552 0.548 2.164 0.040 2.204 fork 16 0.552 0.548 8.640 0.148 8.788 fork 128 0.556 0.548 69.676 0.956 70.632 fork 256 0.876 0.636 212.468 2.108 214.576 fork 512 2.276 0.672 997.324 4.260 1001.584 fork 1020 13.564 0.680 11586.436 6.088 11592.523 Still contention at 512 and 1020. Contention at 1020 is down by a third. 256 still has a slight bit of contention. After this patch the counter threshold will be set to 125 which reduces contention significantly: fork 128 0.560 0.548 69.776 0.932 70.708 fork 256 0.636 0.556 143.460 2.036 145.496 fork 512 0.640 0.548 284.244 4.236 288.480 fork 1020 1.500 0.588 1326.152 8.892 1335.044 [akpm@osdl.org: !SMP build fix] Signed-off-by: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mmzone.h | 1 + mm/vmstat.c | 139 +++++++++++++++++++++++++++++++++++------ 2 files changed, 120 insertions(+), 20 deletions(-) diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index 656b588a9f9..f45163c528e 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -77,6 +77,7 @@ struct per_cpu_pages { struct per_cpu_pageset { struct per_cpu_pages pcp[2]; /* 0: hot. 1: cold */ #ifdef CONFIG_SMP + s8 stat_threshold; s8 vm_stat_diff[NR_VM_ZONE_STAT_ITEMS]; #endif } ____cacheline_aligned_in_smp; diff --git a/mm/vmstat.c b/mm/vmstat.c index 3799a0f7543..c1b5f4106b3 100644 --- a/mm/vmstat.c +++ b/mm/vmstat.c @@ -12,6 +12,7 @@ #include #include #include +#include void __get_zone_counts(unsigned long *active, unsigned long *inactive, unsigned long *free, struct pglist_data *pgdat) @@ -114,17 +115,72 @@ EXPORT_SYMBOL(vm_stat); #ifdef CONFIG_SMP -#define STAT_THRESHOLD 32 +static int calculate_threshold(struct zone *zone) +{ + int threshold; + int mem; /* memory in 128 MB units */ + + /* + * The threshold scales with the number of processors and the amount + * of memory per zone. More memory means that we can defer updates for + * longer, more processors could lead to more contention. + * fls() is used to have a cheap way of logarithmic scaling. + * + * Some sample thresholds: + * + * Threshold Processors (fls) Zonesize fls(mem+1) + * ------------------------------------------------------------------ + * 8 1 1 0.9-1 GB 4 + * 16 2 2 0.9-1 GB 4 + * 20 2 2 1-2 GB 5 + * 24 2 2 2-4 GB 6 + * 28 2 2 4-8 GB 7 + * 32 2 2 8-16 GB 8 + * 4 2 2 <128M 1 + * 30 4 3 2-4 GB 5 + * 48 4 3 8-16 GB 8 + * 32 8 4 1-2 GB 4 + * 32 8 4 0.9-1GB 4 + * 10 16 5 <128M 1 + * 40 16 5 900M 4 + * 70 64 7 2-4 GB 5 + * 84 64 7 4-8 GB 6 + * 108 512 9 4-8 GB 6 + * 125 1024 10 8-16 GB 8 + * 125 1024 10 16-32 GB 9 + */ + + mem = zone->present_pages >> (27 - PAGE_SHIFT); + + threshold = 2 * fls(num_online_cpus()) * (1 + fls(mem)); + + /* + * Maximum threshold is 125 + */ + threshold = min(125, threshold); + + return threshold; +} /* - * Determine pointer to currently valid differential byte given a zone and - * the item number. - * - * Preemption must be off + * Refresh the thresholds for each zone. */ -static inline s8 *diff_pointer(struct zone *zone, enum zone_stat_item item) +static void refresh_zone_stat_thresholds(void) { - return &zone_pcp(zone, smp_processor_id())->vm_stat_diff[item]; + struct zone *zone; + int cpu; + int threshold; + + for_each_zone(zone) { + + if (!zone->present_pages) + continue; + + threshold = calculate_threshold(zone); + + for_each_online_cpu(cpu) + zone_pcp(zone, cpu)->stat_threshold = threshold; + } } /* @@ -133,17 +189,16 @@ static inline s8 *diff_pointer(struct zone *zone, enum zone_stat_item item) void __mod_zone_page_state(struct zone *zone, enum zone_stat_item item, int delta) { - s8 *p; + struct per_cpu_pageset *pcp = zone_pcp(zone, smp_processor_id()); + s8 *p = pcp->vm_stat_diff + item; long x; - p = diff_pointer(zone, item); x = delta + *p; - if (unlikely(x > STAT_THRESHOLD || x < -STAT_THRESHOLD)) { + if (unlikely(x > pcp->stat_threshold || x < -pcp->stat_threshold)) { zone_page_state_add(x, zone, item); x = 0; } - *p = x; } EXPORT_SYMBOL(__mod_zone_page_state); @@ -172,10 +227,12 @@ EXPORT_SYMBOL(mod_zone_page_state); * No overflow check is necessary and therefore the differential can be * incremented or decremented in place which may allow the compilers to * generate better code. - * * The increment or decrement is known and therefore one boundary check can * be omitted. * + * NOTE: These functions are very performance sensitive. Change only + * with care. + * * Some processors have inc/dec instructions that are atomic vs an interrupt. * However, the code must first determine the differential location in a zone * based on the processor number and then inc/dec the counter. There is no @@ -185,13 +242,16 @@ EXPORT_SYMBOL(mod_zone_page_state); */ static void __inc_zone_state(struct zone *zone, enum zone_stat_item item) { - s8 *p = diff_pointer(zone, item); + struct per_cpu_pageset *pcp = zone_pcp(zone, smp_processor_id()); + s8 *p = pcp->vm_stat_diff + item; (*p)++; - if (unlikely(*p > STAT_THRESHOLD)) { - zone_page_state_add(*p + STAT_THRESHOLD / 2, zone, item); - *p = -STAT_THRESHOLD / 2; + if (unlikely(*p > pcp->stat_threshold)) { + int overstep = pcp->stat_threshold / 2; + + zone_page_state_add(*p + overstep, zone, item); + *p = -overstep; } } @@ -204,13 +264,16 @@ EXPORT_SYMBOL(__inc_zone_page_state); void __dec_zone_page_state(struct page *page, enum zone_stat_item item) { struct zone *zone = page_zone(page); - s8 *p = diff_pointer(zone, item); + struct per_cpu_pageset *pcp = zone_pcp(zone, smp_processor_id()); + s8 *p = pcp->vm_stat_diff + item; (*p)--; - if (unlikely(*p < -STAT_THRESHOLD)) { - zone_page_state_add(*p - STAT_THRESHOLD / 2, zone, item); - *p = STAT_THRESHOLD /2; + if (unlikely(*p < - pcp->stat_threshold)) { + int overstep = pcp->stat_threshold / 2; + + zone_page_state_add(*p - overstep, zone, item); + *p = overstep; } } EXPORT_SYMBOL(__dec_zone_page_state); @@ -515,6 +578,10 @@ static int zoneinfo_show(struct seq_file *m, void *arg) pageset->pcp[j].high, pageset->pcp[j].batch); } +#ifdef CONFIG_SMP + seq_printf(m, "\n vm stats threshold: %d", + pageset->stat_threshold); +#endif } seq_printf(m, "\n all_unreclaimable: %u" @@ -603,3 +670,35 @@ struct seq_operations vmstat_op = { #endif /* CONFIG_PROC_FS */ +#ifdef CONFIG_SMP +/* + * Use the cpu notifier to insure that the thresholds are recalculated + * when necessary. + */ +static int __cpuinit vmstat_cpuup_callback(struct notifier_block *nfb, + unsigned long action, + void *hcpu) +{ + switch (action) { + case CPU_UP_PREPARE: + case CPU_UP_CANCELED: + case CPU_DEAD: + refresh_zone_stat_thresholds(); + break; + default: + break; + } + return NOTIFY_OK; +} + +static struct notifier_block __cpuinitdata vmstat_notifier = + { &vmstat_cpuup_callback, NULL, 0 }; + +int __init setup_vmstat(void) +{ + refresh_zone_stat_thresholds(); + register_cpu_notifier(&vmstat_notifier); + return 0; +} +module_init(setup_vmstat) +#endif From ddac7c7e3a0fe9cfdcef0de24476b8d69f8cf3e7 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 31 Aug 2006 21:27:36 -0700 Subject: [PATCH 093/115] [PATCH] md: Fix issues with referencing rdev in md/raid1 We need to be careful when referencing mirrors[i].rdev. It can disappear under us at various times. So: fix a couple of problem places. comment a couple of non-problem places move an 'atomic_add' which deferences rdev down a little way to some where where it is sure to not be NULL. Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/md/raid1.c | 57 ++++++++++++++++++++++++++++------------------ 1 file changed, 35 insertions(+), 22 deletions(-) diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c index 87bfe9e7d8c..3b4d69c0562 100644 --- a/drivers/md/raid1.c +++ b/drivers/md/raid1.c @@ -930,10 +930,13 @@ static void status(struct seq_file *seq, mddev_t *mddev) seq_printf(seq, " [%d/%d] [", conf->raid_disks, conf->working_disks); - for (i = 0; i < conf->raid_disks; i++) + rcu_read_lock(); + for (i = 0; i < conf->raid_disks; i++) { + mdk_rdev_t *rdev = rcu_dereference(conf->mirrors[i].rdev); seq_printf(seq, "%s", - conf->mirrors[i].rdev && - test_bit(In_sync, &conf->mirrors[i].rdev->flags) ? "U" : "_"); + rdev && test_bit(In_sync, &rdev->flags) ? "U" : "_"); + } + rcu_read_unlock(); seq_printf(seq, "]"); } @@ -975,7 +978,6 @@ static void error(mddev_t *mddev, mdk_rdev_t *rdev) static void print_conf(conf_t *conf) { int i; - mirror_info_t *tmp; printk("RAID1 conf printout:\n"); if (!conf) { @@ -985,14 +987,17 @@ static void print_conf(conf_t *conf) printk(" --- wd:%d rd:%d\n", conf->working_disks, conf->raid_disks); + rcu_read_lock(); for (i = 0; i < conf->raid_disks; i++) { char b[BDEVNAME_SIZE]; - tmp = conf->mirrors + i; - if (tmp->rdev) + mdk_rdev_t *rdev = rcu_dereference(conf->mirrors[i].rdev); + if (rdev) printk(" disk %d, wo:%d, o:%d, dev:%s\n", - i, !test_bit(In_sync, &tmp->rdev->flags), !test_bit(Faulty, &tmp->rdev->flags), - bdevname(tmp->rdev->bdev,b)); + i, !test_bit(In_sync, &rdev->flags), + !test_bit(Faulty, &rdev->flags), + bdevname(rdev->bdev,b)); } + rcu_read_unlock(); } static void close_sync(conf_t *conf) @@ -1008,20 +1013,20 @@ static int raid1_spare_active(mddev_t *mddev) { int i; conf_t *conf = mddev->private; - mirror_info_t *tmp; /* * Find all failed disks within the RAID1 configuration - * and mark them readable + * and mark them readable. + * Called under mddev lock, so rcu protection not needed. */ for (i = 0; i < conf->raid_disks; i++) { - tmp = conf->mirrors + i; - if (tmp->rdev - && !test_bit(Faulty, &tmp->rdev->flags) - && !test_bit(In_sync, &tmp->rdev->flags)) { + mdk_rdev_t *rdev = conf->mirrors[i].rdev; + if (rdev + && !test_bit(Faulty, &rdev->flags) + && !test_bit(In_sync, &rdev->flags)) { conf->working_disks++; mddev->degraded--; - set_bit(In_sync, &tmp->rdev->flags); + set_bit(In_sync, &rdev->flags); } } @@ -1237,7 +1242,7 @@ static void sync_request_write(mddev_t *mddev, r1bio_t *r1_bio) /* ouch - failed to read all of that. * Try some synchronous reads of other devices to get * good data, much like with normal read errors. Only - * read into the pages we already have so they we don't + * read into the pages we already have so we don't * need to re-issue the read request. * We don't need to freeze the array, because being in an * active sync request, there is no normal IO, and @@ -1257,6 +1262,10 @@ static void sync_request_write(mddev_t *mddev, r1bio_t *r1_bio) s = PAGE_SIZE >> 9; do { if (r1_bio->bios[d]->bi_end_io == end_sync_read) { + /* No rcu protection needed here devices + * can only be removed when no resync is + * active, and resync is currently active + */ rdev = conf->mirrors[d].rdev; if (sync_page_io(rdev->bdev, sect + rdev->data_offset, @@ -1463,6 +1472,11 @@ static void raid1d(mddev_t *mddev) s = PAGE_SIZE >> 9; do { + /* Note: no rcu protection needed here + * as this is synchronous in the raid1d thread + * which is the thread that might remove + * a device. If raid1d ever becomes multi-threaded.... + */ rdev = conf->mirrors[d].rdev; if (rdev && test_bit(In_sync, &rdev->flags) && @@ -1486,7 +1500,6 @@ static void raid1d(mddev_t *mddev) d = conf->raid_disks; d--; rdev = conf->mirrors[d].rdev; - atomic_add(s, &rdev->corrected_errors); if (rdev && test_bit(In_sync, &rdev->flags)) { if (sync_page_io(rdev->bdev, @@ -1509,9 +1522,11 @@ static void raid1d(mddev_t *mddev) s<<9, conf->tmppage, READ) == 0) /* Well, this device is dead */ md_error(mddev, rdev); - else + else { + atomic_add(s, &rdev->corrected_errors); printk(KERN_INFO "raid1:%s: read error corrected (%d sectors at %llu on %s)\n", mdname(mddev), s, (unsigned long long)(sect + rdev->data_offset), bdevname(rdev->bdev, b)); + } } } } else { @@ -1787,19 +1802,17 @@ static sector_t sync_request(mddev_t *mddev, sector_t sector_nr, int *skipped, i for (i=0; iraid_disks; i++) { bio = r1_bio->bios[i]; if (bio->bi_end_io == end_sync_read) { - md_sync_acct(conf->mirrors[i].rdev->bdev, nr_sectors); + md_sync_acct(bio->bi_bdev, nr_sectors); generic_make_request(bio); } } } else { atomic_set(&r1_bio->remaining, 1); bio = r1_bio->bios[r1_bio->read_disk]; - md_sync_acct(conf->mirrors[r1_bio->read_disk].rdev->bdev, - nr_sectors); + md_sync_acct(bio->bi_bdev, nr_sectors); generic_make_request(bio); } - return nr_sectors; } From 202af6d50155f8594ba01d8e0a54d77facc0152a Mon Sep 17 00:00:00 2001 From: Paul Fulghum Date: Thu, 31 Aug 2006 21:27:36 -0700 Subject: [PATCH 094/115] [PATCH] synclink_gt: fix receive tty error handling Fix receive tty error handling in synclink_gt driver. Adrian reported compiler warning for incorrect bit test against char variable. I determined these and other device specific error bits were incorrectly defined. Signed-off-by: Paul Fulghum Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/synclink_gt.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/drivers/char/synclink_gt.c b/drivers/char/synclink_gt.c index b2dbbdb1bf8..2f07b085536 100644 --- a/drivers/char/synclink_gt.c +++ b/drivers/char/synclink_gt.c @@ -391,8 +391,8 @@ static MGSL_PARAMS default_params = { #define DESC_LIST_SIZE 4096 #define MASK_PARITY BIT1 -#define MASK_FRAMING BIT2 -#define MASK_BREAK BIT3 +#define MASK_FRAMING BIT0 +#define MASK_BREAK BIT14 #define MASK_OVERRUN BIT4 #define GSR 0x00 /* global status */ @@ -1800,17 +1800,17 @@ static void rx_async(struct slgt_info *info) stat = 0; - if ((status = *(p+1) & (BIT9 + BIT8))) { - if (status & BIT9) + if ((status = *(p+1) & (BIT1 + BIT0))) { + if (status & BIT1) icount->parity++; - else if (status & BIT8) + else if (status & BIT0) icount->frame++; /* discard char if tty control flags say so */ if (status & info->ignore_status_mask) continue; - if (status & BIT9) + if (status & BIT1) stat = TTY_PARITY; - else if (status & BIT8) + else if (status & BIT0) stat = TTY_FRAME; } if (tty) { From 30f3174d1c506db2c6d2c1dddc9c064e741d6b76 Mon Sep 17 00:00:00 2001 From: john stultz Date: Thu, 31 Aug 2006 21:27:37 -0700 Subject: [PATCH 095/115] [PATCH] Fix faulty HPET clocksource usage (fix for bug #7062) Apparently some systems export valid HPET addresses, but hpet_enable() fails. Then when the HPET clocksource starts up, it only checks for a valid HPET address, and the result is a system where time does not advance. See http://bugme.osdl.org/show_bug.cgi?id=7062 for details. This patch just makes sure we better check that the HPET is functional before registering the HPET clocksource. Signed-off-by: John Stultz Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/hpet.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/i386/kernel/hpet.c b/arch/i386/kernel/hpet.c index c6737c35815..17647a530b2 100644 --- a/arch/i386/kernel/hpet.c +++ b/arch/i386/kernel/hpet.c @@ -35,7 +35,7 @@ static int __init init_hpet_clocksource(void) void __iomem* hpet_base; u64 tmp; - if (!hpet_address) + if (!is_hpet_enabled()) return -ENODEV; /* calculate the hpet address: */ From 35df17c57cecb08f0120fb18926325f1093dc429 Mon Sep 17 00:00:00 2001 From: Shailabh Nagar Date: Thu, 31 Aug 2006 21:27:38 -0700 Subject: [PATCH 096/115] [PATCH] task delay accounting fixes Cleanup allocation and freeing of tsk->delays used by delay accounting. This solves two problems reported for delay accounting: 1. oops in __delayacct_blkio_ticks http://www.uwsg.indiana.edu/hypermail/linux/kernel/0608.2/1844.html Currently tsk->delays is getting freed too early in task exit which can cause a NULL tsk->delays to get accessed via reading of /proc//stats. The patch fixes this problem by freeing tsk->delays closer to when task_struct itself is freed up. As a result, it also eliminates the use of tsk->delays_lock which was only being used (inadequately) to safeguard access to tsk->delays while a task was exiting. 