Update memcpy() to add two new feature sections: one for aligning the
destination before copying and one for copying using aligned load
and store doubles.
These new feature sections will only affect Power6 and Cell because
the CPU feature bit was only added to these two processors.
Power6 gets its best performance in memcpy() when aligning neither the
source nor the destination, while Cell gets its best performance when
just the destination is aligned. But in order to save on CPU feature
bits we can use the previously added CPU_FTR_CP_USE_DCBTZ feature bit
to differentiate between Power6 and Cell (because CPU_FTR_CP_USE_DCBTZ
was added to Cell but not Power6).
The first feature section acts to nop out the branch that takes us to
the code that aligns us to an eight byte boundary for the destination.
We only want to nop out this branch on Power6.
So the ALT_FTR_SECTION_END() for this feature section creates a test
mask of the two feature bits ORed together and provides an expected
result of just CPU_FTR_UNALIGNED_LD_STD, thus we nop out the branch
if we're on a CPU that has CPU_FTR_UNALIGNED_LD_STD set and
CPU_FTR_CP_USE_DCBTZ unset.
For the second feature section added, if we're on a CPU that has the
CPU_FTR_UNALIGNED_LD_STD bit set then we don't want to do the copy
with aligned loads and stores (and the appropriate shifting left and
right instructions), so we want to nop out the branch to
.Lsrc_unaligned.
The andi. used for this branch is moved to just above the branch
because this allows us to nop out both instructions with just one
feature section which gives us better performance and doesn't hurt
readability which two separate feature sections did.
Moving the andi. to just above the branch doesn't have any noticeable
negative effect on the remaining 64bit processors (the ones that
didn't have this feature bit added).
On Cell this simple modification results in an improvement to measured
memcpy() bandwidth of up to 50% in the hot cache case and up to 15% in
the cold cache case.
On Power6 we get memory bandwidth results that are up to three times
faster in the hot cache case and up to 50% faster in the cold cache
case.
Commit 2a9294369b ("powerpc: Add new CPU
feature: CPU_FTR_CP_USE_DCBTZ") was where CPU_FTR_CP_USE_DCBTZ was
added.
To say that Cell gets its best performance in memcpy() with just the
destination aligned is true but only for the reason that the indirect
shift and rotate instructions, sld and srd, are microcoded on Cell.
This means that either the destination or the source can be aligned,
but not both, and seeing as we get better performance with the
destination aligned we choose this option.
While we're at it make a one line change from cmpldi r1,... to
cmpldi cr1,... for consistency.
Signed-off-by: Mark Nelson <markn@au1.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
After Becky's work we can almost have different DMA offsets
between on-chip devices and PCI. Almost because there's a
problem with the non-coherent DMA code that basically ignores
the programmed offset to use the global one for everything.
This fixes it.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
This new copy_4K_page() function was originally tuned for the best
performance on the Cell processor, but after testing on more 64bit
powerpc chips it was found that with a small modification it either
matched the performance offered by the current mainline version or
bettered it by a small amount.
It was found that on a Cell-based QS22 blade the amount of system
time measured when compiling a 2.6.26 pseries_defconfig decreased
by 4%. Using the same test, a 4-way 970MP machine saw a decrease of
2% in system time. No noticeable change was seen on Power4, Power5
or Power6.
The 4096 byte page is copied in thirty-two 128 byte strides. An
initial setup loop executes dcbt instructions for the whole source
page and dcbz instructions for the whole destination page. To do
this, the cache line size is retrieved from ppc64_caches.
A new CPU feature bit, CPU_FTR_CP_USE_DCBTZ, (introduced in the
previous patch) is used to make the modification to this new copy
routine - on Power4, 970 and Cell the feature bit is set so the
setup loop is executed, but on all other 64bit chips the setup
loop is nop'ed out.
Signed-off-by: Mark Nelson <markn@au1.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Now that arch/ppc is gone and CONFIG_PPC_MERGE is always set, remove
the dead code associated with !CONFIG_PPC_MERGE from arch/powerpc
and include/asm-powerpc.
