riscv: mm: fix regression due to update_mmu_cache change
Subsystems (2)
PATCH 1/1 riscv: mm: fix regression due to update_mmu_cache change
From: Sergey Matyukevich <[email protected]>
This is a partial revert of the commit 4bd1d80efb5a ("riscv: mm: notify remote harts about mmu cache updates"). Original commit included two loosely related changes serving the same purpose of fixing stale TLB entries causing user-space application crash:
- introduce deferred per-ASID TLB flush for CPUs not running the task
- switch to per-ASID TLB flush on all CPUs running the task in update_mmu_cache
According to report and discussion in [1], the second part caused a regression on Renesas RZ/Five SoC. For now restore the old behavior of the update_mmu_cache.
[1] https://lore.kernel.org/linux-riscv/[email protected]/
--- arch/riscv/include/asm/pgtable.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
arch/riscv/include/asm/pgtable.h +1 -1
@@ -415,7 +415,7 @@ static inline void update_mmu_cache(struct vm_area_struct *vma,* Relying on flush_tlb_fix_spurious_fault would suffice, but * the extra traps reduce performance. So, eagerly SFENCE.VMA. */ -flush_tlb_page (vma, address); +local_flush_tlb_page (address); }#define __HAVE_ARCH_UPDATE_MMU_TLB
Signature · 1 line
2.39.0
Quoted 9 lines · Sergey Matyukevich
From: Sergey Matyukevich <[email protected]>
This is a partial revert of the commit
4bd1d80efb5a("riscv: mm: notify remote harts about mmu cache updates"). Original commit included two loosely related changes serving the same purpose of fixing stale TLB entries causing user-space application crash:
- introduce deferred per-ASID TLB flush for CPUs not running the task
- switch to per-ASID TLB flush on all CPUs running the task inupdate_mmu_cache
Quoted 5 lines
According to report and discussion in [1], the second part caused a regression on Renesas RZ/Five SoC. For now restore the old behavior of the
update_mmu_cache.
If you respin for another reason, can you convert this into a "regular"
Link: trailer, so that it can be parsed with git's trailer functionality? IOW, like so:
Otherwise, glad to see you two get this sorted out, even if it is just a revert.
Thanks,
Conor.
Quoted 23 lines
Fixes 4bd1d80efb5a ("riscv: mm: notify remote harts about mmu cache updates")Reported-by "Lad, Prabhakar" <[email protected]>Signed-off-by Sergey Matyukevich <[email protected]>--- arch/riscv/include/asm/pgtable.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)arch/riscv/include/asm/pgtable.h +1
-1@@ -415,7 +415,7 @@ static inline void update_mmu_cache(struct vm_area_struct *vma,* Relying on flush_tlb_fix_spurious_fault would suffice, but * the extra traps reduce performance. So, eagerly SFENCE.VMA. */ -flush_tlb_page (vma, address); +local_flush_tlb_page (address); }#define __HAVE_ARCH_UPDATE_MMU_TLBSignature · 1 line
2.39.0
This patch was applied to riscv/linux.git (for-next) by Palmer Dabbelt <[email protected]>:
Hello:
This patch was applied to riscv/linux.git (for-next) by Palmer Dabbelt <[email protected]>:
Quoted 10 lines · On Mon, 30 Jan 2023 00:18:18 +0300 you wrote:
From: Sergey Matyukevich <[email protected]>
This is a partial revert of the commit
4bd1d80efb5a("riscv: mm: notify remote harts about mmu cache updates"). Original commit included two loosely related changes serving the same purpose of fixing stale TLB entries causing user-space application crash:
- introduce deferred per-ASID TLB flush for CPUs not running the task
- switch to per-ASID TLB flush on all CPUs running the task inupdate_mmu_cache[...]
Here is the summary with links:
- riscv: mm: fix regression due to update_mmu_cache change
https://git.kernel.org/riscv/c/b49f700668ff
You are awesome, thank you!
Signature · 2 lines
Deet-doot-dot, I am a bot. https://korg.docs.kernel.org/patchwork/pwbot.html