arm64-revert-tlb-rcu_table_free.patch: revert 5e5f6dc1

which causes lockups on arm64 machines.
This commit is contained in:
Kyle McMartin 2015-03-13 12:38:27 -04:00
parent 62a6befde2
commit 088b1f27a9
2 changed files with 443 additions and 0 deletions

View File

@ -0,0 +1,436 @@
diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
index 1b8e973..31042fe 100644
--- a/arch/arm64/Kconfig
+++ b/arch/arm64/Kconfig
@@ -65,7 +65,6 @@ config ARM64
select HAVE_PERF_EVENTS
select HAVE_PERF_REGS
select HAVE_PERF_USER_STACK_DUMP
- select HAVE_RCU_TABLE_FREE
select HAVE_SYSCALL_TRACEPOINTS
select IRQ_DOMAIN
select MODULES_USE_ELF_RELA
@@ -119,9 +118,6 @@ config GENERIC_CALIBRATE_DELAY
config ZONE_DMA
def_bool y
-config HAVE_GENERIC_RCU_GUP
- def_bool y
-
config ARCH_DMA_ADDR_T_64BIT
def_bool y
diff --git a/arch/arm64/include/asm/pgtable.h b/arch/arm64/include/asm/pgtable.h
index 800ec0e..e2e79ac 100644
--- a/arch/arm64/include/asm/pgtable.h
+++ b/arch/arm64/include/asm/pgtable.h
@@ -243,16 +243,6 @@ static inline void set_pte_at(struct mm_struct *mm, unsigned long addr,
#define __HAVE_ARCH_PTE_SPECIAL
-static inline pte_t pud_pte(pud_t pud)
-{
- return __pte(pud_val(pud));
-}
-
-static inline pmd_t pud_pmd(pud_t pud)
-{
- return __pmd(pud_val(pud));
-}
-
static inline pte_t pmd_pte(pmd_t pmd)
{
return __pte(pmd_val(pmd));
@@ -275,13 +265,7 @@ static inline pgprot_t mk_sect_prot(pgprot_t prot)
#ifdef CONFIG_TRANSPARENT_HUGEPAGE
#define pmd_trans_huge(pmd) (pmd_val(pmd) && !(pmd_val(pmd) & PMD_TABLE_BIT))
#define pmd_trans_splitting(pmd) pte_special(pmd_pte(pmd))
-#ifdef CONFIG_HAVE_RCU_TABLE_FREE
-#define __HAVE_ARCH_PMDP_SPLITTING_FLUSH
-struct vm_area_struct;
-void pmdp_splitting_flush(struct vm_area_struct *vma, unsigned long address,
- pmd_t *pmdp);
-#endif /* CONFIG_HAVE_RCU_TABLE_FREE */
-#endif /* CONFIG_TRANSPARENT_HUGEPAGE */
+#endif
#define pmd_dirty(pmd) pte_dirty(pmd_pte(pmd))
#define pmd_young(pmd) pte_young(pmd_pte(pmd))
@@ -302,7 +286,7 @@ void pmdp_splitting_flush(struct vm_area_struct *vma, unsigned long address,
#define pfn_pmd(pfn,prot) (__pmd(((phys_addr_t)(pfn) << PAGE_SHIFT) | pgprot_val(prot)))
#define mk_pmd(page,prot) pfn_pmd(page_to_pfn(page),prot)
-#define pud_write(pud) pte_write(pud_pte(pud))
+#define pmd_page(pmd) pfn_to_page(__phys_to_pfn(pmd_val(pmd) & PHYS_MASK))
#define pud_pfn(pud) (((pud_val(pud) & PUD_MASK) & PHYS_MASK) >> PAGE_SHIFT)
#define set_pmd_at(mm, addr, pmdp, pmd) set_pte_at(mm, addr, (pte_t *)pmdp, pmd_pte(pmd))
@@ -407,8 +391,6 @@ static inline pmd_t *pmd_offset(pud_t *pud, unsigned long addr)