2. Possible memory leak in kernel/delayacct.c http://www.uwsg.indiana.edu/hypermail/linux/kernel/0608.2/1389.html The patch cleans up tsk->delays allocations after a bad fork which was missing earlier. The patch has been tested to fix the problems listed above and stress tested with rapid calls to delay accounting's taskstats command interface (which is the other path that can access the same data, besides the /proc interface causing the oops above). Signed-off-by: Shailabh Nagar Cc: Balbir Singh Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/delayacct.h | 10 +++++++--- include/linux/sched.h | 1 - kernel/delayacct.c | 16 ---------------- kernel/exit.c | 1 - kernel/fork.c | 6 ++++-- 5 files changed, 11 insertions(+), 23 deletions(-) diff --git a/include/linux/delayacct.h b/include/linux/delayacct.h index 11487b6e712..561e2a77805 100644 --- a/include/linux/delayacct.h +++ b/include/linux/delayacct.h @@ -59,10 +59,14 @@ static inline void delayacct_tsk_init(struct task_struct *tsk) __delayacct_tsk_init(tsk); } -static inline void delayacct_tsk_exit(struct task_struct *tsk) +/* Free tsk->delays. Called from bad fork and __put_task_struct + * where there's no risk of tsk->delays being accessed elsewhere + */ +static inline void delayacct_tsk_free(struct task_struct *tsk) { if (tsk->delays) - __delayacct_tsk_exit(tsk); + kmem_cache_free(delayacct_cache, tsk->delays); + tsk->delays = NULL; } static inline void delayacct_blkio_start(void) @@ -101,7 +105,7 @@ static inline void delayacct_init(void) {} static inline void delayacct_tsk_init(struct task_struct *tsk) {} -static inline void delayacct_tsk_exit(struct task_struct *tsk) +static inline void delayacct_tsk_free(struct task_struct *tsk) {} static inline void delayacct_blkio_start(void) {} diff --git a/include/linux/sched.h b/include/linux/sched.h index 6674fc1e51b..34ed0d99b1b 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -994,7 +994,6 @@ struct task_struct { */ struct pipe_inode_info *splice_pipe; #ifdef CONFIG_TASK_DELAY_ACCT - spinlock_t delays_lock; struct task_delay_info *delays; #endif }; diff --git a/kernel/delayacct.c b/kernel/delayacct.c index 57ca3730205..36752f124c6 100644 --- a/kernel/delayacct.c +++ b/kernel/delayacct.c @@ -41,24 +41,11 @@ void delayacct_init(void) void __delayacct_tsk_init(struct task_struct *tsk) { - spin_lock_init(&tsk->delays_lock); - /* No need to acquire tsk->delays_lock for allocation here unless - __delayacct_tsk_init called after tsk is attached to tasklist - */ tsk->delays = kmem_cache_zalloc(delayacct_cache, SLAB_KERNEL); if (tsk->delays) spin_lock_init(&tsk->delays->lock); } -void __delayacct_tsk_exit(struct task_struct *tsk) -{ - struct task_delay_info *delays = tsk->delays; - spin_lock(&tsk->delays_lock); - tsk->delays = NULL; - spin_unlock(&tsk->delays_lock); - kmem_cache_free(delayacct_cache, delays); -} - /* * Start accounting for a delay statistic using * its starting timestamp (@start) @@ -118,8 +105,6 @@ int __delayacct_add_tsk(struct taskstats *d, struct task_struct *tsk) struct timespec ts; unsigned long t1,t2,t3; - spin_lock(&tsk->delays_lock); - /* Though tsk->delays accessed later, early exit avoids * unnecessary returning of other data */ @@ -161,7 +146,6 @@ int __delayacct_add_tsk(struct taskstats *d, struct task_struct *tsk) spin_unlock(&tsk->delays->lock); done: - spin_unlock(&tsk->delays_lock); return 0; } diff --git a/kernel/exit.c b/kernel/exit.c index dba194a8d41..a4c19a52ce4 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -908,7 +908,6 @@ fastcall NORET_TYPE void do_exit(long code) audit_free(tsk); taskstats_exit_send(tsk, tidstats, group_dead, mycpu); taskstats_exit_free(tidstats); - delayacct_tsk_exit(tsk); exit_mm(tsk); diff --git a/kernel/fork.c b/kernel/fork.c index aa36c43783c..f9b014e3e70 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -117,6 +117,7 @@ void __put_task_struct(struct task_struct *tsk) security_task_free(tsk); free_uid(tsk->user); put_group_info(tsk->group_info); + delayacct_tsk_free(tsk); if (!profile_handoff_task(tsk)) free_task(tsk); @@ -1011,7 +1012,7 @@ static struct task_struct *copy_process(unsigned long clone_flags, retval = -EFAULT; if (clone_flags & CLONE_PARENT_SETTID) if (put_user(p->pid, parent_tidptr)) - goto bad_fork_cleanup; + goto bad_fork_cleanup_delays_binfmt; INIT_LIST_HEAD(&p->children); INIT_LIST_HEAD(&p->sibling); @@ -1277,7 +1278,8 @@ bad_fork_cleanup_policy: bad_fork_cleanup_cpuset: #endif cpuset_exit(p); -bad_fork_cleanup: +bad_fork_cleanup_delays_binfmt: + delayacct_tsk_free(p); if (p->binfmt) module_put(p->binfmt->module); bad_fork_cleanup_put_domain: From d742eae8e2655a960b2e07f3f622d78546c14031 Mon Sep 17 00:00:00 2001 From: "Bill Huey (hui" Date: Thu, 31 Aug 2006 21:27:39 -0700 Subject: [PATCH 097/115] [PATCH] xtensa: ptrace: EXIT_ZOMBIE fix We're testing the wrong task_struct field. Acked-by: Chris Zankel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/xtensa/kernel/ptrace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/xtensa/kernel/ptrace.c b/arch/xtensa/kernel/ptrace.c index 5064d938396..9aea23cc0dc 100644 --- a/arch/xtensa/kernel/ptrace.c +++ b/arch/xtensa/kernel/ptrace.c @@ -212,7 +212,7 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data) */ case PTRACE_KILL: ret = 0; - if (child->state == EXIT_ZOMBIE) /* already dead */ + if (child->exit_state == EXIT_ZOMBIE) /* already dead */ break; child->exit_code = SIGKILL; child->ptrace &= ~PT_SINGLESTEP; From a9aa141cfc2e08470bba3b9a8328bc50ac457488 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Thu, 31 Aug 2006 21:27:40 -0700 Subject: [PATCH 098/115] [PATCH] x86: increase MAX_MP_BUSSES on default arch Vitezslav Samel reports that an HP DL380 g4 fails using the default arch due to the ISA bus having an ID of 32. It would have worked OK with the generic arch - for some reason the default arch doesn't support as many busses. So bump that up to support 256 busses, but leave it at 32 if we're building a tiny system to save a bit of memory. Cc: Vitezslav Samel Acked-by: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/asm-i386/mach-default/mach_mpspec.h | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/include/asm-i386/mach-default/mach_mpspec.h b/include/asm-i386/mach-default/mach_mpspec.h index 6b5dadcf1d0..51c9a977593 100644 --- a/include/asm-i386/mach-default/mach_mpspec.h +++ b/include/asm-i386/mach-default/mach_mpspec.h @@ -3,6 +3,10 @@ #define MAX_IRQ_SOURCES 256 +#if CONFIG_BASE_SMALL == 0 +#define MAX_MP_BUSSES 256 +#else #define MAX_MP_BUSSES 32 +#endif #endif /* __ASM_MACH_MPSPEC_H */ From ef16b5194f3233a11851180cd82eafb76542047d Mon Sep 17 00:00:00 2001 From: Olaf Hering Date: Thu, 31 Aug 2006 21:27:41 -0700 Subject: [PATCH 099/115] [PATCH] exit early in floppy_init when no floppy exists modprobe -v floppy on a Apple G5 writes incorrect stuff to dmesg: Floppy drive(s): fd0 is 2.88M The reason is that the legacy io check happens very late, when part of the floppy stuff is already initialized. check_legacy_ioport() returns either -ENODEV right away, or it walks the device-tree looking for a floppy node. Signed-off-by: Olaf Hering Acked-by: Benjamin Herrenschmidt Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/block/floppy.c | 12 +++++------- 1 file changed, 5 insertions(+), 7 deletions(-) diff --git a/drivers/block/floppy.c b/drivers/block/floppy.c index 5109fa37c66..ad1d7065a1b 100644 --- a/drivers/block/floppy.c +++ b/drivers/block/floppy.c @@ -4177,6 +4177,11 @@ static int __init floppy_init(void) int i, unit, drive; int err, dr; +#if defined(CONFIG_PPC_MERGE) + if (check_legacy_ioport(FDC1)) + return -ENODEV; +#endif + raw_cmd = NULL; for (dr = 0; dr < N_DRIVE; dr++) { @@ -4234,13 +4239,6 @@ static int __init floppy_init(void) } use_virtual_dma = can_use_virtual_dma & 1; -#if defined(CONFIG_PPC_MERGE) - if (check_legacy_ioport(FDC1)) { - del_timer(&fd_timeout); - err = -ENODEV; - goto out_unreg_region; - } -#endif fdc_state[0].address = FDC1; if (fdc_state[0].address == -1) { del_timer(&fd_timeout); From 290995fc3c06c0548ae303dd0b1371a8727f4fad Mon Sep 17 00:00:00 2001 From: "Ian E. Morgan" Date: Thu, 31 Aug 2006 21:27:42 -0700 Subject: [PATCH 100/115] [PATCH] SBC8360: module_param() permission fixes The last argument of module_param is permissions, not default value. Acked-by: Wim Van Sebroeck Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/watchdog/sbc8360.