Chromium Code Reviews
|
|
Chromium Code Reviews
Issue
1889483003:
[iOS/GN] Enable running tests on simulator. (Closed)
Description[iOS/GN] Enable running tests on simulator.
This runs the tests on gn bots that build for simulator on 64-bit
devices only (as the build is only 64-bit currently).
BUG=459705
Committed: https://crrev.com/6214d1dd6de159103a75ebc266d6cd0ceb54f709
Cr-Commit-Position: refs/heads/master@{#388743}
Patch Set 1 #Patch Set 2 : Enable to tests on iPad by switching to "iPad Air" #Patch Set 3 : Explicitly list all test data files #Patch Set 4 : Sync //net/BUILD.gn with net/net.gyp #Patch Set 5 : Rebase on top of crrev.com/1889483003 #Patch Set 6 : Rebase on top of latest patchset #Patch Set 7 : Rebase on origin/master #
Messages
Total messages: 26 (11 generated)
sdefresne@chromium.org changed reviewers: + dpranke@chromium.org, smut@google.com
Please take a look. The application fails to start on iPad on my machine too, so I've filled https://bugs.chromium.org/p/chromium/issues/detail?id=603180 to fix this (I've never tried to run them previously) but I'd like to have the tests run on iPhone soon to avoid regression.
lgtm; but how come ios_web_unittests and net_unittests are failing on GN? iPad is probably because 32-bit.
lgtm; but how come ios_web_unittests and net_unittests are failing on GN? iPad is probably because 32-bit.
what @smut said ... we need to either fix the tests that are failing before landing this, or list the tests explicitly (separately) to just list the ones that are expected to work for now while we're fixing the rest. lgtm if you do either of those two things :).
Description was changed from ========== [iOS/GN] Enable running tests on simulator. This runs the tests on gn bots that build for simulator. The tests are only run on iPhone and not iPad currently as the application fails at startup on iPad with the following error: Simulator failed to start: "The operation couldn’t be completed. (FBSOpenApplicationErrorDomain error 1.)" (FBSOpenApplicationErrorDomain:1) BUG=459705 ========== to ========== [iOS/GN] Enable running tests on simulator. This runs the tests on gn bots that build for simulator on 64-bit devices only (as the build is only 64-bit currently). BUG=459705 ==========
On 2016/04/14 at 00:46:43, dpranke wrote: > what @smut said ... we need to either fix the tests that are failing before landing this, or list the tests explicitly (separately) to just list the ones that are expected to work for now while we're fixing the rest. > > lgtm if you do either of those two things :). Yes, the iPad error were due to 32/64 bit issues, switched to "iPad Air" that is a Retina 64-bit devices. Better to have some iPad tests than none. Regarding the failing tests however they are strange. They are due to file not found, for example this one: /private/tmp/tmpaF5feG/Library/Developer/CoreSimulator/Devices/B9D47F46-0D27-47A6-B27E-53517F575781/data/Containers/Bundle/Application/2FEF079D-191A-4CF3-8C2D-E6236D21F1AB/net_unittests.app/net/data/verify_signed_data_unittest/rsa-pss-sha256-salt10-using-pss-key-with-params.pem The file is correctly copied into the application bundle when I build locally, so two things can happen: 1. the file is copied in the bundle on the bots too, but lost when somehow before the test is run, 2. ninja thinks it does not have to copy the file. The first option is unlikely if the tests is run on the same machine as the builder, or if the code used is the same as the one used for gyp swarming bots (otherwise we would have the same issue with gyp). The second is possible because we do not explicitly list all the individual files in the BUILD.gn but only the containing directory. As the timestamp of a directory is only updated when files are added/deleted directly in it, this can probably confuse ninja. Patch 3 will try to list all files individually instead of just the containing directories to check whether my analysis is correct.
