Re: [PATCH] perf tools riscv: Add support for riscv lookup_binutils_path
From: Ian Rogers
Date: Mon Mar 13 2023 - 12:28:26 EST
On Sun, Mar 12, 2023 at 5:53 AM Paran Lee <p4ranlee@xxxxxxxxx> wrote:
>
>
>
> 23. 3. 12. 15:27에 Ian Rogers 이(가) 쓴 글:
> > On Sat, Mar 11, 2023 at 3:22 AM paranlee <p4ranlee@xxxxxxxxx> wrote:
> >>
> >> Add to know RISC-V binutils path.
> >> Secondarily, edit the code block with alphabetical order.
> >>
> >> Signed-off-by: Paran Lee <p4ranlee@xxxxxxxxx>
> >> ---
> >> tools/perf/arch/common.c | 51 +++++++++++++++++++++++++++-------------
> >> 1 file changed, 35 insertions(+), 16 deletions(-)
> >>
> >> diff --git a/tools/perf/arch/common.c b/tools/perf/arch/common.c
> >> index 59dd875fd5e4..058527ededdd 100644
> >> --- a/tools/perf/arch/common.c
> >> +++ b/tools/perf/arch/common.c
> >> @@ -29,11 +29,23 @@ const char *const arm_triplets[] = {
> >> };
> >>
> >> const char *const arm64_triplets[] = {
> >> + "aarch64-unknown-linux-",
> >
> > Modifying ARM64 behavior should be a separate change.
> >
> >> "aarch64-linux-android-",
> >> "aarch64-linux-gnu-",
> >> NULL
> >> };
> >>
> >> +const char *const mips_triplets[] = {
> >> + "mips-unknown-linux-gnu-",
> >> + "mipsel-linux-android-",
> >> + "mips-linux-gnu-",
> >> + "mips64-linux-gnu-",
> >> + "mips64el-linux-gnuabi64-",
> >> + "mips64-linux-gnuabi64-",
> >> + "mipsel-linux-gnu-",
> >> + NULL
> >> +};
> >> +
> >
> > This will affect the blame history. It should probably be its own change too.
>
> Thank you for review! I agree. So I would split the patch.
>
> >> -
> >> static bool lookup_path(char *name)
> >> {
> >> bool found = false;
> >> @@ -164,18 +179,22 @@ static int perf_env__lookup_binutils_path(struct perf_env *env,
> >> path_list = arm_triplets;
> >> else if (!strcmp(arch, "arm64"))
> >> path_list = arm64_triplets;
> >> + else if (!strcmp(arch, "mips"))
> >> + path_list = mips_triplets;
> >> else if (!strcmp(arch, "powerpc"))
> >> path_list = powerpc_triplets;
> >> - else if (!strcmp(arch, "sh"))
> >> - path_list = sh_triplets;
> >> + else if (!strcmp(arch, "riscv32"))
> >> + path_list = riscv32_triplets;
> >> + else if (!strcmp(arch, "riscv64"))
> >> + path_list = riscv64_triplets;
> >> else if (!strcmp(arch, "s390"))
> >> - path_list = s390_triplets;
> >> + path_list = s390_triplets;
> >
> > whitespace issue?
>
> I tried to correct the alphabetical order because it was vaguely sorted.
> And I'll try to work on blame history on each arch code block as well.
>
> >> + else if (!strcmp(arch, "sh"))
> >> + path_list = sh_triplets;
> >> else if (!strcmp(arch, "sparc"))
> >> path_list = sparc_triplets;
> >> else if (!strcmp(arch, "x86"))
> >> path_list = x86_triplets;
> >> - else if (!strcmp(arch, "mips"))
> >> - path_list = mips_triplets;
> >> else {
> >> ui__error("binutils for %s not supported.\n", arch);
> >> goto out_error;
> >
> > I think in general we need to revamp this code. Two things that I see
> > that are missing are (1) support for perf config and (2) addr2line
> > should be configurable, you may want llvm-addr2line. Adding RISC-V is
> > of course important too :-)
> >
> > Thanks,
> > Ian
>
> May I ask documentation or hint that I can help work with?
>
> P.S.
>
> I'm interested in the Google Summer Of code perf category this year,
> especially the part about risc-v architecture, I recently purchased a
> development board and would like to be able to test perf on a Sifive U74
> CPU based environment.
> But I've only used perf with command tool and don't know much about the
> internals, so if there is a roadmap for perf development or
> contribution, I have interest in perf internals both kernel and user side.
> May I ask information to apply?
> I am developing Linux Security Driver drivers for a security company.
>
> BR
> Paran Lee
Hi Paran,
Thanks for being interested in GSoC with Linux perf. Here is what I
posted on the mailing list:
https://lore.kernel.org/linux-perf-users/CAP-5=fWxF6in4vQyGuh=0kpAYEXAYZN_KobXCY=TX2oxssZ+HQ@xxxxxxxxxxxxxx/
Applications are ultimately sent to:
https://summerofcode.withgoogle.com/
and the entry requirements are there. I believe they are less strict
than previously.
Wrt the patch, could you reply to Conor's response.
Thanks,
Ian