mozjs102 (102.15.1-1~deb12u1) bookworm; urgency=medium
* New upstream release (Closes: #1037158)
- CVE-2023-4046: Incorrect value used during WASM compilation
- CVE-2023-37202: Potential use-after-free from compartment mismatch
in SpiderMonkey
- CVE-2023-37211: Memory safety bugs
- CVE-2023-34416: Memory safety bugs
* Update debian/upstream/signing-key.asc per upstream rotation
-- Jeremy Bícha <jbicha@ubuntu.com> Tue, 19 Sep 2023 12:35:27 -0400
mozjs102 (102.11.0-1) unstable; urgency=high
* New upstream release (LP: #2018905)
- CVE-2023-32205: Browser prompts could have been obscured by popups
- CVE-2023-32206: Crash in RLBox Expat driver
- CVE-2023-32207: Potential permissions request bypass via clickjacking
- CVE-2023-32211: Content process crash due to invalid wasm code
- CVE-2023-32212: Potential spoof due to obscured address bar
- CVE-2023-32213: Potential memory corruption in FileReader::DoReadData()
- CVE-2023-32214: Potential DoS via exposed protocol handlers
- CVE-2023-32215: Memory safety bugs
-- Jeremy Bícha <jbicha@ubuntu.com> Mon, 08 May 2023 11:59:12 -0400
mozjs102 (102.10.0-1) unstable; urgency=high
* New upstream release (LP: #2015880)
- CVE-2023-29536: Invalid free from JavaScript code
- CVE-2023-29548: Incorrect optimization result on ARM64
- CVE-2023-29550: Memory safety bugs
- CVE-2023-29535: Potential Memory Corruption following Garbage Collector
compaction
-- Jeremy Bícha <jbicha@ubuntu.com> Mon, 10 Apr 2023 19:48:42 -0400
mozjs102 (102.9.0-1) unstable; urgency=high
[ Jeremy Bicha ]
* New upstream release
- CVE-2023-25751: Incorrect code generation during JIT compilation
[ John Paul Adrian Glaubitz ]
* Disable large-arraybuffers/base.js on all big-endian targets
(Closes: #1020700)
-- Jeremy Bicha <jbicha@ubuntu.com> Mon, 13 Mar 2023 11:03:53 -0400
mozjs102 (102.8.0-1) unstable; urgency=medium
* New upstream release
* Drop Python 3.11 patch: applied in new release
-- Jeremy Bicha <jbicha@ubuntu.com> Wed, 15 Feb 2023 08:57:21 -0500
mozjs102 (102.7.0-1) unstable; urgency=medium
* New upstream release
* Cherry-pick patch to fix build with Python 3.11 (Closes: #1028716)
* Re-export upstream signing key without extra signatures
* Update standards version to 4.6.2, no changes needed
-- Jeremy Bicha <jbicha@ubuntu.com> Tue, 17 Jan 2023 15:57:16 -0500
mozjs102 (102.6.0-1) unstable; urgency=medium
* New upstream release
-- Jeremy Bicha <jbicha@ubuntu.com> Mon, 12 Dec 2022 17:16:33 -0500
mozjs102 (102.5.0-1) unstable; urgency=medium
* New upstream release
-- Jeremy Bicha <jbicha@ubuntu.com> Tue, 15 Nov 2022 12:28:41 -0500
mozjs102 (102.4.0-1) unstable; urgency=medium
* New upstream release
* Drop ARMv7 atomic operations patch: applied in new release
-- Jeremy Bicha <jbicha@ubuntu.com> Mon, 17 Oct 2022 14:03:07 -0400
mozjs102 (102.3.0-1) unstable; urgency=medium
* Team upload
[ Jeremy Bicha ]
* New upstream release
[ Simon McVittie ]
* d/p/jit-Only-use-ARMv7-atomic-operations-on-ARMv7-or-higher.patch:
* Add patch to avoid using ARMv7 atomic operations on armel
* d/p/js-Enable-JIT-by-default-on-ARMv7-or-up-but-not-ARMv6-or-.patch:
Attempt to solve FTBFS on armel by disabling JIT
* Update Lintian overrides
* d/copyright: Remove patterns that no longer match anything
* Standards-Version: 4.6.1 (no changes required)
[ Adrian Bunk ]
* d/p/js-arm-i386-no-simd.patch:
Avoid SIMD (SSE/NEON) on 32-bit ARM and x86. It is not in the baseline
of our i386/armel/armhf ports.