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
On 32-bit architectures PAGE_ALIGN() truncates 64-bit values to the 32-bit
boundary. For example:
u64 val = PAGE_ALIGN(size);
always returns a value < 4GB even if size is greater than 4GB.
The problem resides in PAGE_MASK definition (from include/asm-x86/page.h for
example):
#define PAGE_SHIFT 12
#define PAGE_SIZE (_AC(1,UL) << PAGE_SHIFT)
#define PAGE_MASK (~(PAGE_SIZE-1))
...
#define PAGE_ALIGN(addr) (((addr)+PAGE_SIZE-1)&PAGE_MASK)
The "~" is performed on a 32-bit value, so everything in "and" with
PAGE_MASK greater than 4GB will be truncated to the 32-bit boundary.
Using the ALIGN() macro seems to be the right way, because it uses
typeof(addr) for the mask.
Also move the PAGE_ALIGN() definitions out of include/asm-*/page.h in
include/linux/mm.h.
See also lkml discussion: http://lkml.org/lkml/2008/6/11/237
[akpm@linux-foundation.org: fix drivers/media/video/uvc/uvc_queue.c]
[akpm@linux-foundation.org: fix v850]
[akpm@linux-foundation.org: fix powerpc]
[akpm@linux-foundation.org: fix arm]
[akpm@linux-foundation.org: fix mips]
[akpm@linux-foundation.org: fix drivers/media/video/pvrusb2/pvrusb2-dvb.c]
[akpm@linux-foundation.org: fix drivers/mtd/maps/uclinux.c]
[akpm@linux-foundation.org: fix powerpc]
Signed-off-by: Andrea Righi <righi.andrea@gmail.com>
Cc: <linux-arch@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Replace ifdef clutter with the PPC_LONG and PPC_LONG_ALIGN macros
for readability.
No change to the generated code.
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
__WARN() is not defined for all configs, use WARN_ON(1) instead.
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
To allow for a single kernel image on e500 v1/v2/mc we need to fixup lwsync
at runtime. On e500v1/v2 lwsync causes an illop so we need to patch up
the code. We default to 'sync' since that is always safe and if the cpu
is capable we will replace 'sync' with 'lwsync'.
We introduce CPU_FTR_LWSYNC as a way to determine at runtime if this is
needed. This flag could be moved elsewhere since we dont really use it
for the normal CPU_FTR purpose.
Finally we only store the relative offset in the fixup section to keep it
as small as possible rather than using a full fixup_entry.
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
We need to use PPC_LCMPI otherwise we get compile errors like:
arch/powerpc/lib/feature-fixups-test.S: Assembler messages:
arch/powerpc/lib/feature-fixups-test.S:142: Error: Unrecognized opcode: `cmpdi'
arch/powerpc/lib/feature-fixups-test.S:149: Error: Unrecognized opcode: `cmpdi'
arch/powerpc/lib/feature-fixups-test.S:164: Error: Unrecognized opcode: `cmpdi'
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
On PowerPC processors with non-coherent cache architectures the DMA
subsystem calls invalidate_dcache_range() before performing a DMA read
operation. If the address and length of the DMA buffer are not aligned
to a cache-line boundary this can result in memory outside of the DMA
buffer being invalidated in the cache. If this memory has an
uncommitted store then the data will be lost and a subsequent read of
that address will result in an old value being returned from main memory.
Only when the DMA buffer starts on a cache-line boundary and is an exact
mutiple of the cache-line size can invalidate_dcache_range() be called,
otherwise flush_dcache_range() must be called. flush_dcache_range()
will first flush uncommitted writes, and then invalidate the cache.
Signed-off-by: Andrew Lewis <andrew-lewis at netspace.net.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
This commit adds tests of the feature fixup code, they are run during
boot if CONFIG_FTR_FIXUP_SELFTEST=y. Some of the tests manually invoke
the patching routines to check their behaviour, and others use the
macros and so are patched during the normal patching done during boot.
Because we have two sets of macros with different names, we use a macro
to generate the test of the macros, very niiiice.
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
This commit adds the logic to patch alternative sections. This is fairly
straightforward, except for branches. Relative branches that jump from
inside the else section to outside of it need to be translated as they're
moved, otherwise they will jump to the wrong location.
