> On 28/04/26 23:23, Jose E. Marchesi wrote:
>> External email: Use caution opening links or attachments
>> 
>>> On 23/04/26 14:27, Jose E. Marchesi wrote:
>>>> External email: Use caution opening links or attachments
>>>>
>>>>> Hi all,
>>>>>
>>>>> Per prior discussions, here is the specification we are proposing for
>>>>> version 4 of the AutoFDO GCOV profile format. This is a complete re-design
>>>>> focused around being a backwards-compatible and extensible format which
>>>>> can be partially read by the compiler.
>>>> Could you please send me (off list) a sample data file, as complete
>>>> as
>>>> possible?  Thanks!
>>>
>>> Hi Jose,
>>>
>>> Just curious as to why you need an example? No problems with sharing it,
>>> is there any particular part of the design you are interested in taking a
>>> look at? Meaningful profiles are rather unwieldy due to their size.
>> I intend to review the proposed format by writing a GNU poke pickle
>> for
>> it, as complete as possible, including constraints etc.  On one hand, it
>> is a good way to catch potential pitfalls in the designed encoding, data
>> integrity, etc.  On the other, you get an easy to update pickle for
>> free, to poke at data, dump to the textual representation, construct
>> test cases, and the like.
>> Having a sample binary will help.  It doesn't have to contain large
>> sets
>> of data, but the more of the spec it exercises, the better.
>> For reference, you cand find a pickle for gcov streams at
>> https://cgit.git.savannah.gnu.org/cgit/poke.git/tree/pickles/gcov.pk
>
> Hi Jose,
>
> Apologies for the late reply. I've attached some files to this
> mail that try to exhibit most of the features in the format:
>
> - example.c: Simple testcase
> - example.gcov: GCOV v3 profile
> - example.gcov.v4: GCOV v4 profile
> - example.gcov.v4.compact: GCOV v4 compact profile
>
> The testcase was compiled with -O2 -g -fno-omit-frame-pointer.

Thank you.
I will see what can I do and get back to you :)

Reply via email to