* d/rules: Build with -Wl,--allow-multiple-definition on armel
(works around mozjs102's equivalent of mozjs91 bug #1017979)
* Go back to using the default gcc on armel (Closes: #1018047)
* Exclude basic/bug-1649234-1.js test on armel (Closes: #1017961)
-- Simon McVittie <smcv@debian.org> Sat, 24 Sep 2022 20:53:33 +0100
mozjs102 (102.2.0-2) unstable; urgency=medium
[ Simon McVittie ]
* d/rules: Pre-create debian/build/dist/cppunittests/ as a directory
* d/test.sh: Explicitly mark js executable as executable
* Build using gcc-11 on armel
(works around mozjs102's equivalent of mozjs91 bug #1017979)
* Add patch to avoid "impossible constraints" on armhf (Closes: #1017962)
-- Jeremy Bicha <jbicha@ubuntu.com> Wed, 24 Aug 2022 11:39:02 -0400
mozjs102 (102.2.0-1) unstable; urgency=medium
[ Jeremy Bicha ]
* New upstream release
* Drop 2 cherry-picked patches applied in new release
[ Mike Homney ]
* Add patch to remove workaround for old libstdc++ problem,
which now causes problems with GCC 12 on arm
-- Jeremy Bicha <jbicha@ubuntu.com> Mon, 22 Aug 2022 11:55:51 -0400
mozjs102 (102.1.0-2) unstable; urgency=medium
* Rebuild after NEW acceptance
-- Jeremy Bicha <jbicha@ubuntu.com> Tue, 16 Aug 2022 07:34:33 -0400
mozjs102 (102.1.0-1) unstable; urgency=medium
* New upstream release
* Refresh patches
* Cherry-pick 2 patches recommended by GNOME
* Copy patch from Debian Firefox to allow building with older cargo
-- Jeremy Bicha <jbicha@ubuntu.com> Mon, 08 Aug 2022 16:21:11 -0400
mozjs91 (91.12.0-1) unstable; urgency=medium
* New upstream release
-- Jeremy Bicha <jbicha@ubuntu.com> Mon, 08 Aug 2022 13:50:42 -0400
mozjs91 (91.10.0-1) unstable; urgency=medium
* New upstream release
-- Jeremy Bicha <jbicha@ubuntu.com> Wed, 01 Jun 2022 10:07:16 -0400
mozjs91 (91.8.0-1) unstable; urgency=medium
* New upstream release
* Drop Bug-1687417-MIPS32 patch: applied in new release
-- Jeremy Bicha <jbicha@ubuntu.com> Thu, 21 Apr 2022 15:34:55 -0400
mozjs91 (91.7.0-5) unstable; urgency=medium
* Upload to unstable
-- Jeremy Bicha <jbicha@ubuntu.com> Wed, 23 Mar 2022 10:45:50 -0400
mozjs91 (91.7.0-4) experimental; urgency=medium
* Team upload
* Build with vendored copy of ICU until the system copy is >= 69.1.
This decouples the GNOME Shell 42 (libmutter-10.so.0) transition from
the ICU 70 (or possibly 71) transition.
* d/p/series:
Don't apply patches to skip tests that failed with system ICU
(the patches are still present, and we should reapply them when
the system copy catches up)
* d/p/vendored-ICU/tests-Skip-comparison-of-tzdata-version-number.patch:
Add patch to fix test failure with vendored ICU
* Don't rebuild ICU data from first principles every time.
