linux_dsm_epyc7002/arch/x86
Ma Ling 59daa706fb x86, mem: Optimize memcpy by avoiding memory false dependece
All read operations after allocation stage can run speculatively,
all write operation will run in program order, and if addresses are
different read may run before older write operation, otherwise wait
until write commit. However CPU don't check each address bit,
so read could fail to recognize different address even they
are in different page.For example if rsi is 0xf004, rdi is 0xe008,
in following operation there will generate big performance latency.
1. movq (%rsi),	%rax
2. movq %rax,	(%rdi)
3. movq 8(%rsi), %rax
4. movq %rax,	8(%rdi)

If %rsi and rdi were in really the same meory page, there are TRUE
read-after-write dependence because instruction 2 write 0x008 and
instruction 3 read 0x00c, the two address are overlap partially.
Actually there are in different page and no any issues,
but without checking each address bit CPU could think they are
in the same page, and instruction 3 have to wait for instruction 2
to write data into cache from write buffer, then load data from cache,
the cost time read spent is equal to mfence instruction. We may avoid it by
tuning operation sequence as follow.

1. movq 8(%rsi), %rax
2. movq %rax,	8(%rdi)
3. movq (%rsi),	%rax
4. movq %rax,	(%rdi)

Instruction 3 read 0x004, instruction 2 write address 0x010, no any
dependence.  At last on Core2 we gain 1.83x speedup compared with
original instruction sequence.  In this patch we first handle small
size(less 20bytes), then jump to different copy mode. Based on our
micro-benchmark small bytes from 1 to 127 bytes, we got up to 2X
improvement, and up to 1.5X improvement for 1024 bytes on Corei7.  (We
use our micro-benchmark, and will do further test according to your
requirment)

Signed-off-by: Ma Ling <ling.ma@intel.com>
LKML-Reference: <1277753065-18610-1-git-send-email-ling.ma@intel.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2010-08-23 14:56:41 -07:00
..
boot x86, setup: move isdigit.h to ctype.h, header files on top. 2010-08-02 21:07:20 -07:00
configs defconfig reduction 2010-08-14 22:26:53 +02:00
crypto
ia32 Mark arguments to certain syscalls as being const 2010-08-13 16:53:13 -07:00
include/asm Merge branch 'x86-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip 2010-08-20 14:25:08 -07:00
kernel Merge branch 'kvm-updates/2.6.36' of git://git.kernel.org/pub/scm/virt/kvm/kvm 2010-08-22 11:27:36 -07:00
kvm Merge branch 'kvm-updates/2.6.36' of git://git.kernel.org/pub/scm/virt/kvm/kvm 2010-08-22 11:27:36 -07:00
lguest
lib x86, mem: Optimize memcpy by avoiding memory false dependece 2010-08-23 14:56:41 -07:00
math-emu
mm x86: don't send SIGBUS for kernel page faults 2010-08-13 09:49:20 -07:00
oprofile oprofile: add support for Intel processor model 30 2010-08-05 12:17:38 +02:00
pci x86/PCI: use for_each_pci_dev() 2010-07-30 09:47:33 -07:00
power update email address 2010-07-19 10:56:54 +02:00
tools
vdso Merge branches 'x86-cleanups-for-linus', 'x86-vmware-for-linus', 'x86-mtrr-for-linus', 'x86-apic-for-linus', 'x86-fpu-for-linus' and 'x86-vdso-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip 2010-08-06 16:22:59 -07:00
video
xen Merge branch 'stable/xen-swiotlb-0.8.6' of git://git.kernel.org/pub/scm/linux/kernel/git/konrad/xen 2010-08-12 09:09:41 -07:00
.gitignore
Kbuild
Kconfig Replace Configure with Enable in description of MAXSMP 2010-08-21 12:38:58 -07:00
Kconfig.cpu
Kconfig.debug
Makefile
Makefile_32.cpu