mm: Page fault enhancements

PATCH Superseded Peter Xu 1 participant 9 messages Feb 20, 2020

Subsystems (20)

Memory management Arm (32-bit) x86 MIPS +16 more subsystems

lore

Share & tools

lore

Peter Xu

author reviewer

2020-02-20 14:54 UTC #0 lore raw

v6 cover mm: Page fault enhancements

This is v6 of the series. It is majorly a rebase to 5.6-rc2, nothing else to be expected (plus some tests after the rebase). Instead of rewrite the cover letter I decided to use what we have for v5.

Adding extra CCs for both Bobby Powers <[email protected]> and Brian Geffon <[email protected]>.

Any review comment is appreciated. Thanks,

=============== v5 cover letter ==================

This is v5 of the series. As Matthew suggested, I split the previous patch "mm: Return faster for non-fatal signals in user mode faults" into a few smaller ones:

1. One patch to introduce fatal_signal_pending(), and use it in archs that can directly apply

2. A few more patches to let the rest archs to use the new helper. With that we can have an unified entry for signal detection

3. One last patch to change fatal_signal_pending() to detect userspace non-fatal signal

Nothing should have changed in the rest patches. Because the fault retry patches will depend on the previous ones, I decided to simply repost all the patches.

Here's the new patchset layout:

Patch 1-2: cleanup, and potential bugfix of hugetlbfs on fault retry

Patch 3-9: let page fault to respond to non-fatal signals faster

Patch 10: remove the userfaultfd NOPAGE emulation

Patch 11-14: allow page fault to retry more than once

Patch 15-16: let gup code to use FAULT_FLAG_KILLABLE too

I would really appreciate any review comments for the series, especially for the first two patches which IMHO are even not related to this patchset and they should either cleanup or fix things.

Smoke tested on x86 only.

Thanks,

v5:
- split "mm: Return faster for non-fatal signals in user mode faults" into a few more patches, let all archs to use an unified entry for fast signal handling (fatal_signal_pending)

v4:
- use lore.kernel.org for all the links in commit messages [Kirill]
- one more patch ("mm/gup: Fix __get_user_pages() on fault retry of hugetlb") to fix hugetlb path on fault retry
- one more patch ("mm/gup: Allow to react to fatal signals") to:
- use down_read_killable() properly [Linus]
- pass in FAULT_FLAG_KILLABLE for all GUP [Linus]
- one more patch ("mm/userfaultfd: Honor FAULT_FLAG_KILLABLE in fault path") to let handle_userfaultfd() respect FAULT_FLAG_KILLABLE. Should have no functional change after previous two new patches.

v3:
- check fatal signals in __get_user_page_locked() [Linus]
- add r-bs

v2:
- resent previous version, rebase only

=============== v1 cover letter ==================

This series is split out of userfaultfd-wp series to only cover the general page fault changes, since it seems to make sense itself.

Basically it does two things:

(a) Allows the page fault handlers to be more interactive on not only SIGKILL, but also the rest of userspace signals (especially for user-mode faults), and,

(b) Allows the page fault retry (VM_FAULT_RETRY) to happen for more than once.

I'm keeping the CC list as in uffd-wp v5, hopefully I'm not sending too much spams...

And, instead of writting again the cover letter, I'm just copy-pasting my previous link here which has more details on why we do this:

https://patchwork.kernel.org/cover/10691991/

The major change from that latest version should be that we introduced a new page fault flag FAULT_FLAG_INTERRUPTIBLE as suggested by Linus
[1] to represents that we would like the fault handler to respond to non-fatal signals. Also, we're more careful now on when to do the immediate return of the page fault for such signals. For example, now we'll only check against signal_pending() for user-mode page faults and we keep the kernel-mode page fault patch untouched for it. More information can be found in separate patches.

The patchset is only lightly tested on x86.

All comments are greatly welcomed. Thanks,

[1] https://lkml.org/lkml/2019/6/25/1382

Peter Xu (16):
mm/gup: Rename "nonblocking" to "locked" where proper mm/gup: Fix __get_user_pages() on fault retry of hugetlb
mm: Introduce fault_signal_pending()
x86/mm: Use helper fault_signal_pending()
arc/mm: Use helper fault_signal_pending() arm64/mm: Use helper fault_signal_pending() powerpc/mm: Use helper fault_signal_pending() sh/mm: Use helper fault_signal_pending()
mm: Return faster for non-fatal signals in user mode faults
userfaultfd: Don't retake mmap_sem to emulate NOPAGE
mm: Introduce FAULT_FLAG_DEFAULT
mm: Introduce FAULT_FLAG_INTERRUPTIBLE
mm: Allow VM_FAULT_RETRY for multiple times mm/gup: Allow VM_FAULT_RETRY for multiple times mm/gup: Allow to react to fatal signals mm/userfaultfd: Honor FAULT_FLAG_KILLABLE in fault path