The version of its build system that went upstream in Mozilla
generates the data file when a new version is imported, and just
converts it from little-endian to big-endian during build if
necessary. Doing this resolves two FTBFS issues:
- regenerating the data file on a big-endian system breaks
build system assumptions in this area, so not doing that fixes
s390x and others;
- some parts of the vendored ICU library that are needed by
icu_sources_data.py, but not needed by a normal build, don't
link libatomic often enough for armel, so not doing that fixes
armel
* Remove downstream patches for icu_sources_data.py, since we no
longer run it.
-- Simon McVittie <smcv@debian.org> Tue, 22 Mar 2022 13:04:52 +0000
mozjs91 (91.7.0-3) experimental; urgency=medium
* Team upload
* Re-upload to experimental, to avoid the experimental version being
superseded by the version in unstable, which can't be built until
after the icu 70 transition (unless we start building with the
bundled icu).
-- Simon McVittie <smcv@debian.org> Wed, 09 Mar 2022 12:12:22 +0000
mozjs91 (91.7.0-2) unstable; urgency=medium
* Include changes from 91.6.0-2 that were accidentally dropped
-- Jeremy Bicha <jeremy.bicha@canonical.com> Tue, 08 Mar 2022 16:30:35 -0500
mozjs91 (91.7.0-1) unstable; urgency=medium
* New upstream release
* Release to unstable
-- Jeremy Bicha <jeremy.bicha@canonical.com> Tue, 08 Mar 2022 15:40:50 -0500
mozjs91 (91.6.0-2) experimental; urgency=medium
Team upload
[ Marco Trevisan (Treviño) ]
* d/p/jstests-skip-fp-evaluation-order-test-in-some-archs.patch:
Skip fp-evaluation-order.js on all ppc64 architectures
[ Simon McVittie ]
* d/p/Bug-1687417-MIPS32-nojit-silence-an-assert.patch:
Add patch from Firefox 94 to fix compilation on mipsel.
Thanks to Adrian Bunk
* d/p/Fix-math_private.h-for-i386-FTBFS.patch:
Add patch from firefox-esr to fix compilation on i386.
This partially resolves FTBFS on i386, although there is also a test
failure in test262/built-ins/Date/UTC/fp-evaluation-order.js which is
not yet resolved.
Thanks to Mike Hommey
* d/p/jstests-skip-fp-evaluation-order-test-in-some-archs.patch:
Add i386 to the architectures where fp-evaluation-order.js is skipped.
Reduces severity of: #1006152
* d/rules: Skip another test on i386.
The result differs very slightly from what was expected, presumably
because as long as our baseline doesn't guarantee SSE, we have no way
to disable i387 extended precision.
* d/p/tests-Accept-out-of-memory-as-close-enough-to-allocation-.patch
Add patch to fix FTBFS on machines with relatively limited RAM
(Closes: #1006196)
-- Simon McVittie <smcv@debian.org> Mon, 21 Feb 2022 21:48:05 +0000
mozjs91 (91.6.0-1) experimental; urgency=medium
[ Marco Trevisan (Treviño) ]
* New upstream release
* debian/patches: Refresh
* debian/patches: Skip fp-evauation-order tests in some archs
* debian/rules: Skip large arraybuffers tests in s390x
[ Jeremy Bicha ]
* Revert "debian/patches/series: Disable system-ICU related patches"
* Revert "debian/rules: Disable building with system ICU for now"
* Disable non262-Intl-DateTimeFormat-timeZone_version.js-Update-ver.patch
* Revert "debian/control: Remove dependency on libicu, given it's disabled for now"
-- Marco Trevisan (Treviño) <marco@ubuntu.com> Tue, 15 Feb 2022 19:56:29 +0100
mozjs91 (91.5.1-1) experimental; urgency=medium
[ Marco Trevisan (Treviño) ]
* New upstream release (Closes: #1004739)
* debian/patches: Refresh
* debian/control: Bump dependency on libicu-dev 69.1
* debian/tests.sh: Update python3 path for running JIT tests
* debian/patches: Move system-ICU related ones to different topic