return (pmd_t *)pud_page_vaddr(*pud) + pmd_index(addr);
}
-#define pud_page(pud) pfn_to_page(__phys_to_pfn(pud_val(pud) & PHYS_MASK))
-
#endif /* CONFIG_ARM64_PGTABLE_LEVELS > 2 */
#if CONFIG_ARM64_PGTABLE_LEVELS > 3
@@ -443,8 +425,6 @@ static inline pud_t *pud_offset(pgd_t *pgd, unsigned long addr)
return (pud_t *)pgd_page_vaddr(*pgd) + pud_index(addr);
}
-#define pgd_page(pgd) pfn_to_page(__phys_to_pfn(pgd_val(pgd) & PHYS_MASK))
-
#endif /* CONFIG_ARM64_PGTABLE_LEVELS > 3 */
#define pgd_ERROR(pgd) __pgd_error(__FILE__, __LINE__, pgd_val(pgd))
diff --git a/arch/arm64/include/asm/tlb.h b/arch/arm64/include/asm/tlb.h
index c028fe3..62731ef 100644
--- a/arch/arm64/include/asm/tlb.h
+++ b/arch/arm64/include/asm/tlb.h
@@ -19,44 +19,84 @@
#ifndef __ASM_TLB_H
#define __ASM_TLB_H
-#include <linux/pagemap.h>
-#include <linux/swap.h>
-
-#ifdef CONFIG_HAVE_RCU_TABLE_FREE
-
-#define tlb_remove_entry(tlb, entry) tlb_remove_table(tlb, entry)
-static inline void __tlb_remove_table(void *_table)
-{
- free_page_and_swap_cache((struct page *)_table);
-}
-#else
-#define tlb_remove_entry(tlb, entry) tlb_remove_page(tlb, entry)
-#endif /* CONFIG_HAVE_RCU_TABLE_FREE */
+#define __tlb_remove_pmd_tlb_entry __tlb_remove_pmd_tlb_entry
#include <asm-generic/tlb.h>
+/*
+ * There's three ways the TLB shootdown code is used:
+ * 1. Unmapping a range of vmas. See zap_page_range(), unmap_region().
+ * tlb->fullmm = 0, and tlb_start_vma/tlb_end_vma will be called.
+ * 2. Unmapping all vmas. See exit_mmap().
+ * tlb->fullmm = 1, and tlb_start_vma/tlb_end_vma will be called.
+ * Page tables will be freed.
+ * 3. Unmapping argument pages. See shift_arg_pages().
+ * tlb->fullmm = 0, but tlb_start_vma/tlb_end_vma will not be called.
+ */
static inline void tlb_flush(struct mmu_gather *tlb)
{
if (tlb->fullmm) {
flush_tlb_mm(tlb->mm);
- } else {
+ } else if (tlb->end > 0) {
struct vm_area_struct vma = { .vm_mm = tlb->mm, };
flush_tlb_range(&vma, tlb->start, tlb->end);
+ tlb->start = TASK_SIZE;
+ tlb->end = 0;
+ }
+}
+
+static inline void tlb_add_flush(struct mmu_gather *tlb, unsigned long addr)
+{
+ if (!tlb->fullmm) {
+ tlb->start = min(tlb->start, addr);
+ tlb->end = max(tlb->end, addr + PAGE_SIZE);
}
}
+/*
+ * Memorize the range for the TLB flush.
+ */
+static inline void __tlb_remove_tlb_entry(struct mmu_gather *tlb, pte_t *ptep,
+ unsigned long addr)
+{
+ tlb_add_flush(tlb, addr);
+}
+
+/*
+ * In the case of tlb vma handling, we can optimise these away in the
+ * case where we're doing a full MM flush. When we're doing a munmap,
+ * the vmas are adjusted to only cover the region to be torn down.