 arch/alpha/mm/fault.c           |  6 +--
 arch/arc/mm/fault.c             | 35 +++++--------
 arch/arm/mm/fault.c             |  7 +--
 arch/arm64/mm/fault.c           | 26 +++------
 arch/hexagon/mm/vm_fault.c      |  5 +-
 arch/ia64/mm/fault.c            |  5 +-
 arch/m68k/mm/fault.c            |  7 +--
 arch/microblaze/mm/fault.c      |  5 +-
 arch/mips/mm/fault.c            |  5 +-
 arch/nds32/mm/fault.c           |  5 +-
 arch/nios2/mm/fault.c           |  7 +--
 arch/openrisc/mm/fault.c        |  5 +-
 arch/parisc/mm/fault.c          |  8 ++-
 arch/powerpc/mm/fault.c         | 20 ++-----
 arch/riscv/mm/fault.c           |  9 +---
 arch/s390/mm/fault.c            | 10 ++--
 arch/sh/mm/fault.c              | 13 +++--
 arch/sparc/mm/fault_32.c        |  5 +-
 arch/sparc/mm/fault_64.c        |  5 +-
 arch/um/kernel/trap.c           |  3 +-
 arch/unicore32/mm/fault.c       |  8 ++-
 arch/x86/mm/fault.c             | 30 +++++------
 arch/xtensa/mm/fault.c          |  5 +-
 drivers/gpu/drm/ttm/ttm_bo_vm.c | 12 +++--
 fs/userfaultfd.c                | 62 ++++++++++------------
 include/linux/mm.h              | 81 ++++++++++++++++++++++++----
 include/linux/sched/signal.h    | 14 +++++
 mm/filemap.c                    |  2 +-
 mm/gup.c                        | 93 +++++++++++++++++++++------------
 mm/hugetlb.c                    | 17 +++---
 mm/internal.h                   |  6 +--
 31 files changed, 281 insertions(+), 240 deletions(-)

Signature · 1 line

2.24.1

Peter Xu

author reviewer

2020-02-20 14:54 UTC #1 lore raw

PATCH v6 1/16 mm/gup: Rename "nonblocking" to "locked" where proper

There's plenty of places around __get_user_pages() that has a parameter "nonblocking" which does not really mean that "it won't block" (because it can really block) but instead it shows whether the mmap_sem is released by up_read() during the page fault handling mostly when VM_FAULT_RETRY is returned.

We have the correct naming in e.g. get_user_pages_locked() or get_user_pages_remote() as "locked", however there're still many places that are using the "nonblocking" as name.

Renaming the places to "locked" where proper to better suite the functionality of the variable. While at it, fixing up some of the comments accordingly.

Reviewed-by Mike Rapoport <[email protected]>

Reviewed-by Jerome Glisse <[email protected]>

Reviewed-by David Hildenbrand <[email protected]>

Signed-off-by Peter Xu <[email protected]>

---
 mm/gup.c     | 44 +++++++++++++++++++++-----------------------
 mm/hugetlb.c |  8 ++++----
 2 files changed, 25 insertions(+), 27 deletions(-)

mm/gup.c +21 ~~-23~~

index 1b521e0ac1de..1b4411bd0042 100644
@@ -630,12 +630,12 @@ static int get_gate_page(struct mm_struct *mm, unsigned long address,
 }
 
 /*
- * mmap_sem must be held on entry.  If @nonblocking != NULL and
- * *@flags does not include FOLL_NOWAIT, the mmap_sem may be released.
- * If it is, *@nonblocking will be set to 0 and -EBUSY returned.
+ * mmap_sem must be held on entry.  If @locked != NULL and *@flags
+ * does not include FOLL_NOWAIT, the mmap_sem may be released.  If it
+ * is, *@locked will be set to 0 and -EBUSY returned.
  */
 static int faultin_page(struct task_struct *tsk, struct vm_area_struct *vma,
-		unsigned long address, unsigned int *flags, int *nonblocking)
+		unsigned long address, unsigned int *flags, int *locked)
 {
 	unsigned int fault_flags = 0;
 	vm_fault_t ret;

@@ -647,7 +647,7 @@ static int faultin_page(struct task_struct *tsk, struct vm_area_struct *vma,
 		fault_flags |= FAULT_FLAG_WRITE;
 	if (*flags & FOLL_REMOTE)
 		fault_flags |= FAULT_FLAG_REMOTE;
-	if (nonblocking)
+	if (locked)
 		fault_flags |= FAULT_FLAG_ALLOW_RETRY;
 	if (*flags & FOLL_NOWAIT)
 		fault_flags |= FAULT_FLAG_ALLOW_RETRY | FAULT_FLAG_RETRY_NOWAIT;

@@ -673,8 +673,8 @@ static int faultin_page(struct task_struct *tsk, struct vm_area_struct *vma,
 	}
 
 	if (ret & VM_FAULT_RETRY) {
-		if (nonblocking && !(fault_flags & FAULT_FLAG_RETRY_NOWAIT))
-			*nonblocking = 0;
+		if (locked && !(fault_flags & FAULT_FLAG_RETRY_NOWAIT))
+			*locked = 0;
 		return -EBUSY;
 	}