* jstests: Skip property-escapes tests, unsupported by system ICU
* tests: Ignore POSIX/posix mismatch in tests
* tests: Use debian provided locale format to check formatted CN dates
* tests: Adapt subtags test to debian ICU computed values
* tests: Disable failing DateTimeFormat tests with system ICU
* DateTimeFormat: Use normal parenthesis instead of ICU "├ ┤"
* debian/patches: Also skip Date-time-zones-imported test with embedded ICU
* non262/Intl/DateTimeFormat/timeZone_version.js: Update version to match source
* debian/control: Remove dependency on libicu, given it's disabled for now
* debian/patches/series: Disable system-ICU related patches
[ Jeremy Bicha ]
* debian/rules: Disable building with system ICU for now
-- Jeremy Bicha <jeremy.bicha@canonical.com> Sat, 05 Feb 2022 14:18:20 -0500
mozjs78 (78.15.0-2) unstable; urgency=medium
* Team upload
* d/p/Fix-armhf-build-for-GCC-11-627.patch:
Add patch from cc-rs to fix FTBFS on armhf with gcc-11 build-time
configuration changes (Closes: #1001314)
-- Simon McVittie <smcv@debian.org> Wed, 08 Dec 2021 09:51:24 +0000
mozjs78 (78.15.0-1) unstable; urgency=medium
* New upstream release
- Final scheduled 78 ESR release. No mozjs changes since 78.13.0
* debian/rules: Drop overrides not needed with dh 13
-- Jeremy Bicha <jbicha@debian.org> Fri, 08 Oct 2021 07:42:32 -0400
mozjs78 (78.13.0-1) unstable; urgency=medium
* New upstream release
* Drop Bug-1644600 patch: applied in new release
* debian/upstream/signing-key.asc: Update from 91.0esr
* Release to unstable
-- Jeremy Bicha <jbicha@debian.org> Sat, 28 Aug 2021 08:46:19 -0400
mozjs78 (78.10.1-1) experimental; urgency=medium
* New upstream release
* debian/control: Bump standards version (no change required)
* debian/upstream/signing-key.asc: Use minimal key only
* debian/gbp-repack-mozjs-source.sh: Also filter out *.dll files
* debian/source/lintian-overrides: Ignore more warnings on shipped js files
* debian/patches:
- Refresh
- Move JSStructuredCloneData destructor to private impl.
Without this linking against mozjs won't work when structured cloning is
used.
* d/p/tests-Adapt-formatted-strings-results-to-system-ICU.patch:
Ignore system tz name check. This may fail when not on GMT.
* d/p/time-zone-path-test-Update-for-what-our-system-ICU-return.patch:
Protect from false negative.
We did update all the tests to use the system zoneinfo path, but we did
not include the case were an error should be reported.
Without doing this we may not check the point of the test (two colons)
because the test was failing due to the wrong zoneinfo path
-- Marco Trevisan (Treviño) <marco@ubuntu.com> Tue, 04 May 2021 19:41:03 +0200
mozjs78 (78.4.0-2) unstable; urgency=medium
* Team upload
* Use rustc -Clto=thin on all architectures.
Previously dwz only failed to process the debug symbols without this
flag on riscv64, but it seems the same problem now appears on amd64,
and perhaps other architectures. (Closes: #978354)
-- Simon McVittie <smcv@debian.org> Sun, 27 Dec 2020 21:13:20 +0000
mozjs78 (78.4.0-1) unstable; urgency=medium
* Team upload
* New upstream release
* Drop patches that came from upstream
-- Simon McVittie <smcv@debian.org> Wed, 04 Nov 2020 11:01:47 +0000
mozjs78 (78.3.0-2) unstable; urgency=medium
* Team upload
* d/p/Bug-1642332-Avoid-static-initializer-for-Number.NaN-r-jan.patch:
Add patch from upstream to fix Number.NaN being incorrect on mipsen
-- Simon McVittie <smcv@debian.org> Fri, 25 Sep 2020 19:45:45 +0100
mozjs78 (78.3.0-1) unstable; urgency=medium
* Team upload
[ Julien Cristau ]
* Update configure flag --disable-ion to --disable-jit on mipsen to
fix FTBFS.
