Remove an assertion that was overly restrictive and hard to get
correct under all circumstances (i.e., in the presence of incorrect)
code. This matches the code for *Named types in that specific switch.
Fixes#71284.
Change-Id: Ifccf8b73dc70cac9cb1c8b24946d16851d511454
Reviewed-on: https://go-review.googlesource.com/c/go/+/643255
Reviewed-by: Robert Findley <rfindley@google.com>
Auto-Submit: Robert Griesemer <gri@google.com>
Reviewed-by: Robert Griesemer <gri@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Commit generated by update.bash.
For #22487.
Change-Id: Ie5002725b9add75fd22b29ea3a7addbe0151b25c
Reviewed-on: https://go-review.googlesource.com/c/go/+/642020
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
Auto-Submit: Tobias Klauser <tobias.klauser@gmail.com>
Reviewed-by: Ian Lance Taylor <iant@google.com>
If a wasmexport function is called from the host before
initializing the Go Wasm module, currently it will likely fail
with a bounds error, because the uninitialized SP is 0, and any
SP decrement will make it out of bounds.
As at least some Wasm runtime doesn't call _initialize by default,
This error can be common. And the bounds error looks confusing to
the users. Therefore, we detect this case and emit a clearer error.
Fixes#71240.
Updates #65199.
Change-Id: I107095f08c76cdceb7781ab0304218eab7029ab6
Reviewed-on: https://go-review.googlesource.com/c/go/+/643115
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Michael Knyszek <mknyszek@google.com>
These keys are off-spec, but have historically been accepted by
ParsePKCS1PrivateKey.
Thanks to Philippe Antoine (Catena cyber) for reporting this issue.
Fixes#71216
Fixes CVE-2025-22865
Change-Id: I6a6a46564156fa32e29e8d6acbec3fbac47c7352
Reviewed-on: https://go-internal-review.googlesource.com/c/go/+/1820
Reviewed-by: Tatiana Bradley <tatianabradley@google.com>
Reviewed-by: Damien Neil <dneil@google.com>
Commit-Queue: Roland Shoemaker <bracewell@google.com>
(cherry picked from commit 36c6c8b6957e155770461fd710aea9477ef3bc88)
Reviewed-on: https://go-internal-review.googlesource.com/c/go/+/1841
Reviewed-on: https://go-review.googlesource.com/c/go/+/643102
Auto-Submit: Michael Knyszek <mknyszek@google.com>
Reviewed-by: Michael Pratt <mpratt@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Store netrc lines into the credential map backward so that earlier lines
take priority over later lines. This matches Go 1.23 netrc lookup which
stopped at the first match it found.
Additionally, this fixes a security issue related to domain parsing
which could have allowed servers to read credentials belonging to other
servers. The fix was to switch from using path.Dir(currentPrefix) to
strings.Cut(currentPrefix, "/")
Thanks to Juho Forsén of Mattermost for reporting this issue.
Fixes#71249
Fixes CVE-2024-45340
Change-Id: I175a00d6d7f4d31c9e4d79b7cf1c2a0ad35b2781
Reviewed-on: https://go-internal-review.googlesource.com/c/go/+/1781
Reviewed-by: Tatiana Bradley <tatianabradley@google.com>
Commit-Queue: Roland Shoemaker <bracewell@google.com>
Reviewed-by: Roland Shoemaker <bracewell@google.com>
Reviewed-by: Damien Neil <dneil@google.com>
(cherry picked from commit 76833d221aa3ccc978b6f41bd24e26babf771375)
Reviewed-on: https://go-internal-review.googlesource.com/c/go/+/1840
Reviewed-on: https://go-review.googlesource.com/c/go/+/643101
Reviewed-by: Michael Pratt <mpratt@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Auto-Submit: Michael Knyszek <mknyszek@google.com>
When an HTTP redirect changes the host of a request, we drop
sensitive headers such as Authorization from the redirected request.