@@ -751,7 +751,7 @@ static int check_vma_flags(struct vm_area_struct *vma, unsigned long gup_flags)
  *		only intends to ensure the pages are faulted in.
  * @vmas:	array of pointers to vmas corresponding to each page.
  *		Or NULL if the caller does not require them.
- * @nonblocking: whether waiting for disk IO or mmap_sem contention
+ * @locked:     whether we're still with the mmap_sem held
  *
  * Returns either number of pages pinned (which may be less than the
  * number requested), or an error. Details about the return value:

@@ -786,13 +786,11 @@ static int check_vma_flags(struct vm_area_struct *vma, unsigned long gup_flags)
  * appropriate) must be called after the page is finished with, and
  * before put_page is called.
  *
- * If @nonblocking != NULL, __get_user_pages will not wait for disk IO
- * or mmap_sem contention, and if waiting is needed to pin all pages,
- * *@nonblocking will be set to 0.  Further, if @gup_flags does not
- * include FOLL_NOWAIT, the mmap_sem will be released via up_read() in
- * this case.
+ * If @locked != NULL, *@locked will be set to 0 when mmap_sem is
+ * released by an up_read().  That can happen if @gup_flags does not
+ * have FOLL_NOWAIT.
  *
- * A caller using such a combination of @nonblocking and @gup_flags
+ * A caller using such a combination of @locked and @gup_flags
  * must therefore hold the mmap_sem for reading only, and recognize
  * when it's been released.  Otherwise, it must be held for either
  * reading or writing and will not be released.

@@ -804,7 +802,7 @@ static int check_vma_flags(struct vm_area_struct *vma, unsigned long gup_flags)
 static long __get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
 		unsigned long start, unsigned long nr_pages,
 		unsigned int gup_flags, struct page **pages,
-		struct vm_area_struct **vmas, int *nonblocking)
+		struct vm_area_struct **vmas, int *locked)
 {
 	long ret = 0, i = 0;
 	struct vm_area_struct *vma = NULL;

@@ -850,7 +848,7 @@ static long __get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
 			if (is_vm_hugetlb_page(vma)) {
 				i = follow_hugetlb_page(mm, vma, pages, vmas,
 						&start, &nr_pages, i,
-						gup_flags, nonblocking);
+						gup_flags, locked);
 				continue;
 			}
 		}

@@ -868,7 +866,7 @@ static long __get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
 		page = follow_page_mask(vma, start, foll_flags, &ctx);
 		if (!page) {
 			ret = faultin_page(tsk, vma, start, &foll_flags,
-					nonblocking);
+					   locked);
 			switch (ret) {
 			case 0:
 				goto retry;

@@ -1129,7 +1127,7 @@ static __always_inline long __get_user_pages_locked(struct task_struct *tsk,
  * @vma:   target vma
  * @start: start address
  * @end:   end address
- * @nonblocking:
+ * @locked: whether the mmap_sem is still held
  *
  * This takes care of mlocking the pages too if VM_LOCKED is set.
  *

@@ -1137,14 +1135,14 @@ static __always_inline long __get_user_pages_locked(struct task_struct *tsk,
  *
  * vma->vm_mm->mmap_sem must be held.
  *
- * If @nonblocking is NULL, it may be held for read or write and will
+ * If @locked is NULL, it may be held for read or write and will
  * be unperturbed.
  *
- * If @nonblocking is non-NULL, it must held for read only and may be
- * released.  If it's released, *@nonblocking will be set to 0.
+ * If @locked is non-NULL, it must held for read only and may be
+ * released.  If it's released, *@locked will be set to 0.
  */
 long populate_vma_page_range(struct vm_area_struct *vma,
-		unsigned long start, unsigned long end, int *nonblocking)
+		unsigned long start, unsigned long end, int *locked)
 {
 	struct mm_struct *mm = vma->vm_mm;
 	unsigned long nr_pages = (end - start) / PAGE_SIZE;

@@ -1179,7 +1177,7 @@ long populate_vma_page_range(struct vm_area_struct *vma,
 	 * not result in a stack expansion that recurses back here.
 	 */
 	return __get_user_pages(current, mm, start, nr_pages, gup_flags,
-				NULL, NULL, nonblocking);
+				NULL, NULL, locked);
 }
 
 /*

mm/hugetlb.c +4 -4

index dd8737a94bec..c84f721db020 100644
@@ -4266,7 +4266,7 @@ int hugetlb_mcopy_atomic_pte(struct mm_struct *dst_mm,
 long follow_hugetlb_page(struct mm_struct *mm, struct vm_area_struct *vma,
 			 struct page **pages, struct vm_area_struct **vmas,
 			 unsigned long *position, unsigned long *nr_pages,
-			 long i, unsigned int flags, int *nonblocking)
+			 long i, unsigned int flags, int *locked)
 {
 	unsigned long pfn_offset;
 	unsigned long vaddr = *position;