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
The current feature section logic only supports nop'ing out code, this means
if you want to choose at runtime between instruction sequences, one or both
cases will have to execute the nop'ed out contents of the other section, eg:
BEGIN_FTR_SECTION
or 1,1,1
END_FTR_SECTION_IFSET(FOO)
BEGIN_FTR_SECTION
or 2,2,2
END_FTR_SECTION_IFCLR(FOO)
and the resulting code will be either,
or 1,1,1
nop
or,
nop
or 2,2,2
For small code segments this is fine, but for larger code blocks and in
performance criticial code segments, it would be nice to avoid the nops.
This commit starts to implement logic to allow the following:
BEGIN_FTR_SECTION
or 1,1,1
FTR_SECTION_ELSE
or 2,2,2
ALT_FTR_SECTION_END_IFSET(FOO)
and the resulting code will be:
or 1,1,1
or,
or 2,2,2
We achieve this by extending the existing FTR macros. The current feature
section semantic just becomes a special case, ie. if the else case is empty
we nop out the default case.
The key limitation is that the size of the else case must be less than or
equal to the size of the default case. If the else case is smaller the
remainder of the section is nop'ed.
We let the linker put the else case code in with the rest of the text,
so that relative branches from the else case are more likley to link,
this has the disadvantage that we can't free the unused else cases.
This commit introduces the required macro and linker script changes, but
does not enable the patching of the alternative sections.
We also need to update two hand-made section entries in reg.h and timex.h
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
The logic to patch CPU feature sections lives in cputable.c, but these days
it's used for CPU features as well as firmware features. Move it into
it's own file for neatness and as preparation for some additions.
While we're moving the code, we pull the loop body logic into a separate
routine, and remove a comment which doesn't apply anymore.
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Acked-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Add tests of the existing code patching routines, as well as the new
routines added in the last commit. The self-tests are run late in boot
when CONFIG_CODE_PATCHING_SELFTEST=y, which depends on DEBUG_KERNEL=y.
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Acked-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
This commit adds some new routines for patching code, which will be used
in a following commit.
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
If you pass a target value to create_branch() which is more than 32MB - 4,
or - 32MB away from the branch site, then it's impossible to create an
immediate branch. The current code doesn't check, which will lead to us
creating a branch to somewhere else - which is bad.
For code that cares to check we return 0, which is easy to check for, and
for code that doesn't at least we'll be creating an illegal instruction,
rather than a branch to some random address.
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Acked-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Currently create_branch() creates a branch instruction for you, and
patches it into the call site. In some circumstances it would be nice
to be able to create the instruction and patch it later, and also some
code might want to check for errors in the branch creation before
doing the patching. A future commit will change create_branch() to
check for errors.
For callers that don't care, replace create_branch() with
patch_branch(), which just creates the branch and patches it directly.
While we're touching all the callers, change to using unsigned int *,
as this seems to match usage better. That allows (and requires) us to
remove the volatile in the definition of vector in powermac/smp.c and
mpc86xx_smp.c, that's correct because now that we're passing vector as
an unsigned int * the compiler knows that it's value might change
across the patch_branch() call.
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Acked-by: Kumar Gala <galak@kernel.crashing.org>
Acked-by: Jon Loeliger <jdl@freescale.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
We currently have a few routines for patching code in asm/system.h, because
they didn't fit anywhere else. I'd like to clean them up a little and add
some more, so first move them into a dedicated C file - they don't need to
be inlined.
While we're moving the code, drop create_function_call(), it's intended
caller never got merged and will be replaced in future with something
different.
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Acked-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
GCC 4.4.x looks to be adding support for generating out-of-line register
saves/restores based on:
http://gcc.gnu.org/ml/gcc-patches/2008-04/msg01678.html
This breaks the kernel if we enable CONFIG_CC_OPTIMIZE_FOR_SIZE. To fix
this we add the use the save/restore code from gcc and simplified it down
for our needs (integer only).
Additionally, we have to link this code into each module. The other
solution was to add EXPORT_SYMBOL() which meant going through the
trampoline which seemed nonsensical for these out-of-line routines.
