From patchwork Mon Apr 22 10:00:43 2013 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: "Aneesh Kumar K.V" X-Patchwork-Id: 238402 Return-Path: X-Original-To: patchwork-incoming@ozlabs.org Delivered-To: patchwork-incoming@ozlabs.org Received: from ozlabs.org (localhost [IPv6:::1]) by ozlabs.org (Postfix) with ESMTP id BBC8F2C13F2 for ; Mon, 22 Apr 2013 20:13:33 +1000 (EST) Received: from e23smtp05.au.ibm.com (e23smtp05.au.ibm.com [202.81.31.147]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (Client CN "e23smtp05.au.ibm.com", Issuer "GeoTrust SSL CA" (not verified)) by ozlabs.org (Postfix) with ESMTPS id 8AE1F2C0352 for ; Mon, 22 Apr 2013 20:02:54 +1000 (EST) Received: from /spool/local by e23smtp05.au.ibm.com with IBM ESMTP SMTP Gateway: Authorized Use Only! Violators will be prosecuted for from ; Mon, 22 Apr 2013 19:57:54 +1000 Received: from d23dlp03.au.ibm.com (202.81.31.214) by e23smtp05.au.ibm.com (202.81.31.211) with IBM ESMTP SMTP Gateway: Authorized Use Only! Violators will be prosecuted; Mon, 22 Apr 2013 19:57:52 +1000 Received: from d23relay03.au.ibm.com (d23relay03.au.ibm.com [9.190.235.21]) by d23dlp03.au.ibm.com (Postfix) with ESMTP id 639EA357804E for ; Mon, 22 Apr 2013 20:02:51 +1000 (EST) Received: from d23av02.au.ibm.com (d23av02.au.ibm.com [9.190.235.138]) by d23relay03.au.ibm.com (8.13.8/8.13.8/NCO v10.0) with ESMTP id r3MA2EqM61997148 for ; Mon, 22 Apr 2013 20:02:14 +1000 Received: from d23av02.au.ibm.com (loopback [127.0.0.1]) by d23av02.au.ibm.com (8.14.4/8.13.1/NCO v10.0 AVout) with ESMTP id r3MA2J5f011355 for ; Mon, 22 Apr 2013 20:02:20 +1000 Received: from skywalker.in.ibm.com ([9.77.207.129]) by d23av02.au.ibm.com (8.14.4/8.13.1/NCO v10.0 AVin) with ESMTP id r3MA1PsV009405; Mon, 22 Apr 2013 20:02:15 +1000 From: "Aneesh Kumar K.V" To: benh@kernel.crashing.org, paulus@samba.org, David Gibson Subject: [PATCH -V6 09/27] powerpc: Reduce PTE table memory wastage Date: Mon, 22 Apr 2013 15:30:43 +0530 Message-Id: <1366624861-24948-10-git-send-email-aneesh.kumar@linux.vnet.ibm.com> X-Mailer: git-send-email 1.7.10 In-Reply-To: <1366624861-24948-1-git-send-email-aneesh.kumar@linux.vnet.ibm.com> References: <1366624861-24948-1-git-send-email-aneesh.kumar@linux.vnet.ibm.com> x-cbid: 13042209-1396-0000-0000-000002D7ADC4 Cc: linuxppc-dev@lists.ozlabs.org, "Aneesh Kumar K.V" X-BeenThere: linuxppc-dev@lists.ozlabs.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , MIME-Version: 1.0 Errors-To: linuxppc-dev-bounces+patchwork-incoming=ozlabs.org@lists.ozlabs.org Sender: "Linuxppc-dev" From: "Aneesh Kumar K.V" We allocate one page for the last level of linux page table. With THP and large page size of 16MB, that would mean we are wasting large part of that page. To map 16MB area, we only need a PTE space of 2K with 64K page size. This patch reduce the space wastage by sharing the page allocated for the last level of linux page table with multiple pmd entries. We call these smaller chunks PTE page fragments and allocated page, PTE page. In order to support systems which doesn't have 64K HPTE support, we also add another 2K to PTE page fragment. The second half of the PTE fragments is used for storing slot and secondary bit information of an HPTE. With this we now have a 4K PTE fragment. We use a simple approach to share the PTE page. On allocation, we bump the PTE page refcount to 16 and share the PTE page with the next 16 pte alloc request. This should help in the node locality of the PTE page fragment, assuming that the immediate pte alloc request will mostly come from the same NUMA node. We don't try to