linux/arch/x86/include/asm/tlbflush.h

176 lines
4.2 KiB
C
Raw Normal View History

#ifndef _ASM_X86_TLBFLUSH_H
#define _ASM_X86_TLBFLUSH_H
#include <linux/mm.h>
#include <linux/sched.h>
#include <asm/processor.h>
#include <asm/special_insns.h>
#ifdef CONFIG_PARAVIRT
#include <asm/paravirt.h>
#else
#define __flush_tlb() __native_flush_tlb()
#define __flush_tlb_global() __native_flush_tlb_global()
#define __flush_tlb_single(addr) __native_flush_tlb_single(addr)
#endif
static inline void __native_flush_tlb(void)
{
native_write_cr3(native_read_cr3());
}
static inline void __native_flush_tlb_global(void)
{
unsigned long flags;
unsigned long cr4;
/*
* Read-modify-write to CR4 - protect it from preemption and
* from interrupts. (Use the raw variant because this code can
* be called from deep inside debugging code.)
*/
raw_local_irq_save(flags);
cr4 = native_read_cr4();
/* clear PGE */
native_write_cr4(cr4 & ~X86_CR4_PGE);
/* write old PGE again and flush TLBs */
native_write_cr4(cr4);
raw_local_irq_restore(flags);
}
static inline void __native_flush_tlb_single(unsigned long addr)
{
asm volatile("invlpg (%0)" ::"r" (addr) : "memory");
}
static inline void __flush_tlb_all(void)
{
if (cpu_has_pge)
__flush_tlb_global();
else
__flush_tlb();
}
static inline void __flush_tlb_one(unsigned long addr)
{
__flush_tlb_single(addr);
}
#define TLB_FLUSH_ALL -1UL
/*
* TLB flushing:
*
* - flush_tlb() flushes the current mm struct TLBs
* - flush_tlb_all() flushes all processes TLBs
* - flush_tlb_mm(mm) flushes the specified mm context TLB's
* - flush_tlb_page(vma, vmaddr) flushes one page
* - flush_tlb_range(vma, start, end) flushes a range of pages
* - flush_tlb_kernel_range(start, end) flushes a range of kernel pages
x86/flush_tlb: try flush_tlb_single one by one in flush_tlb_range x86 has no flush_tlb_range support in instruction level. Currently the flush_tlb_range just implemented by flushing all page table. That is not the best solution for all scenarios. In fact, if we just use 'invlpg' to flush few lines from TLB, we can get the performance gain from later remain TLB lines accessing. But the 'invlpg' instruction costs much of time. Its execution time can compete with cr3 rewriting, and even a bit more on SNB CPU. So, on a 512 4KB TLB entries CPU, the balance points is at: (512 - X) * 100ns(assumed TLB refill cost) = X(TLB flush entries) * 100ns(assumed invlpg cost) Here, X is 256, that is 1/2 of 512 entries. But with the mysterious CPU pre-fetcher and page miss handler Unit, the assumed TLB refill cost is far lower then 100ns in sequential access. And 2 HT siblings in one core makes the memory access more faster if they are accessing the same memory. So, in the patch, I just do the change when the target entries is less than 1/16 of whole active tlb entries. Actually, I have no data support for the percentage '1/16', so any suggestions are welcomed. As to hugetlb, guess due to smaller page table, and smaller active TLB entries, I didn't see benefit via my benchmark, so no optimizing now. My micro benchmark show in ideal scenarios, the performance improves 70 percent in reading. And in worst scenario, the reading/writing performance is similar with unpatched 3.4-rc4 kernel. Here is the reading data on my 2P * 4cores *HT NHM EP machine, with THP 'always': multi thread testing, '-t' paramter is thread number: with patch unpatched 3.4-rc4 ./mprotect -t 1 14ns 24ns ./mprotect -t 2 13ns 22ns ./mprotect -t 4 12ns 19ns ./mprotect -t 8 14ns 16ns ./mprotect -t 16 28ns 26ns ./mprotect -t 32 54ns 51ns ./mprotect -t 128 200ns 199ns Single process with sequencial flushing and memory accessing: with patch unpatched 3.4-rc4 ./mprotect 7ns 11ns ./mprotect -p 4096 -l 8 -n 10240 21ns 21ns [ hpa: http://lkml.kernel.org/r/1B4B44D9196EFF41AE41FDA404FC0A100BFF94@SHSMSX101.ccr.corp.intel.com has additional performance numbers. ] Signed-off-by: Alex Shi <alex.shi@intel.com> Link: http://lkml.kernel.org/r/1340845344-27557-3-git-send-email-alex.shi@intel.com Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2012-06-28 01:02:17 +00:00
* - flush_tlb_others(cpumask, mm, start, end) flushes TLBs on other cpus
*
* ..but the i386 has somewhat limited tlb flushing capabilities,
* and page-granular flushes are available only on i486 and up.
