[PATCH] Print cwd path to AT_FDCWD when -y (--decode-fd=path) flag is used
Renaud Métrich
rmetrich at redhat.com
Mon Nov 29 18:40:35 UTC 2021
Gotcha, thanks a lot. That split was made to be able to partly revert based
on my observations, sorry for the additional work.
On Monday, November 29, 2021, Dmitry V. Levin <ldv at altlinux.org> wrote:
> On Wed, Oct 06, 2021 at 08:36:33PM +0200, Renaud Métrich wrote:
>> Hi,
>>
>> This set of patches enables to resolve the AT_FDCWD dirfd, which is very
>> useful when syscalls perform an operation on the cwd and relative
>> pathname, avoiding the need to search for the pwd, which is sometimes
>> just not available.
>>
>> Also available as a PR: https://github.com/strace/strace/pull/194
>
> This is now merged into master, thanks for the nice feature!
>
>> The patch is split into 4 pieces for now because I found out that the
>> filter on --decode-fds=path isn't used at all in current strace code:
>> paths get decoded as soon as --decode-fds=<something> is used.
>>
>> Since I don't know if this is really the expected behaviour, I kept the
>> commits separated in case we just want decoding when --decode-fd=path is
>> specified only:
>>
>> - 0001 only decodes when --decode-fd=path is specified
>>
>> - 0002 decodes whenever --decode-fd=<something> is specified
>>
>> The other commits are unit tests.
>
> We have a policy that no commit should knowingly break the test suite, so
> when the strace output changes, test suite has to be adjusted in the same
> commit. I did it for you this time, but next time please remember not to
> break the test suite by intermediate commits.
>
>
> --
> ldv
> --
> Strace-devel mailing list
> Strace-devel at lists.strace.io
> https://lists.strace.io/mailman/listinfo/strace-devel
>
--
Renaud
Sent from my phone, sorry for the typos...
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.strace.io/pipermail/strace-devel/attachments/20211129/1b144451/attachment.htm>
More information about the Strace-devel
mailing list