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/char/watchdog/sbc8360.c b/drivers/char/watchdog/sbc8360.c index 1035be5b501..41fc6f80c49 100644 --- a/drivers/char/watchdog/sbc8360.c +++ b/drivers/char/watchdog/sbc8360.c @@ -200,7 +200,7 @@ static int wd_margin = 0xB; static int wd_multiplier = 2; static int nowayout = WATCHDOG_NOWAYOUT; -module_param(timeout, int, 27); +module_param(timeout, int, 0); MODULE_PARM_DESC(timeout, "Index into timeout table (0-63) (default=27 (60s))"); module_param(nowayout, int, 0); MODULE_PARM_DESC(nowayout, @@ -407,7 +407,7 @@ module_exit(sbc8360_exit); MODULE_AUTHOR("Ian E. Morgan "); MODULE_DESCRIPTION("SBC8360 watchdog driver"); MODULE_LICENSE("GPL"); -MODULE_VERSION("1.0"); +MODULE_VERSION("1.01"); MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR); /* end of sbc8360.c */ From 43a1dd502f40fdb644402f64cd06cf8016cd9780 Mon Sep 17 00:00:00 2001 From: Henrik Kretzschmar Date: Thu, 31 Aug 2006 21:27:44 -0700 Subject: [PATCH 101/115] [PATCH] kerneldoc for handle_bad_irq() Adds the description of the parameters from handle_bad_irq(). Signed-off-by: Henrik Kretzschmar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/irq/handle.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/kernel/irq/handle.c b/kernel/irq/handle.c index fc4e906aedb..48a53f68af9 100644 --- a/kernel/irq/handle.c +++ b/kernel/irq/handle.c @@ -20,6 +20,11 @@ /** * handle_bad_irq - handle spurious and unhandled irqs + * @irq: the interrupt number + * @desc: description of the interrupt + * @regs: pointer to a register structure + * + * Handles spurious and unhandled IRQ's. It also prints a debugmessage. */ void fastcall handle_bad_irq(unsigned int irq, struct irq_desc *desc, struct pt_regs *regs) From a9eec55623f5aedfcef745d3a0e0f97f5d4f74be Mon Sep 17 00:00:00 2001 From: Corey Minyard Date: Thu, 31 Aug 2006 21:27:45 -0700 Subject: [PATCH 102/115] [PATCH] IPMI: fix occasional oops on module unload Olaf Kirch of SuSE tracked down a problem where module unloads of the IPMI driver would occasionally result in Oopses. He tracked that down to a variable that wasn't always initialized properly in some situations. This patch initializes that variable. Olaf sent a patch that kzalloc-ed the data, but this structure is large enough that I would perfer to not do that. Thanks Olaf! Signed-off-by: Corey Minyard Cc: Olaf Kirch Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/ipmi/ipmi_msghandler.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/char/ipmi/ipmi_msghandler.c b/drivers/char/ipmi/ipmi_msghandler.c index 0aa5d608fe6..843d34c8627 100644 --- a/drivers/char/ipmi/ipmi_msghandler.c +++ b/drivers/char/ipmi/ipmi_msghandler.c @@ -3428,6 +3428,7 @@ struct ipmi_recv_msg *ipmi_alloc_recv_msg(void) rv = kmalloc(sizeof(struct ipmi_recv_msg), GFP_ATOMIC); if (rv) { + rv->user = NULL; rv->done = free_recv_msg; atomic_inc(&recv_msg_inuse_count); } From 1e5f5e5cd65eec6ce5c24a9c29f3e52673b121a6 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Thu, 31 Aug 2006 21:27:46 -0700 Subject: [PATCH 103/115] [PATCH] schedule obsolete OSS drivers for removal, 2nd round This patch schedules obsolete OSS drivers (with ALSA drivers that support the same hardware) for removal. A rationale of the patch is in http://lkml.org/lkml/2006/7/11/186 Signed-off-by: Adrian Bunk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/feature-removal-schedule.txt | 7 +++++ sound/oss/Kconfig | 30 ++++++++++++++++------ 2 files changed, 29 insertions(+), 8 deletions(-) diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt index 87851efb022..d1cd5f93e02 100644 --- a/Documentation/feature-removal-schedule.txt +++ b/Documentation/feature-removal-schedule.txt @@ -120,6 +120,13 @@ Who: Adrian Bunk --------------------------- +What: drivers depending on OSS_OBSOLETE_DRIVER +When: options in 2.6.20, code in 2.6.22 +Why: OSS drivers with ALSA replacements +Who: Adrian Bunk + +--------------------------- + What: pci_module_init(driver) When: January 2007 Why: Is replaced by pci_register_driver(pci_driver). diff --git a/sound/oss/Kconfig b/sound/oss/Kconfig index 1b7c3dfc2b4..97e38b66558 100644 --- a/sound/oss/Kconfig +++ b/sound/oss/Kconfig @@ -5,6 +5,20 @@ # # Prompt user for primary drivers. +config OSS_OBSOLETE_DRIVER + bool "Obsolete OSS drivers" + depends on SOUND_PRIME + help + This option enables support for obsolete OSS drivers that + are scheduled for removal in the near future since there + are ALSA drivers for the same hardware. + + Please contact Adrian Bunk if you had to + say Y here because your soundcard is not properly supported + by ALSA. + + If unsure, say N. + config SOUND_BT878 tristate "BT878 audio dma" depends on SOUND_PRIME && PCI @@ -23,7 +37,7 @@ config SOUND_BT878 config SOUND_EMU10K1 tristate "Creative SBLive! (EMU10K1)" - depends on SOUND_PRIME && PCI + depends on SOUND_PRIME && PCI && OSS_OBSOLETE_DRIVER ---help--- Say Y or M if you have a PCI sound card using the EMU10K1 chipset, such as the Creative SBLive!, SB PCI512 or Emu-APS. @@ -49,7 +63,7 @@ config MIDI_EMU10K1 config SOUND_FUSION tristate "Crystal SoundFusion (CS4280/461x)" - depends on SOUND_PRIME && PCI + depends on SOUND_PRIME && PCI && OSS_OBSOLETE_DRIVER help This module drives the Crystal SoundFusion devices (CS4280/46xx series) when wired as native sound drivers with AC97 codecs. If @@ -440,7 +454,7 @@ config SOUND_DMAP config SOUND_AD1816 tristate "AD1816(A) based cards (EXPERIMENTAL)" - depends on EXPERIMENTAL && SOUND_OSS + depends on EXPERIMENTAL && SOUND_OSS && OSS_OBSOLETE_DRIVER help Say M here if you have a sound card based on the Analog Devices AD1816(A) chip. @@ -450,21 +464,21 @@ config SOUND_AD1816 config SOUND_AD1889 tristate "AD1889 based cards (AD1819 codec) (EXPERIMENTAL)" - depends on EXPERIMENTAL && SOUND_OSS && PCI + depends on EXPERIMENTAL && SOUND_OSS && PCI && OSS_OBSOLETE_DRIVER help Say M here if you have a sound card based on the Analog Devices AD1889 chip. config SOUND_ADLIB tristate "Adlib Cards" - depends on SOUND_OSS + depends on SOUND_OSS && OSS_OBSOLETE_DRIVER help Includes ASB 64 4D. Information on programming AdLib cards is available at . config SOUND_ACI_MIXER tristate "ACI mixer (miroSOUND PCM1-pro/PCM12/PCM20)" - depends on SOUND_OSS + depends on SOUND_OSS && OSS_OBSOLETE_DRIVER ---help--- ACI (Audio Command Interface) is a protocol used to communicate with the microcontroller on some sound cards produced by miro and @@ -586,7 +600,7 @@ config SOUND_MPU401 config SOUND_NM256 tristate "NM256AV/NM256ZX audio support" - depends on SOUND_OSS + depends on SOUND_OSS && OSS_OBSOLETE_DRIVER help Say M here to include audio support for the NeoMagic 256AV/256ZX chipsets. These are the audio chipsets found in the Sony @@ -706,7 +720,7 @@ config SOUND_YM3812 config SOUND_OPL3SA2 tristate "Yamaha OPL3-SA2 and SA3 based PnP cards" - depends on SOUND_OSS + depends on SOUND_OSS && OSS_OBSOLETE_DRIVER help Say Y or M if you have a card based on one of these Yamaha sound chipsets or the "SAx", which is actually a SA3. Read From 0b1d647a02c5a1b67d45287eeb6cb3b2219c41c3 Mon Sep 17 00:00:00 2001 From: Pavel Mironchik Date: Thu, 31 Aug 2006 21:27:47 -0700 Subject: [PATCH 104/115] [PATCH] dm: work around mempool_alloc, bio_alloc_bioset deadlocks This patch works around a complex dm-related deadlock/livelock down in the mempool allocator. Alasdair said: Several dm targets suffer from this. Mempools are not yet used correctly everywhere in device-mapper: they can get shared when devices are stacked, and some targets share them across multiple instances. I made fixing this one of the prerequisites for this patch: md-dm-reduce-stack-usage-with-stacked-block-devices.patch which in some cases makes people more likely to hit the problem. There's been some progress on this recently with (unfinished) dm-crypt patches at: http://www.kernel.org/pub/linux/kernel/people/agk/patches/2.6/editing/ (dm-crypt-move-io-to-workqueue.patch plus dependencies) and: I've no problems with a temporary workaround like that, but Milan Broz (a new Redhat developer in the Czech Republic) has started reviewing all the mempool usage in device-mapper so I'm expecting we'll soon have a proper fix for this associated problems. [He's back from holiday at the start of next week.] For now, this sad-but-safe little patch will allow the machine to recover. [akpm@osdl.org: rewrote changelog] Cc: Alasdair G Kergon Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/mempool.