[PATCH 3/3] arm64/kprobe: Optimize the performance of patching single-step slot
Catalin Marinas
catalin.marinas at arm.com
Fri Sep 30 01:15:48 PDT 2022
On Fri, Sep 30, 2022 at 09:02:20AM +0800, liaochang (A) wrote:
>
>
> 在 2022/9/30 0:50, Catalin Marinas 写道:
> > On Tue, Sep 27, 2022 at 10:24:35AM +0800, Liao Chang wrote:
> >> Single-step slot would not be used until kprobe is enabled, that means
> >> no race condition occurs on it under SMP, hence it is safe to pacth ss
> >> slot without stopping machine.
> >>
> >> Since I and D caches are coherent within single-step slot from
> >> aarch64_insn_patch_text_nosync(), hence no need to do it again via
> >> flush_icache_range().
> >>
> >> Acked-by: Will Deacon <will at kernel.org>
> >> Acked-by: Masami Hiramatsu (Google) <mhiramat at kernel.org>
> >> Signed-off-by: Liao Chang <liaochang1 at huawei.com>
> >> ---
> >> arch/arm64/kernel/probes/kprobes.c | 27 +++++++++++++++++++++------
> >> 1 file changed, 21 insertions(+), 6 deletions(-)
> >
> > What's your expectation with this series, should the arch maintainers
> > just pick the individual patches?
>
> Yes, or should i split this series into individual patch?
No need to, I can pick the arm64 patch. If the other maintainers don't
merge the patches, you might want to post them again individually as to
avoid confusion.
--
Catalin
More information about the linux-riscv
mailing list