*/
#ifndef CONFIG_SMP
#define flush_tlb() __flush_tlb()
#define flush_tlb_all() __flush_tlb_all()
#define local_flush_tlb() __flush_tlb()
static inline void flush_tlb_mm(struct mm_struct *mm)
{
if (mm == current->active_mm)
__flush_tlb();
}
static inline void flush_tlb_page(struct vm_area_struct *vma,
unsigned long addr)
{
if (vma->vm_mm == current->active_mm)
__flush_tlb_one(addr);
}
static inline void flush_tlb_range(struct vm_area_struct *vma,
unsigned long start, unsigned long end)
{
if (vma->vm_mm == current->active_mm)
__flush_tlb();
}
static inline void flush_tlb_mm_range(struct mm_struct *mm,
x86/tlb: enable tlb flush range support for x86 Not every tlb_flush execution moment is really need to evacuate all TLB entries, like in munmap, just few 'invlpg' is better for whole process performance, since it leaves most of TLB entries for later accessing. This patch also rewrite flush_tlb_range for 2 purposes: 1, split it out to get flush_blt_mm_range function. 2, clean up to reduce line breaking, thanks for Borislav's input. My micro benchmark 'mummap' http://lkml.org/lkml/2012/5/17/59 show that the random memory access on other CPU has 0~50% speed up on a 2P * 4cores * HT NHM EP while do 'munmap'. Thanks Yongjie's testing on this patch: ------------- I used Linux 3.4-RC6 w/ and w/o his patches as Xen dom0 and guest kernel. After running two benchmarks in Xen HVM guest, I found his patches brought about 1%~3% performance gain in 'kernel build' and 'netperf' testing, though the performance gain was not very stable in 'kernel build' testing. Some detailed testing results are below. Testing Environment: Hardware: Romley-EP platform Xen version: latest upstream Linux kernel: 3.4-RC6 Guest vCPU number: 8 NIC: Intel 82599 (10GB bandwidth) In 'kernel build' testing in guest: Command line | performance gain make -j 4 | 3.81% make -j 8 | 0.37% make -j 16 | -0.52% In 'netperf' testing, we tested TCP_STREAM with default socket size 16384 byte as large packet and 64 byte as small packet. I used several clients to add networking pressure, then 'netperf' server automatically generated several threads to response them. I also used large-size packet and small-size packet in the testing. Packet size | Thread number | performance gain 16384 bytes | 4 | 0.02% 16384 bytes | 8 | 2.21% 16384 bytes | 16 | 2.04% 64 bytes | 4 | 1.07% 64 bytes | 8 | 3.31% 64 bytes | 16 | 0.71% Signed-off-by: Alex Shi <alex.shi@intel.com> Link: http://lkml.kernel.org/r/1340845344-27557-8-git-send-email-alex.shi@intel.com Tested-by: Ren, Yongjie <yongjie.ren@intel.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2012-06-28 01:02:22 +00:00
unsigned long start, unsigned long end, unsigned long vmflag)
{
if (mm == current->active_mm)
x86/tlb: enable tlb flush range support for x86 Not every tlb_flush execution moment is really need to evacuate all TLB entries, like in munmap, just few 'invlpg' is better for whole process performance, since it leaves most of TLB entries for later accessing. This patch also rewrite flush_tlb_range for 2 purposes: 1, split it out to get flush_blt_mm_range function. 