Age | Commit message (Collapse) | Author |
|
Previously, we temporarily switched on RESTRICT="mirror" for
an ebuild before updating the manifest. This allows us to fetch
distfiles from their original locations (rather than the default,
which is to require them to be fetched from mirrors). We don't
actually need this, so this change removes the code that does
this.
BUG=None
TEST=unit tests, also tested on rust-1.60.0 uprev
Change-Id: I5f29ffad83a5826dbe523db4657d9ea17c43bcff
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3594132
Commit-Queue: Bob Haarman <inglorion@chromium.org>
Reviewed-by: George Burgess <gbiv@chromium.org>
Tested-by: Bob Haarman <inglorion@chromium.org>
|
|
When making some changes, I got some complaints from the formatters.
Fixing the formatting first, before making edits.
BUG=None
TEST=./run_tests_for.py rust_tools/rust_uprev.py
Change-Id: I4802b40dfeb3348cf2c060737992250b75136e0e
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3788604
Commit-Queue: Bob Haarman <inglorion@chromium.org>
Reviewed-by: Michael Benfield <mbenfield@google.com>
Tested-by: Bob Haarman <inglorion@chromium.org>
|
|
VS Code (and possibly other editors) like to pass `--version` and
possibly other arguments, so forward them to `rust-analyzer`.
BUG=b:240341002
TEST=Build, test on VS Code
Change-Id: Icadf05088c4220f058fcdce68489d266368ab1e7
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3785120
Commit-Queue: George Burgess <gbiv@chromium.org>
Auto-Submit: Michael Benfield <mbenfield@google.com>
Reviewed-by: George Burgess <gbiv@chromium.org>
Tested-by: Michael Benfield <mbenfield@google.com>
|
|
It's well past 2019. Let's remove this.
Style changes are courtesy of `yapf`.
BUG=None
TEST=None
Change-Id: Ic70f01775958fc61790901028569945f76eeb763
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3785491
Tested-by: George Burgess <gbiv@chromium.org>
Commit-Queue: George Burgess <gbiv@chromium.org>
Auto-Submit: George Burgess <gbiv@chromium.org>
Reviewed-by: Jordan Abrahams-Whitehead <ajordanr@google.com>
|
|
This argument is not actually used anymore. Technically
it was never actually needed and it was passed in
other ways to the patch_manager.py even before the
refactoring. This just removes it from one function's
arguments.
It's hard to clean the forwarded patch_metadata_file arg
in every situation, so let's just do it for this function
that we introduced.
BUG=None
TEST=./run_tests_for llvm_tools/
Change-Id: I5c3f2a217df082cd51362aa70c8b585ad5135e47
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3780680
Tested-by: Jordan Abrahams-Whitehead <ajordanr@google.com>
Commit-Queue: Jordan Abrahams-Whitehead <ajordanr@google.com>
Reviewed-by: Manoj Gupta <manojgupta@chromium.org>
|
|
At present, update_chromeos_llvm_hash.py will keep
trying to apply patches from separate llvm subprojects,
to the same LLVM dir. This causes patches to stack,
so that the second time it applies the same patch
it then fails to apply.
This commit fixes the issue by cleaning up the git
directory on every stage of the loop.
This commit also fixes some type casts.
BUG=b:239279349, b:239280701
TEST=./update_chromeos_llvm_hash_unittest.py
TEST=./update_packages_and_run_tests.py \
--extra_change_lists 1394249 1986966 \
--chroot_path $CROS_ROOT \
--llvm_version google3 cq
Change-Id: I6cc853d80c47fefaba4ff0b5133787b716177567
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3780679
Commit-Queue: Jordan Abrahams-Whitehead <ajordanr@google.com>
Tested-by: Jordan Abrahams-Whitehead <ajordanr@google.com>
Reviewed-by: Christopher Di Bella <cjdb@google.com>
Commit-Queue: Christopher Di Bella <cjdb@google.com>
Reviewed-by: Manoj Gupta <manojgupta@chromium.org>
|
|
At present, programs which import the now deleted
llvm_patch_management break. This commit removes the
dependency, and moves the one relevant function from
llvm_patch_management to update_chromeos_llvm_hash.
