| Message ID | 20180823084709.19717-3-npiggin@gmail.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | minor mmu_gather patches | expand |
On Thu, Aug 23, 2018 at 06:47:09PM +1000, Nicholas Piggin wrote: > The generic tlb_end_vma does not call invalidate_range mmu notifier, > and it resets resets the mmu_gather range, which means the notifier > won't be called on part of the range in case of an unmap that spans > multiple vmas. > > ARM64 seems to be the only arch I could see that has notifiers and > uses the generic tlb_end_vma. I have not actually tested it. We only care about notifiers for KVM but I think it only makes use of mmu_notifier_invalidate_range_(start|end) which are not affected by the range reset in mmu_gather. Your patch looks ok from an arm64 perspective (it would be good if Will has a look as well since he was the last to touch this part for arm64).
On Thu, Aug 23, 2018 at 06:47:09PM +1000, Nicholas Piggin wrote: > The generic tlb_end_vma does not call invalidate_range mmu notifier, > and it resets resets the mmu_gather range, which means the notifier > won't be called on part of the range in case of an unmap that spans > multiple vmas. > > ARM64 seems to be the only arch I could see that has notifiers and > uses the generic tlb_end_vma. I have not actually tested it. > > Signed-off-by: Nicholas Piggin <npiggin@gmail.com> > --- > include/asm-generic/tlb.h | 17 +++++++++++++---- > mm/memory.c | 10 ---------- > 2 files changed, 13 insertions(+), 14 deletions(-) I think we only use the notifiers in the KVM code, which appears to leave the ->invalidate_range() callback empty, so that at least explains why we haven't run into problems here. But the change looks correct to me, so: Acked-by: Will Deacon <will.deacon@arm.com> Thanks, Will
On Thu, Aug 23, 2018 at 06:47:09PM +1000, Nicholas Piggin wrote: > The generic tlb_end_vma does not call invalidate_range mmu notifier, > and it resets resets the mmu_gather range, which means the notifier > won't be called on part of the range in case of an unmap that spans > multiple vmas. > > ARM64 seems to be the only arch I could see that has notifiers and > uses the generic tlb_end_vma. I have not actually tested it. > > Signed-off-by: Nicholas Piggin <npiggin@gmail.com> > Acked-by: Will Deacon <will.deacon@arm.com> This patch breaks riscv builds in mainline. Building riscv:defconfig ... failed -------------- Error log: In file included from riscv/include/asm/tlb.h:17:0, from arch/riscv/include/asm/pgalloc.h:19, from riscv/mm/fault.c:30: include/asm-generic/tlb.h: In function 'tlb_flush_mmu_tlbonly': include/asm-generic/tlb.h:147:2: error: implicit declaration of function 'tlb_flush' In file included from arch/riscv/include/asm/pgalloc.h:19:0, from arch/riscv/mm/fault.c:30: arch/riscv/include/asm/tlb.h: At top level: arch/riscv/include/asm/tlb.h:19:20: warning: conflicting types for 'tlb_flush' Guenter
On Fri, Aug 24, 2018 at 06:07:22AM -0700, Guenter Roeck wrote: > On Thu, Aug 23, 2018 at 06:47:09PM +1000, Nicholas Piggin wrote: > > The generic tlb_end_vma does not call invalidate_range mmu notifier, > > and it resets resets the mmu_gather range, which means the notifier > > won't be called on part of the range in case of an unmap that spans > > multiple vmas. > > > > ARM64 seems to be the only arch I could see that has notifiers and > > uses the generic tlb_end_vma. I have not actually tested it. > > > > Signed-off-by: Nicholas Piggin <npiggin@gmail.com> > > Acked-by: Will Deacon <will.deacon@arm.com> > > This patch breaks riscv builds in mainline. Looks very similar to the breakage we hit on arm64. diff below should fix it. Will --->* diff --git a/arch/riscv/include/asm/tlb.h b/arch/riscv/include/asm/tlb.h index c229509288ea..5017060be63c 100644 --- a/arch/riscv/include/asm/tlb.h +++ b/arch/riscv/include/asm/tlb.h @@ -14,11 +14,11 @@ #ifndef _ASM_RISCV_TLB_H #define _ASM_RISCV_TLB_H -#include <asm-generic/tlb.h> - static inline void tlb_flush(struct mmu_gather *tlb) { flush_tlb_mm(tlb->mm); } +#include <asm-generic/tlb.h> + #endif /* _ASM_RISCV_TLB_H */