[ Simon McVittie ]
* d/p/tests-Expect-some-tests-to-fail-on-armel.patch:
Add more tests that are expected to fail on armel, due to softfloat
not having precisely the same behaviour as hardfloat
* New upstream release
- d/p/Bug-1661094-Always-define-PKGCONF_REQUIRES_PRIVATE.patch:
Drop patch, applied upstream
* d/p/Bug-1666646-Set-CodeAlignment-to-16-for-the-none-backend..patch:
Add patch from upstream to fix static assertion failure on mipsel,
powerpc, and other 32-bit architectures with JIT disabled or missing
-- Simon McVittie <smcv@debian.org> Fri, 25 Sep 2020 12:48:00 +0100
mozjs78 (78.2.0-1) unstable; urgency=medium
* Team upload
[ Marco Trevisan (Treviño) ]
* New upstream version 78.2.0 (Closes: #968341)
- Rename source to mozjs78
- debian/gbp.conf: Set debian and upstream branches to match version
- debian/watch: Update main version regex
* debian/control:
- Add back rust and LLVM dependencies:
mozjs can't be compiled anymore without rust built sources, at least not
without some huge changes that aren't suitable for a distro-patch, so
let's add back rustc and cargo as dependencies.
This also implies some libraries checks that are performed using llvm
tools, so including it as well.
- Bump dependency on debhelper 13
- Bump build dependency to ICU 63.1-4 (tests need debian specific version,
see bug #968609)
- Use python3 (and drop python2.7) (Closes: #956334)
- Update Vcs-* flags to match branches
* debian/rules:
- Only use python3
- Don't set ICU_DATA_FILE as mozilla build system already handles it
- Use GNOME suggested configure parameters
- Factorize the code to compute just once the VERSIONED_LIBNAME
- Export MOZJS_VERSION to be used in install files
- Use thin LTO in rust (via RUSTFLAGS) for riscv64
- Ensure rust follows debian compilation settings
- Set CARGO_HOME to a builddir subdirectory
* debian/test.sh:
- Use python3 to run jit tests
- Run again make check-jstests (it was wrongly removed)
- Don't redefine HOME. This is now handled by default by debhelper 13
* debian/*.install: Use debhelper 13 variables sobstitutions to match files
* debian/README.source: Add doc to explain how we generate the orig file
* debian/source/lintian-overrides:
- Ignore missing signature on source tarball
- Ignore errors on files with very long lines
* debian/gbp.conf:
- Remove filters as they are now handled by the repack script
* debian/gbp-repack-mozjs-source.sh:
- Only support gbp versions that repack the orig file using the content
generated by the postunpack hook
- Add ability to copy more files into the git archive from firefox
sources, and include the ICU sources to allow builds without system ICU.
- Add ability to filter files, and remove Files-Excluded from
debian/copyright (in case gbp imports an orig file without uscan).
- Delete empty directories in mozjs sources (as git will ignore them)
* debian/libmozjs-78-0.docs: Include NOTICE file from 3rd party pkcs11
* debian/patches: Refresh for mozjs78
* d/p/Bug-1545437-Add-options-to-specify-Rust-target-name.patch:
- Refreshed to apply in new version, we still need to set RUST_TARGET
* d/p/Bug-1661094-Always-define-PKGCONF_REQUIRES_PRIVATE.patch:
- Ensure we generate a pkg-config file with proper syntax
* d/p/Remove-unused-LLVM-and-Rust-build-dependencies.patch:
- Dropped, mozjs now strictly depends on rust (and use llvm tools to check)
* d/p/tests-Adapt-formatted-strings-results-to-system-ICU.patch:
- Sync expected test results as per latest system ICU
* d/p/Add-riscv64-support.patch:
- Ensure that we find the proper rust target in all the riscv64 variants
- Ensure we compile with -mabi=lp64d, importing some cc-rs fixes.
* d/p/icu_sources_data.py-Decouple-from-Mozilla-build-system.patch,
d/p/icu_sources_data-Write-command-output-to-our-stderr.patch:
- Add back icu sources data patches, they were wrongly removed, even if
they aren't currently used, they might be useful if internal ICU is
wanted for backport reasons.