Some other developer quality of life additions have been added,
such as type hints, to get the new patch_manager implementation
to validate static checks.
BUG=b:239280701, b:239279349, b:237870186
TEST=get_upstream_patch.py ...
TEST=run_tests_for llvm_tools/*
Change-Id: I2a7880c9f3b285109f6b115ad8b22eccbf0b714f
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3777524
Commit-Queue: Jordan Abrahams-Whitehead <ajordanr@google.com>
Reviewed-by: Christopher Di Bella <cjdb@google.com>
Tested-by: Jordan Abrahams-Whitehead <ajordanr@google.com>
Reviewed-by: George Burgess <gbiv@chromium.org>
|
|
This is a wrapper program enabling use of `rust-analyzer` running in the
chroot with an editor outside the chroot.
BUG=b:235120448
TEST=tested with Neovim
Change-Id: I64287071ce6cc26c6848b6fb09743f6df9fac311
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3715832
Reviewed-by: George Burgess <gbiv@chromium.org>
Commit-Queue: Michael Benfield <mbenfield@google.com>
Tested-by: Michael Benfield <mbenfield@google.com>
Reviewed-by: Michael Benfield <mbenfield@google.com>
|
|
Update chromeos-kernel-4.4
Update chromeos-kernel-4.14
Update chromeos-kernel-4.19
Update chromeos-kernel-5.4
Update chromeos-kernel-5.10
BUG=None
TEST=Verified in kernel-release-afdo-verify-orchestrator
Change-Id: Ic6bea362338745eff5f67b50207b14c1447f36ac
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3765945
Commit-Queue: Manoj Gupta <manojgupta@chromium.org>
Reviewed-by: Manoj Gupta <manojgupta@chromium.org>
Tested-by: Denis Nikitin <denik@chromium.org>
Auto-Submit: Denis Nikitin <denik@chromium.org>
|
|
This code is not used by anything on the ChromeOS side,
and is just a wrapper around patch_manager.py itself.
The README implies that it auto-fills the command line
arguments, but this is somewhat unneccesary, and if we
want to add support for this workflow, it should likely
be in patch_manager.py itself as a subcommand.
This commit removes llvm_patch_management.py and
llvm_patch_management_unittest.py, and also
removes it from the README.md.
Also removes other incorrect flags from the README.md.
BUG=b:188465085
TEST=./run_tests_for.py llvm_tools/*
Change-Id: Ibf5f77977f70a8b7334e111a92a8fec5be462201
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3765939
Reviewed-by: Manoj Gupta <manojgupta@chromium.org>
Tested-by: Jordan Abrahams-Whitehead <ajordanr@google.com>
Commit-Queue: Jordan Abrahams-Whitehead <ajordanr@google.com>
Reviewed-by: George Burgess <gbiv@chromium.org>
|
|
At present, patch_manager.py and related files
had lots of dead code (>50% of the lines were not used).
This commit removes them, and removes the tests which
tested that unused code. Test concepts that were useful
were either already covered, or have been added as well
in this commit.
This commit also cleans up some imports and moves some
functions for better access to testing.
BUG=b:188465085
TEST=Preupload tests pass
TEST=llvm builds successfully locally with copied llvm_tools/
Change-Id: I202952a1f5d53a1182c1c6c51b8072573c5cbc5d
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3761449
Commit-Queue: Jordan Abrahams-Whitehead <ajordanr@google.com>
Tested-by: Jordan Abrahams-Whitehead <ajordanr@google.com>
Reviewed-by: George Burgess <gbiv@chromium.org>
Reviewed-by: Manoj Gupta <manojgupta@chromium.org>
|
|
At present, patch_sync intentionally does not
copy over version changes to existing patches, it only
copies new patches to the other repositories.