On Fri, Aug 24, 2018 at 02:10:27PM +0100, Will Deacon wrote: > On Fri, Aug 24, 2018 at 06:07:22AM -0700, Guenter Roeck wrote: > > On Thu, Aug 23, 2018 at 06:47:09PM +1000, Nicholas Piggin wrote: > > > The generic tlb_end_vma does not call invalidate_range mmu notifier, > > > and it resets resets the mmu_gather range, which means the notifier > > > won't be called on part of the range in case of an unmap that spans > > > multiple vmas. > > > > > > ARM64 seems to be the only arch I could see that has notifiers and > > > uses the generic tlb_end_vma. I have not actually tested it. > > > > > > Signed-off-by: Nicholas Piggin <npiggin@gmail.com> > > > Acked-by: Will Deacon <will.deacon@arm.com> > > > > This patch breaks riscv builds in mainline. > > Looks very similar to the breakage we hit on arm64. diff below should fix > it. > Unfortunately it doesn't. In file included from ./arch/riscv/include/asm/pgtable.h:26:0, from ./include/linux/memremap.h:7, from ./include/linux/mm.h:27, from arch/riscv/mm/fault.c:23: ./arch/riscv/include/asm/tlb.h: In function ‘tlb_flush’: ./arch/riscv/include/asm/tlb.h:19:18: error: dereferencing pointer to incomplete type ‘struct mmu_gather’ flush_tlb_mm(tlb->mm); ^ ./arch/riscv/include/asm/tlbflush.h:58:35: note: in definition of macro ‘flush_tlb_mm’ sbi_remote_sfence_vma(mm_cpumask(mm)->bits, 0, -1) Note that reverting the offending patch does fix the problem, so there is no secondary problem lurking around. Guenter
On Fri, Aug 24, 2018 at 06:24:19AM -0700, Guenter Roeck wrote: > On Fri, Aug 24, 2018 at 02:10:27PM +0100, Will Deacon wrote: > > On Fri, Aug 24, 2018 at 06:07:22AM -0700, Guenter Roeck wrote: > > > On Thu, Aug 23, 2018 at 06:47:09PM +1000, Nicholas Piggin wrote: > > > > The generic tlb_end_vma does not call invalidate_range mmu notifier, > > > > and it resets resets the mmu_gather range, which means the notifier > > > > won't be called on part of the range in case of an unmap that spans > > > > multiple vmas. > > > > > > > > ARM64 seems to be the only arch I could see that has notifiers and > > > > uses the generic tlb_end_vma. I have not actually tested it. > > > > > > > > Signed-off-by: Nicholas Piggin <npiggin@gmail.com> > > > > Acked-by: Will Deacon <will.deacon@arm.com> > > > > > > This patch breaks riscv builds in mainline. > > > > Looks very similar to the breakage we hit on arm64. diff below should fix > > it. > > > > Unfortunately it doesn't. > > In file included from ./arch/riscv/include/asm/pgtable.h:26:0, > from ./include/linux/memremap.h:7, > from ./include/linux/mm.h:27, > from arch/riscv/mm/fault.c:23: > ./arch/riscv/include/asm/tlb.h: In function ‘tlb_flush’: > ./arch/riscv/include/asm/tlb.h:19:18: error: dereferencing pointer to incomplete type ‘struct mmu_gather’ > flush_tlb_mm(tlb->mm); > ^ Sorry, I was a bit quick of the mark there. You'll need a forward declaration for the paramater type. Here it is with a commit message, although still untested because I haven't got round to setting up a riscv toolchain yet. Will --->8 From adb9be33d68320edcda80d540a97a647792894d2 Mon Sep 17 00:00:00 2001 From: Will Deacon <will.deacon@arm.com> Date: Fri, 24 Aug 2018 14:33:48 +0100 Subject: [PATCH] riscv: tlb: Provide definition of tlb_flush() before including tlb.h As of commit fd1102f0aade ("mm: mmu_notifier fix for tlb_end_vma"), asm-generic/tlb.h now calls tlb_flush() from a static inline