@@ -4337,7 +4337,7 @@ long follow_hugetlb_page(struct mm_struct *mm, struct vm_area_struct *vma,
 				spin_unlock(ptl);
 			if (flags & FOLL_WRITE)
 				fault_flags |= FAULT_FLAG_WRITE;
-			if (nonblocking)
+			if (locked)
 				fault_flags |= FAULT_FLAG_ALLOW_RETRY;
 			if (flags & FOLL_NOWAIT)
 				fault_flags |= FAULT_FLAG_ALLOW_RETRY |

@@ -4354,9 +4354,9 @@ long follow_hugetlb_page(struct mm_struct *mm, struct vm_area_struct *vma,
 				break;
 			}
 			if (ret & VM_FAULT_RETRY) {
-				if (nonblocking &&
+				if (locked &&
 				    !(fault_flags & FAULT_FLAG_RETRY_NOWAIT))
-					*nonblocking = 0;
+					*locked = 0;
 				*nr_pages = 0;
 				/*
 				 * VM_FAULT_RETRY must not return an

Signature · 1 line

2.24.1

Peter Xu

author reviewer

2020-02-20 14:54 UTC #2 lore raw

PATCH v6 2/16 mm/gup: Fix __get_user_pages() on fault retry of hugetlb

In reply to Peter Xu

When follow_hugetlb_page() returns with *locked==0, it means we've got a VM_FAULT_RETRY within the fauling process and we've released the mmap_sem. When that happens, we should stop and bail out.

Signed-off-by Peter Xu <[email protected]>

---
 mm/gup.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

mm/gup.c +10 -0

index 1b4411bd0042..76cb420c0fb7 100644
@@ -849,6 +849,16 @@ static long __get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
 				i = follow_hugetlb_page(mm, vma, pages, vmas,
 						&start, &nr_pages, i,
 						gup_flags, locked);
+				if (locked && *locked == 0) {
+					/*
+					 * We've got a VM_FAULT_RETRY
+					 * and we've lost mmap_sem.
+					 * We must stop here.
+					 */
+					BUG_ON(gup_flags & FOLL_NOWAIT);
+					BUG_ON(ret != 0);
+					goto out;
+				}
 				continue;
 			}
 		}

Signature · 1 line

2.24.1

Peter Xu

author reviewer

2020-02-20 14:54 UTC #3 lore raw

PATCH v6 3/16 mm: Introduce fault_signal_pending()

In reply to Peter Xu

For most architectures, we've got a quick path to detect fatal signal after a handle_mm_fault(). Introduce a helper for that quick path.

It cleans the current codes a bit so we don't need to duplicate the same check across archs. More importantly, this will be an unified place that we handle the signal immediately right after an interrupted page fault, so it'll be much easier for us if we want to change the behavior of handling signals later on for all the archs.

Note that currently only part of the archs are using this new helper, because some archs have their own way to handle signals. In the follow up patches, we'll try to apply this helper to all the rest of archs.

Another note is that the "regs" parameter in the new helper is not used yet. It'll be used very soon. Now we kept it in this patch only to avoid touching all the archs again in the follow up patches.

Signed-off-by Peter Xu <[email protected]>

---
 arch/alpha/mm/fault.c        |  2 +-
 arch/arm/mm/fault.c          |  2 +-
 arch/hexagon/mm/vm_fault.c   |  2 +-
 arch/ia64/mm/fault.c         |  2 +-
 arch/m68k/mm/fault.c         |  2 +-
 arch/microblaze/mm/fault.c   |  2 +-
 arch/mips/mm/fault.c         |  2 +-
 arch/nds32/mm/fault.c        |  2 +-
 arch/nios2/mm/fault.c        |  2 +-
 arch/openrisc/mm/fault.c     |  2 +-
 arch/parisc/mm/fault.c       |  2 +-
 arch/riscv/mm/fault.c        |  2 +-
 arch/s390/mm/fault.c         |  3 +--
 arch/sparc/mm/fault_32.c     |  2 +-
 arch/sparc/mm/fault_64.c     |  2 +-
 arch/unicore32/mm/fault.c    |  2 +-
 arch/xtensa/mm/fault.c       |  2 +-
 include/linux/sched/signal.h | 13 +++++++++++++
 18 files changed, 30 insertions(+), 18 deletions(-)

arch/alpha/mm/fault.c +1 -1

index 741e61ef9d3f..aea33b599037 100644
@@ -150,7 +150,7 @@ do_page_fault(unsigned long address, unsigned long mmcsr,
 	   the fault.  */
 	fault = handle_mm_fault(vma, address, flags);
 
