| Summary: | net-libs/opal-3.6.2 build fails with gcc 4.4.0: 'sprintf' was not declared in this scope | ||
|---|---|---|---|
| Product: | Gentoo Linux | Reporter: | Jose daLuz <jdaluz> |
| Component: | New packages | Assignee: | Mounir Lamouri (volkmar) (RETIRED) <volkmar> |
| Status: | RESOLVED FIXED | ||
| Severity: | normal | CC: | esigra, ikelos, joost.ruis, mgorny, voip+disabled |
| Priority: | Normal | ||
| Version: | 2008.0 | ||
| Hardware: | All | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Package list: | Runtime testing required: | --- | |
| Bug Depends on: | |||
| Bug Blocks: | 249226 | ||
| Attachments: |
Output of paludis --info opal
And here is the backported fix and diff to ebuild |
||
|
Description
Jose daLuz
2009-05-10 21:43:45 UTC
Created attachment 190878 [details]
Output of paludis --info opal
This Fedora build log: http://koji.fedoraproject.org/koji/buildinfo?buildID=99604 indicates gcc 4.4 compilation was fixed: * Tue Feb 10 2009 Peter Robinson <pbrobinson@gmail.com> - 3.5.2-7 - Add patches to fix gcc44 compilation, remove celt until issues are fixed upstream I don't know which patches are responsible but it's a place to start. Same issue with =net-libs/opal-3.6.2 from main tree. Could you change 'Summary' to be less overlay-ish? Well, since this has now hit the main tree, I've removed the version from my overlay. As there's been a report against the version in the main tree, I'm reassigning this... 5:) Thank you for the gift Mike :) It looks like opal-3.6.4 has some patches for gcc-4.4.0. I will wait until I bump opal to fix the bug. (In reply to comment #5) > It looks like opal-3.6.4 has some patches for gcc-4.4.0. I will wait until I > bump opal to fix the bug. Could you precise when it'll be? I think I can provide a patch for this version in a while if that would make any difference right now. (In reply to comment #6) > (In reply to comment #5) > > It looks like opal-3.6.4 has some patches for gcc-4.4.0. I will wait until I > > bump opal to fix the bug. > > Could you precise when it'll be? I think I can provide a patch for this version > in a while if that would make any difference right now. > opal-3.6.4, ptlib-2.6.4 and ekiga-3.2.5 are released so it could be in a few days. But as I'm really busy these days, it could also be in a few weeks. So, if you want to provide gcc-4.4 patch for opal-3.6.2, it can worth the effort. Actually, I think you can backport patches submitted to opal (search gcc in websvn) and test if they are working with opal-3.6.2. In my opinion, as long as gcc-4.4 is masked, fixing gcc-4.4 bugs is not a priority. Created attachment 197111 [details, diff]
And here is the backported fix
Created attachment 197112 [details, diff]
and diff to ebuild
If I recall correctly, there are 2 or 3 patches for gcc-4.4. One for theora for example. Can you check/confirm that ? Only searching for 'gcc' keyword, I found four patches but only this one was essential. Others were there only to remove warnings, so I decided to not waste time on them. With this patch, opal compiles fine. There's problem with ekiga though but it also seems easy to fix. I'll take a look at it in next 16h, I think. I can confirm opal-3.6.2 from Portage builds with this patch. Now to reopen bug 269329 and see if ekiga 3.2.4 will build. (In reply to comment #11) > With this patch, opal compiles fine. There's problem with ekiga though but it > also seems easy to fix. I'll take a look at it in next 16h, I think. > Please see if it's the same problem I ran into in bug 269329. (In reply to comment #13) > (In reply to comment #11) > > With this patch, opal compiles fine. There's problem with ekiga though but it > > also seems easy to fix. I'll take a look at it in next 16h, I think. > > > Please see if it's the same problem I ran into in bug 269329. Yes, that's the one. Patch in the tree. Thanks Michal :) |