Fix a bug where a chain of redirects could result in sensitive
headers being sent to the wrong host:
1. request to a.tld with Authorization header
2. a.tld redirects to b.tld
3. request to b.tld with no Authorization header
4. b.tld redirects to b.tld
3. request to b.tld with Authorization header restored
Thanks to Kyle Seely for reporting this issue.
For #70530Fixes#71212
Fixes CVE-2024-45336
Change-Id: Ia58a2e10d33d6b0cc7220935e771450e5c34de72
Reviewed-on: https://go-internal-review.googlesource.com/c/go/+/1641
Reviewed-by: Roland Shoemaker <bracewell@google.com>
Reviewed-by: Tatiana Bradley <tatianabradley@google.com>
Commit-Queue: Roland Shoemaker <bracewell@google.com>
(cherry picked from commit 2889169b87a61f1218a02994feb80fd3d8bfa87c)
Reviewed-on: https://go-internal-review.googlesource.com/c/go/+/1766
Reviewed-on: https://go-review.googlesource.com/c/go/+/643100
Auto-Submit: Michael Knyszek <mknyszek@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Michael Pratt <mpratt@google.com>
When checking URI constraints, use netip.ParseAddr, which understands
zones, unlike net.ParseIP which chokes on them. This prevents zone IDs
from mistakenly satisfying URI constraints.
Thanks to Juho Forsén of Mattermost for reporting this issue.
For #71156Fixes#71209
Fixes CVE-2024-45341
Change-Id: Iecac2529f3605382d257996e0fb6d6983547e400
Reviewed-on: https://go-internal-review.googlesource.com/c/go/+/1700
Reviewed-by: Tatiana Bradley <tatianabradley@google.com>
Reviewed-by: Damien Neil <dneil@google.com>
(cherry picked from commit 22ca55d396ba801e6ae9b2bd67a059fcb30562fd)
Reviewed-on: https://go-internal-review.googlesource.com/c/go/+/1800
Commit-Queue: Roland Shoemaker <bracewell@google.com>
Reviewed-by: Roland Shoemaker <bracewell@google.com>
Reviewed-on: https://go-review.googlesource.com/c/go/+/643099
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Auto-Submit: Michael Knyszek <mknyszek@google.com>
Reviewed-by: Michael Pratt <mpratt@google.com>
These keys are off-spec, but have historically been accepted by
ParsePKCS1PrivateKey.
Thanks to Philippe Antoine (Catena cyber) for reporting this issue.
Fixes#71216
Fixes CVE-2025-22865
Change-Id: I6a6a46564156fa32e29e8d6acbec3fbac47c7352
Reviewed-on: https://go-internal-review.googlesource.com/c/go/+/1820
Reviewed-by: Tatiana Bradley <tatianabradley@google.com>
Reviewed-by: Damien Neil <dneil@google.com>
Commit-Queue: Roland Shoemaker <bracewell@google.com>
Reviewed-on: https://go-review.googlesource.com/c/go/+/643098
Auto-Submit: Michael Knyszek <mknyszek@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Michael Pratt <mpratt@google.com>
Store netrc lines into the credential map backward so that earlier lines
take priority over later lines. This matches Go 1.23 netrc lookup which
stopped at the first match it found.
Additionally, this fixes a security issue related to domain parsing
which could have allowed servers to read credentials belonging to other
servers. The fix was to switch from using path.Dir(currentPrefix) to
strings.Cut(currentPrefix, "/")
Thanks to Juho Forsén of Mattermost for reporting this issue.
Fixes#71249
Fixes CVE-2024-45340
Change-Id: I175a00d6d7f4d31c9e4d79b7cf1c2a0ad35b2781
Reviewed-on: https://go-internal-review.googlesource.com/c/go/+/1781
Reviewed-by: Tatiana Bradley <tatianabradley@google.com>
Commit-Queue: Roland Shoemaker <bracewell@google.com>
Reviewed-by: Roland Shoemaker <bracewell@google.com>
Reviewed-by: Damien Neil <dneil@google.com>
Reviewed-on: https://go-review.googlesource.com/c/go/+/643097
Reviewed-by: Michael Pratt <mpratt@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Auto-Submit: Michael Knyszek <mknyszek@google.com>
When checking URI constraints, use netip.ParseAddr, which understands
zones, unlike net.ParseIP which chokes on them. This prevents zone IDs
from mistakenly satisfying URI constraints.
