<div dir="ltr">Emil<div><br></div><div>I looked at commit 3689ef32 which started the problem, and it seems that patch removed the line "touch git_sha1.h.tmp". Previously my workflow was "working" because the touch command ensures there's always going to be an empty header file.<div><br></div><div>Was this intentional? For the new git_sha1.h rules, I wonder if it is ok to add a step for git_sha1.h.tmp that does a "touch git_sha1.h" to ensure the file exists?</div></div></div><div class="gmail_extra"><br><div class="gmail_quote">On Thu, Jun 16, 2016 at 12:48 PM, Emil Velikov <span dir="ltr"><<a href="mailto:emil.l.velikov@gmail.com" target="_blank">emil.l.velikov@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><span class="">On 16 June 2016 at 20:47, Emil Velikov <<a href="mailto:emil.l.velikov@gmail.com">emil.l.velikov@gmail.com</a>> wrote:<br>
> On 16 June 2016 at 20:42, Haixia Shi <<a href="mailto:hshi@chromium.org">hshi@chromium.org</a>> wrote:<br>
>> Bisect shows the problem started at commit<br>
>> 3689ef32afdafbb030069e560aac0e563fc29048<br>
>> Author: Emil Velikov <<a href="mailto:emil.velikov@collabora.com">emil.velikov@collabora.com</a>><br>
>> Date: Mon May 30 12:32:05 2016 +0100<br>
>><br>
>> automake: rework the git_sha1.h rule, include in tarball<br>
>><br>
>> As we'll need the file in the release tarball, rework the rule so that<br>
>> the file is regenerated _only_ if we're in a git repository.<br>
>><br>
>> With this in place we can build vulkan (anv) from a release tarball.<br>
>><br>
>> Cc: Jason Ekstrand <<a href="mailto:jason.ekstrand@intel.com">jason.ekstrand@intel.com</a>><br>
>> Cc: Kristian Høgsberg Kristensen <<a href="mailto:krh@bitplanet.net">krh@bitplanet.net</a>><br>
>> Signed-off-by: Emil Velikov <<a href="mailto:emil.velikov@collabora.com">emil.velikov@collabora.com</a>><br>
>><br>
>> I'm not building from a release tarball, but the workflow first copies the<br>
>> source tree to a tmp directory (without the .git directory), so there's no<br>
>> git_sha1.h file, and the file is not generated either.<br>
>><br>
> Hmmm... I wouldn't have imagine this particular workflow. Have you<br>
> considered gitlink and/or git worktree ?<br>
><br>
> If neither of these are an option for you see my earlier email.<br>
><br>
</span>The former seems to be missing a manpage so here's a PDF<br>
<a href="https://media.readthedocs.org/pdf/git-link/latest/git-link.pdf" rel="noreferrer" target="_blank">https://media.readthedocs.org/pdf/git-link/latest/git-link.pdf</a><br>
<span class="HOEnZb"><font color="#888888"><br>
-Emil<br>
</font></span></blockquote></div><br></div>