2, clean up to reduce line breaking, thanks for Borislav's input. My micro benchmark 'mummap' http://lkml.org/lkml/2012/5/17/59 show that the random memory access on other CPU has 0~50% speed up on a 2P * 4cores * HT NHM EP while do 'munmap'. Thanks Yongjie's testing on this patch: ------------- I used Linux 3.4-RC6 w/ and w/o his patches as Xen dom0 and guest kernel. After running two benchmarks in Xen HVM guest, I found his patches brought about 1%~3% performance gain in 'kernel build' and 'netperf' testing, though the performance gain was not very stable in 'kernel build' testing. Some detailed testing results are below. Testing Environment: Hardware: Romley-EP platform Xen version: latest upstream Linux kernel: 3.4-RC6 Guest vCPU number: 8 NIC: Intel 82599 (10GB bandwidth) In 'kernel build' testing in guest: Command line | performance gain make -j 4 | 3.81% make -j 8 | 0.37% make -j 16 | -0.52% In 'netperf' testing, we tested TCP_STREAM with default socket size 16384 byte as large packet and 64 byte as small packet. I used several clients to add networking pressure, then 'netperf' server automatically generated several threads to response them. I also used large-size packet and small-size packet in the testing. Packet size | Thread number | performance gain 16384 bytes | 4 | 0.02% 16384 bytes | 8 | 2.21% 16384 bytes | 16 | 2.04% 64 bytes | 4 | 1.07% 64 bytes | 8 | 3.31% 64 bytes | 16 | 0.71% Signed-off-by: Alex Shi <alex.shi@intel.com> Link: http://lkml.kernel.org/r/1340845344-27557-8-git-send-email-alex.shi@intel.com Tested-by: Ren, Yongjie <yongjie.ren@intel.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2012-06-28 01:02:22 +00:00
__flush_tlb();
}
static inline void native_flush_tlb_others(const struct cpumask *cpumask,
struct mm_struct *mm,
x86/flush_tlb: try flush_tlb_single one by one in flush_tlb_range x86 has no flush_tlb_range support in instruction level. Currently the flush_tlb_range just implemented by flushing all page table. That is not the best solution for all scenarios. In fact, if we just use 'invlpg' to flush few lines from TLB, we can get the performance gain from later remain TLB lines accessing. But the 'invlpg' instruction costs much of time. Its execution time can compete with cr3 rewriting, and even a bit more on SNB CPU. So, on a 512 4KB TLB entries CPU, the balance points is at: (512 - X) * 100ns(assumed TLB refill cost) = X(TLB flush entries) * 100ns(assumed invlpg cost) Here, X is 256, that is 1/2 of 512 entries. But with the mysterious CPU pre-fetcher and page miss handler Unit, the assumed TLB refill cost is far lower then 100ns in sequential access. And 2 HT siblings in one core makes the memory access more faster if they are accessing the same memory. So, in the patch, I just do the change when the target entries is less than 1/16 of whole active tlb entries. Actually, I have no data support for the percentage '1/16', so any suggestions are welcomed. As to hugetlb, guess due to smaller page table, and smaller active TLB entries, I didn't see benefit via my benchmark, so no optimizing now. My micro benchmark show in ideal scenarios, the performance improves 70 percent in reading. And in worst scenario, the reading/writing performance is similar with unpatched 3.4-rc4 kernel. Here is the reading data on my 2P * 4cores *HT NHM EP machine, with THP 'always': multi thread testing, '-t' paramter is thread number: with patch unpatched 3.4-rc4 ./mprotect -t 1 14ns 24ns ./mprotect -t 2 13ns 22ns ./mprotect -t 4 12ns 19ns ./mprotect -t 8 14ns 16ns ./mprotect -t 16 28ns 26ns ./mprotect -t 32 54ns 51ns ./mprotect -t 128 200ns 199ns Single process with sequencial flushing and memory accessing: with patch unpatched 3.4-rc4 ./mprotect 7ns 11ns ./mprotect -p 4096 -l 8 -n 10240 21ns 21ns [ hpa: http://lkml.kernel.org/r/1B4B44D9196EFF41AE41FDA404FC0A100BFF94@SHSMSX101.ccr.corp.intel.com has additional performance numbers. ] Signed-off-by: Alex Shi <alex.shi@intel.com> Link: http://lkml.kernel.org/r/1340845344-27557-3-git-send-email-alex.shi@intel.com Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2012-06-28 01:02:17 +00:00