Thanks to Juho Forsén of Mattermost for reporting this issue.
Fixes#71156
Fixes CVE-2024-45341
Change-Id: Iecac2529f3605382d257996e0fb6d6983547e400
Reviewed-on: https://go-internal-review.googlesource.com/c/go/+/1700
Reviewed-by: Tatiana Bradley <tatianabradley@google.com>
Reviewed-by: Damien Neil <dneil@google.com>
Reviewed-on: https://go-review.googlesource.com/c/go/+/643096
Reviewed-by: Michael Pratt <mpratt@google.com>
Auto-Submit: Michael Knyszek <mknyszek@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
When an HTTP redirect changes the host of a request, we drop
sensitive headers such as Authorization from the redirected request.
Fix a bug where a chain of redirects could result in sensitive
headers being sent to the wrong host:
1. request to a.tld with Authorization header
2. a.tld redirects to b.tld
3. request to b.tld with no Authorization header
4. b.tld redirects to b.tld
3. request to b.tld with Authorization header restored
Thanks to Kyle Seely for reporting this issue.
For #70530
Fixes CVE-2024-45336
Change-Id: Ia58a2e10d33d6b0cc7220935e771450e5c34de72
Reviewed-on: https://go-internal-review.googlesource.com/c/go/+/1641
Reviewed-by: Roland Shoemaker <bracewell@google.com>
Reviewed-by: Tatiana Bradley <tatianabradley@google.com>
Commit-Queue: Roland Shoemaker <bracewell@google.com>
Reviewed-on: https://go-review.googlesource.com/c/go/+/643095
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Michael Pratt <mpratt@google.com>
Auto-Submit: Michael Knyszek <mknyszek@google.com>
Perform minor cleanups in tests to improve printout of diffs
and/or follow modern coding style.
This reduces the amount of diffs between v1 and the v2 prototype.
Change-Id: I019bb9642e2135f2fa3eac6abfa6df91c397aa82
Reviewed-on: https://go-review.googlesource.com/c/go/+/642257
Reviewed-by: Damien Neil <dneil@google.com>
Auto-Submit: Joseph Tsai <joetsai@digital-static.net>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Ian Lance Taylor <iant@google.com>
Newer git versions (at least git 2.47.1) do not send all the matching tags
for a shallow fetch of a specific hash anymore. The go command assumes
that git servers do this. Since that assumption is broken, use the local
copy of the remote refs list to augment the tags sent by the server.
This makes the cmd/go/internal/modfetch tests pass again with newer git.
Fixes#71261.
Change-Id: I9fd4f3fd7beeb68a522938599f8f3acd887d0b26
Reviewed-on: https://go-review.googlesource.com/c/go/+/642437
Reviewed-by: Michael Matloob <matloob@golang.org>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Auto-Submit: Russ Cox <rsc@golang.org>
https://ci.chromium.org/ui/inv/build-8725798219051312433/test-results?sortby=&groupby=
shows a mysterious failure with this stack:
=== RUN BenchmarkAll
BenchmarkAll
panic: runtime error: invalid memory address or nil pointer dereference
[signal SIGSEGV: segmentation violation code=0x1 addr=0x0 pc=0x7c497f]
goroutine 20 gp=0xc000004000 m=7 mp=0xc000182808 [running]:
panic({0x81c5e0?, 0xabc6b0?})
/home/swarming/.swarming/w/ir/x/w/goroot/src/runtime/panic.go:806 +0x168 fp=0xc00c7ffce0 sp=0xc00c7ffc30 pc=0x4ad4c8
runtime.panicmem(...)