-	if ((fault & VM_FAULT_RETRY) && fatal_signal_pending(current))
+	if (fault_signal_pending(fault, regs))
 		return;
 
 	if (unlikely(fault & VM_FAULT_ERROR)) {

arch/arm/mm/fault.c +1 -1

index bd0f4821f7e1..937b81ff8649 100644
@@ -295,7 +295,7 @@ do_page_fault(unsigned long addr, unsigned int fsr, struct pt_regs *regs)
 	 * signal first. We do not need to release the mmap_sem because
 	 * it would already be released in __lock_page_or_retry in
 	 * mm/filemap.c. */
-	if ((fault & VM_FAULT_RETRY) && fatal_signal_pending(current)) {
+	if (fault_signal_pending(fault, regs)) {
 		if (!user_mode(regs))
 			goto no_context;
 		return 0;

arch/hexagon/mm/vm_fault.c +1 -1

index b3bc71680ae4..d19beaf11b4c 100644
@@ -91,7 +91,7 @@ void do_page_fault(unsigned long address, long cause, struct pt_regs *regs)
 
 	fault = handle_mm_fault(vma, address, flags);
 
-	if ((fault & VM_FAULT_RETRY) && fatal_signal_pending(current))
+	if (fault_signal_pending(fault, regs))
 		return;
 
 	/* The most common case -- we are done. */

arch/ia64/mm/fault.c +1 -1

index c2f299fe9e04..211b4f439384 100644
@@ -141,7 +141,7 @@ ia64_do_page_fault (unsigned long address, unsigned long isr, struct pt_regs *re
 	 */
 	fault = handle_mm_fault(vma, address, flags);
 
-	if ((fault & VM_FAULT_RETRY) && fatal_signal_pending(current))
+	if (fault_signal_pending(fault, regs))
 		return;
 
 	if (unlikely(fault & VM_FAULT_ERROR)) {

arch/m68k/mm/fault.c +1 -1

index e9b1d7585b43..a455e202691b 100644
@@ -138,7 +138,7 @@ int do_page_fault(struct pt_regs *regs, unsigned long address,
 	fault = handle_mm_fault(vma, address, flags);
 	pr_debug("handle_mm_fault returns %x\n", fault);
 
-	if ((fault & VM_FAULT_RETRY) && fatal_signal_pending(current))
+	if (fault_signal_pending(fault, regs))
 		return 0;
 
 	if (unlikely(fault & VM_FAULT_ERROR)) {

arch/microblaze/mm/fault.c +1 -1

index e6a810b0c7ad..cdde01dcdfc3 100644
@@ -217,7 +217,7 @@ void do_page_fault(struct pt_regs *regs, unsigned long address,
 	 */
 	fault = handle_mm_fault(vma, address, flags);
 
-	if ((fault & VM_FAULT_RETRY) && fatal_signal_pending(current))
+	if (fault_signal_pending(fault, regs))
 		return;
 
 	if (unlikely(fault & VM_FAULT_ERROR)) {

arch/mips/mm/fault.c +1 -1

index 1e8d00793784..0ee6fafc57bc 100644
@@ -154,7 +154,7 @@ static void __kprobes __do_page_fault(struct pt_regs *regs, unsigned long write,
 	 */
 	fault = handle_mm_fault(vma, address, flags);
 
-	if ((fault & VM_FAULT_RETRY) && fatal_signal_pending(current))
+	if (fault_signal_pending(regs))
 		return;
 
 	perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS, 1, regs, address);

arch/nds32/mm/fault.c +1 -1

index 906dfb25353c..0e63f81eff5b 100644
@@ -214,7 +214,7 @@ void do_page_fault(unsigned long entry, unsigned long addr,
 	 * signal first. We do not need to release the mmap_sem because it
 	 * would already be released in __lock_page_or_retry in mm/filemap.c.
 	 */
-	if ((fault & VM_FAULT_RETRY) && fatal_signal_pending(current)) {
+	if (fault_signal_pending(fault, regs)) {
 		if (!user_mode(regs))
 			goto no_context;
 		return;

arch/nios2/mm/fault.c +1 -1

index 6a2e716b959f..704ace8ca0ee 100644
@@ -133,7 +133,7 @@ asmlinkage void do_page_fault(struct pt_regs *regs, unsigned long cause,
 	 */
 	fault = handle_mm_fault(vma, address, flags);
 
-	if ((fault & VM_FAULT_RETRY) && fatal_signal_pending(current))
+	if (fault_signal_pending(fault, regs))
 		return;
 
 	if (unlikely(fault & VM_FAULT_ERROR)) {

arch/openrisc/mm/fault.c +1 -1

index 5d4d3a9691d0..85c7eb0c0186 100644
@@ -161,7 +161,7 @@ asmlinkage void do_page_fault(struct pt_regs *regs, unsigned long address,
 
 	fault = handle_mm_fault(vma, address, flags);
 
-	if ((fault & VM_FAULT_RETRY) && fatal_signal_pending(current))
+	if (fault_signal_pending(fault, regs))
 		return;
 
 	if (unlikely(fault & VM_FAULT_ERROR)) {

arch/parisc/mm/fault.c +1 -1

index adbd5e2144a3..f9be1d1cb43f 100644
@@ -304,7 +304,7 @@ void do_page_fault(struct pt_regs *regs, unsigned long code,
 
 	fault = handle_mm_fault(vma, address, flags);
 