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/mm/mempool.c b/mm/mempool.c index fe6e05289cc..ccd8cb8cd41 100644 --- a/mm/mempool.c +++ b/mm/mempool.c @@ -238,8 +238,13 @@ repeat_alloc: init_wait(&wait); prepare_to_wait(&pool->wait, &wait, TASK_UNINTERRUPTIBLE); smp_mb(); - if (!pool->curr_nr) - io_schedule(); + if (!pool->curr_nr) { + /* + * FIXME: this should be io_schedule(). The timeout is there + * as a workaround for some DM problems in 2.6.18. + */ + io_schedule_timeout(5*HZ); + } finish_wait(&pool->wait, &wait); goto repeat_alloc; From d565dd3b0824b67a8442df4de83cc44f7c726fc9 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Thu, 31 Aug 2006 21:27:49 -0700 Subject: [PATCH 105/115] [PATCH] powerpc: More via-pmu backlight fixes The via-pmu backlight code (introduced in 2.6.18) has various design issues causing crashes on machines using it like the old Wallstreet powerbook (Michael, the author, never managed to test on these and I just got my hand on one of those old beasts). This fixes them by no longer trying to hijack the backlight device of the frontmost framebuffer (causing that framebuffer to crash) but having it's own local bits instead. Might look weird but it's better that way on those old machines, at least as a last-minute fix for 2.6.18. We might rework the whole thing later. This patch also changes the way it gets notified of sleep and wakeup in order to properly shut the backlight down on sleep and bring it back on wakeup. Signed-off-by: Benjamin Herrenschmidt Cc: "Antonino A. Daplas" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/macintosh/via-pmu-backlight.c | 95 +++++++++++++-------------- drivers/macintosh/via-pmu.c | 12 ++++ 2 files changed, 57 insertions(+), 50 deletions(-) diff --git a/drivers/macintosh/via-pmu-backlight.c b/drivers/macintosh/via-pmu-backlight.c index d3f8d75bcbb..4397fac55ba 100644 --- a/drivers/macintosh/via-pmu-backlight.c +++ b/drivers/macintosh/via-pmu-backlight.c @@ -18,17 +18,48 @@ static struct backlight_properties pmu_backlight_data; static spinlock_t pmu_backlight_lock; static int sleeping; +static u8 bl_curve[FB_BACKLIGHT_LEVELS]; -static int pmu_backlight_get_level_brightness(struct fb_info *info, - int level) +static void pmu_backlight_init_curve(u8 off, u8 min, u8 max) +{ + unsigned int i, flat, count, range = (max - min); + + bl_curve[0] = off; + + for (flat = 1; flat < (FB_BACKLIGHT_LEVELS / 16); ++flat) + bl_curve[flat] = min; + + count = FB_BACKLIGHT_LEVELS * 15 / 16; + for (i = 0; i < count; ++i) + bl_curve[flat + i] = min + (range * (i + 1) / count); +} + +static int pmu_backlight_curve_lookup(int value) +{ + int level = (FB_BACKLIGHT_LEVELS - 1); + int i, max = 0; + + /* Look for biggest value */ + for (i = 0; i < FB_BACKLIGHT_LEVELS; i++) + max = max((int)bl_curve[i], max); + + /* Look for nearest value */ + for (i = 0; i < FB_BACKLIGHT_LEVELS; i++) { + int diff = abs(bl_curve[i] - value); + if (diff < max) { + max = diff; + level = i; + } + } + return level; +} + +static int pmu_backlight_get_level_brightness(int level) { int pmulevel; /* Get and convert the value */ - mutex_lock(&info->bl_mutex); - pmulevel = info->bl_curve[level] * FB_BACKLIGHT_MAX / MAX_PMU_LEVEL; - mutex_unlock(&info->bl_mutex); - + pmulevel = bl_curve[level] * FB_BACKLIGHT_MAX / MAX_PMU_LEVEL; if (pmulevel < 0) pmulevel = 0; else if (pmulevel > MAX_PMU_LEVEL) @@ -39,7 +70,6 @@ static int pmu_backlight_get_level_brightness(struct fb_info *info, static int pmu_backlight_update_status(struct backlight_device *bd) { - struct fb_info *info = class_get_devdata(&bd->class_dev); struct adb_request req; unsigned long flags; int level = bd->props->brightness; @@ -55,7 +85,7 @@ static int pmu_backlight_update_status(struct backlight_device *bd) level = 0; if (level > 0) { - int pmulevel = pmu_backlight_get_level_brightness(info, level); + int pmulevel = pmu_backlight_get_level_brightness(level); pmu_request(&req, NULL, 2, PMU_BACKLIGHT_BRIGHT, pmulevel); pmu_wait_complete(&req); @@ -88,35 +118,19 @@ static struct backlight_properties pmu_backlight_data = { }; #ifdef CONFIG_PM -static int pmu_backlight_sleep_call(struct pmu_sleep_notifier *self, int when) +void pmu_backlight_set_sleep(int sleep) { unsigned long flags; spin_lock_irqsave(&pmu_backlight_lock, flags); - - switch (when) { - case PBOOK_SLEEP_REQUEST: - sleeping = 1; - break; - case PBOOK_WAKE: - sleeping = 0; - break; - } - + sleeping = sleep; spin_unlock_irqrestore(&pmu_backlight_lock, flags); - - return PBOOK_SLEEP_OK; } - -static struct pmu_sleep_notifier pmu_backlight_sleep_notif = { - .notifier_call = pmu_backlight_sleep_call, -}; -#endif +#endif /* CONFIG_PM */ void __init pmu_backlight_init() { struct backlight_device *bd; - struct fb_info *info; char name[10]; int level, autosave; @@ -131,27 +145,14 @@ void __init pmu_backlight_init() !machine_is_compatible("PowerBook1,1")) return; - /* Actually, this is a hack, but I don't know of a better way - * to get the first framebuffer device. - */ - info = registered_fb[0]; - if (!info) { - printk("pmubl: No framebuffer found\n"); - goto error; - } + snprintf(name, sizeof(name), "pmubl"); - snprintf(name, sizeof(name), "pmubl%d", info->node); - - bd = backlight_device_register(name, info, &pmu_backlight_data); + bd = backlight_device_register(name, NULL, &pmu_backlight_data); if (IS_ERR(bd)) { printk("pmubl: Backlight registration failed\n"); goto error; } - - mutex_lock(&info->bl_mutex); - info->bl_dev = bd; - fb_bl_default_curve(info, 0x7F, 0x46, 0x0E); - mutex_unlock(&info->bl_mutex); + pmu_backlight_init_curve(0x7F, 0x46, 0x0E); level = pmu_backlight_data.max_brightness; @@ -161,11 +162,9 @@ void __init pmu_backlight_init() pmu_request(&req, NULL, 2, 0xd9, 0); pmu_wait_complete(&req); - mutex_lock(&info->bl_mutex); - level = pmac_backlight_curve_lookup(info, + level = pmu_backlight_curve_lookup( (req.reply[0] >> 4) * pmu_backlight_data.max_brightness / 15); - mutex_unlock(&info->bl_mutex); } up(&bd->sem); @@ -179,10 +178,6 @@ void __init pmu_backlight_init() pmac_backlight = bd; mutex_unlock(&pmac_backlight_mutex); -#ifdef CONFIG_PM - pmu_register_sleep_notifier(&pmu_backlight_sleep_notif); -#endif - printk("pmubl: Backlight initialized (%s)\n", name); return; diff --git a/drivers/macintosh/via-pmu.c b/drivers/macintosh/via-pmu.c index ea386801e21..14610a63f58 100644 --- a/drivers/macintosh/via-pmu.c +++ b/drivers/macintosh/via-pmu.c @@ -1995,6 +1995,8 @@ restore_via_state(void) out_8(&via[IER], IER_SET | SR_INT | CB1_INT); } +extern void pmu_backlight_set_sleep(int sleep); + static int pmac_suspend_devices(void) { @@ -2032,6 +2034,11 @@ pmac_suspend_devices(void) return -EBUSY; } +#ifdef CONFIG_PMAC_BACKLIGHT + /* Tell backlight code not to muck around with the chip anymore */ + pmu_backlight_set_sleep(1); +#endif + /* Call platform functions marked "on sleep" */ pmac_pfunc_i2c_suspend(); pmac_pfunc_base_suspend(); @@ -2090,6 +2097,11 @@ pmac_wakeup_devices(void) { mdelay(100); +#ifdef CONFIG_PMAC_BACKLIGHT + /* Tell backlight code it can use the chip again */ + pmu_backlight_set_sleep(0); +#endif + /* Power back up system devices (including the PIC) */ device_power_up(); From ca72945d2d0981299035bd9d40a08ff06e74b5b8 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Thu, 31 Aug 2006 21:27:50 -0700 Subject: [PATCH 106/115] [PATCH] powerpc: Fix PowerMac IRQ handling bug The port to genirq & the new powerpc interrupt model in 2.6.18 introduced a bug in the legacy PowerMac PIC code (used on older machines) because of a typo potentially causing hangs due to interrupt storms. This fixes it, along with a performance issue causing us to do spurrious retriggers after masking an interrupt. Signed-off-by: Benjamin Herrenschmidt Cc: "Antonino A. Daplas" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/powerpc/platforms/powermac/pic.