unsigned long start,
unsigned long end)
{
}
static inline void reset_lazy_tlbstate(void)
{
}
static inline void flush_tlb_kernel_range(unsigned long start,
unsigned long end)
{
flush_tlb_all();
}
#else /* SMP */
#include <asm/smp.h>
#define local_flush_tlb() __flush_tlb()
x86/tlb: enable tlb flush range support for x86 Not every tlb_flush execution moment is really need to evacuate all TLB entries, like in munmap, just few 'invlpg' is better for whole process performance, since it leaves most of TLB entries for later accessing. This patch also rewrite flush_tlb_range for 2 purposes: 1, split it out to get flush_blt_mm_range function. 2, clean up to reduce line breaking, thanks for Borislav's input. My micro benchmark 'mummap' http://lkml.org/lkml/2012/5/17/59 show that the random memory access on other CPU has 0~50% speed up on a 2P * 4cores * HT NHM EP while do 'munmap'. Thanks Yongjie's testing on this patch: ------------- I used Linux 3.4-RC6 w/ and w/o his patches as Xen dom0 and guest kernel. After running two benchmarks in Xen HVM guest, I found his patches brought about 1%~3% performance gain in 'kernel build' and 'netperf' testing, though the performance gain was not very stable in 'kernel build' testing. Some detailed testing results are below. Testing Environment: Hardware: Romley-EP platform Xen version: latest upstream Linux kernel: 3.4-RC6 Guest vCPU number: 8 NIC: Intel 82599 (10GB bandwidth) In 'kernel build' testing in guest: Command line | performance gain make -j 4 | 3.81% make -j 8 | 0.37% make -j 16 | -0.52% In 'netperf' testing, we tested TCP_STREAM with default socket size 16384 byte as large packet and 64 byte as small packet. I used several clients to add networking pressure, then 'netperf' server automatically generated several threads to response them. I also used large-size packet and small-size packet in the testing. Packet size | Thread number | performance gain 16384 bytes | 4 | 0.02% 16384 bytes | 8 | 2.21% 16384 bytes | 16 | 2.04% 64 bytes | 4 | 1.07% 64 bytes | 8 | 3.31% 64 bytes | 16 | 0.71% Signed-off-by: Alex Shi <alex.shi@intel.com> Link: http://lkml.kernel.org/r/1340845344-27557-8-git-send-email-alex.shi@intel.com Tested-by: Ren, Yongjie <yongjie.ren@intel.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2012-06-28 01:02:22 +00:00
#define flush_tlb_mm(mm) flush_tlb_mm_range(mm, 0UL, TLB_FLUSH_ALL, 0UL)
#define flush_tlb_range(vma, start, end) \
flush_tlb_mm_range(vma->vm_mm, start, end, vma->vm_flags)
extern void flush_tlb_all(void);
extern void flush_tlb_current_task(void);
extern void flush_tlb_page(struct vm_area_struct *, unsigned long);
x86/tlb: enable tlb flush range support for x86 Not every tlb_flush execution moment is really need to evacuate all TLB entries, like in munmap, just few 'invlpg' is better for whole process performance, since it leaves most of TLB entries for later accessing. This patch also rewrite flush_tlb_range for 2 purposes: 1, split it out to get flush_blt_mm_range function. 