+ */
+static inline void tlb_start_vma(struct mmu_gather *tlb,
+ struct vm_area_struct *vma)
+{
+ if (!tlb->fullmm) {
+ tlb->start = TASK_SIZE;
+ tlb->end = 0;
+ }
+}
+
+static inline void tlb_end_vma(struct mmu_gather *tlb,
+ struct vm_area_struct *vma)
+{
+ if (!tlb->fullmm)
+ tlb_flush(tlb);
+}
+
static inline void __pte_free_tlb(struct mmu_gather *tlb, pgtable_t pte,
unsigned long addr)
{
pgtable_page_dtor(pte);
- tlb_remove_entry(tlb, pte);
+ tlb_add_flush(tlb, addr);
+ tlb_remove_page(tlb, pte);
}
#if CONFIG_ARM64_PGTABLE_LEVELS > 2
static inline void __pmd_free_tlb(struct mmu_gather *tlb, pmd_t *pmdp,
unsigned long addr)
{
- tlb_remove_entry(tlb, virt_to_page(pmdp));
+ tlb_add_flush(tlb, addr);
+ tlb_remove_page(tlb, virt_to_page(pmdp));
}
#endif
@@ -64,8 +104,15 @@ static inline void __pmd_free_tlb(struct mmu_gather *tlb, pmd_t *pmdp,
static inline void __pud_free_tlb(struct mmu_gather *tlb, pud_t *pudp,
unsigned long addr)
{
- tlb_remove_entry(tlb, virt_to_page(pudp));
+ tlb_add_flush(tlb, addr);
+ tlb_remove_page(tlb, virt_to_page(pudp));
}
#endif
+static inline void __tlb_remove_pmd_tlb_entry(struct mmu_gather *tlb, pmd_t *pmdp,
+ unsigned long address)
+{
+ tlb_add_flush(tlb, address);
+}
+
#endif
diff --git a/arch/arm64/mm/flush.c b/arch/arm64/mm/flush.c
index b6f14e8..0d64089 100644
--- a/arch/arm64/mm/flush.c
+++ b/arch/arm64/mm/flush.c
@@ -104,19 +104,3 @@ EXPORT_SYMBOL(flush_dcache_page);
*/
EXPORT_SYMBOL(flush_cache_all);
EXPORT_SYMBOL(flush_icache_range);
-
-#ifdef CONFIG_TRANSPARENT_HUGEPAGE
-#ifdef CONFIG_HAVE_RCU_TABLE_FREE
-void pmdp_splitting_flush(struct vm_area_struct *vma, unsigned long address,
- pmd_t *pmdp)
-{
- pmd_t pmd = pmd_mksplitting(*pmdp);
-
- VM_BUG_ON(address & ~PMD_MASK);
- set_pmd_at(vma->vm_mm, address, pmdp, pmd);
-
- /* dummy IPI to serialise against fast_gup */
- kick_all_cpus_sync();
-}
-#endif /* CONFIG_HAVE_RCU_TABLE_FREE */
-#endif /* CONFIG_TRANSPARENT_HUGEPAGE */
diff --git a/include/asm-generic/tlb.h b/include/asm-generic/tlb.h
index db284bf..5672d7e 100644
--- a/include/asm-generic/tlb.h
+++ b/include/asm-generic/tlb.h
@@ -96,9 +96,10 @@ struct mmu_gather {
#endif
unsigned long start;
unsigned long end;
+ unsigned int need_flush : 1, /* Did free PTEs */
/* we are in the middle of an operation to clear
* a full mm and can make some optimizations */
- unsigned int fullmm : 1,
+ fullmm : 1,
/* we have performed an operation which
* requires a complete flush of the tlb */
need_flush_all : 1;
@@ -127,58 +128,16 @@ static inline void tlb_remove_page(struct mmu_gather *tlb, struct page *page)
tlb_flush_mmu(tlb);
}
-static inline void __tlb_adjust_range(struct mmu_gather *tlb,
- unsigned long address)
-{
- tlb->start = min(tlb->start, address);
- tlb->end = max(tlb->end, address + PAGE_SIZE);
-}
-
-static inline void __tlb_reset_range(struct mmu_gather *tlb)
-{
- if (tlb->fullmm) {
- tlb->start = tlb->end = ~0;
- } else {
- tlb->start = TASK_SIZE;
- tlb->end = 0;
- }
-}
-
-/*
- * In the case of tlb vma handling, we can optimise these away in the
- * case where we're doing a full MM flush. When we're doing a munmap,
- * the vmas are adjusted to only cover the region to be torn down.