reuse the freed PTE page fragment. Hence we could be waisting some space. Signed-off-by: Aneesh Kumar K.V --- arch/powerpc/include/asm/mmu-book3e.h | 4 ++ arch/powerpc/include/asm/mmu-hash64.h | 4 ++ arch/powerpc/include/asm/page.h | 4 ++ arch/powerpc/include/asm/pgalloc-64.h | 82 +++++++---------------- arch/powerpc/kernel/setup_64.c | 4 +- arch/powerpc/mm/mmu_context_hash64.c | 37 +++++++++++ arch/powerpc/mm/pgtable_64.c | 118 +++++++++++++++++++++++++++++++++ 7 files changed, 195 insertions(+), 58 deletions(-) diff --git a/arch/powerpc/include/asm/mmu-book3e.h b/arch/powerpc/include/asm/mmu-book3e.h index 99d43e0..8bd560c 100644 --- a/arch/powerpc/include/asm/mmu-book3e.h +++ b/arch/powerpc/include/asm/mmu-book3e.h @@ -231,6 +231,10 @@ typedef struct { u64 high_slices_psize; /* 4 bits per slice for now */ u16 user_psize; /* page size index */ #endif +#ifdef CONFIG_PPC_64K_PAGES + /* for 4K PTE fragment support */ + void *pte_frag; +#endif } mm_context_t; /* Page size definitions, common between 32 and 64-bit diff --git a/arch/powerpc/include/asm/mmu-hash64.h b/arch/powerpc/include/asm/mmu-hash64.h index 05895cf..de9e577 100644 --- a/arch/powerpc/include/asm/mmu-hash64.h +++ b/arch/powerpc/include/asm/mmu-hash64.h @@ -516,6 +516,10 @@ typedef struct { unsigned long acop; /* mask of enabled coprocessor types */ unsigned int cop_pid; /* pid value used with coprocessors */ #endif /* CONFIG_PPC_ICSWX */ +#ifdef CONFIG_PPC_64K_PAGES + /* for 4K PTE fragment support */ + void *pte_frag; +#endif } mm_context_t; diff --git a/arch/powerpc/include/asm/page.h b/arch/powerpc/include/asm/page.h index 6faa416..c204ad1 100644 --- a/arch/powerpc/include/asm/page.h +++ b/arch/powerpc/include/asm/page.h @@ -372,7 +372,11 @@ void arch_free_page(struct page *page, int order); struct vm_area_struct; +#ifdef CONFIG_PPC_64K_PAGES +typedef pte_t *pgtable_t; +#else typedef struct page *pgtable_t; +#endif #include #endif /* __ASSEMBLY__ */ diff --git a/arch/powerpc/include/asm/pgalloc-64.h b/arch/powerpc/include/asm/pgalloc-64.h index d390123..91acb12 100644 --- a/arch/powerpc/include/asm/pgalloc-64.h +++ b/arch/powerpc/include/asm/pgalloc-64.h @@ -152,6 +152,23 @@ static inline void __pte_free_tlb(struct mmu_gather *tlb, pgtable_t table, } #else /* if CONFIG_PPC_64K_PAGES */ +/* + * we support 16 fragments per PTE page. + */ +#define PTE_FRAG_NR 16 +/* + * We use a 2K PTE page fragment and another 2K for storing + * real_pte_t hash index + */ +#define PTE_FRAG_SIZE_SHIFT 12 +#define PTE_FRAG_SIZE (2 * PTRS_PER_PTE * sizeof(pte_t)) + +extern pte_t *page_table_alloc(struct mm_struct *, unsigned long, int); +extern void page_table_free(struct mm_struct *, unsigned long *, int); +extern void pgtable_free_tlb(struct mmu_gather *tlb, void *table, int shift); +#ifdef CONFIG_SMP +extern void __tlb_remove_table(void *_table); +#endif #define pud_populate(mm, pud, pmd) pud_set(pud, (unsigned long)pmd) @@ -164,90 +181,42 @@ static inline void pmd_populate_kernel(struct mm_struct *mm, pmd_t *pmd, static inline void pmd_populate(struct mm_struct *mm, pmd_t *pmd, pgtable_t pte_page) { - pmd_populate_kernel(mm, pmd, page_address(pte_page)); + pmd_set(pmd, (unsigned long)pte_page); } static inline pgtable_t pmd_pgtable(pmd_t pmd) { - return pmd_page(pmd); + return (pgtable_t)(pmd_val(pmd) & -sizeof(pte_t)*PTRS_PER_PTE); } 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 *)page_table_alloc(mm, address, 1); } static inline pgtable_t pte_alloc_one(struct mm_struct *mm, - unsigned long address) + unsigned long address) { - struct page *page; - pte_t *pte; - - pte = pte_alloc_one_kernel(mm, address); - if (!pte) - return NULL; - page = virt_to_page(pte); - pgtable_page_ctor(page); - return page; + return (pgtable_t)page_table_alloc(mm, address, 0); } static inline void pte_free_kernel(struct mm_struct *mm, pte_t *pte) { - free_page((unsigned long)pte); + page_table_free(mm, (unsigned long *)pte, 1); } static inline void pte_free(struct mm_struct *mm, pgtable_t ptepage) { - pgtable_page_dtor(ptepage); - __free_page(ptepage); + page_table_free(mm, (unsigned long *)ptepage, 0); } -static inline void pgtable_free(void *table, unsigned index_size) -{ - if (!index_size) - free_page((unsigned long)table); - else { - BUG_ON(index_size > MAX_PGTABLE_INDEX_SIZE); - kmem_cache_free(PGT_CACHE(index_size), table); - } -} - -#ifdef CONFIG_SMP -static inline void pgtable_free_tlb(struct mmu_gather *tlb, - void *table, int shift) -{ - unsigned long pgf = (unsigned long)table; - BUG_ON(shift > MAX_PGTABLE_INDEX_SIZE); - pgf |= shift; - tlb_remove_table(tlb, (void *)pgf); -} - -static inline void __tlb_remove_table(void *_table) -{ - void *table = (void *)((unsigned long)_table & ~MAX_PGTABLE_INDEX_SIZE); - unsigned shift = (unsigned long)_table & MAX_PGTABLE_INDEX_SIZE; - - pgtable_free(table, shift); -} -#else /* !CONFIG_SMP */ -static inline void pgtable_free_tlb(struct mmu_gather *tlb, - void *table, int shift) -{ - pgtable_free(table, shift); -} -#endif /* CONFIG_SMP */ - static inline void __pte_free_tlb(struct mmu_gather *tlb, pgtable_t table, unsigned long address) { - struct page *page = page_address(table); - tlb_flush_pgtable(tlb, address); - pgtable_page_dtor(page); - pgtable_free_tlb(tlb, page, 0); + pgtable_free_tlb(tlb, table, 0); } - #endif /* CONFIG_PPC_64K_PAGES */ static inline pmd_t *pmd_alloc_one(struct mm_struct *mm, unsigned long addr) @@ -261,7 +230,6 @@ static inline void pmd_free(struct mm_struct *mm, pmd_t *pmd) kmem_cache_free(PGT_CACHE(PMD_INDEX_SIZE), pmd); } - #define __pmd_free_tlb(tlb, pmd, addr) \ pgtable_free_tlb(tlb, pmd, PMD_INDEX_SIZE) #ifndef CONFIG_PPC_64K_PAGES diff --git a/arch/powerpc/kernel/setup_64.c b/arch/powerpc/kernel/setup_64.c index 75fbaceb..e379d3f 100644 --- a/arch/powerpc/kernel/setup_64.c +++ b/arch/powerpc/kernel/setup_64.c @@ -583,7 +583,9 @@ void __init setup_arch(char **cmdline_p) init_mm.end_code = (unsigned long) _etext; init_mm.end_data = (unsigned long) _edata; init_mm.brk = klimit; - +#ifdef CONFIG_PPC_64K_PAGES + init_mm.context.pte_frag = NULL; +#endif irqstack_early_init(); exc_lvl_early_init(); emergency_stack_init(); diff --git a/arch/powerpc/mm/mmu_context_hash64.c b/arch/powerpc/mm/mmu_context_hash64.c index d1d1b92..178876ae 100644 --- a/arch/powerpc/mm/mmu_context_hash64.c +++ b/arch/powerpc/mm/mmu_context_hash64.c @@ -23,6 +23,7 @@ #include #include +#include #include "icswx.h" @@ -85,6 +86,9 @@ int init_new_context(struct task_struct *tsk, struct mm_struct *mm) spin_lock_init(mm->context.cop_lockp); #endif /* CONFIG_PPC_ICSWX */ +#ifdef CONFIG_PPC_64K_PAGES + mm->context.pte_frag = NULL; +#endif return 0; } @@ -96,13 +100,46 @@ void __destroy_context(int context_id) } EXPORT_SYMBOL_GPL(__destroy_context); +#ifdef CONFIG_PPC_64K_PAGES +static void destroy_pagetable_page(struct mm_struct *mm) +{ + int count; + void *pte_frag; + struct page *page; + + pte_frag = mm->context.pte_frag; + if (!pte_frag) + return; + + page = virt_to_page(pte_frag); + /* drop all the pending references */ + count = ((unsigned long)pte_frag & ~PAGE_MASK) >> PTE_FRAG_SIZE_SHIFT; + /* We allow PTE_FRAG_NR fragments from a PTE page */ + count = atomic_sub_return(PTE_FRAG_NR - count, &page->_count); + if (!count) { + pgtable_page_dtor(page); + free_hot_cold_page(page, 0); + } +} + +#else +static inline void destroy_pagetable_page(struct mm_struct *mm) +{ + return; +} +#endif + + void destroy_context(struct mm_struct *mm) { + #ifdef CONFIG_PPC_ICSWX drop_cop(mm->context.acop, mm); kfree(mm->context.cop_lockp); mm->context.cop_lockp = NULL; #endif /* CONFIG_PPC_ICSWX */ + + destroy_pagetable_page(mm); __destroy_context(mm->context.id); subpage_prot_free(mm); mm->context.id = MMU_NO_CONTEXT; diff --git a/arch/powerpc/mm/pgtable_64.c b/arch/powerpc/mm/pgtable_64.c index 654258f..a854096 100644 --- a/arch/powerpc/mm/pgtable_64.c +++ b/arch/powerpc/mm/pgtable_64.c @@ -337,3 +337,121 @@ EXPORT_SYMBOL(__ioremap_at); EXPORT_SYMBOL(iounmap); EXPORT_SYMBOL(__iounmap); EXPORT_SYMBOL(__iounmap_at); + +#ifdef CONFIG_PPC_64K_PAGES +static pte_t *get_from_cache(struct mm_struct *mm) +{ + void *pte_frag, *ret; + + spin_lock(&mm->page_table_lock); + ret = mm->context.pte_frag; + if (ret) { + pte_frag = ret + PTE_FRAG_SIZE; + /* + * If we have taken up all the fragments mark PTE page NULL + */ + if (((unsigned long)pte_frag & ~PAGE_MASK) == 0) + pte_frag = NULL; + mm->context.pte_frag = pte_frag; + } + spin_unlock(&mm->page_table_lock); + return (pte_t *)ret; +} + +static pte_t *__alloc_for_cache(struct mm_struct *mm, int kernel) +{ + void *ret = NULL; + struct page *page = alloc_page(GFP_KERNEL | __GFP_NOTRACK | + __GFP_REPEAT | __GFP_ZERO); + if (!page) + return NULL; + + ret = page_address(page); + spin_lock(&mm->page_table_lock); + /* + * If we find pgtable_page set, we return + * the allocated page with single fragement + * count. + */ + if (likely(!mm->context.pte_frag)) { + atomic_set(&page->_count, PTE_FRAG_NR); + mm->context.pte_frag = ret + PTE_FRAG_SIZE; + } + spin_unlock(&mm->page_table_lock); + + if (!kernel) + pgtable_page_ctor(page); + + return (pte_t *)ret; +} + +pte_t *page_table_alloc(struct mm_struct *mm, unsigned long vmaddr, int kernel) +{ + pte_t *pte; + + pte = get_from_cache(mm); + if (pte) + return pte; + + return __alloc_for_cache(mm, kernel); +} + +void page_table_free(struct mm_struct *mm, unsigned long *table, int kernel) +{ + struct page *page = virt_to_page(table); + if (put_page_testzero(page)) { + if (!kernel) + pgtable_page_dtor(page); + free_hot_cold_page(page, 0); + } +} + +#ifdef CONFIG_SMP +static void page_table_free_rcu(void *table) +{ + struct page *page = virt_to_page(table); + if (put_page_testzero(page)) { + pgtable_page_dtor(page); + free_hot_cold_page(page, 0); + } +} + +void pgtable_free_tlb(struct mmu_gather *tlb, void *table, int shift) +{ + unsigned long pgf = (unsigned long)table; + + BUG_ON(shift > MAX_PGTABLE_INDEX_SIZE); + pgf |= shift; + tlb_remove_table(tlb, (void *)pgf); +} + +void __tlb_remove_table(void *_table) +{ + void *table = (void *)((unsigned long)_table & ~MAX_PGTABLE_INDEX_SIZE); + unsigned shift = (unsigned long)_table & MAX_PGTABLE_INDEX_SIZE; + + if (!shift) + /* PTE page needs special handling */ + page_table_free_rcu(table); + else { + BUG_ON(shift > MAX_PGTABLE_INDEX_SIZE); + kmem_cache_free(PGT_CACHE(shift), table); + } +} +#else +void pgtable_free_tlb(struct mmu_gather *tlb, void *table, int shift) +{ + if (!shift) { + /* PTE page needs special handling */ + struct page *page = virt_to_page(table); + if (put_page_testzero(page)) { + pgtable_page_dtor(page); + free_hot_cold_page(page, 0); + } + } else { + BUG_ON(shift > MAX_PGTABLE_INDEX_SIZE); + kmem_cache_free(PGT_CACHE(shift), table); + } +} +#endif +#endif /* CONFIG_PPC_64K_PAGES */