function, so we need to make sure that it's declared before #including the asm-generic header in the arch header. Since tlb_flush() is a one-liner for riscv, we can define it before including asm-generic/tlb.h as long as we provide a forward declaration of struct mmu_gather. Reported-by: Guenter Roeck <linux@roeck-us.net> Signed-off-by: Will Deacon <will.deacon@arm.com> --- arch/riscv/include/asm/tlb.h | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/arch/riscv/include/asm/tlb.h b/arch/riscv/include/asm/tlb.h index c229509288ea..a3d1380ad970 100644 --- a/arch/riscv/include/asm/tlb.h +++ b/arch/riscv/include/asm/tlb.h @@ -14,11 +14,13 @@ #ifndef _ASM_RISCV_TLB_H #define _ASM_RISCV_TLB_H -#include <asm-generic/tlb.h> +struct mmu_gather; static inline void tlb_flush(struct mmu_gather *tlb) { flush_tlb_mm(tlb->mm); } +#include <asm-generic/tlb.h> + #endif /* _ASM_RISCV_TLB_H */
On Fri, Aug 24, 2018 at 02:34:27PM +0100, Will Deacon wrote: > On Fri, Aug 24, 2018 at 06:24:19AM -0700, Guenter Roeck wrote: > > On Fri, Aug 24, 2018 at 02:10:27PM +0100, Will Deacon wrote: > > > On Fri, Aug 24, 2018 at 06:07:22AM -0700, Guenter Roeck wrote: > > > > On Thu, Aug 23, 2018 at 06:47:09PM +1000, Nicholas Piggin wrote: > > > > > The generic tlb_end_vma does not call invalidate_range mmu notifier, > > > > > and it resets resets the mmu_gather range, which means the notifier > > > > > won't be called on part of the range in case of an unmap that spans > > > > > multiple vmas. > > > > > > > > > > ARM64 seems to be the only arch I could see that has notifiers and > > > > > uses the generic tlb_end_vma. I have not actually tested it. > > > > > > > > > > Signed-off-by: Nicholas Piggin <npiggin@gmail.com> > > > > > Acked-by: Will Deacon <will.deacon@arm.com> > > > > > > > > This patch breaks riscv builds in mainline. > > > > > > Looks very similar to the breakage we hit on arm64. diff below should fix > > > it. > > > > > > > Unfortunately it doesn't. > > > > In file included from ./arch/riscv/include/asm/pgtable.h:26:0, > > from ./include/linux/memremap.h:7, > > from ./include/linux/mm.h:27, > > from arch/riscv/mm/fault.c:23: > > ./arch/riscv/include/asm/tlb.h: In function ‘tlb_flush’: > > ./arch/riscv/include/asm/tlb.h:19:18: error: dereferencing pointer to incomplete type ‘struct mmu_gather’ > > flush_tlb_mm(tlb->mm); > > ^ > > Sorry, I was a bit quick of the mark there. You'll need a forward > declaration for the paramater type. Here it is with a commit message, > although still untested because I haven't got round to setting up a riscv > toolchain yet. > > Will > > --->8 > > From adb9be33d68320edcda80d540a97a647792894d2 Mon Sep 17 00:00:00 2001 > From: Will Deacon <will.deacon@arm.com> > Date: Fri, 24 Aug 2018 14:33:48 +0100 > Subject: [PATCH] riscv: tlb: Provide definition of tlb_flush() before > including tlb.h > > As of commit fd1102f0aade ("mm: mmu_notifier fix for tlb_end_vma"), > asm-generic/tlb.h now calls tlb_flush() from a static inline function, > so we need to make sure that it's declared before #including the > asm-generic header in the arch header. > > Since tlb_flush() is a one-liner for riscv, we can define it before > including asm-generic/tlb.h as long as we provide a forward declaration > of struct mmu_gather. > > Reported-by: Guenter Roeck <linux@roeck-us.net> > Signed-off-by: Will Deacon <will.deacon@arm.com> > --- > arch/riscv/include/asm/tlb.h | 4 +++- > 1 file changed, 3 insertions(+), 1 deletion(-) > > diff --git a/arch/riscv/include/asm/tlb.h b/arch/riscv/include/asm/tlb.h > index c229509288ea..a3d1380ad970 100644 > --- a/arch/riscv/include/asm/tlb.h > +++ b/arch/riscv/include/asm/tlb.h > @@ -14,11 +14,13 @@ > #ifndef _ASM_RISCV_TLB_H > #define _ASM_RISCV_TLB_H > > -#include <asm-generic/tlb.h> > +struct mmu_gather; > > static inline void tlb_flush(struct mmu_gather *tlb) > { > flush_tlb_mm(tlb->mm); Bah, didn't spot the dereference so this won't work either. You basically just need to copy what I did for arm64 in d475fac95779. Will