This commit changes this behaviour at the request of the
Android toolchain team, so that now version range
changes are also copied over (but nothing else).
BUG=b:237030928
TEST=cargo test
TEST=patch_sync locally with version range changes
Change-Id: I115fa02012cdf663a2b5b5657e769f513241dedd
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3756345
Reviewed-by: George Burgess <gbiv@chromium.org>
Tested-by: Jordan Abrahams-Whitehead <ajordanr@google.com>
Commit-Queue: Jordan Abrahams-Whitehead <ajordanr@google.com>
|
|
These copyright lines should be added, and they're currently
missing. Failing presubmit checks without this change.
BUG=None
TEST=Presubmit checks pass
Change-Id: I9cd3b9e26a4d7284974705aa96cb6050db25e767
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3756344
Auto-Submit: Jordan Abrahams-Whitehead <ajordanr@google.com>
Reviewed-by: George Burgess <gbiv@chromium.org>
Tested-by: Jordan Abrahams-Whitehead <ajordanr@google.com>
Commit-Queue: George Burgess <gbiv@chromium.org>
|
|
Update chromeos-kernel-4.4
Update chromeos-kernel-4.14
Update chromeos-kernel-4.19
Update chromeos-kernel-5.4
Update chromeos-kernel-5.10
BUG=None
TEST=Verified in kernel-release-afdo-verify-orchestrator
Change-Id: I65702c5f9e1342eaba5707546f049df6f7eb5fd6
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3748693
Tested-by: Denis Nikitin <denik@chromium.org>
Commit-Queue: Christopher Di Bella <cjdb@google.com>
Reviewed-by: Christopher Di Bella <cjdb@google.com>
Auto-Submit: Denis Nikitin <denik@chromium.org>
|
|
This CL updates OWNERS.toolchain with recent changes to the ChromeOS
Toolchain Team.
BUG=b:237711871
TEST=presubmit
Change-Id: I66b02236664f04b49cd5a7a81c17ccb5b6fc86e3
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3741464
Tested-by: George Burgess <gbiv@chromium.org>
Reviewed-by: George Burgess <gbiv@chromium.org>
Commit-Queue: George Burgess <gbiv@chromium.org>
|
|
As outlined in the attached bug, we want to be able to crash the
compiler based on a handful of heuristics. Crashing Clang helps get us
self-contained reproducers fairly easily.
This CL provides (off-by-default) functionality to do the above. The
expectation is that a SWE will hack at it to make it work as they need
to.
BUG=b:236736327
TEST=Installed the new wrapper; observed autocrashes.
Change-Id: I76ec753ec37baa5e9b6dab92668081fa7c605725
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3714885
Reviewed-by: Manoj Gupta <manojgupta@chromium.org>
Reviewed-by: Jordan Abrahams-Whitehead <ajordanr@google.com>
Commit-Queue: George Burgess <gbiv@chromium.org>
Tested-by: George Burgess <gbiv@chromium.org>
|
|
Currently, get_upstream_patch does not validate that a patch applies to
the current LLVM state.
Add validation before modifying PATCHES.json.
Move several functions into patch_utils to avoid depending on
patch_manager.