-	if ((fault & VM_FAULT_RETRY) && fatal_signal_pending(current))
+	if (fault_signal_pending(fault, regs))
 		return;
 
 	if (unlikely(fault & VM_FAULT_ERROR)) {

arch/riscv/mm/fault.c +1 -1

index cf7248e07f43..1d3869e9ddef 100644
@@ -117,7 +117,7 @@ asmlinkage void do_page_fault(struct pt_regs *regs)
 	 * signal first. We do not need to release the mmap_sem because it
 	 * would already be released in __lock_page_or_retry in mm/filemap.c.
 	 */
-	if ((fault & VM_FAULT_RETRY) && fatal_signal_pending(tsk))
+	if (fault_signal_pending(fault, regs))
 		return;
 
 	if (unlikely(fault & VM_FAULT_ERROR)) {

arch/s390/mm/fault.c +1 -2

index 7b0bb475c166..179cf92a56e5 100644
@@ -480,8 +480,7 @@ static inline vm_fault_t do_exception(struct pt_regs *regs, int access)
 	 * the fault.
 	 */
 	fault = handle_mm_fault(vma, address, flags);
-	/* No reason to continue if interrupted by SIGKILL. */
-	if ((fault & VM_FAULT_RETRY) && fatal_signal_pending(current)) {
+	if (fault_signal_pending(fault, regs)) {
 		fault = VM_FAULT_SIGNAL;
 		if (flags & FAULT_FLAG_RETRY_NOWAIT)
 			goto out_up;

arch/sparc/mm/fault_32.c +1 -1

index 89976c9b936c..6efbeb227644 100644
@@ -237,7 +237,7 @@ asmlinkage void do_sparc_fault(struct pt_regs *regs, int text_fault, int write,
 	 */
 	fault = handle_mm_fault(vma, address, flags);
 
-	if ((fault & VM_FAULT_RETRY) && fatal_signal_pending(current))
+	if (fault_signal_pending(fault, regs))
 		return;
 
 	if (unlikely(fault & VM_FAULT_ERROR)) {

arch/sparc/mm/fault_64.c +1 -1

index 8b7ddbd14b65..dd1ed6555831 100644
@@ -425,7 +425,7 @@ asmlinkage void __kprobes do_sparc64_fault(struct pt_regs *regs)
 
 	fault = handle_mm_fault(vma, address, flags);
 
-	if ((fault & VM_FAULT_RETRY) && fatal_signal_pending(current))
+	if (fault_signal_pending(fault, regs))
 		goto exit_exception;
 
 	if (unlikely(fault & VM_FAULT_ERROR)) {

arch/unicore32/mm/fault.c +1 -1

index 76342de9cf8c..59d0e6ec2cfc 100644
@@ -250,7 +250,7 @@ static int do_pf(unsigned long addr, unsigned int fsr, struct pt_regs *regs)
 	 * signal first. We do not need to release the mmap_sem because
 	 * it would already be released in __lock_page_or_retry in
 	 * mm/filemap.c. */
-	if ((fault & VM_FAULT_RETRY) && fatal_signal_pending(current))
+	if (fault_signal_pending(fault, regs))
 		return 0;
 
 	if (!(fault & VM_FAULT_ERROR) && (flags & FAULT_FLAG_ALLOW_RETRY)) {

arch/xtensa/mm/fault.c +1 -1

index bee30a77cd70..59515905d4ad 100644
@@ -110,7 +110,7 @@ void do_page_fault(struct pt_regs *regs)
 	 */
 	fault = handle_mm_fault(vma, address, flags);
 
-	if ((fault & VM_FAULT_RETRY) && fatal_signal_pending(current))
+	if (fault_signal_pending(fault, regs))
 		return;
 
 	if (unlikely(fault & VM_FAULT_ERROR)) {

include/linux/sched/signal.h +13 -0

index 88050259c466..4c87ffce64d1 100644
@@ -369,6 +369,19 @@ static inline int signal_pending_state(long state, struct task_struct *p)
 	return (state & TASK_INTERRUPTIBLE) || __fatal_signal_pending(p);
 }
 
+/*
+ * This should only be used in fault handlers to decide whether we
+ * should stop the current fault routine to handle the signals
+ * instead, especially with the case where we've got interrupted with
+ * a VM_FAULT_RETRY.
+ */
+static inline bool fault_signal_pending(unsigned int fault_flags,
+					struct pt_regs *regs)
+{
+	return unlikely((fault_flags & VM_FAULT_RETRY) &&
+			fatal_signal_pending(current));
+}
+
 /*
  * Reevaluate whether the task has signals pending delivery.
  * Wake the task if so.

