mirror of
https://github.com/AuxXxilium/linux_dsm_epyc7002.git
synced 2024-12-28 11:18:45 +07:00
99baac21e4
Nadav reported parallel MADV_DONTNEED on same range has a stale TLB problem and Mel fixed it[1] and found same problem on MADV_FREE[2]. Quote from Mel Gorman: "The race in question is CPU 0 running madv_free and updating some PTEs while CPU 1 is also running madv_free and looking at the same PTEs. CPU 1 may have writable TLB entries for a page but fail the pte_dirty check (because CPU 0 has updated it already) and potentially fail to flush. Hence, when madv_free on CPU 1 returns, there are still potentially writable TLB entries and the underlying PTE is still present so that a subsequent write does not necessarily propagate the dirty bit to the underlying PTE any more. Reclaim at some unknown time at the future may then see that the PTE is still clean and discard the page even though a write has happened in the meantime. I think this is possible but I could have missed some protection in madv_free that prevents it happening." This patch aims for solving both problems all at once and is ready for other problem with KSM, MADV_FREE and soft-dirty story[3]. TLB batch API(tlb_[gather|finish]_mmu] uses [inc|dec]_tlb_flush_pending and mmu_tlb_flush_pending so that when tlb_finish_mmu is called, we can catch there are parallel threads going on. In that case, forcefully, flush TLB to prevent for user to access memory via stale TLB entry although it fail to gather page table entry. I confirmed this patch works with [4] test program Nadav gave so this patch supersedes "mm: Always flush VMA ranges affected by zap_page_range v2" in current mmotm. NOTE: This patch modifies arch-specific TLB gathering interface(x86, ia64, s390, sh, um). It seems most of architecture are straightforward but s390 need to be careful because tlb_flush_mmu works only if mm->context.flush_mm is set to non-zero which happens only a pte entry really is cleared by ptep_get_and_clear and friends. However, this problem never changes the pte entries but need to flush to prevent memory access from stale tlb. [1] http://lkml.kernel.org/r/20170725101230.5v7gvnjmcnkzzql3@techsingularity.net [2] http://lkml.kernel.org/r/20170725100722.2dxnmgypmwnrfawp@suse.de [3] http://lkml.kernel.org/r/BD3A0EBE-ECF4-41D4-87FA-C755EA9AB6BD@gmail.com [4] https://patchwork.kernel.org/patch/9861621/ [minchan@kernel.org: decrease tlb flush pending count in tlb_finish_mmu] Link: http://lkml.kernel.org/r/20170808080821.GA31730@bbox Link: http://lkml.kernel.org/r/20170802000818.4760-7-namit@vmware.com Signed-off-by: Minchan Kim <minchan@kernel.org> Signed-off-by: Nadav Amit <namit@vmware.com> Reported-by: Nadav Amit <namit@vmware.com> Reported-by: Mel Gorman <mgorman@techsingularity.net> Acked-by: Mel Gorman <mgorman@techsingularity.net> Cc: Ingo Molnar <mingo@redhat.com> Cc: Russell King <linux@armlinux.org.uk> Cc: Tony Luck <tony.luck@intel.com> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: "David S. Miller" <davem@davemloft.net> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: Yoshinori Sato <ysato@users.sourceforge.jp> Cc: Jeff Dike <jdike@addtoit.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: Andy Lutomirski <luto@kernel.org> Cc: Hugh Dickins <hughd@google.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Nadav Amit <nadav.amit@gmail.com> Cc: Rik van Riel <riel@redhat.com> Cc: Sergey Senozhatsky <sergey.senozhatsky@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
189 lines
5.3 KiB
C
189 lines
5.3 KiB
C
#ifndef _S390_TLB_H
|
|
#define _S390_TLB_H
|
|
|
|
/*
|
|
* TLB flushing on s390 is complicated. The following requirement
|
|
* from the principles of operation is the most arduous:
|
|
*
|
|
* "A valid table entry must not be changed while it is attached
|
|
* to any CPU and may be used for translation by that CPU except to
|
|
* (1) invalidate the entry by using INVALIDATE PAGE TABLE ENTRY,
|
|
* or INVALIDATE DAT TABLE ENTRY, (2) alter bits 56-63 of a page
|
|
* table entry, or (3) make a change by means of a COMPARE AND SWAP
|
|
* AND PURGE instruction that purges the TLB."
