Two changes happened during the LLVM 11 timeframe that break our Searchfox plugin.
First, the conversion from `llvm::StringRef` to `std::string` became explicit: adcd026838 This is easy enough to fix in a version-agnostic way.
Second, `mangleCXXCtor` no longer exists: 29e1a16be8 Since there isn't a one-size-fits-all fix, I had to use an ifdef. I mostly cargo-culted the change from 29e1a16be8 (diff-dac09655ff6a54658c320a28a6ea297c).
Differential Revision: https://phabricator.services.mozilla.com/D83838
LLVM 11 introduces a hard requirement for SDK 10.12 in order to build for Mac. We want to keep building older LLVMs with 10.11 though, so this patch adds some flexibility so that build-clang can make use of whatever SDK package a particular task pulls from tooltool (but still requesting a deployment target of 10.11).
Differential Revision: https://phabricator.services.mozilla.com/D82621
After bug 1651806, we're trying to caution people against running `mach build $A_SPECIFIC_TARGET` because it's not generally supported. `dumbmake` is a piece of infrastructure that attempts to make this use case a little bit more useable, but it was always supposed to be a stopgap. There doesn't seem to be a need for it any more.
Differential Revision: https://phabricator.services.mozilla.com/D83136
This is where the `compare-mozconfig` test properly belongs, and `mozbuild` tests already re-run every time a `mozboot` file changes.
Differential Revision: https://phabricator.services.mozilla.com/D83130
Add an action that will trigger a task that runs
`mach release push-scriptworker-canary`
to test a new scriptworker deployment.
Differential Revision: https://phabricator.services.mozilla.com/D82821
clang/LLVM's build scripts can turn these on on their own, but explicitly
setting what we want is better than guessing. The change is not huge, maybe
~2-3% on the major shared libraries (`libclang`, `libclang-cpp`, `libLLVM`),
about 1% on the overall `.tar.zst` size, but every little bit counts, right?
Differential Revision: https://phabricator.services.mozilla.com/D82896
2020-06-26 Kevin Jacobs <kjacobs@mozilla.com>
* automation/abi-check/expected-report-libssl3.so.txt, automation/abi-
check/previous-nss-release, lib/nss/nss.h, lib/softoken/softkver.h,
lib/util/nssutil.h:
Set version numbers to 3.55 beta
[332ab7db68ba]
2020-06-25 Kevin Jacobs <kjacobs@mozilla.com>
* tests/all.sh:
Bug 1649190 - Run cipher, sdr, and ocsp tests under standard test
cycle.
[f373809abfc0]
2020-06-15 Kevin Jacobs <kjacobs@mozilla.com>
* gtests/common/testvectors/p256ecdsa-sha256-vectors.h,
gtests/common/testvectors/p384ecdsa-sha384-vectors.h,
gtests/common/testvectors/p521ecdsa-sha512-vectors.h,
gtests/common/testvectors_base/test-structs.h,
gtests/common/wycheproof/genTestVectors.py,
gtests/pk11_gtest/pk11_ecdsa_unittest.cc:
Bug 1649226 - Add Wycheproof ECDSA tests.
[41292ff7f545]
2020-06-30 Benjamin Beurdouche <bbeurdouche@mozilla.com>
* lib/pkcs12/p12d.c:
Bug 1649322 - Fix null pointer passed as argument in
pk11wrap/pk11pbe.c:1246 r=kjacobs
[cc43ebf5bf88]
2020-06-30 Danh <congdanhqx@gmail.com>
* coreconf/arch.mk, coreconf/config.mk, lib/freebl/Makefile:
Bug 1646594 - Enable AVX2 if applicable on x86_64 with make 4.3
r=bbeurdouche
[b579895aceb0]
2020-07-02 Benjamin Beurdouche <bbeurdouche@mozilla.com>
* lib/ssl/ssl3con.c:
Bug 1649316 - Prevent memcmp to be called with a zero length in
ssl/ssl3con.c:6621 r=kjacobs
[8fe9213d0551]
2020-07-02 Alexander Scheel <ascheel@redhat.com>
* lib/cryptohi/secvfy.c:
Bug 1649487 - Fix bad assert in VFY_EndWithSignature. r=jcj
[c9438b528103]
2020-07-06 Dana Keeler <dkeeler@mozilla.com>
* automation/abi-check/expected-report-libnss3.so.txt,
gtests/pk11_gtest/pk11_find_certs_unittest.cc, lib/nss/nss.def,
lib/pk11wrap/pk11cert.c, lib/pk11wrap/pk11pub.h:
Bug 1649633 - add PK11_FindEncodedCertInSlot r=kjacobs,jcj
PK11_FindEncodedCertInSlot can be used to determine the PKCS#11
object handle of an encoded certificate in a given slot. If the
given certificate does not exist in that slot, CK_INVALID_HANDLE is
returned.
[32fe710a942f]
* gtests/pk11_gtest/pk11_find_certs_unittest.cc:
Bug 1649633 - follow-up to make test comparisons in
pk11_find_certs_unittest.cc yoda comparisons r=kjacobs
[424dae31a1c1]
2020-07-07 Kevin Jacobs <kjacobs@mozilla.com>
* gtests/pk11_gtest/pk11_rsapkcs1_unittest.cc, lib/freebl/rsapkcs.c:
Bug 1067214 - Check minimum padding in RSA_CheckSignRecover.
r=rrelyea
This patch adds a check to `RSA_CheckSignRecover` enforcing a
minimum padding length of 8 bytes for PKCS #1 v1.5-formatted
signatures. In practice, RSA key size requirements already ensure
this requirement is met, but smaller (read: broken) key sizes can be
used via configuration overrides, and NSS should just follow the
spec.