/home/swarming/.swarming/w/ir/x/w/goroot/src/runtime/panic.go:262
runtime.sigpanic()
/home/swarming/.swarming/w/ir/x/w/goroot/src/runtime/signal_unix.go:925 +0x359 fp=0xc00c7ffd40 sp=0xc00c7ffce0 pc=0x4af6d9
cmd/api.(*Walker).export(0xc000034100, 0x0)
/home/swarming/.swarming/w/ir/x/w/goroot/src/cmd/api/main_test.go:193 +0x3f fp=0xc00c7ffe08 sp=0xc00c7ffd40 pc=0x7c497f
cmd/api.BenchmarkAll(0xc000214288)
/home/swarming/.swarming/w/ir/x/w/goroot/src/cmd/api/api_test.go:205 +0x207 fp=0xc00c7ffeb0 sp=0xc00c7ffe08 pc=0x7c1c07
testing.(*B).runN(0xc000214288, 0x1)
/home/swarming/.swarming/w/ir/x/w/goroot/src/testing/benchmark.go:202 +0x291 fp=0xc00c7fff78 sp=0xc00c7ffeb0 pc=0x57e611
testing.(*B).run1.func1()
/home/swarming/.swarming/w/ir/x/w/goroot/src/testing/benchmark.go:224 +0x7c fp=0xc00c7fffe0 sp=0xc00c7fff78 pc=0x57f11c
runtime.goexit({})
/home/swarming/.swarming/w/ir/x/w/goroot/src/runtime/asm_amd64.s:1700 +0x1 fp=0xc00c7fffe8 sp=0xc00c7fffe0 pc=0x4b4a61
created by testing.(*B).run1 in goroutine 1
/home/swarming/.swarming/w/ir/x/w/goroot/src/testing/benchmark.go:217 +0x173
So import_ must have returned an error, making pkg nil. Show that error.
Also do the same at the other calls to import_.
Change-Id: Ie782571c4bda3334a86b303f61969cf1cc7d3c32
Reviewed-on: https://go-review.googlesource.com/c/go/+/642438
Auto-Submit: Dmitri Shuralyov <dmitshur@golang.org>
Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Dmitri Shuralyov <dmitshur@golang.org>
Re-enable tests for stack-allocated maps and fast map accessors.
Those are implemented now.
Update #54766
Change-Id: I8c019702bd9fb077b2fe3f7c78e8e9e10d2263a6
Reviewed-on: https://go-review.googlesource.com/c/go/+/642376
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Keith Randall <khr@google.com>
Reviewed-by: Michael Pratt <mpratt@google.com>
Auto-Submit: Keith Randall <khr@golang.org>
Change-Id: I17bc68a2e3a96d0dc3d9ddcad40149df37fc4839
Reviewed-on: https://go-review.googlesource.com/c/go/+/642198
Reviewed-by: Ian Lance Taylor <iant@google.com>
Reviewed-by: Damien Neil <dneil@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Auto-Submit: Ian Lance Taylor <iant@google.com>
Auto-Submit: Damien Neil <dneil@google.com>
This case recently started happening on the builders.
The synctest experiment was recently enabled for some targets (CL 642422).
This caused the list of standard packages to include testing/synctest.
However, BenchmarkAll tests for all configurations;
some did not include testing/synctest. That caused the test to crash.
Change-Id: Icade10af147c2e2bcbac25bf744919083db3e70f
Reviewed-on: https://go-review.googlesource.com/c/go/+/642397
Auto-Submit: Ian Lance Taylor <iant@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Ian Lance Taylor <iant@google.com>
Reviewed-by: Russ Cox <rsc@golang.org>
Change-Id: Ic410375987c0f376d0a975e5a6284de10f08b741
Reviewed-on: https://go-review.googlesource.com/c/go/+/642495
Auto-Submit: Ian Lance Taylor <iant@golang.org>
Reviewed-by: Alan Donovan <adonovan@google.com>
Reviewed-by: Michael Knyszek <mknyszek@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
The existing code assumed the type argument count check in
Checker.instance couldn't fail for generic alias types
(matching the code for generic signatures), but it actually
can.
