Description of problem: The patch rhbz-911314.patch is being conditionally applied on ppc and ppc64 only. However it fails to apply on xulrunner-21. Patch #21 (rhbz-911314.patch): + /usr/bin/cat /builddir/build/SOURCES/rhbz-911314.patch + /usr/bin/patch -p1 -b --suffix .ppc --fuzz=0 patching file js/src/gc/Heap.h Hunk #1 succeeded at 812 (offset 12 lines). patching file js/src/gc/Memory.cpp patching file js/src/gc/Memory.h patching file js/src/jsapi.cpp Hunk #1 succeeded at 1110 (offset -3 lines). patching file js/src/jscntxt.h Hunk #1 succeeded at 1026 (offset 153 lines). patching file js/src/jsgc.cpp Hunk #1 succeeded at 465 (offset -12 lines). Hunk #2 succeeded at 561 (offset -12 lines). Hunk #3 succeeded at 592 (offset -12 lines). Hunk #4 succeeded at 608 (offset -12 lines). Hunk #5 succeeded at 725 (offset -11 lines). Hunk #6 succeeded at 736 (offset -11 lines). Hunk #7 FAILED at 781. Hunk #8 succeeded at 937 (offset 56 lines). Hunk #9 succeeded at 2104 (offset 133 lines). Hunk #10 succeeded at 2172 (offset 133 lines). 1 out of 10 hunks FAILED -- saving rejects to file js/src/jsgc.cpp.rej patching file js/public/HeapAPI.h Hunk #1 succeeded at 11 (offset -2 lines). error: Bad exit status from /var/tmp/rpm-tmp.lWG2uA (%prep) Version-Release number of selected component (if applicable): xulrunner-21.0-3.fc19 How reproducible: Always Steps to Reproduce: 1. ppc-koji build --scratch f19 xulrunner-21.0-3.fc19.src.rpm 2. 3. Actual results: http://ppc.koji.fedoraproject.org/koji/taskinfo?taskID=1112214
Created attachment 750632 [details] rhbz-911314.patch backported to xulrunner-21 This is a backport of rhbz-911314.patch to xulrunner-21. We also need this patch from upstream to fix a different build issue: https://hg.mozilla.org/mozilla-central/raw-rev/fe010206e5ad With these patches xulrunner-21 builds fine on ppc64, though it crashes at runtime inside gfx, which indicates a different issue. Please include these patches to Fedora and release a new xulrunner version. I will file a different bug for the runtime issue.
Gustavo, the patch you attached does not apply: + /usr/bin/cat /home/komat/CVS/xulrunner/rhbz-911314.patch + /usr/bin/patch -p1 -b --suffix .ppc --fuzz=0 patching file js/src/gc/Heap.h Hunk #1 succeeded at 812 (offset 12 lines). patching file js/src/gc/Memory.cpp patching file js/src/gc/Memory.h patching file js/src/jsapi.cpp Hunk #1 succeeded at 1110 (offset -3 lines). patching file js/src/jscntxt.h Hunk #1 succeeded at 1026 (offset 153 lines). patching file js/src/jsgc.cpp Hunk #1 succeeded at 465 (offset -12 lines). Hunk #2 succeeded at 561 (offset -12 lines). Hunk #3 succeeded at 592 (offset -12 lines). Hunk #4 succeeded at 608 (offset -12 lines). Hunk #5 succeeded at 725 (offset -11 lines). Hunk #6 succeeded at 736 (offset -11 lines). Hunk #7 FAILED at 781. Hunk #8 succeeded at 937 (offset 56 lines). Hunk #9 succeeded at 2104 (offset 133 lines). Hunk #10 succeeded at 2172 (offset 133 lines). 1 out of 10 hunks FAILED -- saving rejects to file js/src/jsgc.cpp.rej patching file js/public/HeapAPI.h Hunk #1 succeeded at 11 (offset -2 lines). error: Bad exit status from /var/tmp/rpm-tmp.2KmTPM (%prep)
Created attachment 751226 [details] rhbz-911314.patch backported to xulrunner-21 Sorry Martin, here is the correct patch.
The patch does not apply either but whatever, I updated the broken chunk. It's in brew now. If you want the patch in upstream it should be more friendly to other arches...
I double checked here and my second patch applied just fine. Anyway, thanks for correcting whatever was wrong with it. Regarding getting the patch upstream, I would appreciate if you post your comments on the upstream bugzilla, where the patch originated from: https://bugzilla.mozilla.org/show_bug.cgi?id=840242 I checked out the latest f19 xulrunner code from git and tried to build it. I got the following error: error: Bad source: /builddir/fedora/xulrunner/f19/mozilla-851850.patch: No such file or directory Thanks for your help, Martin.
Martin, we cannot do builds for secondary arches of packages that were not built on primary. Could you please do a build with these fixes on primary? Reopening the bug while we wait for a build. Thanks again
It's built already, right?