2, clean up to reduce line breaking, thanks for Borislav's input. My micro benchmark 'mummap' http://lkml.org/lkml/2012/5/17/59 show that the random memory access on other CPU has 0~50% speed up on a 2P * 4cores * HT NHM EP while do 'munmap'. Thanks Yongjie's testing on this patch: ------------- I used Linux 3.4-RC6 w/ and w/o his patches as Xen dom0 and guest kernel. After running two benchmarks in Xen HVM guest, I found his patches brought about 1%~3% performance gain in 'kernel build' and 'netperf' testing, though the performance gain was not very stable in 'kernel build' testing. Some detailed testing results are below. Testing Environment: Hardware: Romley-EP platform Xen version: latest upstream Linux kernel: 3.4-RC6 Guest vCPU number: 8 NIC: Intel 82599 (10GB bandwidth) In 'kernel build' testing in guest: Command line | performance gain make -j 4 | 3.81% make -j 8 | 0.37% make -j 16 | -0.52% In 'netperf' testing, we tested TCP_STREAM with default socket size 16384 byte as large packet and 64 byte as small packet. I used several clients to add networking pressure, then 'netperf' server automatically generated several threads to response them. I also used large-size packet and small-size packet in the testing. Packet size | Thread number | performance gain 16384 bytes | 4 | 0.02% 16384 bytes | 8 | 2.21% 16384 bytes | 16 | 2.04% 64 bytes | 4 | 1.07% 64 bytes | 8 | 3.31% 64 bytes | 16 | 0.71% Signed-off-by: Alex Shi <alex.shi@intel.com> Link: http://lkml.kernel.org/r/1340845344-27557-8-git-send-email-alex.shi@intel.com Tested-by: Ren, Yongjie <yongjie.ren@intel.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2012-06-28 01:02:22 +00:00
extern void flush_tlb_mm_range(struct mm_struct *mm, unsigned long start,
unsigned long end, unsigned long vmflag);
extern void flush_tlb_kernel_range(unsigned long start, unsigned long end);
#define flush_tlb() flush_tlb_current_task()
void native_flush_tlb_others(const struct cpumask *cpumask,
x86/flush_tlb: try flush_tlb_single one by one in flush_tlb_range x86 has no flush_tlb_range support in instruction level. Currently the flush_tlb_range just implemented by flushing all page table. That is not the best solution for all scenarios. In fact, if we just use 'invlpg' to flush few lines from TLB, we can get the performance gain from later remain TLB lines accessing. But the 'invlpg' instruction costs much of time. Its execution time can compete with cr3 rewriting, and even a bit more on SNB CPU. So, on a 512 4KB TLB entries CPU, the balance points is at: (512 - X) * 100ns(assumed TLB refill cost) = X(TLB flush entries) * 100ns(assumed invlpg cost) Here, X is 256, that is 1/2 of 512 entries. But with the mysterious CPU pre-fetcher and page miss handler Unit, the assumed TLB refill cost is far lower then 100ns in sequential access. And 2 HT siblings in one core makes the memory access more faster if they are accessing the same memory. So, in the patch, I just do the change when the target entries is less than 1/16 of whole active tlb entries. Actually, I have no data support for the percentage '1/16', so any suggestions are welcomed. As to hugetlb, guess due to smaller page table, and smaller active TLB entries, I didn't see benefit via my benchmark, so no optimizing now. My micro benchmark show in ideal scenarios, the performance improves 70 percent in reading. And in worst scenario, the reading/writing performance is similar with unpatched 3.4-rc4 kernel. Here is the reading data on my 2P * 4cores *HT NHM EP machine, with THP 'always': multi thread testing, '-t' paramter is thread number: with patch unpatched 3.4-rc4 ./mprotect -t 1 14ns 24ns ./mprotect -t 2 13ns 22ns ./mprotect -t 4 12ns 19ns ./mprotect -t 