On 2016/04/14 at 10:36:12, sdefresne wrote: > On 2016/04/14 at 00:46:43, dpranke wrote: > > what @smut said ... we need to either fix the tests that are failing before landing this, or list the tests explicitly (separately) to just list the ones that are expected to work for now while we're fixing the rest. > > > > lgtm if you do either of those two things :). > > Yes, the iPad error were due to 32/64 bit issues, switched to "iPad Air" that is a Retina 64-bit devices. Better to have some iPad tests than none. > > Regarding the failing tests however they are strange. They are due to file not found, for example this one: > /private/tmp/tmpaF5feG/Library/Developer/CoreSimulator/Devices/B9D47F46-0D27-47A6-B27E-53517F575781/data/Containers/Bundle/Application/2FEF079D-191A-4CF3-8C2D-E6236D21F1AB/net_unittests.app/net/data/verify_signed_data_unittest/rsa-pss-sha256-salt10-using-pss-key-with-params.pem > > The file is correctly copied into the application bundle when I build locally, so two things can happen: > 1. the file is copied in the bundle on the bots too, but lost when somehow before the test is run, > 2. ninja thinks it does not have to copy the file. > > The first option is unlikely if the tests is run on the same machine as the builder, or if the code used is the same as the one used for gyp swarming bots (otherwise we would have the same issue with gyp). The second is possible because we do not explicitly list all the individual files in the BUILD.gn but only the containing directory. As the timestamp of a directory is only updated when files are added/deleted directly in it, this can probably confuse ninja. > > Patch 3 will try to list all files individually instead of just the containing directories to check whether my analysis is correct. So, my hypothesis is validated. Listing the files explicitly is required to get incremental build working correctly. Since the bots do incremental build, it is critical that we list files individually. I'm going to try to trim down the list to just the files that are required to run the tests successfully and send those as individual CL. I'll keep this CL ready for enabling the tests on the bots, and only land it once all the required individual files are listed in correct BUILD.gn file (and bundle_data object).
Patchset #6 (id:100001) has been deleted
The CQ bit was checked by sdefresne@chromium.org
Flakyness has been fixed, let's try to enable this!
The patchset sent to the CQ was uploaded after l-g-t-m from dpranke@chromium.org, smut@google.com Link to the patchset: https://codereview.chromium.org/1889483003/#ps120001 (title: "Rebase on top of latest patchset")
CQ is trying da patch. Follow status at https://chromium-cq-status.appspot.com/patch-status/1889483003/120001 View timeline at https://chromium-cq-status.appspot.com/patch-timeline/1889483003/120001
The CQ bit was unchecked by commit-bot@chromium.org
Try jobs failed on following builders: linux_chromium_rel_ng on tryserver.chromium.linux (JOB_FAILED, http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)
The CQ bit was checked by sdefresne@chromium.org
The patchset sent to the CQ was uploaded after l-g-t-m from dpranke@chromium.org, smut@google.com Link to the patchset: https://codereview.chromium.org/1889483003/#ps140001 (title: "Rebase on origin/master")
CQ is trying da patch. Follow status at https://chromium-cq-status.appspot.com/patch-status/1889483003/140001 View timeline at https://chromium-cq-status.appspot.com/patch-timeline/1889483003/140001
The CQ bit was unchecked by commit-bot@chromium.org
Try jobs failed on following builders: android_chromium_gn_compile_dbg on tryserver.chromium.android (JOB_FAILED, https://build.chromium.org/p/tryserver.chromium.android/builders/android_chro...) linux_chromium_clobber_rel_ng on tryserver.chromium.linux (JOB_FAILED, http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...) linux_chromium_rel_ng on tryserver.chromium.linux (JOB_FAILED, http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)
The CQ bit was checked by sdefresne@chromium.org
CQ is trying da patch. Follow status at https://chromium-cq-status.appspot.com/patch-status/1889483003/140001 View timeline at https://chromium-cq-status.appspot.com/patch-timeline/1889483003/140001
Sending back to the bots now that goma issue is fixed.
Message was sent while issue was closed.
Committed patchset #7 (id:140001)
Message was sent while issue was closed.
Description was changed from ========== [iOS/GN] Enable running tests on simulator. This runs the tests on gn bots that build for simulator on 64-bit devices only (as the build is only 64-bit currently). BUG=459705 ========== to ========== [iOS/GN] Enable running tests on simulator. This runs the tests on gn bots that build for simulator on 64-bit devices only (as the build is only 64-bit currently). BUG=459705 Committed: https://crrev.com/6214d1dd6de159103a75ebc266d6cd0ceb54f709 Cr-Commit-Position: refs/heads/master@{#388743} ==========
Message was sent while issue was closed.
Patchset 7 (id:??) landed as https://crrev.com/6214d1dd6de159103a75ebc266d6cd0ceb54f709 Cr-Commit-Position: refs/heads/master@{#388743} |
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
