Closed
Bug 953314
Opened 10 years ago
Closed 10 years ago
Unable to run tinderbox js shells on Windows
Categories
(Firefox Build System :: General, defect)
Tracking
(Not tracked)
RESOLVED
FIXED
mozilla29
People
(Reporter: gkw, Assigned: ted)
References
Details
(Keywords: regression)
Attachments
(2 files)
19.08 KB,
image/png
|
Details | |
2.04 KB,
patch
|
glandium
:
review+
|
Details | Diff | Splinter Review |
Download: https://ftp.mozilla.org/pub/mozilla.org/firefox/nightly/2013-12-27-mozilla-central-debug/jsshell-win32.zip You will be unable to run it due to a System Error: "The program can't start because icuind50.dll is missing from your computer. Try reinstalling the program to fix this problem." Is it because icuind50.dll is not packaged together in the zip file? Guessing this is probably related to bug 915735. === Tinderbox Build Bisection Results by autoBisect === The "good" changeset has the timestamp "20131206043732", the hash "bbd9375c5f50", and the reason for the result is: Acceptable exit code 3 The "bad" changeset has the timestamp "20131206051028", the hash "099f65a712a8", and the reason for the result is: Negative exit code -1073741515 Likely regression window: http://hg.mozilla.org/integration/mozilla-inbound/pushloghtml?fromchange=bbd9375c5f50&tochange=099f65a712a8
Flags: needinfo?(ehsan)
Comment 1•10 years ago
|
||
Hmm, I thought <http://mxr.mozilla.org/mozilla-central/source/toolkit/mozapps/installer/packager.mk#105> is supposed to take care of this?
Component: JavaScript Engine → Build Config
Flags: needinfo?(ehsan) → needinfo?(mh+mozilla)
Comment 2•10 years ago
|
||
(In reply to :Ehsan Akhgari (needinfo? me!) from comment #1) > Hmm, I thought > <http://mxr.mozilla.org/mozilla-central/source/toolkit/mozapps/installer/ > packager.mk#105> is supposed to take care of this? It is.
Flags: needinfo?(mh+mozilla)
Comment 3•10 years ago
|
||
From the build log (https://tbpl.mozilla.org/php/getParsedLog.php?id=32490746&tree=Mozilla-Central&full=1): Packaging JavaScript Shell... c:\builds\moz2_slave\m-cen-w32-00000000000000000000\build\toolkit\mozapps\installer\packager.mk:725:0$ rm -f ../../dist/jsshell-win32.zip c:\builds\moz2_slave\m-cen-w32-00000000000000000000\build\toolkit\mozapps\installer\packager.mk:726:0$ zip -9j ../../dist/jsshell-win32.zip ../../dist/bin/js.exe ../../dist/bin/mozglue.dll ../../dist/bin/msvcr100.dll ../../dist/bin/msvcp100.dll ../../dist/bin/nss3.dll adding: js.exe (140 bytes security) (deflated 59%) adding: mozglue.dll (140 bytes security) (deflated 42%) adding: msvcr100.dll (140 bytes security) (deflated 46%) adding: msvcp100.dll (140 bytes security) (deflated 68%) adding: nss3.dll (140 bytes security) (deflated 49%) So, clearly what we have there is not enough. Gary, please find a build system hacker who is interested in working on this. I don't know where to start debugging this.
Reporter | ||
Comment 4•10 years ago
|
||
> Gary, please find a build system hacker who is interested in working on
> this. I don't know where to start debugging this.
Thanks for looking at this, Ehsan. I'll needinfo our build system overlord.
gps, who will be a good person for this? This completely blocks starting up Windows js shells downloaded from tinderbox.
Flags: needinfo?(gps)
Assignee | ||
Comment 5•10 years ago
|
||
Oh man, this is stupid. XP_WIN isn't defined in Makefiles. I can't believe I didn't notice that in review.
Assignee | ||
Comment 6•10 years ago
|
||
This should work on Windows. I changed the semantics here slightly by taking out the XP_UNIX block, but I don't think it will matter in practice
Attachment #8355517 -
Flags: review?(mh+mozilla)
Assignee | ||
Updated•10 years ago
|
Assignee: nobody → ted
Status: NEW → ASSIGNED
Comment 7•10 years ago
|
||
Comment on attachment 8355517 [details] [diff] [review] fix jsshell packaging of ICU libs Review of attachment 8355517 [details] [diff] [review]: ----------------------------------------------------------------- ::: toolkit/mozapps/installer/packager.mk @@ +117,5 @@ > $(DIST)/bin/icuuc$(MOZ_ICU_VERSION).dll \ > $(NULL) > endif # MOZ_DEBUG > else > +ifeq ($(OS_TARGET), Darwin) Fun fact: if you put "else ifeq ($(OS_TARGET), Darwin)" on one line, you only need one endif below.
Updated•10 years ago
|
Attachment #8355517 -
Flags: review?(mh+mozilla) → review+
Reporter | ||
Comment 9•10 years ago
|
||
I helped to land this: https://hg.mozilla.org/integration/mozilla-inbound/rev/d1beadca2983
Target Milestone: --- → mozilla29
Comment 10•10 years ago
|
||
https://hg.mozilla.org/mozilla-central/rev/d1beadca2983
Status: ASSIGNED → RESOLVED
Closed: 10 years ago
Resolution: --- → FIXED
Updated•6 years ago
|
Product: Core → Firefox Build System
You need to log in
before you can comment on or make changes to this bug.
Description
•