BUG=b:227216280
TEST=./get_upstream_patch.py --platform chromiumos --sha [patch SHA]
Change-Id: I97e7d401e7f8fc6d85dbfb9a310e4a77205ef444
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3711269
Reviewed-by: Adrian Dole <adriandole@google.com>
Commit-Queue: Adrian Dole <adriandole@google.com>
Auto-Submit: Adrian Dole <adriandole@google.com>
Reviewed-by: Jordan Abrahams-Whitehead <ajordanr@google.com>
Tested-by: Adrian Dole <adriandole@google.com>
|
|
BUG=None
TEST=manually running the script
Change-Id: I53fa857210b5f13edb318aba407b945a5de98f0d
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3715833
Reviewed-by: George Burgess <gbiv@chromium.org>
Tested-by: Michael Benfield <mbenfield@google.com>
Commit-Queue: Michael Benfield <mbenfield@google.com>
|
|
Update chromeos-kernel-4.4
Update chromeos-kernel-4.14
Update chromeos-kernel-4.19
Update chromeos-kernel-5.4
Update chromeos-kernel-5.10
BUG=None
TEST=Verified in kernel-release-afdo-verify-orchestrator
Change-Id: I3e39be51d01147f17059f4c94df085185eea1561
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3722453
Reviewed-by: Denis Nikitin <denik@chromium.org>
Commit-Queue: Denis Nikitin <denik@chromium.org>
Tested-by: Manoj Gupta <manojgupta@chromium.org>
Auto-Submit: Manoj Gupta <manojgupta@chromium.org>
Commit-Queue: Manoj Gupta <manojgupta@chromium.org>
|
|
With CL:3710850, pie will be defaults for clang and GCC
(cross-compiles). No need to manage them separately.
BUG=b:190047257
TEST=go test
Change-Id: Icf8e74d6a31c5de678ec9a6a5c321a17a0154d37
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3722436
Reviewed-by: Jordan Abrahams-Whitehead <ajordanr@google.com>
Reviewed-by: George Burgess <gbiv@chromium.org>
Auto-Submit: Manoj Gupta <manojgupta@chromium.org>
Tested-by: Manoj Gupta <manojgupta@chromium.org>
Commit-Queue: George Burgess <gbiv@chromium.org>
|
|
Update chromeos-kernel-4.4
Update chromeos-kernel-4.14
Update chromeos-kernel-4.19
Update chromeos-kernel-5.4
Update chromeos-kernel-5.10
BUG=None
TEST=Verified in kernel-release-afdo-verify-orchestrator
Change-Id: Icde18d30e128cfb4fbc26414ec747403392541a0
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3708110
Tested-by: Denis Nikitin <denik@chromium.org>
Commit-Queue: Michael Benfield <mbenfield@google.com>
Reviewed-by: Denis Nikitin <denik@chromium.org>
|
|
Previous restructing did not handle the bisection
case, as it was considerably more complicated than
other failure modes for patch_manager.py
This commit changes the operation of bisection in
patch_manager.py quite drastically, as it now
assumes that the git bisection process can occur
outside of patch_manager.py, rather than internally.
BUG=b:188465085
TEST=./patch_manager_unittests.py
TEST=./patch_manager.py --failure_mode bisect <...> \
# With a patch that applied far too late
Change-Id: I156373dc13bcbd3d297f8537fb6fb77fe9f0e9a5
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3704539
Commit-Queue: Jordan Abrahams-Whitehead <ajordanr@google.com>
Tested-by: Jordan Abrahams-Whitehead <ajordanr@google.com>
Reviewed-by: George Burgess <gbiv@chromium.org>
|
|
We should print out what we actually did, instead of just
exiting silently when everything goes well. This is
informative to the user running patch_manager.py code.
BUG=b:188465085
TEST=./patch_manager_unittest.py
Change-Id: Idcb203d0c3e28d1b00e7b9503334863b7844d033
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3706059
Reviewed-by: George Burgess <gbiv@chromium.org>
Tested-by: Jordan Abrahams-Whitehead <ajordanr@google.com>
Commit-Queue: Jordan Abrahams-Whitehead <ajordanr@google.com>
|
|
At present, I made a mistake where patch_utils expected
certain keys to exist. However, the original patch_manager.py
didn't. This was further broken by the fact that patch_sync.py
would avoid serializing the 'platforms' field if empty, causing
a back and forth between patch_utils.py and patch_sync.
Update the unittests to match and verify we're doing the correct
thing.
