From a145abf12c9f7d30d8c330c9d8a97428cbf0589b Mon Sep 17 00:00:00 2001 From: "Aneesh Kumar K.V" Date: Wed, 8 Jun 2016 19:55:51 +0530 Subject: powerpc/mm/radix: Flush page walk cache when freeing page table Even though a tlb_flush() does a flush with invalidate all cache, we can end up doing an RCU page table free before calling tlb_flush(). That means we can have page walk cache entries even after we free the page table pages. This can result in us doing wrong page table walk. Avoid this by doing pwc flush on every page table free. We can't batch the pwc flush, because the rcu call back function where we free the page table pages doesn't have information of the mmu gather. Thus we have to do a pwc on every page table page freed. Note: I also removed the dummy tlb_flush_pgtable call functions for hash 32. Fixes: 1a472c9dba6b ("powerpc/mm/radix: Add tlbflush routines") Signed-off-by: Aneesh Kumar K.V Signed-off-by: Michael Ellerman --- arch/powerpc/include/asm/book3s/64/pgalloc.h | 16 +++++++++++++++- 1 file changed, 15 insertions(+), 1 deletion(-) (limited to 'arch/powerpc/include/asm/book3s/64/pgalloc.h') diff --git a/arch/powerpc/include/asm/book3s/64/pgalloc.h b/arch/powerpc/include/asm/book3s/64/pgalloc.h index 488279edb1f0..26eb2cb80c4e 100644 --- a/arch/powerpc/include/asm/book3s/64/pgalloc.h +++ b/arch/powerpc/include/asm/book3s/64/pgalloc.h @@ -110,6 +110,11 @@ static inline void pud_populate(struct mm_struct *mm, pud_t *pud, pmd_t *pmd) static inline void __pud_free_tlb(struct mmu_gather *tlb, pud_t *pud, unsigned long address) { + /* + * By now all the pud entries should be none entries. So go + * ahead and flush the page walk cache + */ + flush_tlb_pgtable(tlb, address); pgtable_free_tlb(tlb, pud, PUD_INDEX_SIZE); } @@ -127,6 +132,11 @@ static inline void pmd_free(struct mm_struct *mm, pmd_t *pmd) static inline void __pmd_free_tlb(struct mmu_gather *tlb, pmd_t *pmd, unsigned long address) { + /* + * By now all the pud entries should be none entries. So go + * ahead and flush the page walk cache + */ + flush_tlb_pgtable(tlb, address); return pgtable_free_tlb(tlb, pmd, PMD_CACHE_INDEX); } @@ -198,7 +208,11 @@ static inline void pte_free(struct mm_struct *mm, pgtable_t ptepage) static inline void __pte_free_tlb(struct mmu_gather *tlb, pgtable_t table, unsigned long address) { - tlb_flush_pgtable(tlb, address); + /* + * By now all the pud entries should be none entries. So go + * ahead and flush the page walk cache + */ + flush_tlb_pgtable(tlb, address); pgtable_free_tlb(tlb, table, 0); } -- cgit From 32d6bd9059f265f617f6502c68dfbcae7e515add Mon Sep 17 00:00:00 2001 From: Michal Hocko Date: Fri, 24 Jun 2016 14:48:47 -0700 Subject: tree wide: get rid of __GFP_REPEAT for order-0 allocations part I This is the third version of the patchset previously sent [1]. I have basically only rebased it on top of 4.7-rc1 tree and dropped "dm: get rid of superfluous gfp flags" which went through dm tree. I am sending it now because it is tree wide and chances for conflicts are reduced considerably when we want to target rc2. I plan to send the next step and rename the flag and move to a better semantic later during this release cycle so we will have a new semantic ready for 4.8 merge window hopefully. Motivation: While working on something unrelated I've checked the current usage of __GFP_REPEAT in the tree. It seems that a majority of the usage is and always has been bogus because __GFP_REPEAT has always been about costly high order allocations while we are using it for order-0 or very small orders very often. It seems that a big pile of them is just a copy&paste when a code has been adopted from one arch to another. I think it makes some sense to get rid of them because they are just making the semantic more unclear. Please note that GFP_REPEAT is documented as * __GFP_REPEAT: Try hard to allocate the memory, but the allocation attempt * _might_ fail. This depends upon the particular VM implementation. while !costly requests have basically nofail semantic. So one could reasonably expect that order-0 request with __GFP_REPEAT will not loop for ever. This is not implemented right now though. I would like to move on with __GFP_REPEAT and define a better semantic for it. $ git grep __GFP_REPEAT origin/master | wc -l 111 $ git grep __GFP_REPEAT | wc -l 36 So we are down to the third after this patch series. The remaining places really seem to be relying on __GFP_REPEAT due to large allocation requests. This still needs some double checking which I will do later after all the simple ones are sorted out. I am touching a lot of arch specific code here and I hope I got it right but as a matter of fact I even didn't compile test for some archs as I do not have cross compiler for them. Patches should be quite trivial to review for stupid compile mistakes though. The tricky parts are usually hidden by macro definitions and thats where I would appreciate help from arch