Signature · 1 line

2.24.1

Peter Xu

author reviewer

2020-02-20 15:01 UTC #4 lore raw

PATCH v6 4/16 x86/mm: Use helper fault_signal_pending()

In reply to Peter Xu

Let's move the fatal signal check even earlier so that we can directly use the new fault_signal_pending() in x86 mm code.

Signed-off-by Peter Xu <[email protected]>

---
 arch/x86/mm/fault.c | 28 +++++++++++++---------------
 1 file changed, 13 insertions(+), 15 deletions(-)

arch/x86/mm/fault.c +13 ~~-15~~

index fa4ea09593ab..6a00bc8d047f 100644
@@ -1442,27 +1442,25 @@ void do_user_addr_fault(struct pt_regs *regs,
 	fault = handle_mm_fault(vma, address, flags);
 	major |= fault & VM_FAULT_MAJOR;
 
+	/* Quick path to respond to signals */
+	if (fault_signal_pending(fault, regs)) {
+		if (!user_mode(regs))
+			no_context(regs, hw_error_code, address, SIGBUS,
+				   BUS_ADRERR);
+		return;
+	}
+
 	/*
 	 * If we need to retry the mmap_sem has already been released,
 	 * and if there is a fatal signal pending there is no guarantee
 	 * that we made any progress. Handle this case first.
 	 */
-	if (unlikely(fault & VM_FAULT_RETRY)) {
+	if (unlikely((fault & VM_FAULT_RETRY) &&
+		     (flags & FAULT_FLAG_ALLOW_RETRY))) {
 		/* Retry at most once */
-		if (flags & FAULT_FLAG_ALLOW_RETRY) {
-			flags &= ~FAULT_FLAG_ALLOW_RETRY;
-			flags |= FAULT_FLAG_TRIED;
-			if (!fatal_signal_pending(tsk))
-				goto retry;
-		}
-
-		/* User mode? Just return to handle the fatal exception */
-		if (flags & FAULT_FLAG_USER)
-			return;
-
-		/* Not returning to user mode? Handle exceptions or die: */
-		no_context(regs, hw_error_code, address, SIGBUS, BUS_ADRERR);
-		return;
+		flags &= ~FAULT_FLAG_ALLOW_RETRY;
+		flags |= FAULT_FLAG_TRIED;
+		goto retry;
 	}
 
 	up_read(&mm->mmap_sem);

Signature · 1 line

2.24.1

Peter Xu

author reviewer

2020-02-20 15:02 UTC #5 lore raw

PATCH v6 4/16 x86/mm: Use helper fault_signal_pending()

In reply to Peter Xu

Let's move the fatal signal check even earlier so that we can directly use the new fault_signal_pending() in x86 mm code.

Signed-off-by Peter Xu <[email protected]>

---
 arch/x86/mm/fault.c | 28 +++++++++++++---------------
 1 file changed, 13 insertions(+), 15 deletions(-)

arch/x86/mm/fault.c +13 ~~-15~~

index fa4ea09593ab..6a00bc8d047f 100644
@@ -1442,27 +1442,25 @@ void do_user_addr_fault(struct pt_regs *regs,
 	fault = handle_mm_fault(vma, address, flags);
 	major |= fault & VM_FAULT_MAJOR;
 
+	/* Quick path to respond to signals */
+	if (fault_signal_pending(fault, regs)) {
+		if (!user_mode(regs))
+			no_context(regs, hw_error_code, address, SIGBUS,
+				   BUS_ADRERR);
+		return;
+	}
+
 	/*
 	 * If we need to retry the mmap_sem has already been released,
 	 * and if there is a fatal signal pending there is no guarantee
 	 * that we made any progress. Handle this case first.
 	 */
-	if (unlikely(fault & VM_FAULT_RETRY)) {
+	if (unlikely((fault & VM_FAULT_RETRY) &&
+		     (flags & FAULT_FLAG_ALLOW_RETRY))) {
 		/* Retry at most once */
-		if (flags & FAULT_FLAG_ALLOW_RETRY) {
-			flags &= ~FAULT_FLAG_ALLOW_RETRY;
-			flags |= FAULT_FLAG_TRIED;
-			if (!fatal_signal_pending(tsk))
-				goto retry;
-		}
-
-		/* User mode? Just return to handle the fatal exception */
-		if (flags & FAULT_FLAG_USER)
-			return;
-
-		/* Not returning to user mode? Handle exceptions or die: */
-		no_context(regs, hw_error_code, address, SIGBUS, BUS_ADRERR);
-		return;
+		flags &= ~FAULT_FLAG_ALLOW_RETRY;
+		flags |= FAULT_FLAG_TRIED;
+		goto retry;
 	}
 
 	up_read(&mm->mmap_sem);

Signature · 1 line

2.24.1

Peter Xu

author reviewer

2020-02-20 15:08 UTC #6 lore raw

PATCH v6 5/16 arc/mm: Use helper fault_signal_pending()

In reply to Peter Xu

Let ARC to use the new helper fault_signal_pending() by moving the signal check out of the retry logic as standalone. This should also helps to simplify the code a bit.

