From 5f954c3426190f7ae432a09abd62164d5d14c709 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Fri, 24 Jul 2009 12:39:49 +0200 Subject: [PATCH 1/5] [S390] hibernation: fix lowcore handling Our swsusp_arch_suspend() backend implementation disables prefixing by setting the contents of the prefix register to 0. However afterwards common code functions are called which might access percpu data structures. Since the lowcore contains e.g. the percpu base pointer this isn't a good idea. So fix this by copying the hibernating cpu's lowcore to absolute address zero. Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- arch/s390/kernel/early.c | 4 ++-- arch/s390/power/swsusp_asm64.S | 13 ++++++++----- 2 files changed, 10 insertions(+), 7 deletions(-) diff --git a/arch/s390/kernel/early.c b/arch/s390/kernel/early.c index f9b144049dc..8d15314381e 100644 --- a/arch/s390/kernel/early.c +++ b/arch/s390/kernel/early.c @@ -210,7 +210,7 @@ static noinline __init void detect_machine_type(void) machine_flags |= MACHINE_FLAG_VM; } -static void early_pgm_check_handler(void) +static __init void early_pgm_check_handler(void) { unsigned long addr; const struct exception_table_entry *fixup; @@ -222,7 +222,7 @@ static void early_pgm_check_handler(void) S390_lowcore.program_old_psw.addr = fixup->fixup | PSW_ADDR_AMODE; } -void setup_lowcore_early(void) +static noinline __init void setup_lowcore_early(void) { psw_t psw; diff --git a/arch/s390/power/swsusp_asm64.S b/arch/s390/power/swsusp_asm64.S index 76d688da32f..e27bd316489 100644 --- a/arch/s390/power/swsusp_asm64.S +++ b/arch/s390/power/swsusp_asm64.S @@ -40,11 +40,11 @@ swsusp_arch_suspend: /* Store prefix register on stack */ stpx __SF_EMPTY(%r15) - /* Setup base register for lowcore (absolute 0) */ - llgf %r1,__SF_EMPTY(%r15) + /* Save prefix register contents for lowcore */ + llgf %r4,__SF_EMPTY(%r15) /* Get pointer to save area */ - aghi %r1,0x1000 + lghi %r1,0x1000 /* Store registers */ mvc 0x318(4,%r1),__SF_EMPTY(%r15) /* move prefix to lowcore */ @@ -79,8 +79,11 @@ swsusp_arch_suspend: xc __SF_EMPTY(4,%r15),__SF_EMPTY(%r15) spx __SF_EMPTY(%r15) - /* Setup lowcore */ - brasl %r14,setup_lowcore_early + lghi %r2,0 + lghi %r3,2*PAGE_SIZE + lghi %r5,2*PAGE_SIZE +1: mvcle %r2,%r4,0 + jo 1b /* Save image */ brasl %r14,swsusp_save From c63b196afcf22405527abe4c2c57926a5bbd6fc9 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Fri, 24 Jul 2009 12:39:50 +0200 Subject: [PATCH 2/5] [S390] hibernation: fix register corruption on machine checks swsusp_arch_suspend() actually saves all cpu register contents on hibernation. Machine checks must be disabled since swsusp_arch_suspend() stores register contents to their lowcore save areas. That's the same place where register contents on machine checks would be saved. To avoid register corruption disable machine checks. We must also disable machine checks in the new psw mask for program checks, since swsusp_arch_suspend() may generate program checks. Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- arch/s390/power/swsusp.c | 36 ++++++++++++++++++++++------------ arch/s390/power/swsusp_asm64.S | 22 +-------------------- 2 files changed, 25 insertions(+), 33 deletions(-) diff --git a/arch/s390/power/swsusp.c b/arch/s390/power/swsusp.c index e6a4fe9f5f2..bd1f5c6b0b8 100644 --- a/arch/s390/power/swsusp.c +++ b/arch/s390/power/swsusp.c @@ -7,24 +7,36 @@ * */ +#include -/* - * save CPU registers before creating a hibernation image and before - * restoring the memory state from it - */ void save_processor_state(void) { - /* implentation contained in the - * swsusp_arch_suspend function + /* swsusp_arch_suspend() actually saves all cpu register contents. + * Machine checks must be disabled since swsusp_arch_suspend() stores + * register contents to their lowcore save areas. That's the same + * place where register contents on machine checks would be saved. + * To avoid register corruption disable machine checks. + * We must also disable machine checks in the new psw mask for + * program checks, since swsusp_arch_suspend() may generate program + * checks. Disabling machine checks for all other new psw masks is + * just paranoia. */ + local_mcck_disable(); + /* Disable lowcore protection */ + __ctl_clear_bit(0,28); + S390_lowcore.external_new_psw.mask &= ~PSW_MASK_MCHECK; + S390_lowcore.svc_new_psw.mask &= ~PSW_MASK_MCHECK; + S390_lowcore.io_new_psw.mask &= ~PSW_MASK_MCHECK; + S390_lowcore.program_new_psw.mask &= ~PSW_MASK_MCHECK; } -/* - * restore the contents of CPU registers - */ void restore_processor_state(void) { - /* implentation contained in the - * swsusp_arch_resume function - */ + S390_lowcore.external_new_psw.mask |= PSW_MASK_MCHECK; + S390_lowcore.svc_new_psw.mask |= PSW_MASK_MCHECK; + S390_lowcore.io_new_psw.mask |= PSW_MASK_MCHECK; + S390_lowcore.program_new_psw.mask |= PSW_MASK_MCHECK; + /* Enable lowcore protection */ + __ctl_set_bit(0,28); + local_mcck_enable(); } diff --git a/arch/s390/power/swsusp_asm64.S b/arch/s390/power/swsusp_asm64.S index e27bd316489..b26df5c5933 100644 --- a/arch/s390/power/swsusp_asm64.S +++ b/arch/s390/power/swsusp_asm64.S @@ -32,11 +32,6 @@ swsusp_arch_suspend: /* Deactivate DAT */ stnsm __SF_EMPTY(%r15),0xfb - /* Switch off lowcore protection */ - stctg %c0,%c0,__SF_EMPTY(%r15) - ni __SF_EMPTY+4(%r15),0xef - lctlg %c0,%c0,__SF_EMPTY(%r15) - /* Store prefix register on stack */ stpx __SF_EMPTY(%r15) @@ -88,11 +83,6 @@ swsusp_arch_suspend: /* Save image */ brasl %r14,swsusp_save - /* Switch on lowcore protection */ - stctg %c0,%c0,__SF_EMPTY(%r15) - oi __SF_EMPTY+4(%r15),0x10 - lctlg %c0,%c0,__SF_EMPTY(%r15) - /* Restore prefix register and return */ lghi %r1,0x1000 spx 0x318(%r1) @@ -120,11 +110,6 @@ swsusp_arch_resume: /* Deactivate DAT */ stnsm __SF_EMPTY(%r15),0xfb - /* Switch off lowcore protection */ - stctg %c0,%c0,__SF_EMPTY(%r15) - ni __SF_EMPTY+4(%r15),0xef - lctlg %c0,%c0,__SF_EMPTY(%r15) - /* Set prefix page to zero */ xc __SF_EMPTY(4,%r15),__SF_EMPTY(%r15) spx __SF_EMPTY(%r15) @@ -178,7 +163,7 @@ swsusp_arch_resume: /* Load old stack */ lg %r15,0x2f8(%r13) - /* Pointer to save arae */ + /* Pointer to save area */ lghi %r13,0x1000 #ifdef CONFIG_SMP @@ -190,11 +175,6 @@ swsusp_arch_resume: /* Restore prefix register */ spx 0x318(%r13) - /* Switch on lowcore protection */ - stctg %c0,%c0,__SF_EMPTY(%r15) - oi __SF_EMPTY+4(%r15),0x10 - lctlg %c0,%c0,__SF_EMPTY(%r15) - /* Activate DAT */ stosm __SF_EMPTY(%r15),0x04 From 3a6ba4600d6fb913ddb0dd08843ad75405795883 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Fri, 24 Jul 2009 12:39:51 +0200 Subject: [PATCH 3/5] [S390] vdso: fix per cpu area allocation vdso per cpu area allocation in smp_prepare_cpus() happens with GFP_KERNEL but irqs disabled. Triggers this one: Badness at kernel/lockdep.c:2280 Modules linked in: CPU: 0 Not tainted 2.6.30 #2 Process swapper (pid: 1, task: 000000003fe88000, ksp: 000000003fe87eb8) Krnl PSW : 0400c00180000000 0000000000083360 (lockdep_trace_alloc+0xec/0xf8) [...] Call Trace: ([<00000000000832b6>] lockdep_trace_alloc+0x42/0xf8) [<00000000000b1880>] __alloc_pages_internal+0x3e8/0x5c4 [<00000000000b1b4a>] __get_free_pages+0x3a/0xb0 [<0000000000026546>] vdso_alloc_per_cpu+0x6a/0x18c [<00000000005eff82>] smp_prepare_cpus+0x322/0x594 [<00000000005e8232>] kernel_init+0x76/0x398 [<000000000001bb1e>] kernel_thread_starter+0x6/0xc [<000000000001bb18>] kernel_thread_starter+0x0/0xc Fix this by moving the allocation out of the irqs disabled section. Reported-by: Christian Borntraeger Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- arch/s390/kernel/smp.