Finally, we add some checks to prom_init_check.sh to ignore the
out-of-line save/restore functions.
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
This fixes a few more miscellaneous compile problems with ARCH=ppc.
1. Don't compile devres.c on ARCH=ppc, it doesn't have ioremap_flags.
2. Include <asm/irq.h> in setup.c for the __DO_IRQ_CANON definition.
3. Include <linux/proc_fs.h> in residual.c for the
definition of create_proc_read_entry.
4. Fix xchg_ptr to be a static inline to eliminate a compiler warning.
Signed-off-by: Paul Mackerras <paulus@samba.org>
We provide an ioremap_flags, so this provides a corresponding
devm_ioremap_prot. The slight name difference is at Ben
Herrenschmidt's request as he plans on changing ioremap_flags to
ioremap_prot in the future.
Signed-off-by: Emil Medve <Emilian.Medve@Freescale.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Acked-by: Tejun Heo <htejun@gmail.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
The rheap allocation function, rh_alloc, could call kmalloc with GFP_KERNEL.
This can sleep, which means you couldn't hold a spinlock while called rh_alloc.
Change all kmalloc calls to use GFP_ATOMIC so that it won't sleep. This is
safe because only small blocks are allocated.
Signed-off-by: Timur Tabi <timur@freescale.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
We have an assembly version of strncmp for the bootwrapper, but not
for the kernel, so we end up using the C version in the kernel. This
takes the strncmp code from the bootup and copies it to the kernel
proper, adding two instructions so it copes correctly with len==0.
Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Instead of having in the makefile all the option that
requires rheap, we define a configuration symbol
and when needed we make sure it's selected.
Signed-off-by: Sylvain Munaut <tnt@246tNt.com>
Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
Theses can be useful in modules too. So we export them.
Signed-off-by: Sylvain Munaut <tnt@246tNt.com>
Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
This patch introduces zalloc_maybe_bootmem and uses it so that we don't
have to mark a whole (largish) routine as __init_ref_ok.
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Linus made this suggestion for the x86 merge and this starts the process
for powerpc. We assume that CONFIG_PPC64 implies CONFIG_PPC_MERGE and
CONFIG_PPC_STD_MMU_32 implies CONFIG_PPC_STD_MMU.
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Create a helper function (alloc_maybe_bootmem) that is marked __init_refok
to limit the chances of mistakenly referring to other __init routines.
WARNING: vmlinux.o(.text+0x2a9c4): Section mismatch: reference to .init.text:.__alloc_bootmem (between '.update_dn_pci_info' and '.pci_dn_reconfig_notifier')
WARNING: vmlinux.o(.text+0x36430): Section mismatch: reference to .init.text:.__alloc_bootmem (between '.mpic_msi_init_allocator' and '.find_ht_magic_addr')
WARNING: vmlinux.o(.text+0x5e804): Section mismatch: reference to .init.text:.__alloc_bootmem (between '.celleb_setup_phb' and '.celleb_fake_pci_write_config')
WARNING: vmlinux.o(.text+0x5e8e8): Section mismatch: reference to .init.text:.__alloc_bootmem (between '.celleb_setup_phb' and '.celleb_fake_pci_write_config')
WARNING: vmlinux.o(.text+0x5e968): Section mismatch: reference to .init.text:.__alloc_bootmem (between '.celleb_setup_phb' and '.celleb_fake_pci_write_config')
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
The patch adds fragments caused by rh_alloc_align() back to free list, instead
of allocating the whole chunk of memory. This will greatly improve memory
utilization managed by rheap.
It solves MURAM not enough problem with 3 UCCs enabled on MPC8323.
Signed-off-by: Li Yang <leoli@freescale.com>
Acked-by: Joakim Tjernlund <joakim.tjernlund@transmode.se>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
When an rheap is created, the caller can specify the alignment to use. In
rh_alloc_align(), if a free block is found that is the exact size needed
(including extra space for alignment), that configured alignment value is not
used to align the pointer. Instead, the default alignment is used. If the
default alignment is smaller than the configured alignment, then the returned
value will not be aligned correctly.