BUG=None
TEST=./patch_utils_unittest.py
TEST=./patch_manager_unittest.py
Change-Id: Ib02c9d552848831f395b006de9a28ea4292b82f7
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3704542
Tested-by: Jordan Abrahams-Whitehead <ajordanr@google.com>
Reviewed-by: George Burgess <gbiv@chromium.org>
Commit-Queue: Jordan Abrahams-Whitehead <ajordanr@google.com>
|
|
This is an attempt to gradually replace some of the
behavior of patch_manager.py with a more extendable
structure.
Instead of having the same code handle every patch_manager.py
mode, instead we can have a dispatch for supported modes,
and the fallback to legacy when we can't separate the
existing behavior.
This does not change the external API of patch_manager.py
at all. All unittests still pass, and we're still
applying patches correctly as expected.
BUG=b:188465085, b:227216280
TEST=./patch_manager_unittest.py
TEST=cp patch_manager.py patch_utils.py ${llvm_files}/patch_manager/ \
&& sudo emerge llvm
Change-Id: I43d26d4e903140ce2e490624aaac15d0bae898cd
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3661358
Tested-by: Jordan Abrahams-Whitehead <ajordanr@google.com>
Reviewed-by: George Burgess <gbiv@chromium.org>
Commit-Queue: Jordan Abrahams-Whitehead <ajordanr@google.com>
|
|
This allows us to provide more information about the compiler being
wrapped to the wrapper. The intended use of this is to embed hashes of
the compiler being wrapped into the wrapper itself, so as the compiler
changes, the wrapper's SHA changes with it.
While I'm in the area, fix up a comment, and apply this flag to our
manual wrapper installation script.
BUG=b:222321317
TEST=`emerge llvm`
Change-Id: I20380722e1e539d51fe7ea708c63dad696d84c87
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3696434
Reviewed-by: Denis Nikitin <denik@chromium.org>
Tested-by: George Burgess <gbiv@chromium.org>
Commit-Queue: George Burgess <gbiv@chromium.org>
|
|
This adds the json_to_patch_entries function
which abstracts away a common pattern to convert
the PATCHES.json file contents into patch_entries
eagerly.
Fixes some typing issues too that have presented
themselves in later CLs.
BUG=None
TEST=./patch_utils_unittest.py
Change-Id: I5ecb1aedbbfb4f04176c021ff976da417319c17e
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3699193
Reviewed-by: George Burgess <gbiv@chromium.org>
Tested-by: Jordan Abrahams-Whitehead <ajordanr@google.com>
Commit-Queue: Jordan Abrahams-Whitehead <ajordanr@google.com>
|
|
This CL adds a flag to clang tidy invocations to prevent supressed lints
in header files.
BUG=b:187790543
TEST=None
Change-Id: I9977818894899fc6c28f0c3d4121326ca854e1b3
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3694553
Commit-Queue: Ryan Beltran <ryanbeltran@chromium.org>
Tested-by: Ryan Beltran <ryanbeltran@chromium.org>
Reviewed-by: George Burgess <gbiv@chromium.org>
|
|
This allows a utility function which can write to an
arbitrary file without risking an incomplete write error,
causing issues and creating an invalid edit.
This function ensures that the file is only swapped if the
file write was successful, otherwise the file is deleted.
BUG=None
TEST=./patch_utils_unittest.py
Change-Id: Iedc3297b0e59d216f027e6ff125f92bc4d088c7d
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3685569
Reviewed-by: George Burgess <gbiv@chromium.org>
Commit-Queue: Jordan Abrahams-Whitehead <ajordanr@google.com>
Tested-by: Jordan Abrahams-Whitehead <ajordanr@google.com>
|
|
This is just some general clean up of
patch_manager.py before any structural changes.
Mostly sorting imports, re-doing some typing.
BUG=None
TEST=./patch_manager_unittest.py
Change-Id: I96d6db4efb20ed4b934a39defe9171de5e19d450
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3673493
Commit-Queue: Jordan Abrahams-Whitehead <ajordanr@google.com>
Reviewed-by: George Burgess <gbiv@chromium.org>
Tested-by: Jordan Abrahams-Whitehead <ajordanr@google.com>
|
|
This introduces the patch_utils.py library, which
contains various PATCHES.json and patch_manager
utilities that will be useful for future patch manager
restructuring.