[e5324bd5a885]
2020-07-08 Kevin Jacobs <kjacobs@mozilla.com>
* gtests/ssl_gtest/libssl_internals.c,
gtests/ssl_gtest/libssl_internals.h,
gtests/ssl_gtest/ssl_record_unittest.cc,
gtests/ssl_gtest/tls_agent.cc, gtests/ssl_gtest/tls_agent.h,
lib/ssl/dtls13con.c, lib/ssl/dtls13con.h, lib/ssl/ssl3con.c,
lib/ssl/ssl3prot.h, lib/ssl/sslspec.h, lib/ssl/sslt.h,
lib/ssl/tls13con.c, lib/ssl/tls13exthandle.c:
Bug 1647752 - Update DTLS 1.3 implementation to draft-38. r=mt
This patch updates DTLS 1.3 to draft-38. Specifically:
# `ssl_ct_ack` value changes from 25 to 26. # AEAD limits in
`tls13_UnprotectRecord` enforce a maximum of 2^36-1 (as we only
support GCM/ChaCha20 AEADs) decryption failures before the
connection is closed. # Post-handshake authentication will no longer
be negotiated in DTLS 1.3. This allows us to side-step the more
convoluted state machine requirements.
[132a87fc8689]
2020-07-09 Benjamin Beurdouche <bbeurdouche@mozilla.com>
* lib/pk11wrap/pk11pbe.c, lib/pkcs12/p12d.c:
Bug 1649322 - Fix null pointer passed as argument in
pk11wrap/pk11pbe.c:1246 r=kjacobs
This is a fixup patch that reverts https://hg.mozilla.org/projects/n
ss/rev/cc43ebf5bf88355837c5fafa2f3c46e37626707a and adds a null
check around the memcpy in question.
[80bea0e22b20]
2020-07-09 J.C. Jones <jjones@mozilla.com>
* lib/softoken/pkcs11.c:
Bug 1651520 - slotLock race in NSC_GetTokenInfo r=kjacobs
Basically, NSC_GetTokenInfo doesn't lock slot->slotLock before
accessing slot after obtaining it, even though slotLock is defined
as its lock. [0]
[0] https://searchfox.org/nss/rev/a412e70e55218aaf670f1f10322fa734d8
a9fbde/lib/softoken/pkcs11i.h#320-321
[58c2abd7404e] [tip]
Differential Revision: https://phabricator.services.mozilla.com/D82466
This is just re-arranging the deckchairs in preparation for adding an
"update programs" build project. Here "update programs" refers to the
various standalone binaries we produce in order to update Firefox.
There's not strong conceptual coherence between them; rather, it's the
pieces that the Install/Update team generally work on.
Differential Revision: https://phabricator.services.mozilla.com/D82643
The current setup sets /some/ flags via CLANGCL_ASFLAGS (handling of x86
is notably missing, for instance), and uses "clang-cl" as the assembler,
assuming it's in $PATH.
This can be simplified by just using `CC`, which will contain the full
path to "clang-cl" and the right flags for the targets, which makes
CLANGCL_ASFLAGS unnecessary.
Differential Revision: https://phabricator.services.mozilla.com/D82660
This is a partial revert of bug 1647628.
The ImageBridgeChild's thread is used to dispatch synchronous tasks by its consumers.
While the background taskqueue is using a single thread threadpool, this would prevent many places to also use a background taskqueue that could end up calling the ImageBridgeChild.
Differential Revision: https://phabricator.services.mozilla.com/D82395
This patch includes content from the following places.
- Lina's "Getting Rusty: How to ship an XPCOM component in Firefox" slide deck.
- Zibi's "Rust <--> C/C++ FFI for newbies" gist.
It also links to Emilio's "FFI patterns #1 - Complex Rust data structures
exposed seamlessly to C++" blog post. I was going to include that content, but
it's very long, so I have omitted it for now.
Differential Revision: https://phabricator.services.mozilla.com/D81963
Source history does not give any good clues about why this line was added in the first place. In any case, LLVM trunk currently has build bustage when threads are disabled. We could work around the bustage and/or wait for a fix, but it seems like threads are a good thing to have in general nowadays. Maybe this could help with LTO build times.
Differential Revision: https://phabricator.services.mozilla.com/D82447
The logic was apparently initially copied from the compositor thread which too got migrated to nsThread in bug 1634253, so we re-use that same logic again.
This allows to remove all use of the Google's MessageLoop.
Differential Revision: https://phabricator.services.mozilla.com/D81974
There's a small race that can happen when the remote decoder gets shutdown during xpcom shutdown; that would cause GetCurrentSerialEventTarget to return null. Leading to an assertion failure in ActorLifecycleProxy thread-safety check when PRemoteDecoderManagerParent gets destroyed.
So we use a background taskqueue instead and cleanup a bit the threading code in there allowed thanks to the TaskQueue ability to not require an explicit shutdown.
Differential Revision: https://phabricator.services.mozilla.com/D81287
Because jar_maker is not in the libs tier, we also rename the libs-%
targets for l10n repacks to l10n-%, which make it clearer what they are
for.
And because multilocale.txt is both a GeneratedFile and a file that is
generated (and installed) via manual build rules, keeping it in the misc
target actually breaks building in toolkit/locales during l10n repacks,
so move it to libs for now.
Differential Revision: https://phabricator.services.mozilla.com/D81766