[ Dimitri John Ledkov ]
* Disable CET protection in mozjs, until after JIT is ported upstream.
[ Simon McVittie ]
* d/copyright: Update
* d/p/Skip-another-timezone-related-test-that-fails-with-system.patch:
Add a patch to skip another timezone-related test with system ICU
-- Simon McVittie <smcv@debian.org> Thu, 27 Aug 2020 21:59:39 +0100
mozjs68 (68.9.0-1) unstable; urgency=medium
[ Marco Trevisan (Treviño) ]
* New upstream release
[ Simon McVittie ]
* Remove rust and LLVM dependencies for better portability.
Thanks to John Paul Adrian Glaubitz (Closes: #959144)
* Refresh patch series with gbp-pq-style metadata
* Install as libmozjs-68.so.0, not libmozjs-68.so.68.
The SONAME is libmozjs-68.so.0, so we need to install the
libmozjs-68.so.0 symbolic link. In native installations, ldconfig
would create libmozjs-68.so.0 for us, but when using mozjs68 to
cross-compile other packages, that won't happen. (Closes: #959792)
* d/rules: Justify why we skip basic/bug653153.js on i386.
That test seems to be failing on i386 due to extended precision in the
i387 FPU. Upstream always builds with -mfpmath=sse and does not support
use of i387, but that isn't currently allowed in Debian due to our i386
baseline being 20 years old (in particular no SSE and no SSE2).
* d/rules: Avoid parsing d/changelog
* d/rules: Make variable names less confusing.
Avoid the DEB_ prefix in our own variables, to avoid mixing up
our DEB_UPSTREAM_VERSION with dpkg's DEB_VERSION_UPSTREAM.
[ Marco Trevisan (Treviño) ]
* debian/rules: Properly parse DEB_VERSION_UPSTREAM using shell
* d/p/Bug-1545437-Add-options-to-specify-Rust-target-name.patch:
- Dropped, included upstream
-- Marco Trevisan (Treviño) <marco@ubuntu.com> Wed, 03 Jun 2020 22:16:07 +0200
mozjs68 (68.7.0-1) unstable; urgency=medium
[ Marco Trevisan (Treviño) ]
* New upstream release
[ William Grant ]
* Add riscv64 support (Closes: #932893)
-- Jeremy Bicha <jbicha@debian.org> Tue, 28 Apr 2020 21:12:05 -0400
mozjs68 (68.6.0-2) unstable; urgency=medium
* control: Use clang and LLVM 10. The default clang is not installable
currently due to bug #954826 - work around this by explicitly picking a
different version.
-- Iain Lane <laney@debian.org> Fri, 27 Mar 2020 11:04:21 +0000
mozjs68 (68.6.0-1) experimental; urgency=medium
[ Marco Trevisan (Treviño) ]
* New upstream release with security fixes:
- https://www.mozilla.org/en-US/firefox/68.6.0/releasenotes/
[ Simon McVittie ]
* d/rules: Swap --host and --target.
The Mozilla build system's terminology is not the same as Autoconf's,
and previous versions of the mozjs Debian packages got the translation
wrong.