maintainers. [1] http://lkml.kernel.org/r/1461849846-27209-1-git-send-email-mhocko@kernel.org This patch (of 19): __GFP_REPEAT has a rather weak semantic but since it has been introduced around 2.6.12 it has been ignored for low order allocations. Yet we have the full kernel tree with its usage for apparently order-0 allocations. This is really confusing because __GFP_REPEAT is explicitly documented to allow allocation failures which is a weaker semantic than the current order-0 has (basically nofail). Let's simply drop __GFP_REPEAT from those places. This would allow to identify place which really need allocator to retry harder and formulate a more specific semantic for what the flag is supposed to do actually. Link: http://lkml.kernel.org/r/1464599699-30131-2-git-send-email-mhocko@kernel.org Signed-off-by: Michal Hocko Cc: "David S. Miller" Cc: "H. Peter Anvin" Cc: "James E.J. Bottomley" Cc: "Theodore Ts'o" Cc: Andy Lutomirski Cc: Benjamin Herrenschmidt Cc: Catalin Marinas Cc: Chen Liqin Cc: Chris Metcalf [for tile] Cc: Guan Xuetao Cc: Heiko Carstens Cc: Helge Deller Cc: Ingo Molnar Cc: Jan Kara Cc: John Crispin Cc: Lennox Wu Cc: Ley Foon Tan Cc: Martin Schwidefsky Cc: Matt Fleming Cc: Ralf Baechle Cc: Rich Felker Cc: Russell King Cc: Thomas Gleixner Cc: Vineet Gupta Cc: Will Deacon Cc: Yoshinori Sato Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/powerpc/include/asm/book3s/64/pgalloc.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/powerpc/include/asm/book3s/64/pgalloc.h') diff --git a/arch/powerpc/include/asm/book3s/64/pgalloc.h b/arch/powerpc/include/asm/book3s/64/pgalloc.h index 488279edb1f0..049b80359db6 100644 --- a/arch/powerpc/include/asm/book3s/64/pgalloc.h +++ b/arch/powerpc/include/asm/book3s/64/pgalloc.h @@ -151,7 +151,7 @@ static inline pgtable_t pmd_pgtable(pmd_t pmd) static inline pte_t *pte_alloc_one_kernel(struct mm_struct *mm, unsigned long address) { - return (pte_t *)__get_free_page(GFP_KERNEL | __GFP_REPEAT | __GFP_ZERO); + return (pte_t *)__get_free_page(GFP_KERNEL | __GFP_ZERO); } static inline pgtable_t pte_alloc_one(struct mm_struct *mm, -- cgit From 2379a23e34b58520dfc8f4909f116a08393138e4 Mon Sep 17 00:00:00 2001 From: Michal Hocko Date: Fri, 24 Jun 2016 14:49:12 -0700 Subject: powerpc: get rid of superfluous __GFP_REPEAT __GFP_REPEAT has a rather weak semantic but since it has been introduced around 2.6.12 it has been ignored for low order allocations. {pud,pmd}_alloc_one are allocating from {PGT,PUD}_CACHE initialized in pgtable_cache_init which doesn't have larger than sizeof(void *) << 12 size and that fits into !costly allocation request size. PGALLOC_GFP is used only in radix__pgd_alloc which uses either order-0 or order-4 requests. The first one doesn't need the flag while the second does. Drop __GFP_REPEAT from PGALLOC_GFP and add it for the order-4 one. This means that this flag has never been actually useful here because it has always been used only for PAGE_ALLOC_COSTLY requests. Link: http://lkml.kernel.org/r/1464599699-30131-12-git-send-email-mhocko@kernel.org Signed-off-by: Michal Hocko Cc: Benjamin Herrenschmidt Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/powerpc/include/asm/book3s/64/pgalloc.h | 10 ++++------ 1 file changed, 4 insertions(+), 6 deletions(-) (limited to 'arch/powerpc/include/asm/book3s/64/pgalloc.h') diff --git a/arch/powerpc/include/asm/book3s/64/pgalloc.h b/arch/powerpc/include/asm/book3s/64/pgalloc.h index 049b80359db6..d14fcf82c00c 100644 --- a/arch/powerpc/include/asm/book3s/64/pgalloc.h +++ b/arch/powerpc/include/asm/book3s/64/pgalloc.h @@ -41,7 +41,7 @@ extern struct kmem_cache *pgtable_cache[]; pgtable_cache[(shift) - 1]; \ }) -#define PGALLOC_GFP GFP_KERNEL | __GFP_NOTRACK | __GFP_REPEAT | __GFP_ZERO +#define PGALLOC_GFP GFP_KERNEL | __GFP_NOTRACK | __GFP_ZERO extern pte_t *pte_fragment_alloc(struct mm_struct *, unsigned long, int); extern void pte_fragment_free(unsigned long *, int); @@ -56,7 +56,7 @@ static inline pgd_t *radix__pgd_alloc(struct mm_struct *mm) return (pgd_t *)__get_free_page(PGALLOC_GFP); #else struct page *page; - page = alloc_pages(PGALLOC_GFP, 4); + page = alloc_pages(PGALLOC_GFP | __GFP_REPEAT, 4); if (!page) return NULL; return (pgd_t *) page_address(page); @@ -93,8 +93,7 @@ static inline void pgd_populate(struct mm_struct *mm, pgd_t *pgd, pud_t *pud) static inline pud_t *pud_alloc_one(struct mm_struct *mm, unsigned long addr) { - return kmem_cache_alloc(PGT_CACHE(PUD_INDEX_SIZE), - GFP_KERNEL|__GFP_REPEAT); + return kmem_cache_alloc(PGT_CACHE(PUD_INDEX_SIZE), GFP_KERNEL); } static inline void pud_free(struct mm_struct *mm, pud_t *pud) @@ -115,8 +114,7 @@ static inline void __pud_free_tlb(struct mmu_gather *tlb, pud_t *pud, static inline pmd_t *pmd_alloc_one(struct mm_struct *mm, unsigned long addr) { - return kmem_cache_alloc(PGT_CACHE(PMD_CACHE_INDEX), - GFP_KERNEL|__GFP_REPEAT); + return kmem_cache_alloc(PGT_CACHE(PMD_CACHE_INDEX), GFP_KERNEL); } static inline void pmd_free(struct mm_struct *mm, pmd_t *pmd) -- cgit