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/arch/powerpc/platforms/powermac/pic.c b/arch/powerpc/platforms/powermac/pic.c index 060789e31c6..39f7ddb554e 100644 --- a/arch/powerpc/platforms/powermac/pic.c +++ b/arch/powerpc/platforms/powermac/pic.c @@ -87,8 +87,8 @@ static void __pmac_retrigger(unsigned int irq_nr) static void pmac_mask_and_ack_irq(unsigned int virq) { unsigned int src = irq_map[virq].hwirq; - unsigned long bit = 1UL << (virq & 0x1f); - int i = virq >> 5; + unsigned long bit = 1UL << (src & 0x1f); + int i = src >> 5; unsigned long flags; spin_lock_irqsave(&pmac_pic_lock, flags); @@ -175,7 +175,7 @@ static void pmac_mask_irq(unsigned int virq) spin_lock_irqsave(&pmac_pic_lock, flags); __clear_bit(src, ppc_cached_irq_mask); - __pmac_set_irq_mask(src, 0); + __pmac_set_irq_mask(src, 1); spin_unlock_irqrestore(&pmac_pic_lock, flags); } From 7931e2a92b4de23093362308087daa8aabd3848a Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Thu, 31 Aug 2006 21:27:50 -0700 Subject: [PATCH 107/115] [PATCH] ALSA: ac97: correct some Mic mixer elements Revert the mixer element names of some Mic controls to the state of 2.6.17. This should fix the name mismatch in alsactl. Signed-off-by: Takashi Iwai Cc: Jaroslav Kysela Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- sound/pci/ac97/ac97_codec.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/sound/pci/ac97/ac97_codec.c b/sound/pci/ac97/ac97_codec.c index 0abf2808d59..51e83d7a839 100644 --- a/sound/pci/ac97/ac97_codec.c +++ b/sound/pci/ac97/ac97_codec.c @@ -573,7 +573,7 @@ AC97_SINGLE("PC Speaker Playback Volume", AC97_PC_BEEP, 1, 15, 1) }; static const struct snd_kcontrol_new snd_ac97_controls_mic_boost = - AC97_SINGLE("Mic Boost (+20dB) Switch", AC97_MIC, 6, 1, 0); + AC97_SINGLE("Mic Boost (+20dB)", AC97_MIC, 6, 1, 0); static const char* std_rec_sel[] = {"Mic", "CD", "Video", "Aux", "Line", "Mix", "Mix Mono", "Phone"}; @@ -615,7 +615,7 @@ AC97_SINGLE("Simulated Stereo Enhancement", AC97_GENERAL_PURPOSE, 14, 1, 0), AC97_SINGLE("3D Control - Switch", AC97_GENERAL_PURPOSE, 13, 1, 0), AC97_SINGLE("Loudness (bass boost)", AC97_GENERAL_PURPOSE, 12, 1, 0), AC97_ENUM("Mono Output Select", std_enum[2]), -AC97_ENUM("Mic Select Capture Switch", std_enum[3]), +AC97_ENUM("Mic Select", std_enum[3]), AC97_SINGLE("ADC/DAC Loopback", AC97_GENERAL_PURPOSE, 7, 1, 0) }; From 1678df37be8abbb381becdc40242ed915e775550 Mon Sep 17 00:00:00 2001 From: John Keller Date: Thu, 31 Aug 2006 21:27:51 -0700 Subject: [PATCH 108/115] [PATCH] sgiioc4: fixup use of mmio ops Fix some bugs in the patch that converted the IOC4 driver from port IO ops to memio ops. http://marc.theaimsgroup.com/?l=linux-ide&m=114895892231438&w=2 Problems fixed are: - Call to default_hwif_mmiops() was not being done until _after_ first IO operation, resulting in the first IO operation being done as a port IO op, instead of memio. - request_region() calls needed to be request_mem_region() - Incomplete error case handling. - Non-usage of ioremap() and __iomem. Signed-off-by: John Keller Signed-off-by: Jeremy Higdon Cc: Alan Cox Cc: Bartlomiej Zolnierkiewicz Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/ide/pci/sgiioc4.c | 60 +++++++++++++++++++++++++++------------ 1 file changed, 42 insertions(+), 18 deletions(-) diff --git a/drivers/ide/pci/sgiioc4.c b/drivers/ide/pci/sgiioc4.c index e125032bb40..d8a0d87df73 100644 --- a/drivers/ide/pci/sgiioc4.c +++ b/drivers/ide/pci/sgiioc4.c @@ -367,12 +367,13 @@ sgiioc4_INB(unsigned long port) static void __devinit ide_dma_sgiioc4(ide_hwif_t * hwif, unsigned long dma_base) { + void __iomem *virt_dma_base; int num_ports = sizeof (ioc4_dma_regs_t); printk(KERN_INFO "%s: BM-DMA at 0x%04lx-0x%04lx\n", hwif->name, dma_base, dma_base + num_ports - 1); - if (!request_region(dma_base, num_ports, hwif->name)) { + if (!request_mem_region(dma_base, num_ports, hwif->name)) { printk(KERN_ERR "%s(%s) -- ERROR, Addresses 0x%p to 0x%p " "ALREADY in use\n", @@ -381,13 +382,21 @@ ide_dma_sgiioc4(ide_hwif_t * hwif, unsigned long dma_base) goto dma_alloc_failure; } - hwif->dma_base = dma_base; + virt_dma_base = ioremap(dma_base, num_ports); + if (virt_dma_base == NULL) { + printk(KERN_ERR + "%s(%s) -- ERROR, Unable to map addresses 0x%lx to 0x%lx\n", + __FUNCTION__, hwif->name, dma_base, dma_base + num_ports - 1); + goto dma_remap_failure; + } + hwif->dma_base = (unsigned long) virt_dma_base; + hwif->dmatable_cpu = pci_alloc_consistent(hwif->pci_dev, IOC4_PRD_ENTRIES * IOC4_PRD_BYTES, &hwif->dmatable_dma); if (!hwif->dmatable_cpu) - goto dma_alloc_failure; + goto dma_pci_alloc_failure; hwif->sg_max_nents = IOC4_PRD_ENTRIES; @@ -411,6 +420,12 @@ dma_base2alloc_failure: printk(KERN_INFO "Changing from DMA to PIO mode for Drive %s\n", hwif->name); +dma_pci_alloc_failure: + iounmap(virt_dma_base); + +dma_remap_failure: + release_mem_region(dma_base, num_ports); + dma_alloc_failure: /* Disable DMA because we couldnot allocate any DMA maps */ hwif->autodma = 0; @@ -607,18 +622,15 @@ ide_init_sgiioc4(ide_hwif_t * hwif) hwif->ide_dma_lostirq = &sgiioc4_ide_dma_lostirq; hwif->ide_dma_timeout = &__ide_dma_timeout; - /* - * The IOC4 uses MMIO rather than Port IO. - * It also needs special workarounds for INB. - */ - default_hwif_mmiops(hwif); hwif->INB = &sgiioc4_INB; } static int __devinit sgiioc4_ide_setup_pci_device(struct pci_dev *dev, ide_pci_device_t * d) { - unsigned long base, ctl, dma_base, irqport; + unsigned long cmd_base, dma_base, irqport; + unsigned long bar0, cmd_phys_base, ctl; + void __iomem *virt_base; ide_hwif_t *hwif; int h; @@ -636,23 +648,32 @@ sgiioc4_ide_setup_pci_device(struct pci_dev *dev, ide_pci_device_t * d) } /* Get the CmdBlk and CtrlBlk Base Registers */ - base = pci_resource_start(dev, 0) + IOC4_CMD_OFFSET; - ctl = pci_resource_start(dev, 0) + IOC4_CTRL_OFFSET; - irqport = pci_resource_start(dev, 0) + IOC4_INTR_OFFSET; + bar0 = pci_resource_start(dev, 0); + virt_base = ioremap(bar0, pci_resource_len(dev, 0)); + if (virt_base == NULL) { + printk(KERN_ERR "%s: Unable to remap BAR 0 address: 0x%lx\n", + d->name, bar0); + return -ENOMEM; + } + cmd_base = (unsigned long) virt_base + IOC4_CMD_OFFSET; + ctl = (unsigned long) virt_base + IOC4_CTRL_OFFSET; + irqport = (unsigned long) virt_base + IOC4_INTR_OFFSET; dma_base = pci_resource_start(dev, 0) + IOC4_DMA_OFFSET; - if (!request_region(base, IOC4_CMD_CTL_BLK_SIZE, hwif->name)) { + cmd_phys_base = bar0 + IOC4_CMD_OFFSET; + if (!request_mem_region(cmd_phys_base, IOC4_CMD_CTL_BLK_SIZE, + hwif->name)) { printk(KERN_ERR - "%s : %s -- ERROR, Port Addresses " + "%s : %s -- ERROR, Addresses " "0x%p to 0x%p ALREADY in use\n", - __FUNCTION__, hwif->name, (void *) base, - (void *) base + IOC4_CMD_CTL_BLK_SIZE); + __FUNCTION__, hwif->name, (void *) cmd_phys_base, + (void *) cmd_phys_base + IOC4_CMD_CTL_BLK_SIZE); return -ENOMEM; } - if (hwif->io_ports[IDE_DATA_OFFSET] != base) { + if (hwif->io_ports[IDE_DATA_OFFSET] != cmd_base) { /* Initialize the IO registers */ - sgiioc4_init_hwif_ports(&hwif->hw, base, ctl, irqport); + sgiioc4_init_hwif_ports(&hwif->hw, cmd_base, ctl, irqport); memcpy(hwif->io_ports, hwif->hw.io_ports, sizeof (hwif->io_ports)); hwif->noprobe = !hwif->io_ports[IDE_DATA_OFFSET]; @@ -665,6 +686,9 @@ sgiioc4_ide_setup_pci_device(struct pci_dev *dev, ide_pci_device_t * d) hwif->cds = (struct ide_pci_device_s *) d; hwif->gendev.parent = &dev->dev;/* setup proper ancestral information */ + /* The IOC4 uses MMIO rather than Port IO. */ + default_hwif_mmiops(hwif); + /* Initializing chipset IRQ Registers */ hwif->OUTL(0x03, irqport + IOC4_INTR_SET * 4); From 3b98b087fc2daab67518d2baa8aef19a6ad82723 Mon Sep 17 00:00:00 2001 From: Nishanth Aravamudan Date: Thu, 31 Aug 2006 21:27:53 -0700 Subject: [PATCH 109/115] [PATCH] fix NUMA interleaving for huge pages Since vma->vm_pgoff is in units of smallpages, VMAs for huge pages have the lower HPAGE_SHIFT - PAGE_SHIFT bits always cleared, which results in badd offsets to the interleave functions. Take this difference from small pages into account when calculating the offset. This does add a 0-bit shift into the small-page path (via alloc_page_vma()), but I think that is negligible. Also add a BUG_ON to prevent the offset from growing due to a negative right-shift, which probably shouldn't be allowed anyways. Tested on an 8-memory node ppc64 NUMA box and got the interleaving I expected. Signed-off-by: Nishanth Aravamudan Signed-off-by: Adam Litke Cc: Andi Kleen Acked-by: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/mempolicy.