In particular, patch_manager.py doesn't explain why its
patches fail, or give any information as to what is wrong with
its patch applications.
patch_utils.py provides the PatchEntry class, which is a self
contained object which can provide this diagnostic information.
This module will later be incorporated into patch_manager.py
and get_upstream_patches.py
BUG=b:188465085, b:227216280
TEST=./patch_utils_unittest.py
Change-Id: I6f6e24e6449ea68f6751fbcad14fca76c1bbaec8
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3648887
Tested-by: Jordan Abrahams-Whitehead <ajordanr@google.com>
Reviewed-by: George Burgess <gbiv@chromium.org>
Commit-Queue: Jordan Abrahams-Whitehead <ajordanr@google.com>
|
|
We don't need the ".cros" suffix anymore, and its presence
causes the ping check for at least one host to fail.
BUG=b:231402615
TEST=Checking lab machines should pass after this
Change-Id: I620db48c6f92837c12f86886eb9e87168dfc6ed7
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3661356
Reviewed-by: Denis Nikitin <denik@chromium.org>
Tested-by: Bob Haarman <inglorion@chromium.org>
Commit-Queue: Bob Haarman <inglorion@chromium.org>
|
|
Update chromeos-kernel-4.4
Update chromeos-kernel-4.14
Update chromeos-kernel-4.19
Update chromeos-kernel-5.4
Update chromeos-kernel-5.10
BUG=None
TEST=Verified in kernel-release-afdo-verify-orchestrator
Change-Id: I882dfa16066e0048c1ed3640dcbb29c697786e23
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3656068
Commit-Queue: Manoj Gupta <manojgupta@chromium.org>
Auto-Submit: Denis Nikitin <denik@chromium.org>
Tested-by: Denis Nikitin <denik@chromium.org>
Reviewed-by: Manoj Gupta <manojgupta@chromium.org>
|
|
Currently in get_upstream_patch.py, it's possible for
the "until" value to be earlier than "from", which means
the patch can never apply. This CL reports this as an error
now.
Applies some pre-upload autofixes as well, and adds
minor documentation for the --differential flag to describe
some non-obvious behavior.
BUG=None
TEST=get_upstream_patch.py with invalid patch versions
Change-Id: Ie8338f8b9b27f0c41f0e350706a3131e5daab1b2
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3653987
Reviewed-by: Manoj Gupta <manojgupta@chromium.org>
Commit-Queue: Jordan Abrahams-Whitehead <ajordanr@google.com>
Tested-by: Jordan Abrahams-Whitehead <ajordanr@google.com>
Reviewed-by: George Burgess <gbiv@chromium.org>
|
|
Clean up code which handles ".cros" suffix.
Crosfleet, ssh and cros shell don't require .cros suffix in remotes.
Snappy device is back. Add it to lab machines checks.
BUG=b:231402615
TEST=tested locally
Change-Id: I6e9a308428de05b8e84891933bdc19c55e18d08e
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3651281
Commit-Queue: Denis Nikitin <denik@chromium.org>
Reviewed-by: Manoj Gupta <manojgupta@chromium.org>
Tested-by: Denis Nikitin <denik@chromium.org>
|
|
Update chromeos-kernel-4.4
Update chromeos-kernel-4.14
Update chromeos-kernel-4.19
Update chromeos-kernel-5.4
Update chromeos-kernel-5.10
BUG=None
TEST=Verified in kernel-release-afdo-verify-orchestrator
Change-Id: If3f7a828c656a6534f7cf93aa7beed5666e1ed71
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3647287
Commit-Queue: Jordan Abrahams-Whitehead <ajordanr@google.com>
Auto-Submit: Denis Nikitin <denik@chromium.org>
Reviewed-by: Jordan Abrahams-Whitehead <ajordanr@google.com>
Tested-by: Denis Nikitin <denik@chromium.org>
|
|
To avoid cases like the attached bug, simply log when a dir doesn't
exist. Doesn't seem bad to consider this a success.