Adjust the code accordingly and document that the result of
Checker.instance may be invalid.
Review all call sites of Checker.instance and make sure we
handle the failure case, or document the code accordingly
(in the case of generic signatures).
When reporting an type argument count error, use the alias
name rather than the alias string representation to match
the error we get for a non-alias type.
While at it, update the manual.go template for ease of use.
Fixes#71198.
Change-Id: I6d19ec6418440e9b49574a2d7dd9825e0af6c2fc
Reviewed-on: https://go-review.googlesource.com/c/go/+/641857
Reviewed-by: Robert Findley <rfindley@google.com>
Reviewed-by: Robert Griesemer <gri@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Auto-Submit: Robert Griesemer <gri@google.com>
Change-Id: If4d3b3965762c8979d304a82493c9eb1068ee13c
Reviewed-on: https://go-review.googlesource.com/c/go/+/642037
Reviewed-by: Robert Griesemer <gri@google.com>
TryBot-Bypass: Robert Griesemer <gri@google.com>
Reviewed-by: Michael Knyszek <mknyszek@google.com>
Auto-Submit: Robert Griesemer <gri@google.com>
The presence of a syntax error in the input immediately unmarshaling
before unmarshaling into the underlying value.
Otherwise, semantic errors are generally lazily reported and
allow unmarshaling to continue on.
Change-Id: Icf1cfc684e415312d9c8bf739c396ede15299d7d
Reviewed-on: https://go-review.googlesource.com/c/go/+/642295
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Ian Lance Taylor <iant@google.com>
Auto-Submit: Joseph Tsai <joetsai@digital-static.net>
Reviewed-by: Michael Knyszek <mknyszek@google.com>
Even if an error occurs during unmarshal, check the resulting Go value.
The documented API specifies no guarantees on how much of a Go value
will be populated when an error occurs and the "json" package
is technically not bounded by the Go compatibility agreement
to ensure this behavior never changes.
However, there is still value in running checks for
what exactly what is partially mutated in the event of an error
even if this is not guaranteed behavior.
Change-Id: I6e923a31f77768a14c4adfb0d37dbeee5807a4a2
Reviewed-on: https://go-review.googlesource.com/c/go/+/642275
Auto-Submit: Joseph Tsai <joetsai@digital-static.net>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Michael Knyszek <mknyszek@google.com>
Reviewed-by: Ian Lance Taylor <iant@google.com>
We can't coalesce a non-WB store with a subsequent Move, as the
result of the store might be the source of the move.
There's a simple codegen test. Not sure how we might do a real test,
as all the repro's I've come up with are very expensive and unreliable.
Fixes#71228
Change-Id: If18bf181a266b9b90964e2591cd2e61a7168371c
Reviewed-on: https://go-review.googlesource.com/c/go/+/642197
Reviewed-by: Keith Randall <khr@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: David Chase <drchase@google.com>
Adds ACVP test coverage for the hmacDRBG algorithm based on the NIST
spec:
https://pages.nist.gov/ACVP/draft-vassilev-acvp-drbg.html#section-7.2
The HMAC DRBG algorithm in our fips module is a minimal implementation
tailored for use for generating ECDSA nonces and so lives in
crypto/internal/fips140/ecdsa.
In order to be testable by crypto/internal/fips140test this changeset
exports a ecdsa.TestingOnlyNewDrbg() constructor to support the ACVP use-case.
All FIPS-compatible SHA2 and SHA3 digests are tested.
The ACVP capability registration is customized to match the limited
capabilities of our ecdsa-focused impl. Most notably:
* reseedImplemented is false - we expect this impl to be invoked
only once or twice per instantiation and do not support explicit
reseeding.
* predResistanceEnabled is false - this requires reseeding.
* Per mode:
* derFuncEnabled is always false - this is only used by ctrDRBG.
* additionalInputLen is 0 for all modes - this is only used with
preResistanceEnabled.