|
|
*
|
|
* The modification of a pte of an active mm struct therefore is
|
|
* a two step process: i) invalidate the pte, ii) store the new pte.
|
|
* This is true for the page protection bit as well.
|
|
* The only possible optimization is to flush at the beginning of
|
|
* a tlb_gather_mmu cycle if the mm_struct is currently not in use.
|
|
*
|
|
* Pages used for the page tables is a different story. FIXME: more
|
|
*/
|
|
|
|
#include <linux/mm.h>
|
|
#include <linux/pagemap.h>
|
|
#include <linux/swap.h>
|
|
#include <asm/processor.h>
|
|
#include <asm/pgalloc.h>
|
|
#include <asm/tlbflush.h>
|
|
|
|
struct mmu_gather {
|
|
struct mm_struct *mm;
|
|
struct mmu_table_batch *batch;
|
|
unsigned int fullmm;
|
|
unsigned long start, end;
|
|
};
|
|
|
|
struct mmu_table_batch {
|
|
struct rcu_head rcu;
|
|
unsigned int nr;
|
|
void *tables[0];
|
|
};
|
|
|
|
#define MAX_TABLE_BATCH \
|
|
((PAGE_SIZE - sizeof(struct mmu_table_batch)) / sizeof(void *))
|
|
|
|
extern void tlb_table_flush(struct mmu_gather *tlb);
|
|
extern void tlb_remove_table(struct mmu_gather *tlb, void *table);
|
|
|
|
static inline void
|
|
arch_tlb_gather_mmu(struct mmu_gather *tlb, struct mm_struct *mm,
|
|
unsigned long start, unsigned long end)
|
|
{
|
|
tlb->mm = mm;
|
|
tlb->start = start;
|
|
tlb->end = end;
|
|
tlb->fullmm = !(start | (end+1));
|
|
tlb->batch = NULL;
|
|
}
|
|
|
|
static inline void tlb_flush_mmu_tlbonly(struct mmu_gather *tlb)
|
|
{
|
|
__tlb_flush_mm_lazy(tlb->mm);
|
|
}
|
|
|
|
static inline void tlb_flush_mmu_free(struct mmu_gather *tlb)
|
|
{
|
|
tlb_table_flush(tlb);
|
|
}
|
|
|
|
|
|
static inline void tlb_flush_mmu(struct mmu_gather *tlb)
|
|
{
|
|
tlb_flush_mmu_tlbonly(tlb);
|
|
tlb_flush_mmu_free(tlb);
|
|
}
|
|
|
|
static inline void
|
|
arch_tlb_finish_mmu(struct mmu_gather *tlb,
|
|
unsigned long start, unsigned long end, bool force)
|
|
{
|
|
if (force) {
|
|
tlb->start = start;
|
|
tlb->end = end;
|
|
}
|
|
|
|
tlb_flush_mmu(tlb);
|
|
}
|
|
|
|
/*
|
|
* Release the page cache reference for a pte removed by
|
|
* tlb_ptep_clear_flush. In both flush modes the tlb for a page cache page
|
|
* has already been freed, so just do free_page_and_swap_cache.