Signed-off-by Peter Xu <[email protected]>

---
 arch/arc/mm/fault.c | 34 +++++++++++++---------------------
 1 file changed, 13 insertions(+), 21 deletions(-)

arch/arc/mm/fault.c +13 ~~-21~~

index fb86bc3e9b35..6eb821a59b49 100644
@@ -133,29 +133,21 @@ void do_page_fault(unsigned long address, struct pt_regs *regs)
 
 	fault = handle_mm_fault(vma, address, flags);
 
+	/* Quick path to respond to signals */
+	if (fault_signal_pending(fault, regs)) {
+		if (!user_mode(regs))
+			goto no_context;
+		return;
+	}
+
 	/*
-	 * Fault retry nuances
+	 * Fault retry nuances, mmap_sem already relinquished by core mm
 	 */
-	if (unlikely(fault & VM_FAULT_RETRY)) {
-
-		/*
-		 * If fault needs to be retried, handle any pending signals
-		 * first (by returning to user mode).
-		 * mmap_sem already relinquished by core mm for RETRY case
-		 */
-		if (fatal_signal_pending(current)) {
-			if (!user_mode(regs))
-				goto no_context;
-			return;
-		}
-		/*
-		 * retry state machine
-		 */
-		if (flags & FAULT_FLAG_ALLOW_RETRY) {
-			flags &= ~FAULT_FLAG_ALLOW_RETRY;
-			flags |= FAULT_FLAG_TRIED;
-			goto retry;
-		}
+	if (unlikely((fault & VM_FAULT_RETRY) &&
+		     (flags & FAULT_FLAG_ALLOW_RETRY))) {
+		flags &= ~FAULT_FLAG_ALLOW_RETRY;
+		flags |= FAULT_FLAG_TRIED;
+		goto retry;
 	}
 
 bad_area:

Signature · 1 line

2.24.1

Peter Xu

author reviewer

2020-02-20 15:11 UTC #7 lore raw

PATCH v6 5/16 arc/mm: Use helper fault_signal_pending()

In reply to Peter Xu

Let ARC to use the new helper fault_signal_pending() by moving the signal check out of the retry logic as standalone. This should also helps to simplify the code a bit.

Signed-off-by Peter Xu <[email protected]>

---
 arch/arc/mm/fault.c | 34 +++++++++++++---------------------
 1 file changed, 13 insertions(+), 21 deletions(-)

arch/arc/mm/fault.c +13 ~~-21~~

index fb86bc3e9b35..6eb821a59b49 100644
@@ -133,29 +133,21 @@ void do_page_fault(unsigned long address, struct pt_regs *regs)
 
 	fault = handle_mm_fault(vma, address, flags);
 
+	/* Quick path to respond to signals */
+	if (fault_signal_pending(fault, regs)) {
+		if (!user_mode(regs))
+			goto no_context;
+		return;
+	}
+
 	/*
-	 * Fault retry nuances
+	 * Fault retry nuances, mmap_sem already relinquished by core mm
 	 */
-	if (unlikely(fault & VM_FAULT_RETRY)) {
-
-		/*
-		 * If fault needs to be retried, handle any pending signals
-		 * first (by returning to user mode).
-		 * mmap_sem already relinquished by core mm for RETRY case
-		 */
-		if (fatal_signal_pending(current)) {
-			if (!user_mode(regs))
-				goto no_context;
-			return;
-		}
-		/*
-		 * retry state machine
-		 */
-		if (flags & FAULT_FLAG_ALLOW_RETRY) {
-			flags &= ~FAULT_FLAG_ALLOW_RETRY;
-			flags |= FAULT_FLAG_TRIED;
-			goto retry;
-		}
+	if (unlikely((fault & VM_FAULT_RETRY) &&
+		     (flags & FAULT_FLAG_ALLOW_RETRY))) {
+		flags &= ~FAULT_FLAG_ALLOW_RETRY;
+		flags |= FAULT_FLAG_TRIED;
+		goto retry;
 	}
 
 bad_area:

Signature · 1 line

2.24.1

Peter Xu

author reviewer

2020-02-20 15:19 UTC #8 lore raw

In reply to Peter Xu

Quoted 3 lines · Peter Xu

This is v6 of the series. It is majorly a rebase to 5.6-rc2, nothing else to be expected (plus some tests after the rebase). Instead of rewrite the cover letter I decided to use what we have for v5.

My mail server fails again for me during the middle of sending this series. I wanted to fix this manually but it could have made it even a bit more messed up... I'll try to fix this all and resend the series. Sorry. Please ignore this whole thread for now.

There is also another standalone patch:

Subject: [PATCH v6 04/16] x86/mm: Use helper fault_signal_pending()
Message-Id: <[email protected]>

Please ignore that too; that's also some fallout of this misery...

Thanks,

Signature · 1 line

Peter Xu