mirror of
https://github.com/AuxXxilium/linux_dsm_epyc7002.git
synced 2024-11-26 23:20:58 +07:00
2b4a08150e
The generic TLB flush functions kept upto 506 pages per CPU to avoid too frequent IPIs. This value was done for the L1 cache of older x86 CPUs, but with modern CPUs it does not make much sense anymore. TLB flushing is slow enough that using the L2 cache is fine. This patch increases the flush array on x86-64 to cache 5350 pages. That is roughly 20MB with 4K pages. It speeds up large munmaps in multithreaded processes on SMP considerably. The cost is roughly 42k of memory per CPU, which is reasonable. I only increased it on x86-64 for now, but it would probably make sense to increase it everywhere. Embedded architectures with SMP may keep it smaller to save some memory per CPU. Signed-off-by: Andi Kleen <ak@suse.de> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
129 lines
3.4 KiB
C
129 lines
3.4 KiB
C
#ifndef _X8664_TLBFLUSH_H
|
|
#define _X8664_TLBFLUSH_H
|
|
|
|
#include <linux/config.h>
|
|
#include <linux/mm.h>
|
|
#include <asm/processor.h>
|
|
|
|
#define __flush_tlb() \
|
|
do { \
|
|
unsigned long tmpreg; \
|
|
\
|
|
__asm__ __volatile__( \
|
|
"movq %%cr3, %0; # flush TLB \n" \
|
|
"movq %0, %%cr3; \n" \
|
|
: "=r" (tmpreg) \
|
|
:: "memory"); \
|
|
} while (0)
|
|
|
|
/*
|
|
* Global pages have to be flushed a bit differently. Not a real
|
|
* performance problem because this does not happen often.
|
|
*/
|
|
#define __flush_tlb_global() \
|
|
do { \
|
|
unsigned long tmpreg, cr4, cr4_orig; \
|
|
\
|
|
__asm__ __volatile__( \
|
|
"movq %%cr4, %2; # turn off PGE \n" \
|
|
"movq %2, %1; \n" \
|
|
"andq %3, %1; \n" \
|
|
"movq %1, %%cr4; \n" \
|
|
"movq %%cr3, %0; # flush TLB \n" \
|
|
"movq %0, %%cr3; \n" \
|
|
"movq %2, %%cr4; # turn PGE back on \n" \
|
|
: "=&r" (tmpreg), "=&r" (cr4), "=&r" (cr4_orig) \
|
|
: "i" (~X86_CR4_PGE) \
|
|
: "memory"); \
|
|
} while (0)
|
|
|
|
extern unsigned long pgkern_mask;
|
|
|
|
#define __flush_tlb_all() __flush_tlb_global()
|
|
|
|
#define __flush_tlb_one(addr) \
|
|
__asm__ __volatile__("invlpg %0": :"m" (*(char *) addr))
|
|
|
|
|
|
/*
|
|
* 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
|
|
* - flush_tlb_pgtables(mm, start, end) flushes a range of page tables
|
|
*
|
|
* x86-64 can only flush individual pages or full VMs. For a range flush
|
|
* we always do the full VM. Might be worth trying if for a small
|
|
* range a few INVLPGs in a row are a win.
|
|
*/
|
|
|
|
#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();
|
|
}
|
|
|
|
#else
|
|
|
|
#include <asm/smp.h>
|
|
|
|
#define local_flush_tlb() \
|
|
__flush_tlb()
|
|
|
|
extern void flush_tlb_all(void);
|
|
extern void flush_tlb_current_task(void);
|
|
extern void flush_tlb_mm(struct mm_struct *);
|
|
extern void flush_tlb_page(struct vm_area_struct *, unsigned long);
|
|
|
|
#define flush_tlb() flush_tlb_current_task()
|
|
|
|
static inline void flush_tlb_range(struct vm_area_struct * vma, unsigned long start, unsigned long end)
|
|
{
|
|
flush_tlb_mm(vma->vm_mm);
|
|
}
|
|
|
|
#define TLBSTATE_OK 1
|
|
#define TLBSTATE_LAZY 2
|
|
|
|
/* Roughly an IPI every 20MB with 4k pages for freeing page table
|
|
ranges. Cost is about 42k of memory for each CPU. */
|
|
#define ARCH_FREE_PTE_NR 5350
|
|
|
|
#endif
|
|
|
|
#define flush_tlb_kernel_range(start, end) flush_tlb_all()
|
|
|
|
static inline void flush_tlb_pgtables(struct mm_struct *mm,
|
|
unsigned long start, unsigned long end)
|
|
{
|
|
/* x86_64 does not keep any page table caches in a software TLB.
|
|
The CPUs do in their hardware TLBs, but they are handled
|
|
by the normal TLB flushing algorithms. */
|
|
}
|
|
|
|
#endif /* _X8664_TLBFLUSH_H */
|