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c index 2270730f535..be2cae08340 100644 --- a/arch/s390/kernel/smp.c +++ b/arch/s390/kernel/smp.c @@ -687,13 +687,14 @@ void __init smp_prepare_cpus(unsigned int max_cpus) #ifndef CONFIG_64BIT if (MACHINE_HAS_IEEE) lowcore->extended_save_area_addr = (u32) save_area; -#else - if (vdso_alloc_per_cpu(smp_processor_id(), lowcore)) - BUG(); #endif set_prefix((u32)(unsigned long) lowcore); local_mcck_enable(); local_irq_enable(); +#ifdef CONFIG_64BIT + if (vdso_alloc_per_cpu(smp_processor_id(), &S390_lowcore)) + BUG(); +#endif for_each_possible_cpu(cpu) if (cpu != smp_processor_id()) smp_create_idle(cpu); From 1277580fe5dfb5aef84854bdb7983657df00b920 Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Fri, 24 Jul 2009 12:39:52 +0200 Subject: [PATCH 4/5] [S390] vdso: clock_gettime of CLOCK_THREAD_CPUTIME_ID with noexec=on The combination of noexec=on and a clock_gettime call with clock id CLOCK_THREAD_CPUTIME_ID is broken. The vdso code switches to the access register mode to get access to the per-cpu data structure to execute the magic ectg instruction. After the ectg instruction the code always switches back to the primary mode but for noexec=on the correct mode is the secondary mode. The effect of the bug is that the user space program looses the access to all mappings without PROT_EXEC, e.g. the stack. The problem is fixed by restoring the mode that has been active before the switch to the access register mode. Signed-off-by: Martin Schwidefsky --- arch/s390/kernel/vdso64/clock_gettime.S | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/arch/s390/kernel/vdso64/clock_gettime.S b/arch/s390/kernel/vdso64/clock_gettime.S index 79dbfee831e..49106c6e6f8 100644 --- a/arch/s390/kernel/vdso64/clock_gettime.S +++ b/arch/s390/kernel/vdso64/clock_gettime.S @@ -88,10 +88,17 @@ __kernel_clock_gettime: llilh %r4,0x0100 sar %a4,%r4 lghi %r4,0 + epsw %r5,0 sacf 512 /* Magic ectg instruction */ .insn ssf,0xc80100000000,__VDSO_ECTG_BASE(4),__VDSO_ECTG_USER(4),4 - sacf 0 - sar %a4,%r2 + tml %r5,0x4000 + jo 11f + tml %r5,0x8000 + jno 10f + sacf 256 + j 11f +10: sacf 0 +11: sar %a4,%r2 algr %r1,%r0 /* r1 = cputime as TOD value */ mghi %r1,1000 /* convert to nanoseconds */ srlg %r1,%r1,12 /* r1 = cputime in nanosec */ From 8d406c6de2e67b5bae3c43b62b492c4ff63afb92 Mon Sep 17 00:00:00 2001 From: Felix Beck Date: Fri, 24 Jul 2009 12:39:53 +0200 Subject: [PATCH 5/5] [S390] zcrypt: fix scheduling of hrtimer ap_poll_timer Every time a request is enqueued or there is some work outstanding from the ap_tasklet, the ap_poll_timer is scheduled again. Unfortunately it was permanently called. It looked as if it was started in the past and thus imediately expired. This has been changed. First it is checked if the hrtimer is already expired. Then the expiring time is forwarded and the timer restarted. Signed-off-by: Felix Beck Signed-off-by: Martin Schwidefsky --- drivers/s390/crypto/ap_bus.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/drivers/s390/crypto/ap_bus.c b/drivers/s390/crypto/ap_bus.c index 727a809636d..ed3dcdea7fe 100644 --- a/drivers/s390/crypto/ap_bus.c +++ b/drivers/s390/crypto/ap_bus.c @@ -1145,12 +1145,17 @@ ap_config_timeout(unsigned long ptr) */ static inline void ap_schedule_poll_timer(void) { + ktime_t hr_time; if (ap_using_interrupts() || ap_suspend_flag) return; if (hrtimer_is_queued(&ap_poll_timer)) return; - hrtimer_start(&ap_poll_timer, ktime_set(0, poll_timeout), - HRTIMER_MODE_ABS); + if (ktime_to_ns(hrtimer_expires_remaining(&ap_poll_timer)) <= 0) { + hr_time = ktime_set(0, poll_timeout); + hrtimer_forward_now(&ap_poll_timer, hr_time); + hrtimer_restart(&ap_poll_timer); + } + return; } /**