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/mm/mempolicy.c b/mm/mempolicy.c index e07e27e846a..a9963ceddd6 100644 --- a/mm/mempolicy.c +++ b/mm/mempolicy.c @@ -1176,7 +1176,15 @@ static inline unsigned interleave_nid(struct mempolicy *pol, if (vma) { unsigned long off; - off = vma->vm_pgoff; + /* + * for small pages, there is no difference between + * shift and PAGE_SHIFT, so the bit-shift is safe. + * for huge pages, since vm_pgoff is in units of small + * pages, we need to shift off the always 0 bits to get + * a useful offset. + */ + BUG_ON(shift < PAGE_SHIFT); + off = vma->vm_pgoff >> (shift - PAGE_SHIFT); off += (addr - vma->vm_start) >> shift; return offset_il_node(pol, vma, off); } else From 4cfb04a9d3d6865d932d02616b27b89d3a634026 Mon Sep 17 00:00:00 2001 From: Badari Pulavarty Date: Thu, 31 Aug 2006 21:27:54 -0700 Subject: [PATCH 110/115] [PATCH] manage-jbd-its-own-slab fix Missed a place where I forgot to convert kfree() to kmem_cache_free() as part of jbd-manage-its-own-slab changes. Signed-off-by: Badari Pulavarty Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/jbd/transaction.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/jbd/transaction.c b/fs/jbd/transaction.c index de2e4cbbf79..f5169a96260 100644 --- a/fs/jbd/transaction.c +++ b/fs/jbd/transaction.c @@ -727,7 +727,7 @@ done: out: if (unlikely(frozen_buffer)) /* It's usually NULL */ - kfree(frozen_buffer); + jbd_slab_free(frozen_buffer, bh->b_size); JBUFFER_TRACE(jh, "exit"); return error; From a930363881c225fb52824145d1ba8f1a8c447dd8 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Thu, 31 Aug 2006 21:27:54 -0700 Subject: [PATCH 111/115] [PATCH] backlight last round of fixes Fix some more problems (inverted use of semaphores in some places). He also moved my checks into within the protected section which is better. Signed-off-by: Michael Hanselmann Signed-off-by: Benjamin Herrenschmidt Cc: "Antonino A. Daplas" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/macintosh/via-pmu-backlight.c | 4 ++-- drivers/video/aty/aty128fb.c | 19 ++++++++++--------- drivers/video/aty/atyfb_base.c | 19 ++++++++++--------- drivers/video/aty/radeon_backlight.c | 4 ++-- drivers/video/nvidia/nv_backlight.c | 18 ++++++++++-------- drivers/video/riva/fbdev.c | 19 ++++++++++--------- 6 files changed, 44 insertions(+), 39 deletions(-) diff --git a/drivers/macintosh/via-pmu-backlight.c b/drivers/macintosh/via-pmu-backlight.c index 4397fac55ba..a82f313d9dc 100644 --- a/drivers/macintosh/via-pmu-backlight.c +++ b/drivers/macintosh/via-pmu-backlight.c @@ -167,11 +167,11 @@ void __init pmu_backlight_init() pmu_backlight_data.max_brightness / 15); } - up(&bd->sem); + down(&bd->sem); bd->props->brightness = level; bd->props->power = FB_BLANK_UNBLANK; bd->props->update_status(bd); - down(&bd->sem); + up(&bd->sem); mutex_lock(&pmac_backlight_mutex); if (!pmac_backlight) diff --git a/drivers/video/aty/aty128fb.c b/drivers/video/aty/aty128fb.c index 106d428b72c..276a21530b9 100644 --- a/drivers/video/aty/aty128fb.c +++ b/drivers/video/aty/aty128fb.c @@ -1800,14 +1800,15 @@ static struct backlight_properties aty128_bl_data = { static void aty128_bl_set_power(struct fb_info *info, int power) { - if (info->bl_dev == NULL) - return; - mutex_lock(&info->bl_mutex); - up(&info->bl_dev->sem); - info->bl_dev->props->power = power; - __aty128_bl_update_status(info->bl_dev); - down(&info->bl_dev->sem); + + if (info->bl_dev) { + down(&info->bl_dev->sem); + info->bl_dev->props->power = power; + __aty128_bl_update_status(info->bl_dev); + up(&info->bl_dev->sem); + } + mutex_unlock(&info->bl_mutex); } @@ -1842,11 +1843,11 @@ static void aty128_bl_init(struct aty128fb_par *par) 219 * FB_BACKLIGHT_MAX / MAX_LEVEL); mutex_unlock(&info->bl_mutex); - up(&bd->sem); + down(&bd->sem); bd->props->brightness = aty128_bl_data.max_brightness; bd->props->power = FB_BLANK_UNBLANK; bd->props->update_status(bd); - down(&bd->sem); + up(&bd->sem); #ifdef CONFIG_PMAC_BACKLIGHT mutex_lock(&pmac_backlight_mutex); diff --git a/drivers/video/aty/atyfb_base.c b/drivers/video/aty/atyfb_base.c index 510e4ea296e..19a71f04578 100644 --- a/drivers/video/aty/atyfb_base.c +++ b/drivers/video/aty/atyfb_base.c @@ -2199,14 +2199,15 @@ static struct backlight_properties aty_bl_data = { static void aty_bl_set_power(struct fb_info *info, int power) { - if (info->bl_dev == NULL) - return; - mutex_lock(&info->bl_mutex); - up(&info->bl_dev->sem); - info->bl_dev->props->power = power; - __aty_bl_update_status(info->bl_dev); - down(&info->bl_dev->sem); + + if (info->bl_dev) { + down(&info->bl_dev->sem); + info->bl_dev->props->power = power; + __aty_bl_update_status(info->bl_dev); + up(&info->bl_dev->sem); + } + mutex_unlock(&info->bl_mutex); } @@ -2237,11 +2238,11 @@ static void aty_bl_init(struct atyfb_par *par) 0xFF * FB_BACKLIGHT_MAX / MAX_LEVEL); mutex_unlock(&info->bl_mutex); - up(&bd->sem); + down(&bd->sem); bd->props->brightness = aty_bl_data.max_brightness; bd->props->power = FB_BLANK_UNBLANK; bd->props->update_status(bd); - down(&bd->sem); + up(&bd->sem); #ifdef CONFIG_PMAC_BACKLIGHT mutex_lock(&pmac_backlight_mutex); diff --git a/drivers/video/aty/radeon_backlight.c b/drivers/video/aty/radeon_backlight.c index 1755dddf189..585eb7b9e63 100644 --- a/drivers/video/aty/radeon_backlight.c +++ b/drivers/video/aty/radeon_backlight.c @@ -195,11 +195,11 @@ void radeonfb_bl_init(struct radeonfb_info *rinfo) 217 * FB_BACKLIGHT_MAX / MAX_RADEON_LEVEL); mutex_unlock(&rinfo->info->bl_mutex); - up(&bd->sem); + down(&bd->sem); bd->props->brightness = radeon_bl_data.max_brightness; bd->props->power = FB_BLANK_UNBLANK; bd->props->update_status(bd); - down(&bd->sem); + up(&bd->sem); #ifdef CONFIG_PMAC_BACKLIGHT mutex_lock(&pmac_backlight_mutex); diff --git a/drivers/video/nvidia/nv_backlight.c b/drivers/video/nvidia/nv_backlight.c index 14c37c42191..5b75ae4e945 100644 --- a/drivers/video/nvidia/nv_backlight.c +++ b/drivers/video/nvidia/nv_backlight.c @@ -112,13 +112,15 @@ static struct backlight_properties nvidia_bl_data = { void nvidia_bl_set_power(struct fb_info *info, int power) { - if (info->bl_dev == NULL) - return; mutex_lock(&info->bl_mutex); - up(&info->bl_dev->sem); - info->bl_dev->props->power = power; - __nvidia_bl_update_status(info->bl_dev); - down(&info->bl_dev->sem); + + if (info->bl_dev) { + down(&info->bl_dev->sem); + info->bl_dev->props->power = power; + __nvidia_bl_update_status(info->bl_dev); + up(&info->bl_dev->sem); + } + mutex_unlock(&info->bl_mutex); } @@ -153,11 +155,11 @@ void nvidia_bl_init(struct nvidia_par *par) 0x534 * FB_BACKLIGHT_MAX / MAX_LEVEL); mutex_unlock(&info->bl_mutex); - up(&bd->sem); + down(&bd->sem); bd->props->brightness = nvidia_bl_data.max_brightness; bd->props->power = FB_BLANK_UNBLANK; bd->props->update_status(bd); - down(&bd->sem); + up(&bd->sem); #ifdef CONFIG_PMAC_BACKLIGHT mutex_lock(&pmac_backlight_mutex); diff --git a/drivers/video/riva/fbdev.c b/drivers/video/riva/fbdev.c index 90363943bd5..8ddb47a56b0 100644 --- a/drivers/video/riva/fbdev.c +++ b/drivers/video/riva/fbdev.c @@ -354,14 +354,15 @@ static struct backlight_properties riva_bl_data = { static void riva_bl_set_power(struct fb_info *info, int power) { - if (info->bl_dev == NULL) - return; - mutex_lock(&info->bl_mutex); - up(&info->bl_dev->sem); - info->bl_dev->props->power = power; - __riva_bl_update_status(info->bl_dev); - down(&info->bl_dev->sem); + + if (info->bl_dev) { + down(&info->bl_dev->sem); + info->bl_dev->props->power = power; + __riva_bl_update_status(info->bl_dev); + up(&info->bl_dev->sem); + } + mutex_unlock(&info->bl_mutex); } @@ -396,11 +397,11 @@ static void riva_bl_init(struct riva_par *par) 0x534 * FB_BACKLIGHT_MAX / MAX_LEVEL); mutex_unlock(&info->bl_mutex); - up(&bd->sem); + down(&bd->sem); bd->props->brightness = riva_bl_data.max_brightness; bd->props->power = FB_BLANK_UNBLANK; bd->props->update_status(bd); - down(&bd->sem); + up(&bd->sem); #ifdef CONFIG_PMAC_BACKLIGHT mutex_lock(&pmac_backlight_mutex); From a188ad2bc7dbfa16ccdcaa8d43ade185b969baff Mon Sep 17 00:00:00 2001 From: "George G. Davis" Date: Sat, 2 Sep 2006 18:43:20 +0100 Subject: [PATCH 112/115] [ARM] 3762/1: Fix ptrace cache coherency bug for ARM1136 VIPT nonaliasing Harvard caches Patch from George G. Davis Resolve ARM1136 VIPT non-aliasing cache coherency issues observed when using ptrace to set breakpoints and cleanup copy_{to,from}_user_page() while we're here as requested by Russell King because "it's also far too heavy on non-v6 CPUs". NOTES: 1. Only access_process_vm() calls copy_{to,from}_user_page(). 2. access_process_vm() calls get_user_pages() to pin down the "page". 3. get_user_pages() calls flush_dcache_page(page) which ensures cache coherency between kernel and userspace mappings of "page". However flush_dcache_page(page) may not invalidate I-Cache over this range for all cases, specifically, I-Cache is not invalidated for the VIPT non-aliasing case. So memory is consistent between kernel and user space mappings of "page" but I-Cache may still be hot over this range. IOW, we don't have to worry about flush_cache_page() before memcpy(). 4. Now, for the copy_to_user_page() case, after memcpy(), we must flush the caches so memory is consistent with kernel cache entries and invalidate the I-Cache if this mm region is executable. We don't need to do anything after memcpy() for the copy_from_user_page() case since kernel cache entries will be invalidated via the same process above if we access "page" again. The flush_ptrace_access() function (borrowed from SPARC64 implementation) is added to handle cache flushing after memcpy() for the copy_to_user_page() case. Signed-off-by: George G. Davis Signed-off-by: Russell King --- arch/arm/mm/flush.c | 26 ++++++++++++++++++++++++++ include/asm-arm/cacheflush.h | 18 +++++++++++++++--- 2 files changed, 41 insertions(+), 3 deletions(-) diff --git a/arch/arm/mm/flush.c b/arch/arm/mm/flush.c index b103e56806b..d438ce41cdd 100644 --- a/arch/arm/mm/flush.c +++ b/arch/arm/mm/flush.c @@ -87,6 +87,32 @@ void flush_cache_page(struct vm_area_struct *vma, unsigned long user_addr, unsig if (cache_is_vipt_aliasing()) flush_pfn_alias(pfn, user_addr); } + +void flush_ptrace_access(struct vm_area_struct *vma, struct page *page, + unsigned long uaddr, void *kaddr, + unsigned long len, int write) +{ + if (cache_is_vivt()) { + if (cpu_isset(smp_processor_id(), vma->vm_mm->cpu_vm_mask)) { + unsigned long addr = (unsigned long)kaddr; + __cpuc_coherent_kern_range(addr, addr + len); + } + return; + } + + if (cache_is_vipt_aliasing()) { + flush_pfn_alias(page_to_pfn(page), uaddr); + return; + } + + /* VIPT non-aliasing cache */ + if (cpu_isset(smp_processor_id(), vma->vm_mm->cpu_vm_mask) && + vma->vm_flags | VM_EXEC) { + unsigned long addr = (unsigned long)kaddr; + /* only flushing the kernel mapping on non-aliasing VIPT */ + __cpuc_coherent_kern_range(addr, addr + len); + } +} #else #define flush_pfn_alias(pfn,vaddr) do { } while (0) #endif diff --git a/include/asm-arm/cacheflush.h b/include/asm-arm/cacheflush.h index fe0c744e026..e4a2569c636 100644 --- a/include/asm-arm/cacheflush.h +++ b/include/asm-arm/cacheflush.h @@ -247,14 +247,12 @@ extern void dmac_flush_range(unsigned long, unsigned long); */ #define copy_to_user_page(vma, page, vaddr, dst, src, len) \ do { \ - flush_cache_page(vma, vaddr, page_to_pfn(page));\ memcpy(dst, src, len); \ - flush_dcache_page(page); \ + flush_ptrace_access(vma, page, vaddr, dst, len, 1);\ } while (0) #define copy_from_user_page(vma, page, vaddr, dst, src, len) \ do { \ - flush_cache_page(vma, vaddr, page_to_pfn(page));\ memcpy(dst, src, len); \ } while (0) @@ -285,10 +283,24 @@ flush_cache_page(struct vm_area_struct *vma, unsigned long user_addr, unsigned l __cpuc_flush_user_range(addr, addr + PAGE_SIZE, vma->vm_flags); } } + +static inline void +flush_ptrace_access(struct vm_area_struct *vma, struct page *page, + unsigned long uaddr, void *kaddr, + unsigned long len, int write) +{ + if (cpu_isset(smp_processor_id(), vma->vm_mm->cpu_vm_mask)) { + unsigned long addr = (unsigned long)kaddr; + __cpuc_coherent_kern_range(addr, addr + len); + } +} #else extern void flush_cache_mm(struct mm_struct *mm); extern void flush_cache_range(struct vm_area_struct *vma, unsigned long start, unsigned long end); extern void flush_cache_page(struct vm_area_struct *vma, unsigned long user_addr, unsigned long pfn); +extern void flush_ptrace_access(struct vm_area_struct *vma, struct page *page, + unsigned long uaddr, void *kaddr, + unsigned long len, int write); #endif /* From 3b6362b833b9f7a9d4222cf1bb35f99c411abb31 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Sat, 2 Sep 2006 21:22:16 +0400 Subject: [PATCH 113/115] [PATCH] eligible_child: remove an obsolete ->tgid check It is not possible to find a sub-thread in ->children/->ptrace_children lists, ptrace_attach() does not allow to attach to sub-threads. Even if it was possible to ptrace the task from the same thread group, we can't allow to release ->group_leader while there are others (ptracer) threads in the same group. Signed-off-by: Oleg Nesterov Signed-off-by: Linus Torvalds --- kernel/exit.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/exit.c b/kernel/exit.c index a4c19a52ce4..d891883420f 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -1053,7 +1053,7 @@ static int eligible_child(pid_t pid, int options, struct task_struct *p) * Do not consider thread group leaders that are * in a non-empty thread group: */ - if (current->tgid != p->tgid && delay_group_leader(p)) + if (delay_group_leader(p)) return 2; if (security_task_wait(p)) From 8f2779491abe57b8f0f27d96b6a3edd5a8775a58 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Mon, 4 Sep 2006 10:36:15 +1000 Subject: [PATCH 114/115] [PATCH] powerpc: Fix typo in powermac platform functions New sparse caught that typo which could have caused erratic hardware behaviour on some machines if the platform functions are used by the firmware to change bits in some FCR registers. Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Linus Torvalds --- arch/powerpc/platforms/powermac/pfunc_base.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/powerpc/platforms/powermac/pfunc_base.c b/arch/powerpc/platforms/powermac/pfunc_base.c index 6d66359ec8c..aacfa59595d 100644 --- a/arch/powerpc/platforms/powermac/pfunc_base.c +++ b/arch/powerpc/platforms/powermac/pfunc_base.c @@ -256,7 +256,7 @@ static struct pmf_handlers macio_mmio_handlers = { .write_reg32 = macio_do_write_reg32, .read_reg32 = macio_do_read_reg32, .write_reg8 = macio_do_write_reg8, - .read_reg32 = macio_do_read_reg8, + .read_reg8 = macio_do_read_reg8, .read_reg32_msrx = macio_do_read_reg32_msrx, .read_reg8_msrx = macio_do_read_reg8_msrx, .write_reg32_slm = macio_do_write_reg32_slm, From c336923b668fdcf0312efbec3b44895d713f4d81 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 3 Sep 2006 19:19:48 -0700 Subject: [PATCH 115/115] Linux 2.6.18-rc6 --- Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Makefile b/Makefile index 33559b56644..c9ed7b42873 100644 --- a/Makefile +++ b/Makefile @@ -1,7 +1,7 @@ VERSION = 2 PATCHLEVEL = 6 SUBLEVEL = 18 -EXTRAVERSION = -rc5 +EXTRAVERSION = -rc6 NAME=Crazed Snow-Weasel # *DOCUMENTATION*