BUG=b:232843376
TEST=None
Change-Id: I291d1b1ca4007d2402a6707c83ce483470099c84
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3651935
Auto-Submit: George Burgess <gbiv@chromium.org>
Reviewed-by: Jordan Abrahams-Whitehead <ajordanr@google.com>
Commit-Queue: Jordan Abrahams-Whitehead <ajordanr@google.com>
Tested-by: George Burgess <gbiv@chromium.org>
|
|
For sanitizers like `return`, `builtin`, etc., we have no reason to also
drop FORTIFY checks.
BUG=b:231357370
TEST=`go test`
Change-Id: I1e349a4f6743e549b7bc0899a307b10683bb42e9
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3651188
Reviewed-by: Manoj Gupta <manojgupta@chromium.org>
Commit-Queue: George Burgess <gbiv@chromium.org>
Tested-by: George Burgess <gbiv@chromium.org>
|
|
`-Wl,-z -Wl,defs` should be passed as separate flags.
BUG=b:231357370
TEST=go test
Change-Id: Iacd0be0c74df0cc7bd8607473a859c36c5ef4f06
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3651187
Commit-Queue: Manoj Gupta <manojgupta@chromium.org>
Auto-Submit: George Burgess <gbiv@chromium.org>
Reviewed-by: Manoj Gupta <manojgupta@chromium.org>
Tested-by: George Burgess <gbiv@chromium.org>
|
|
During the libxcrypt build, both "-Wl,-z,defs" and its
split equivalent "-Wl,-z -Wl,defs" are used which create
problems because only the unified version is supported.
This adds support for filtering the flags seprately
and two tests to verify the functionality.
BUG=b:187795307
TEST=Local builds with cross-*/libxcrypt; CQ.
Change-Id: If48499f5c8e552e28c7cefd2d959e40f4757a88f
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3641843
Tested-by: Adrian Ratiu <adrian.ratiu@collabora.corp-partner.google.com>
Reviewed-by: Manoj Gupta <manojgupta@chromium.org>
Reviewed-by: George Burgess <gbiv@chromium.org>
Commit-Queue: Manoj Gupta <manojgupta@chromium.org>
Tested-by: Manoj Gupta <manojgupta@chromium.org>
|
|
BUG=b:231402615
TEST=ping <dut>
Change-Id: I40b3c47f9c5209501cb2a15457a4c640887e8733
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3642336
Tested-by: Denis Nikitin <denik@chromium.org>
Commit-Queue: Denis Nikitin <denik@chromium.org>
Reviewed-by: Ryan Beltran <ryanbeltran@chromium.org>
|
|
Pre-upload checks in chromiumos-overlay complained about these files not
being `go fmt`'ed; do that.
This is apparently an artifact of a build tag migration upstream:
https://go.googlesource.com/proposal/+/master/design/draft-gobuild.md
...so leave both in here for a while.
BUG=b:232114933
TEST=go test
Change-Id: I8f8e0896d3e3ce37d155035d93412f8c1c89b1b7
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3639686
Tested-by: George Burgess <gbiv@chromium.org>
Commit-Queue: Jordan Abrahams-Whitehead <ajordanr@google.com>
Reviewed-by: Jordan Abrahams-Whitehead <ajordanr@google.com>
Auto-Submit: George Burgess <gbiv@chromium.org>
|
|
Wimplicit-int has been promoted to be an default error
even with Wno-error. Disable the error by default since it is breaking
a lot of packages.