The other capability values are chosen based on Table 4:
https://pages.nist.gov/ACVP/draft-vassilev-acvp-drbg.html#section-7.4
Updates #69642
Change-Id: Ia58979d691f912e2ed739a05efb719f580fbbf89
Reviewed-on: https://go-review.googlesource.com/c/go/+/639775
Reviewed-by: Michael Pratt <mpratt@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Auto-Submit: Filippo Valsorda <filippo@golang.org>
Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
Reviewed-by: Filippo Valsorda <filippo@golang.org>
Adds ACVP test coverage for ML-KEM based on the NIST spec:
https://pages.nist.gov/ACVP/draft-celi-acvp-ml-kem.html
Notably we need to update the BoringSSL module version because the
acvptool was only recently updated to support testing ML-KEM.
A few non-test updates are also required for the
crypto/internal/fips140/mlkem package:
* For keyGen tests a new ExpandedBytes768() function is added that
converts a DecapsualtionKey768 struct into the expanded NIST
serialization. The existing Bytes() function returns the
key's seed, while ACVP testing requires the more cumbersome format.
* For decap tests a new TestingOnlyNewDecapsulationKey768()
constructor is added to produce a DecapsulationKey768 struct from the
expanded FIPS 203 serialization provided by the ACVP test vector. The
pre-existing NewDecapsulationKey768() function expects a seed as
input.
The generate1024.go helper is updated to translate the above changes to
the generated mlkem1024.go implementation.
Both of these new functions are exclusively for ACVP usage and so not
present in the public mlkem API. End users should always prefer to work
with seeds.
Updates #69642
Change-Id: I79784f8a8db00a2ddefdcece4b8de50b033c8f69
Reviewed-on: https://go-review.googlesource.com/c/go/+/637439
Reviewed-by: Carlos Amedee <carlos@golang.org>
Reviewed-by: Filippo Valsorda <filippo@golang.org>
Reviewed-by: David Chase <drchase@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Could have fixed this some other ways, including inside the FIPS 140-3
module, but this is small and self-contained, clearly not affecting
production non-toy key sizes. This late in the freeze, a surgical fix
felt best.
Fixes#71185
Change-Id: I6a6a465641357c9d6b076c8a520b221be4210ed5
Reviewed-on: https://go-review.googlesource.com/c/go/+/641755
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Roland Shoemaker <roland@golang.org>
Auto-Submit: Filippo Valsorda <filippo@golang.org>
Reviewed-by: Russ Cox <rsc@golang.org>
Heap profiles hide "runtime" frames like runtime.mapassign. This broke
in 1.24 because the map implementation moved to internal/runtime/maps,
and runtime/pprof only considered literal "runtime." when looking for
runtime frames.
It would be nice to use cmd/internal/objabi.PkgSpecial to find runtime
packages, but that is hidden away in cmd.
Fixes#71174.
Change-Id: I6a6a636cb42aa17539e47da16854bd3fd8cb1bfe
Reviewed-on: https://go-review.googlesource.com/c/go/+/641775
Auto-Submit: Michael Pratt <mpratt@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Michael Knyszek <mknyszek@google.com>
The Mmap function returns a Data struct containing a slice with the
mapped contents of the file. Before this change, on Windows, the slice
contained the contents of all the pages of the mapping, including past
the end of the file. Re-slice the slice to the length of the file (if
if the slice is longer) so that the slice contains only the data in the
file.
For #71059
Change-Id: I389b752505b6fa1252b5c6d836a37bc7e662a45d
Reviewed-on: https://go-review.googlesource.com/c/go/+/640155
Reviewed-by: Russ Cox <rsc@golang.org>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Both fips140only and the service indicator checks in
crypto/internal/fips140/... expect to type assert to
crypto/internal/fips140/{sha256,sha512,sha3}.Digest.
However, crypto/sha3 returns a wrapper concrete type around sha3.Digest.
Add a new fips140hash.Unwrap function to turn the wrapper into the
underlying sha3.Digest, and use it consistently before calling into
fips140only or the FIPS 140-3 module.