8 14ns 16ns ./mprotect -t 16 28ns 26ns ./mprotect -t 32 54ns 51ns ./mprotect -t 128 200ns 199ns Single process with sequencial flushing and memory accessing: with patch unpatched 3.4-rc4 ./mprotect 7ns 11ns ./mprotect -p 4096 -l 8 -n 10240 21ns 21ns [ hpa: http://lkml.kernel.org/r/1B4B44D9196EFF41AE41FDA404FC0A100BFF94@SHSMSX101.ccr.corp.intel.com has additional performance numbers. ] Signed-off-by: Alex Shi <alex.shi@intel.com> Link: http://lkml.kernel.org/r/1340845344-27557-3-git-send-email-alex.shi@intel.com Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2012-06-28 01:02:17 +00:00
struct mm_struct *mm,
unsigned long start, unsigned long end);
#define TLBSTATE_OK 1
#define TLBSTATE_LAZY 2
struct tlb_state {
struct mm_struct *active_mm;
int state;
};
DECLARE_PER_CPU_SHARED_ALIGNED(struct tlb_state, cpu_tlbstate);
static inline void reset_lazy_tlbstate(void)
{
this_cpu_write(cpu_tlbstate.state, 0);
this_cpu_write(cpu_tlbstate.active_mm, &init_mm);
}
#endif /* SMP */
#ifndef CONFIG_PARAVIRT
x86/flush_tlb: try flush_tlb_single one by one in flush_tlb_range x86 has no flush_tlb_range support in instruction level. Currently the flush_tlb_range just implemented by flushing all page table. That is not the best solution for all scenarios. In fact, if we just use 'invlpg' to flush few lines from TLB, we can get the performance gain from later remain TLB lines accessing. But the 'invlpg' instruction costs much of time. Its execution time can compete with cr3 rewriting, and even a bit more on SNB CPU. So, on a 512 4KB TLB entries CPU, the balance points is at: (512 - X) * 100ns(assumed TLB refill cost) = X(TLB flush entries) * 100ns(assumed invlpg cost) Here, X is 256, that is 1/2 of 512 entries. But with the mysterious CPU pre-fetcher and page miss handler Unit, the assumed TLB refill cost is far lower then 100ns in sequential access. And 2 HT siblings in one core makes the memory access more faster if they are accessing the same memory. So, in the patch, I just do the change when the target entries is less than 1/16 of whole active tlb entries. Actually, I have no data support for the percentage '1/16', so any suggestions are welcomed. As to hugetlb, guess due to smaller page table, and smaller active TLB entries, I didn't see benefit via my benchmark, so no optimizing now. My micro benchmark show in ideal scenarios, the performance improves 70 percent in reading. And in worst scenario, the reading/writing performance is similar with unpatched 3.4-rc4 kernel. Here is the reading data on my 2P * 4cores *HT NHM EP machine, with THP 'always': multi thread testing, '-t' paramter is thread number: with patch unpatched 3.4-rc4 ./mprotect -t 1 14ns 24ns ./mprotect -t 2 13ns 22ns ./mprotect -t 4 12ns 19ns ./mprotect -t 8 14ns 16ns ./mprotect -t 16 28ns 26ns ./mprotect -t 32 54ns 51ns ./mprotect -t 128 200ns 199ns Single process with sequencial flushing and memory accessing: with patch unpatched 3.4-rc4 ./mprotect 7ns 11ns ./mprotect -p 4096 -l 8 -n 10240 21ns 21ns [ hpa: http://lkml.kernel.org/r/1B4B44D9196EFF41AE41FDA404FC0A100BFF94@SHSMSX101.ccr.corp.intel.com has additional performance numbers. ] Signed-off-by: Alex Shi <alex.shi@intel.com> Link: http://lkml.kernel.org/r/1340845344-27557-3-git-send-email-alex.shi@intel.com Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2012-06-28 01:02:17 +00:00
#define flush_tlb_others(mask, mm, start, end) \
native_flush_tlb_others(mask, mm, start, end)
#endif
#endif /* _ASM_X86_TLBFLUSH_H */