Signed-off-by: Timur Tabi <timur@freescale.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
The rheap allocation functions return a pointer, but the actual value is based
on how the heap was initialized, and so it can be anything, e.g. an offset
into a buffer. A ulong is a better representation of the value returned by
the allocation functions.
This patch changes all of the relevant rheap functions to use a unsigned long
integers instead of a pointer. In case of an error, the value returned is
a negative error code that has been cast to an unsigned long. The caller can
use the IS_ERR_VALUE() macro to check for this.
All code which calls the rheap functions is updated accordingly. Macros
IS_MURAM_ERR() and IS_DPERR(), have been deleted in favor of IS_ERR_VALUE().
Also added error checking to rh_attach_region().
Signed-off-by: Timur Tabi <timur@freescale.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
For 32-bit systems, powerpc still relies on the 4level-fixup.h hack,
to pretend that the generic pagetable handling stuff is 3-levels
rather than 4. This patch removes this, instead using the newer
pgtable-nopmd.h to handle the elision of both the pud and pmd
pagetable levels (ppc32 pagetables are actually 2 levels).
This removes a little extraneous code, and makes it more easily
compared to the 64-bit pagetable code.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Emulate a few more instructions in software - especially useful during
singlestepping (xmon/kprobes).
Instructions emulated with this patch are mfcr/mtcr rX, mfxer/mtxer rX,
mflr/mtlr rX, mfctr/mtctr rX and mr rA,rB.
Signed-off-by: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
mtocrf is a faster single-field mtcrf (move to condition register
fields) instruction available in POWER4 and later processors. It can
make quite a difference in performance on some implementations, so use
it for CONFIG_POWER4_ONLY builds.
Signed-off-by: Olof Johansson <olof@lixom.net>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Use ARRAY_SIZE macro already defined in linux/kernel.h
Signed-off-by: Ahmed S. Darwish <darwish.07@gmail.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Added kprobes to ppc32 platforms that have use single_step_exception. This
excludes 4xx and anything Book-E since their debug mechanisms for single stepping
are completely different.
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
The grow() function in the rheap library allocates a larger array of blocks,
copies the contents of the old blocks array to the newly allocated array and
fixes the list_head pointers after the copy. At the end, the new blocks must
be enqueued to the empty_list of the rh_info_t structure. This patch fixes
a bug where the code was indexing past the end of the array when enqueueing
blocks. The UCC ethernet driver, which uses the rheap allocator, experiences
kernel panics because of this bug.
Signed-off-by: Ionut Nicu <ionut.nicu@freescale.com>
Signed-off-by: Timur Tabi <timur@freescale.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Including support for non-coherent cache, some mm-related things +
relevant field in Kconfig and Makefiles. Also included rheap.o compilation
if 8xx is defined.
Non-coherent mapping were refined and renamed according to Cristoph
Hellwig. Orphaned functions were cleaned up.
[Also removed arch/ppc/kernel/dma-mapping.c, because otherwise
compiling with ARCH=ppc for a non DMA-cache-coherent platform ends up
with two copies of __dma_alloc_coherent etc.
-- paulus.]
Signed-off-by: Vitaly Bordug <vbordug@ru.mvista.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
On powerpc, probing on emulate_step function will crash 2.6.18.1 when
it is triggered.
When kprobe is triggered, emulate_step() is on its kernel path and
will cause recursive kprobe fault. And branch_taken() is called
in emulate_step(). This disallows kprobes on both of them.
Signed-off-by: Paul Mackerras <paulus@samba.org>
Honor alignment parameter in the rheap allocator. This is needed by
qe_lib.
Remove compile warning.
Signed-off-by: Pantelis Antoniou <pantelis@embeddedalley.com>
Signed-off-by: Li Yang <leoli@freescale.com>
Acked-by: Kumar Galak <galak@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
This moves the cpm2 common code and PIC stuff to the powerpc. Most of the files
were just copied from ppc/, with minor tuning to make it compile, and, subsequently, work.
Signed-off-by: Vitaly Bordug <vbordug@ru.mvista.com>
Mark Nelson
Sylvain Munaut