In crypto/rsa, also made the fips140only checks apply consistently after
the Go+BoringCrypto shims, so we can instantiate the hash, and avoid
having to wrap the New function. Note that fips140=only is incompatible
with Go+BoringCrypto.
Fixes#70879
Change-Id: I6a6a4656ec55c3e13f6cbfadb9cf89c0f9183bdc
Reviewed-on: https://go-review.googlesource.com/c/go/+/640855
Auto-Submit: Filippo Valsorda <filippo@golang.org>
Reviewed-by: Roland Shoemaker <roland@golang.org>
Reviewed-by: Russ Cox <rsc@golang.org>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
In particular, cover the behavior of unmarshaling a JSON string
into a Number type regardless of whether the `string` option
is specified or not.
Change-Id: Ibc55f16860442240bcfeea1fd51aaa76f7e50f67
Reviewed-on: https://go-review.googlesource.com/c/go/+/641416
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Auto-Submit: Joseph Tsai <joetsai@digital-static.net>
Reviewed-by: Damien Neil <dneil@google.com>
Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
The constants appeared badly ordered and grouped in the godoc before
const (
CiphertextSize1024 = 1568
EncapsulationKeySize1024 = 1568
)
const (
SharedKeySize = 32
SeedSize = 64
CiphertextSize768 = 1088
EncapsulationKeySize768 = 1184
)
while now they are a single group with the good size first
const (
SharedKeySize = 32
SeedSize = 64
CiphertextSize768 = 1088
EncapsulationKeySize768 = 1184
CiphertextSize1024 = 1568
EncapsulationKeySize1024 = 1568
)
No code changes.
Change-Id: I6a6a4656961b1e8c8bca3992aafa33e0575af8a2
Reviewed-on: https://go-review.googlesource.com/c/go/+/640997
Auto-Submit: Filippo Valsorda <filippo@golang.org>
Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Russ Cox <rsc@golang.org>
Reviewed-by: Daniel McCarney <daniel@binaryparadox.net>
The original implementation of the type checkers accepted any boolean
result type for yield, but the compiler's front-end had a problem with
it (#71131).
As a temporary fix (for 1.24), adjust the type checkers to insist on the
spec's literal wording and avoid the compiler panic.
Fixes#71131.
For #71164.
Change-Id: Ie25f9a892e58b5e489d399b0bce2d0af55dc3c48
Reviewed-on: https://go-review.googlesource.com/c/go/+/640599
Reviewed-by: Robert Griesemer <gri@google.com>
Auto-Submit: Robert Griesemer <gri@google.com>
Reviewed-by: Tim King <taking@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
New is called to get a Hash which can then be rejected with an error
(instead of a panic) from fips140only.ApprovedHash.
Also, it's reasonable to call New().Size() and then not use the hash.
Change-Id: I6a6a4656c43528d169c4b28c8b6de48448236d4f
Reviewed-on: https://go-review.googlesource.com/c/go/+/641317
Auto-Submit: Filippo Valsorda <filippo@golang.org>
Reviewed-by: Daniel McCarney <daniel@binaryparadox.net>
Reviewed-by: Russ Cox <rsc@golang.org>
Reviewed-by: Roland Shoemaker <roland@golang.org>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Having a global lock on the random state (used only in FIPS-140 mode)
introduces contention in concurrent programs. Use an approximately
per-P random state instead, using sync.Pool to manage per-P state.
This code is important to land for the Go 1.24 release because it is
part of the FIPS-140 module that will be validated and certified,
so it will live for a long time. We otherwise wouldn't be able to
correct this contention for at least a year, perhaps more.
At the same time, the code is only used in the FIPS-140 mode,
so there is no risk to normal programs.
Fixes#71155.
