Re: [PATCH dwarves] btf_encoder: verify 0 address DWARF variables are really in ELF section

[Alan Maguire <alan.maguire@oracle.com>]

On 26/01/2025 20:04, Cong Wang wrote:
> On Sat, Jan 25, 2025 at 8:55 PM Cong Wang <xiyou.wangcong@gmail.com> wrote:
>>
>> Hi Alan,
>>
>> On Tue, Dec 17, 2024 at 2:36 AM Alan Maguire <alan.maguire@oracle.com> wrote:
>>>
>>> We use the DWARF location information to match a variable with its
>>> associated ELF section.  In the case of per-CPU variables their
>>> ELF section address range starts at 0, so any 0 address variables will
>>> appear to belong in that ELF section.  However, for "discard" sections
>>> DWARF encodes the associated variables with address location 0 so
>>> we need to double-check that address 0 variables really are in the
>>> associated section by checking the ELF symbol table.
>>>
>>> This resolves an issue exposed by CONFIG_DEBUG_FORCE_WEAK_PER_CPU=y
>>> kernel builds where __pcpu_* dummary variables in a .discard section
>>> get misclassified as belonging in the per-CPU variable section since
>>> they specify location address 0.
>>
>> It is _not_ your patch's fault, but I got this segfault which prevents me from
>> testing this patch. (It also happens after reverting your patch.)
> 
> Never mind, I managed to workaround this issue by a clean build.
> 
> And I tested your patch, it works for me with CONFIG_DEBUG_FORCE_WEAK_PER_CPU=y.
> 
> Tested-by: Cong Wang <cong.wang@bytedance.com>
> 
> Thanks a lot!

Thanks for verifying the fix! You didn't happen to get a coredump or
backtrace for the earlier segmentation fault by any chance? Just want to
make sure there aren't other issues lurking here. Thanks again!

Alan