On Fri, Aug 24, 2018 at 02:50:48PM +0100, Will Deacon wrote: > > > > Sorry, I was a bit quick of the mark there. You'll need a forward > > declaration for the paramater type. Here it is with a commit message, > > although still untested because I haven't got round to setting up a riscv > > toolchain yet. > > Still doesn't work. CC mm/memory.o In file included from ./arch/riscv/include/asm/pgtable.h:26:0, from ./include/linux/memremap.h:7, from ./include/linux/mm.h:27, from arch/riscv/mm/fault.c:23: ./arch/riscv/include/asm/tlb.h: In function ‘tlb_flush’: ./arch/riscv/include/asm/tlb.h:21:18: error: dereferencing pointer to incomplete type ‘struct mmu_gather’ flush_tlb_mm(tlb->mm); Problem is that struct mmu_gather is dereferenced in tlb_flush(). Guenter > > Will > > > > --->8 > > > > From adb9be33d68320edcda80d540a97a647792894d2 Mon Sep 17 00:00:00 2001 > > From: Will Deacon <will.deacon@arm.com> > > Date: Fri, 24 Aug 2018 14:33:48 +0100 > > Subject: [PATCH] riscv: tlb: Provide definition of tlb_flush() before > > including tlb.h > > > > As of commit fd1102f0aade ("mm: mmu_notifier fix for tlb_end_vma"), > > asm-generic/tlb.h now calls tlb_flush() from a static inline function, > > so we need to make sure that it's declared before #including the > > asm-generic header in the arch header. > > > > Since tlb_flush() is a one-liner for riscv, we can define it before > > including asm-generic/tlb.h as long as we provide a forward declaration > > of struct mmu_gather. > > > > Reported-by: Guenter Roeck <linux@roeck-us.net> > > Signed-off-by: Will Deacon <will.deacon@arm.com> > > --- > > arch/riscv/include/asm/tlb.h | 4 +++- > > 1 file changed, 3 insertions(+), 1 deletion(-) > > > > diff --git a/arch/riscv/include/asm/tlb.h b/arch/riscv/include/asm/tlb.h > > index c229509288ea..a3d1380ad970 100644 > > --- a/arch/riscv/include/asm/tlb.h > > +++ b/arch/riscv/include/asm/tlb.h > > @@ -14,11 +14,13 @@ > > #ifndef _ASM_RISCV_TLB_H > > #define _ASM_RISCV_TLB_H > > > > -#include <asm-generic/tlb.h> > > +struct mmu_gather; > > > > static inline void tlb_flush(struct mmu_gather *tlb) > > { > > flush_tlb_mm(tlb->mm); > > Bah, didn't spot the dereference so this won't work either. You basically > just need to copy what I did for arm64 in d475fac95779. > > Will
On 08/24/2018 06:50 AM, Will Deacon wrote: >> -#include <asm-generic/tlb.h> >> +struct mmu_gather; >> >> static inline void tlb_flush(struct mmu_gather *tlb) >> { >> flush_tlb_mm(tlb->mm); > > Bah, didn't spot the dereference so this won't work either. You basically > just need to copy what I did for arm64 in d475fac95779. > Yes, this seems to work. It doesn't really need "struct mmu_gather;" - I assume that is included from elsewhere - but I added it to be safe. Can you send a full patch, or do you want me to do it ? Thanks, Guenter --- diff --git a/arch/riscv/include/asm/tlb.h b/arch/riscv/include/asm/tlb.h index c229509288ea..439dc7072e05 100644 --- a/arch/riscv/include/asm/tlb.h +++ b/arch/riscv/include/asm/tlb.h @@ -14,6 +14,10 @@ #ifndef _ASM_RISCV_TLB_H #define _ASM_RISCV_TLB_H +struct mmu_gather; + +static void tlb_flush(struct mmu_gather *tlb); + #include <asm-generic/tlb.h> static inline void tlb_flush(struct mmu_gather *tlb)
