Please test and stabilize =dev-vcs/mercurial-1.7, unfortunately we (me and Dirkjan) are aware of some test failures (it heavily depends on other packages, like test failures in bzr import, environment, arch, etc). Please *do* report those issues, but unless it's something like most of them failing please *do* stabilize mercurial. It's not optimal, but we really, really need newer mercurial stabilized on most arches.
amd64 ok
Tested on x86, all good here.
amd64 done. Thanks Agostino
compile and run nicely on x86 but tests are failing. (see attached emerge --log)
Created attachment 257341 [details] tests failure on x86 the build.log with failing tests
x86 done. Thanks everyone!
Testing on SPARC, Mercurial 1.5 failed 4 out of 347 tests. Testing Mercurial 1.7 generated 16 failures from 377 tests. These failures are: Failed test-bheads.t: output changed and returned error code -15 Failed test-keyword.t: output changed and returned error code -15 Failed test-mq.t: output changed and returned error code -15 Failed test-command-template.t: output changed and returned error code -15 Failed test-fetch.t: output changed and returned error code -15 Failed test-import.t: output changed and returned error code -15 Failed test-log.t: output changed and returned error code -15 Failed test-mq-header-from.t: output changed and returned error code -15 Failed test-push-warn.t: output changed and returned error code -15 Failed test-mq-guards.t: output changed and returned error code -15 Failed test-mv-cp-st-diff.t: output changed and returned error code -15 Failed test-rename-merge2.t: output changed and returned error code -15 Failed test-revset.t: output changed and returned error code -15 Failed test-merge-tools.t: output changed and returned error code -15 Failed test-mq-header-date.t: output changed and returned error code -15 Failed test-pull-pull-corruption2.t: output changed
Alex: could you attach build.log? If those are timeouts please see bug 348854.
Created attachment 257531 [details] tests failure on SPARC
Alex: yes, all of those failures seem to be timeouts. It's an issue with how many tests are run at the same time, bug 348854.
Small correction: all except test-pull-pull-corruption2.t: ERROR: /var/tmp/portage/dev-vcs/mercurial-1.7/work/mercurial-1.7/tests/test-pull-pull-corruption2.t output changed --- /var/tmp/portage/dev-vcs/mercurial-1.7/work/mercurial-1.7/tests/test-pull-pull-corruption2.t +++ /var/tmp/portage/dev-vcs/mercurial-1.7/work/mercurial-1.7/tests/test-pull-pull-corruption2.t.err @@ -16,6 +16,7 @@ $ hg init version2 $ hg -R version2 pull source1 & $ sleep 1 + $ hg clone --pull -U version2 corrupted pulling from source1 requesting all changes adding changesets @@ -23,7 +24,6 @@ adding file changes added 10 changesets with 10 changes to 1 files (run 'hg update' to get a working copy) - $ hg clone --pull -U version2 corrupted requesting all changes adding changesets adding manifests
Created attachment 258017 [details] dev-vcs:mercurial-1.7:20101223-190044.log buildlog on ARM build log on arm. Krzysztof, is there anything else than the timeouts?
Markus: all of failures (even those not marked with timeout) look to me like the effects of -j4. Dirkjan: -j4 has more side effects than only timeouts: ERROR: /var/tmp/portage/dev-vcs/mercurial-1.7/work/mercurial-1.7/tests/test-hgweb.t output changed --- /var/tmp/portage/dev-vcs/mercurial-1.7/work/mercurial-1.7/tests/test-hgweb.t +++ /var/tmp/portage/dev-vcs/mercurial-1.7/work/mercurial-1.7/tests/test-hgweb.t.err @@ -9,421 +9,72 @@ adding da/foo adding foo $ hg serve -n test -p $HGPORT -d --pid-file=hg.pid -A access.log -E errors.log + abort: cannot start server at ':$HGPORT': Address already in use + abort: child process failed to start + [255] $ cat hg.pid >> $DAEMON_PIDS + cat: hg.pid: No such file or directory + [1]
I've removed -j4 from 1.7, 1.7.1 and 1.7.2 - this should help with test failures.
arm stable
Test failures on SPARC are now much less: Failed test-mq-header-date.t: output changed and returned error code -15 Failed test-mq-header-from.t: output changed and returned error code -15 Failed test-mq.t: output changed and returned error code -15 Failed test-rename-merge2.t: output changed and returned error code -15 # Ran 376 tests, 12 skipped, 4 failed. Could definitely consider this a candidate for stabilisation on SPARC.
(In reply to comment #16) > Test failures on SPARC are now much less: > > Failed test-mq-header-date.t: output changed and returned error code -15 > Failed test-mq-header-from.t: output changed and returned error code -15 > Failed test-mq.t: output changed and returned error code -15 > Failed test-rename-merge2.t: output changed and returned error code -15 > # Ran 376 tests, 12 skipped, 4 failed. > > Could definitely consider this a candidate for stabilisation on SPARC. Awesome! It's definitively better than previous stable anyway :) As a side note: please full build log or just tests output - that way we can check why there are test failure.
ppc stable
alpha/ia64/sparc stable
ppc64 stable
Stable for HPPA and closing.