* d/rules: Ask moz.configure to give us more information
* d/rules: Print complete config.log if configure fails
* d/rules: Forcibly set appropriate values for the Rust targets,
fixing FTBFS on armel (Closes: #954774)
-- Iain Lane <laney@debian.org> Thu, 26 Mar 2020 17:52:32 +0000
mozjs68 (68.5.0-2) experimental; urgency=medium
* rules, test.sh: Allow tests to be excluded by name and exclude
basic/bug653153.js. Without more extensive patching, this test can't be
skipped conditionally based on dpkg architecture. It fails on i386 because
of a new test that was added in mozjs68 - the behaviour is the same as it
was previously. This has been reported upstream:
https://bugzilla.mozilla.org/show_bug.cgi?id=1621900
-- Iain Lane <laney@debian.org> Thu, 12 Mar 2020 12:14:53 +0000
mozjs68 (68.5.0-1) experimental; urgency=medium
[ Marco Trevisan (Treviño) ]
* New upstream release (from firefox-68.5.0 ESR tarball)
* debian/watch,
debian/upstream/signing-key.asc:
- Set URI to point to latest Firefox ESR
- Ensure we check the tarball signature with upstream key
* debian/gbp.conf:
- Filter undeeded files for SpiderMonkey from firefox tarball
- Only export to upstream/latest the mozjs tarball content as
generated by upstream js/src/make-source-package.sh tool using
the wrapper debian/gbp-repack-mozjs-source.sh as postunpack tool
* debian/copyright: Dropped Files-Excluded, as we rely on gbp.conf now
* debian/control:
- Bump Standards-Version to 4.5.0 (no changes required)
* debian/rules:
- Set python2.7 as default
- compile with --enable-unaligned-private-values (as requested by gjs)
- Don't hardcode the upstream version and use versioned lib
- Don't compile with -fno-schedule-insns(2) (not needed anymore with gcc9)
- Use default gcc optimizations in all the architectures, a part for ia64
* debian/patches:
- Refreshed
- d/p/tests-Skip-a-test-on-s390x.patch:
+ The tests now correctly pass under s390
- d/p/tests-Adapt-formatted-strings-results-to-system-ICU.patch,
- d/p/Skip-some-i18n-tests-because-we-are-now-using-system-ICU.patch:
+ Adapt ICU-related tests to system instead of skipping them
- d/p/Skip-time-zone-tests-that-fails-with-system-ICU.patch:
+ Skip time-zone tests that fails with system ICU
* debian/tests.sh:
- Set HOME directory to a temp directory (may be used by cache)
- Run basic JIT tests
[ Iain Lane ]
* debian/control:
- BD on rustc and cargo, now required for the build
- BD on python2-dev instead of python-dev
- BD on clang, libclang-dev and llvm
- Use the new debhelper-compat method of specifying compat level,
and bump to 12
* debian/rules:
- Don't explicitly enable PIE
- Explicitly use gcc/g++
* debian/patches:
- d/p/sdk-mozjs68-import-a-patch-to-workaround-arm-compilation-.patch:
+ Fix ARM build (as suggested by gjs upstream)
- d/p/TestingFunctions-Update-ICU-s-default-tz-when-setting-TZ.patch:
+ Update the ICU default timezone when setting it directly via setenv
- d/p/time-zone-path-test-Update-for-what-our-system-ICU-return.patch:
+ Fixing some values being tested for to correspond to our ICU
-- Marco Trevisan (Treviño) <marco@ubuntu.com> Mon, 24 Feb 2020 13:11:22 +0000
mozjs60 (60.8.0-2) unstable; urgency=medium
* Release to unstable
-- Jeremy Bicha <jbicha@debian.org> Tue, 01 Oct 2019 07:07:32 -0400
mozjs60 (60.8.0-1) experimental; urgency=medium
[ Iain Lane ]
* New upstream release (re-imported to respect Files-Excluded from
debian/copyright).
[ Philip Chimento ]
* d/p: Manually refreshed patches due to style reformatting of Mozilla
codebase having been applied upstream since the last update
* d/p: Dropped tests-snans-be.patch, already applied upstream
* d/p: Dropped Bug-1560064-Barrier-Set-MovableCellHasher-JSObject-visibi,
already applied upstream
* d/p: Added patches from https://bugzilla.mozilla.org/show_bug.cgi?id=1375074
and https://bugzilla.mozilla.org/show_bug.cgi?id=1445907 to fix JIT
crashes on aarch64, based on the suggestion in
https://bugzilla.mozilla.org/show_bug.cgi?id=1527311 (LP: #1842927)
-- Iain Lane <laney@debian.org> Mon, 09 Sep 2019 16:05:19 +0100
# Older entries have been removed from this changelog.
# To read the complete changelog use `apt changelog libmozjs-102-dev`.
Generated by dwww version 1.15 on Wed Jun 26 14:45:11 CEST 2024.