On Fri, Aug 24, 2018 at 07:06:51AM -0700, Guenter Roeck wrote: > On 08/24/2018 06:50 AM, Will Deacon wrote: > > >>-#include <asm-generic/tlb.h> > >>+struct mmu_gather; > >> static inline void tlb_flush(struct mmu_gather *tlb) > >> { > >> flush_tlb_mm(tlb->mm); > > > >Bah, didn't spot the dereference so this won't work either. You basically > >just need to copy what I did for arm64 in d475fac95779. > > > > Yes, this seems to work. It doesn't really need "struct mmu_gather;" - > I assume that is included from elsewhere - but I added it to be safe. struct mmu_gather comes in via asm-generic/tlb.h. > Can you send a full patch, or do you want me to do it ? I'm evidently incapable of writing code today, so please go ahead :) Will
On Fri, Aug 24, 2018 at 03:25:33PM +0100, Will Deacon wrote: > On Fri, Aug 24, 2018 at 07:06:51AM -0700, Guenter Roeck wrote: > > On 08/24/2018 06:50 AM, Will Deacon wrote: > > > > >>-#include <asm-generic/tlb.h> > > >>+struct mmu_gather; > > >> static inline void tlb_flush(struct mmu_gather *tlb) > > >> { > > >> flush_tlb_mm(tlb->mm); > > > > > >Bah, didn't spot the dereference so this won't work either. You basically > > >just need to copy what I did for arm64 in d475fac95779. > > > > > > > Yes, this seems to work. It doesn't really need "struct mmu_gather;" - > > I assume that is included from elsewhere - but I added it to be safe. > > struct mmu_gather comes in via asm-generic/tlb.h. > From linux/mm.h, really, which happens to be included before asm/tlb.h is included (see arch/riscv/include/asm/pgalloc.h). I kept it to be future-proof. > > Can you send a full patch, or do you want me to do it ? > > I'm evidently incapable of writing code today, so please go ahead :) > Done. We'll see if I am any better. Guenter
On Fri, 24 Aug 2018 06:50:48 PDT (-0700), Will Deacon wrote: > On Fri, Aug 24, 2018 at 02:34:27PM +0100, Will Deacon wrote: >> On Fri, Aug 24, 2018 at 06:24:19AM -0700, Guenter Roeck wrote: >> > On Fri, Aug 24, 2018 at 02:10:27PM +0100, Will Deacon wrote: >> > > On Fri, Aug 24, 2018 at 06:07:22AM -0700, Guenter Roeck wrote: >> > > > On Thu, Aug 23, 2018 at 06:47:09PM +1000, Nicholas Piggin wrote: >> > > > > The generic tlb_end_vma does not call invalidate_range mmu notifier, >> > > > > and it resets resets the mmu_gather range, which means the notifier >> > > > > won't be called on part of the range in case of an unmap that spans >> > > > > multiple vmas. >> > > > > >> > > > > ARM64 seems to be the only arch I could see that has notifiers and >> > > > > uses the generic tlb_end_vma. I have not actually tested it. >> > > > > >> > > > > Signed-off-by: Nicholas Piggin <npiggin@gmail.com> >> > > > > Acked-by: Will Deacon <will.deacon@arm.com> >> > > > >> > > > This patch breaks riscv builds in mainline. >> > > >> > > Looks very similar to the breakage we hit on arm64. diff below should fix >> > > it. >> > > >> > >> > Unfortunately it doesn't. >> > >> > In file included from ./arch/riscv/include/asm/pgtable.h:26:0, >> > from ./include/linux/memremap.h:7, >> > from ./include/linux/mm.h:27, >> > from arch/riscv/mm/fault.c:23: >> > ./arch/riscv/include/asm/tlb.h: In function ‘tlb_flush’: >> > ./arch/riscv/include/asm/tlb.h:19:18: error: dereferencing pointer to incomplete type ‘struct mmu_gather’ >> > flush_tlb_mm(tlb->mm); >> > ^ >> >> Sorry, I was a bit quick of the mark there. You'll need a forward >> declaration for the paramater type. Here it is with a commit message, >> although still untested because I haven't got round to setting up a riscv >> toolchain yet. FWIW, Arnd built them last time he updated the cross tools so you should be able to get GCC 8.1.0 for RISC-V from there. I use this make.cross script that I stole from the Intel 0-day robot https://github.com/palmer-dabbelt/home/blob/master/.local/src/local-scripts/make.cross.bash If I'm reading it correctly the tools come from here http://cdn.kernel.org/pub/tools/crosstool/files/bin/x86_64/8.1.0/x86_64-gcc-8.1.0-nolibc-riscv64-linux.tar.gz