Re: [PATCH v2] kallsyms: strip LTO-only suffixes from promoted global functions

[Yonghong Song <yhs@meta.com>]

On 7/7/23 3:27 PM, Nick Desaulniers wrote:
> On Wed, Jun 28, 2023 at 11:19 AM Yonghong Song <yhs@fb.com> wrote:
>>
>> Commit 6eb4bd92c1ce ("kallsyms: strip LTO suffixes from static functions")
>> stripped all function/variable suffixes started with '.' regardless
>> of whether those suffixes are generated at LTO mode or not. In fact,
>> as far as I know, in LTO mode, when a static function/variable is
>> promoted to the global scope, '.llvm.<...>' suffix is added.
>>
>> The existing mechanism breaks live patch for a LTO kernel even if
>> no <symbol>.llvm.<...> symbols are involved. For example, for the following
>> kernel symbols:
>>    $ grep bpf_verifier_vlog /proc/kallsyms
>>    ffffffff81549f60 t bpf_verifier_vlog
>>    ffffffff8268b430 d bpf_verifier_vlog._entry
>>    ffffffff8282a958 d bpf_verifier_vlog._entry_ptr
>>    ffffffff82e12a1f d bpf_verifier_vlog.__already_done
>> 'bpf_verifier_vlog' is a static function. '_entry', '_entry_ptr' and
>> '__already_done' are static variables used inside 'bpf_verifier_vlog',
>> so llvm promotes them to file-level static with prefix 'bpf_verifier_vlog.'.
>> Note that the func-level to file-level static function promotion also
>> happens without LTO.
>>
>> Given a symbol name 'bpf_verifier_vlog', with LTO kernel, current mechanism will
>> return 4 symbols to live patch subsystem which current live patching
>> subsystem cannot handle it. With non-LTO kernel, only one symbol
>> is returned.
>>
>> In [1], we have a lengthy discussion, the suggestion is to separate two
>> cases:
>>    (1). new symbols with suffix which are generated regardless of whether
>>         LTO is enabled or not, and
>>    (2). new symbols with suffix generated only when LTO is enabled.
>>
>> The cleanup_symbol_name() should only remove suffixes for case (2).
>> Case (1) should not be changed so it can work uniformly with or without LTO.
>>
>> This patch removed LTO-only suffix '.llvm.<...>' so live patching and
>> tracing should work the same way for non-LTO kernel.
>> The cleanup_symbol_name() in scripts/kallsyms.c is also changed to have the same
>> filtering pattern so both kernel and kallsyms tool have the same
>> expectation on the order of symbols.
>>
>>   [1] https://lore.kernel.org/live-patching/20230615170048.2382735-1-song@kernel.org/T/#u
>>
>> Fixes: 6eb4bd92c1ce ("kallsyms: strip LTO suffixes from static functions")
>> Reported-by: Song Liu <song@kernel.org>
>> Signed-off-by: Yonghong Song <yhs@fb.com>
> 
> Thanks for the patch and improving live patch with LTO.  Looking back
> at the internal report that resulted in
> commit 6eb4bd92c1ce ("kallsyms: strip LTO suffixes from static functions")
> your version was what I originally had.  I did not leave a comment as
> to why I changed it when I sent it 2 years ago, and no longer recall
> the reason.
> 
> 
> Reviewed-by: Nick Desaulniers <ndesaulniers@google.com>

Nick, thanks for the review. I am not sure how this patch could
be merged into the mainline. I checked your patch
    6eb4bd92c1ce ("kallsyms: strip LTO suffixes from static functions")
and it looks like the patch was merged by Kees Cook.

I added Kees in the 'To' list. Kees, could you help merge
this patch if you are okay with the change?

Thanks!

Yonghong

> 
>> ---
>>   kernel/kallsyms.c  | 5 ++---
>>   scripts/kallsyms.c | 6 +++---
>>   2 files changed, 5 insertions(+), 6 deletions(-)
>>
>> Changelogs:
>>    v1 -> v2:
>>      . add 'Reported-by: Song Liu <song@kernel.org>'
>>      . also fix in scripts/kallsyms.c.
>>
>> diff --git a/kernel/kallsyms.c b/kernel/kallsyms.c
>> index 77747391f49b..4874508bb950 100644
>> --- a/kernel/kallsyms.c
>> +++ b/kernel/kallsyms.c
>> @@ -174,11 +174,10 @@ static bool cleanup_symbol_name(char *s)
>>           * LLVM appends various suffixes for local functions and variables that
>>           * must be promoted to global scope as part of LTO.  This can break
>>           * hooking of static functions with kprobes. '.' is not a valid
>> -        * character in an identifier in C. Suffixes observed:
>> +        * character in an identifier in C. Suffixes only in LLVM LTO observed:
>>           * - foo.llvm.[0-9a-f]+
>> -        * - foo.[0-9a-f]+
>>           */
>> -       res = strchr(s, '.');
>> +       res = strstr(s, ".llvm.");
>>          if (res) {
>>                  *res = '\0';
>>                  return true;
>> diff --git a/scripts/kallsyms.c b/scripts/kallsyms.c
>> index 0d2db41177b2..13af6d0ff845 100644
>> --- a/scripts/kallsyms.c
>> +++ b/scripts/kallsyms.c
>> @@ -346,10 +346,10 @@ static void cleanup_symbol_name(char *s)
>>           * ASCII[_]   = 5f
>>           * ASCII[a-z] = 61,7a
>>           *
>> -        * As above, replacing '.' with '\0' does not affect the main sorting,
>> -        * but it helps us with subsorting.
>> +        * As above, replacing the first '.' in ".llvm." with '\0' does not
>> +        * affect the main sorting, but it helps us with subsorting.
>>           */
>> -       p = strchr(s, '.');
>> +       p = strstr(s, ".llvm.");
>>          if (p)
>>                  *p = '\0';
>>   }
>> --
>> 2.34.1
>>
> 
>