- */
-#ifndef tlb_start_vma
-#define tlb_start_vma(tlb, vma) do { } while (0)
-#endif
-
-#define __tlb_end_vma(tlb, vma) \
- do { \
- if (!tlb->fullmm && tlb->end) { \
- tlb_flush(tlb); \
- __tlb_reset_range(tlb); \
- } \
- } while (0)
-
-#ifndef tlb_end_vma
-#define tlb_end_vma __tlb_end_vma
-#endif
-
-#ifndef __tlb_remove_tlb_entry
-#define __tlb_remove_tlb_entry(tlb, ptep, address) do { } while (0)
-#endif
-
/**
* tlb_remove_tlb_entry - remember a pte unmapping for later tlb invalidation.
*
- * Record the fact that pte's were really unmapped by updating the range,
- * so we can later optimise away the tlb invalidate. This helps when
- * userspace is unmapping already-unmapped pages, which happens quite a lot.
+ * Record the fact that pte's were really umapped in ->need_flush, so we can
+ * later optimise away the tlb invalidate. This helps when userspace is
+ * unmapping already-unmapped pages, which happens quite a lot.
*/
#define tlb_remove_tlb_entry(tlb, ptep, address) \
do { \
- __tlb_adjust_range(tlb, address); \
+ tlb->need_flush = 1; \
__tlb_remove_tlb_entry(tlb, ptep, address); \
} while (0)
@@ -192,27 +151,27 @@ static inline void __tlb_reset_range(struct mmu_gather *tlb)
#define tlb_remove_pmd_tlb_entry(tlb, pmdp, address) \
do { \
- __tlb_adjust_range(tlb, address); \
+ tlb->need_flush = 1; \
__tlb_remove_pmd_tlb_entry(tlb, pmdp, address); \
} while (0)
#define pte_free_tlb(tlb, ptep, address) \
do { \
- __tlb_adjust_range(tlb, address); \
+ tlb->need_flush = 1; \
__pte_free_tlb(tlb, ptep, address); \
} while (0)
#ifndef __ARCH_HAS_4LEVEL_HACK
#define pud_free_tlb(tlb, pudp, address) \
do { \
- __tlb_adjust_range(tlb, address); \
+ tlb->need_flush = 1; \
__pud_free_tlb(tlb, pudp, address); \
} while (0)
#endif
#define pmd_free_tlb(tlb, pmdp, address) \
do { \
- __tlb_adjust_range(tlb, address); \
+ tlb->need_flush = 1; \
__pmd_free_tlb(tlb, pmdp, address); \
} while (0)
diff --git a/mm/memory.c b/mm/memory.c
index 411144f..2772d7a 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -220,6 +220,9 @@ void tlb_gather_mmu(struct mmu_gather *tlb, struct mm_struct *mm, unsigned long
/* Is it from 0 to ~0? */
tlb->fullmm = !(start | (end+1));
tlb->need_flush_all = 0;
+ tlb->start = start;
+ tlb->end = end;
+ tlb->need_flush = 0;
tlb->local.next = NULL;
tlb->local.nr = 0;
tlb->local.max = ARRAY_SIZE(tlb->__pages);
@@ -229,28 +232,23 @@ void tlb_gather_mmu(struct mmu_gather *tlb, struct mm_struct *mm, unsigned long
#ifdef CONFIG_HAVE_RCU_TABLE_FREE
tlb->batch = NULL;
#endif
-
- __tlb_reset_range(tlb);
}
static void tlb_flush_mmu_tlbonly(struct mmu_gather *tlb)
{
- if (!tlb->end)
- return;
-
+ tlb->need_flush = 0;
tlb_flush(tlb);
mmu_notifier_invalidate_range(tlb->mm, tlb->start, tlb->end);
#ifdef CONFIG_HAVE_RCU_TABLE_FREE
tlb_table_flush(tlb);
#endif
- __tlb_reset_range(tlb);
}
static void tlb_flush_mmu_free(struct mmu_gather *tlb)