I use the make.cross script as it makes it super easy to test my across-the-tree patches on other people's ports. >> >> Will >> >> --->8 >> >> From adb9be33d68320edcda80d540a97a647792894d2 Mon Sep 17 00:00:00 2001 >> From: Will Deacon <will.deacon@arm.com> >> Date: Fri, 24 Aug 2018 14:33:48 +0100 >> Subject: [PATCH] riscv: tlb: Provide definition of tlb_flush() before >> including tlb.h >> >> As of commit fd1102f0aade ("mm: mmu_notifier fix for tlb_end_vma"), >> asm-generic/tlb.h now calls tlb_flush() from a static inline function, >> so we need to make sure that it's declared before #including the >> asm-generic header in the arch header. >> >> Since tlb_flush() is a one-liner for riscv, we can define it before >> including asm-generic/tlb.h as long as we provide a forward declaration >> of struct mmu_gather. >> >> Reported-by: Guenter Roeck <linux@roeck-us.net> >> Signed-off-by: Will Deacon <will.deacon@arm.com> >> --- >> arch/riscv/include/asm/tlb.h | 4 +++- >> 1 file changed, 3 insertions(+), 1 deletion(-) >> >> diff --git a/arch/riscv/include/asm/tlb.h b/arch/riscv/include/asm/tlb.h >> index c229509288ea..a3d1380ad970 100644 >> --- a/arch/riscv/include/asm/tlb.h >> +++ b/arch/riscv/include/asm/tlb.h >> @@ -14,11 +14,13 @@ >> #ifndef _ASM_RISCV_TLB_H >> #define _ASM_RISCV_TLB_H >> >> -#include <asm-generic/tlb.h> >> +struct mmu_gather; >> >> static inline void tlb_flush(struct mmu_gather *tlb) >> { >> flush_tlb_mm(tlb->mm); > > Bah, didn't spot the dereference so this won't work either. You basically > just need to copy what I did for arm64 in d475fac95779. > > Will
diff --git a/include/asm-generic/tlb.h b/include/asm-generic/tlb.h index e811ef7b8350..79cb76b89c26 100644 --- a/include/asm-generic/tlb.h +++ b/include/asm-generic/tlb.h @@ -15,6 +15,7 @@ #ifndef _ASM_GENERIC__TLB_H #define _ASM_GENERIC__TLB_H +#include <linux/mmu_notifier.h> #include <linux/swap.h> #include <asm/pgalloc.h> #include <asm/tlbflush.h> @@ -138,6 +139,16 @@ static inline void __tlb_reset_range(struct mmu_gather *tlb) } } +static inline void tlb_flush_mmu_tlbonly(struct mmu_gather *tlb) +{ + if (!tlb->end) + return; + + tlb_flush(tlb); + mmu_notifier_invalidate_range(tlb->mm, tlb->start, tlb->end); + __tlb_reset_range(tlb); +} + static inline void tlb_remove_page_size(struct mmu_gather *tlb, struct page *page, int page_size) { @@ -186,10 +197,8 @@ static inline void tlb_remove_check_page_size_change(struct mmu_gather *tlb, #define __tlb_end_vma(tlb, vma) \ do { \ - if (!tlb->fullmm && tlb->end) { \ - tlb_flush(tlb); \ - __tlb_reset_range(tlb); \ - } \ + if (!tlb->fullmm) \ + tlb_flush_mmu_tlbonly(tlb); \ } while (0) #ifndef tlb_end_vma diff --git a/mm/memory.c b/mm/memory.c index 7c58310734eb..c4a7b6cb17c8 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -238,16 +238,6 @@ void arch_tlb_gather_mmu(struct mmu_gather *tlb, struct mm_struct *mm, __tlb_reset_range(tlb); } -static void tlb_flush_mmu_tlbonly(struct mmu_gather *tlb) -{ - if (!tlb->end) - return; - - tlb_flush(tlb); - mmu_notifier_invalidate_range(tlb->mm, tlb->start, tlb->end); - __tlb_reset_range(tlb); -} - static void tlb_flush_mmu_free(struct mmu_gather *tlb) { struct mmu_gather_batch *batch;
The generic tlb_end_vma does not call invalidate_range mmu notifier, and it resets resets the mmu_gather range, which means the notifier won't be called on part of the range in case of an unmap that spans multiple vmas. ARM64 seems to be the only arch I could see that has notifiers and uses the generic tlb_end_vma. I have not actually tested it. Signed-off-by: Nicholas Piggin <npiggin@gmail.com> --- include/asm-generic/tlb.h | 17 +++++++++++++---- mm/memory.c | 10 ---------- 2 files changed, 13 insertions(+), 14 deletions(-)