Re: [PATCH] perf machine: Don't ignore _etext when not a text symbol
From: Arnaldo Carvalho de Melo
Date: Wed Jan 08 2025 - 15:15:10 EST
On Wed, Jan 08, 2025 at 10:15:24AM +0100, Christophe Leroy wrote:
> Depending on how vmlinux.lds is written, _etext might be the very
> first data symbol instead of the very last text symbol.
>
> Don't require it to be a text symbol, accept any symbol type.
I'm adding a Link:
Link: https://lore.kernel.org/all/752a31b0-4370-4f52-b7cc-45f0078c1d6c@xxxxxxxxxx
To give more context as where this has been observed, and also add a
snippet of your explanation there, this:
----
# grep -e _stext -e _etext -e _edata /proc/kallsyms
c0000000 T _stext
c08b8000 D _etext
So there is no _edata and _etext is not text
$ ppc-linux-objdump -x vmlinux | grep -e _stext -e _etext -e _edata
c0000000 g .head.text 00000000 _stext
c08b8000 g .rodata 00000000 _etext
c1378000 g .sbss 00000000 _edata
----
Thanks,
- Arnaldo
> Fixes: ed9adb2035b5 ("perf machine: Read also the end of the kernel")
> Signed-off-by: Christophe Leroy <christophe.leroy@xxxxxxxxxx>
> ---
> tools/perf/util/machine.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c
> index 27d5345d2b30..9be2f4479f52 100644
> --- a/tools/perf/util/machine.c
> +++ b/tools/perf/util/machine.c
> @@ -1003,7 +1003,7 @@ static int machine__get_running_kernel_start(struct machine *machine,
>
> err = kallsyms__get_symbol_start(filename, "_edata", &addr);
> if (err)
> - err = kallsyms__get_function_start(filename, "_etext", &addr);
> + err = kallsyms__get_symbol_start(filename, "_etext", &addr);
> if (!err)
> *end = addr;
>
> --
> 2.47.0