{
struct mmu_gather_batch *batch;
- for (batch = &tlb->local; batch && batch->nr; batch = batch->next) {
+ for (batch = &tlb->local; batch; batch = batch->next) {
free_pages_and_swap_cache(batch->pages, batch->nr);
batch->nr = 0;
}
@@ -259,6 +257,8 @@ static void tlb_flush_mmu_free(struct mmu_gather *tlb)
void tlb_flush_mmu(struct mmu_gather *tlb)
{
+ if (!tlb->need_flush)
+ return;
tlb_flush_mmu_tlbonly(tlb);
tlb_flush_mmu_free(tlb);
}
@@ -293,7 +293,7 @@ int __tlb_remove_page(struct mmu_gather *tlb, struct page *page)
{
struct mmu_gather_batch *batch;
- VM_BUG_ON(!tlb->end);
+ VM_BUG_ON(!tlb->need_flush);
batch = tlb->active;
batch->pages[batch->nr++] = page;
@@ -360,6 +360,8 @@ void tlb_remove_table(struct mmu_gather *tlb, void *table)
{
struct mmu_table_batch **batch = &tlb->batch;
+ tlb->need_flush = 1;
+
/*
* When there's less then two users of this mm there cannot be a
* concurrent page-table walk.
@@ -1161,8 +1163,20 @@ again:
arch_leave_lazy_mmu_mode();
/* Do the actual TLB flush before dropping ptl */
- if (force_flush)
+ if (force_flush) {
+ unsigned long old_end;
+
+ /*
+ * Flush the TLB just for the previous segment,
+ * then update the range to be the remaining
+ * TLB range.
+ */
+ old_end = tlb->end;
+ tlb->end = addr;
tlb_flush_mmu_tlbonly(tlb);
+ tlb->start = addr;
+ tlb->end = old_end;
+ }
pte_unmap_unlock(start_pte, ptl);
/*

View File

@ -660,6 +660,7 @@ Patch26168: HID-multitouch-add-support-of-clickpads.patch
# git clone ssh://git.fedorahosted.org/git/kernel-arm64.git, git diff master...devel # git clone ssh://git.fedorahosted.org/git/kernel-arm64.git, git diff master...devel
Patch30000: kernel-arm64.patch Patch30000: kernel-arm64.patch
Patch30001: arm64-revert-tlb-rcu_table_free.patch
# END OF PATCH DEFINITIONS # END OF PATCH DEFINITIONS
@ -1422,8 +1423,10 @@ ApplyPatch HID-multitouch-add-support-of-clickpads.patch
%if 0%{?aarch64patches} %if 0%{?aarch64patches}
ApplyPatch kernel-arm64.patch ApplyPatch kernel-arm64.patch
ApplyPatch arm64-revert-tlb-rcu_table_free.patch
%ifnarch aarch64 # this is stupid, but i want to notice before secondary koji does. %ifnarch aarch64 # this is stupid, but i want to notice before secondary koji does.
ApplyPatch kernel-arm64.patch -R ApplyPatch kernel-arm64.patch -R
ApplyPatch arm64-revert-tlb-rcu_table_free.patch -R
%endif %endif
%endif %endif
@ -2277,6 +2280,10 @@ fi
# #
# #
%changelog %changelog
* Fri Mar 13 2015 Kyle McMartin <kyle@fedoraproject.org>
- arm64-revert-tlb-rcu_table_free.patch: revert 5e5f6dc1 which
causes lockups on arm64 machines.
* Fri Mar 13 2015 Josh Boyer <jwboyer@fedoraproject.org> - 4.0.0-0.rc3.git2.1 * Fri Mar 13 2015 Josh Boyer <jwboyer@fedoraproject.org> - 4.0.0-0.rc3.git2.1
- Linux v4.0-rc3-148-gc202baf017ae - Linux v4.0-rc3-148-gc202baf017ae
- Add patch to support clickpads (rhbz 1201532) - Add patch to support clickpads (rhbz 1201532)