|
|
*/
|
|
static inline bool __tlb_remove_page(struct mmu_gather *tlb, struct page *page)
|
|
{
|
|
free_page_and_swap_cache(page);
|
|
return false; /* avoid calling tlb_flush_mmu */
|
|
}
|
|
|
|
static inline void tlb_remove_page(struct mmu_gather *tlb, struct page *page)
|
|
{
|
|
free_page_and_swap_cache(page);
|
|
}
|
|
|
|
static inline bool __tlb_remove_page_size(struct mmu_gather *tlb,
|
|
struct page *page, int page_size)
|
|
{
|
|
return __tlb_remove_page(tlb, page);
|
|
}
|
|
|
|
static inline void tlb_remove_page_size(struct mmu_gather *tlb,
|
|
struct page *page, int page_size)
|
|
{
|
|
return tlb_remove_page(tlb, page);
|
|
}
|
|
|
|
/*
|
|
* pte_free_tlb frees a pte table and clears the CRSTE for the
|
|
* page table from the tlb.
|
|
*/
|
|
static inline void pte_free_tlb(struct mmu_gather *tlb, pgtable_t pte,
|
|
unsigned long address)
|
|
{
|
|
page_table_free_rcu(tlb, (unsigned long *) pte, address);
|
|
}
|
|
|
|
/*
|
|
* pmd_free_tlb frees a pmd table and clears the CRSTE for the
|
|
* segment table entry from the tlb.
|
|
* If the mm uses a two level page table the single pmd is freed
|
|
* as the pgd. pmd_free_tlb checks the asce_limit against 2GB
|
|
* to avoid the double free of the pmd in this case.
|
|
*/
|
|
static inline void pmd_free_tlb(struct mmu_gather *tlb, pmd_t *pmd,
|
|
unsigned long address)
|
|
{
|
|
if (tlb->mm->context.asce_limit <= (1UL << 31))
|
|
return;
|
|
pgtable_pmd_page_dtor(virt_to_page(pmd));
|
|
tlb_remove_table(tlb, pmd);
|
|
}
|
|
|
|
/*
|
|
* p4d_free_tlb frees a pud table and clears the CRSTE for the
|
|
* region second table entry from the tlb.
|
|
* If the mm uses a four level page table the single p4d is freed
|
|
* as the pgd. p4d_free_tlb checks the asce_limit against 8PB
|
|
* to avoid the double free of the p4d in this case.
|
|
*/
|
|
static inline void p4d_free_tlb(struct mmu_gather *tlb, p4d_t *p4d,
|
|
unsigned long address)
|
|
{
|
|
if (tlb->mm->context.asce_limit <= (1UL << 53))
|
|
return;
|
|
tlb_remove_table(tlb, p4d);
|
|
}
|
|
|
|
/*
|
|
* pud_free_tlb frees a pud table and clears the CRSTE for the
|
|
* region third table entry from the tlb.
|
|
* If the mm uses a three level page table the single pud is freed
|
|
* as the pgd. pud_free_tlb checks the asce_limit against 4TB
|
|
* to avoid the double free of the pud in this case.
|
|
*/
|
|
static inline void pud_free_tlb(struct mmu_gather *tlb, pud_t *pud,
|
|
unsigned long address)
|
|
{
|
|
if (tlb->mm->context.asce_limit <= (1UL << 42))
|
|
return;
|
|
tlb_remove_table(tlb, pud);
|
|
}
|
|
|
|
#define tlb_start_vma(tlb, vma) do { } while (0)
|
|
#define tlb_end_vma(tlb, vma) do { } while (0)
|
|
#define tlb_remove_tlb_entry(tlb, ptep, addr) do { } while (0)
|
|
#define tlb_remove_pmd_tlb_entry(tlb, pmdp, addr) do { } while (0)
|
|
#define tlb_migrate_finish(mm) do { } while (0)
|
|
#define tlb_remove_huge_tlb_entry(h, tlb, ptep, address) \
|
|
tlb_remove_tlb_entry(tlb, ptep, address)
|
|
|
|
#define tlb_remove_check_page_size_change tlb_remove_check_page_size_change
|
|
static inline void tlb_remove_check_page_size_change(struct mmu_gather *tlb,
|
|
unsigned int page_size)
|
|
{
|
|
}
|
|
|
|
#endif /* _S390_TLB_H */
|