BUG=b:231987783
TEST=cq
Change-Id: If88877cb63cdd5c392fcf05c2940751547466213
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3642338
Tested-by: Manoj Gupta <manojgupta@chromium.org>
Reviewed-by: George Burgess <gbiv@chromium.org>
Commit-Queue: George Burgess <gbiv@chromium.org>
Auto-Submit: Manoj Gupta <manojgupta@chromium.org>
|
|
BUG=b:232114933
TEST=go test
Change-Id: Iffc686c47c21017a31b9fb69e1f2764390fe29f3
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3639683
Reviewed-by: Jordan Abrahams-Whitehead <ajordanr@google.com>
Commit-Queue: George Burgess <gbiv@chromium.org>
Tested-by: George Burgess <gbiv@chromium.org>
|
|
This CL sorts clang flags, and puts ones shared across all CrOS
configurations into their own textual blocks. This should make factoring
these out into their own function not require golden updates, which
makes verifying the CL that actually _does_ that simpler.
BUG=b:232114933
TEST=go test
Change-Id: I7dc6110d680505d4ad2af98709730e85c386ae5d
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3639682
Reviewed-by: Jordan Abrahams-Whitehead <ajordanr@google.com>
Tested-by: George Burgess <gbiv@chromium.org>
Commit-Queue: George Burgess <gbiv@chromium.org>
|
|
Since order doesn't matter, make this sorted. Otherwise, these are all
shared, so no point in repeating them.
BUG=b:232114933
TEST=go test
Change-Id: Ide3c620de9018fed3fe8d898b21b3d5d4f42700f
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3639281
Reviewed-by: Jordan Abrahams-Whitehead <ajordanr@google.com>
Commit-Queue: George Burgess <gbiv@chromium.org>
Tested-by: George Burgess <gbiv@chromium.org>
|
|
We do nothing with these but immediately deref them. It's simpler (and
should be infinitesimally faster) to just have values here.
BUG=None
TEST=go test
Change-Id: I6df8eda8f36032e856f9abad3e090c62c9d6beb0
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3639684
Commit-Queue: George Burgess <gbiv@chromium.org>
Reviewed-by: Jordan Abrahams-Whitehead <ajordanr@google.com>
Tested-by: George Burgess <gbiv@chromium.org>
|
|
Wimplicit-function-declaration has been promoted to be an default error
even with Wno-error. Disable the error by default since it is breaking
a lot of packages.
BUG=b:230345382
TEST=cq
Change-Id: Ib46b0b47a3e1dea3797739370a4d9cf92a4d25f9
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3637117
Tested-by: Manoj Gupta <manojgupta@chromium.org>
Auto-Submit: Manoj Gupta <manojgupta@chromium.org>
Reviewed-by: George Burgess <gbiv@chromium.org>
Commit-Queue: Manoj Gupta <manojgupta@chromium.org>
|
|
This CL removes -checks=*` from the Clang Tidy flags when
WITH_TIDY=tricium which allows projects to provide their own
configuration files for clang tidy.
BUG=b:187790543
TEST=manually tested
Change-Id: Ic6893a09146f071874f21f7bedb04a1f85e83284
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3635256
Reviewed-by: George Burgess <gbiv@chromium.org>
Commit-Queue: Ryan Beltran <ryanbeltran@chromium.org>
Tested-by: Ryan Beltran <ryanbeltran@chromium.org>
|
|
This requires a few things:
- Running `isort` to begin with
- Syncing with Chromite's `yapf` config, so `yapf` and `isort` don't
disagree
- ...Which itself requires upgrading `yapf`, since these disagree with
`yapf`'s current configuration in toolchain-utils. The most recent
chromite configuration requires v0.31.0, and depot_tools only has
v0.27.0.
This CL does all of these.
BUG=b:231985625
TEST=Ran on a few files in toolchain-utils
Change-Id: I43414abea7b75790b129e78708903ed90dae6ab0
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/toolchain-utils/+/3636138
Reviewed-by: Jordan Abrahams-Whitehead <ajordanr@google.com>
Commit-Queue: George Burgess <gbiv@chromium.org>
Tested-by: George Burgess <gbiv@chromium.org>
|