Change-Id: I6b779f15ddfdf232f608f5cda08f75906e58114f
Reviewed-on: https://go-review.googlesource.com/c/go/+/641097
Reviewed-by: Austin Clements <austin@google.com>
Reviewed-by: Filippo Valsorda <filippo@golang.org>
Reviewed-by: Michael Knyszek <mknyszek@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
The backslashes on the windows paths will be escaped, so when checking
for them in the regular expression we'd have to have quadruple
backslashes '\\\\'. Since it's difficult to escape $GOCACHEPROG properly
for both json and regexp, just check for a string that ends in
cacheprog$GOEXE. We already check that the proper value is reported in
go env and go env -changed, and the json test case is mostly useful to
verify that GOCACHEPROG shows up in the json output.
For #71059
Change-Id: I52d49de61f2309a139f84c4d232b4cd94546ec8c
Cq-Include-Trybots: luci.golang.try:gotip-windows-amd64-longtest,gotip-linux-amd64-longtest
Reviewed-on: https://go-review.googlesource.com/c/go/+/641375
Reviewed-by: Dmitri Shuralyov <dmitshur@golang.org>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Sam Thanawalla <samthanawalla@google.com>
Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
Currently injectglist emits all the trace events before actually calling
casgstatus on each goroutine. This is a problem, since tracing can
observe an inconsistent state (gstatus does not match tracer's 'emitted
an event' state).
This change fixes the problem by having injectglist do what every other
scheduler function does, and that's wrap each call to casgstatus in
traceAcquire/traceRelease.
Fixes#70883.
Change-Id: I857e96cec01688013597e8efc0c4c3d0b72d3a70
Reviewed-on: https://go-review.googlesource.com/c/go/+/638558
Reviewed-by: Michael Pratt <mpratt@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Change-Id: I3e7a8498514da1d278acd566d526fdf6278f7d41
Reviewed-on: https://go-review.googlesource.com/c/go/+/638916
Reviewed-by: Dmitri Shuralyov <dmitshur@golang.org>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Ian Lance Taylor <iant@google.com>
Auto-Submit: Dmitri Shuralyov <dmitshur@golang.org>
Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
For #67434
Change-Id: Iebcfc2559a62405fea7e3ceff8cf6c2f50b61def
Reviewed-on: https://go-review.googlesource.com/c/go/+/641176
Reviewed-by: Ian Lance Taylor <iant@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Auto-Submit: Damien Neil <dneil@google.com>
On Windows, we can't open a file that's already been opened. Before this
change, we'd try to write an index entry if mmapping the entry failed.
But that could happen either if the file doesn't exist or if there was a
problem mmapping an already opened file. Pass through information about
whether the file was actually opened so that we don't try to write to an
already opened file.
For #71059
Change-Id: I6adabe1093fed9ec37e7fafb13384c102786cbce
Reviewed-on: https://go-review.googlesource.com/c/go/+/640577
Reviewed-by: Sam Thanawalla <samthanawalla@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
These break if the tools are run with GODEBUG=fips140=only,
which happens if someone sets that during 'go test' (and a test
binary must be built).
The easiest fix is to make the tools compatible with this GODEBUG
by just using sha256 as the underlying hash always. Just in case,
I made the wrappers select different sections of the hash, but
none of the call sites really care.
This CL is for the Go 1.24 release, but a follow-up during the Go 1.25
dev cycle could change all the usage sites to only use Sum32/New32.
For #70514Fixes#70878
Change-Id: Id5fea779c83df51d1680dbe561e0949c56e8d1e5
Reviewed-on: https://go-review.googlesource.com/c/go/+/641096
Reviewed-by: Keith Randall <khr@golang.org>
Auto-Submit: Russ Cox <rsc@golang.org>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Keith Randall <khr@google.com>
Allow a gocacheprog to not respond to close. The intention of the code
is that after we send the close message we'd ignore errors reading from
the cacheprog's stdout. But before this change if a cacheprog
did not respond to close and we got an EOF reading from the cacheprog's
stdout we'd just ignore all pending requests. The send operation would
then block forever waiting for a response. With this change, we close
all response channels for pending responses if there's an error reading
from the cacheprog's stdout while we're closing. The receives from the
response channels would then proceed (but now have to handle a nil
value). Then the send operation would return and the (*ProgCache).Close
function can proceed.
Fixes#70848
Change-Id: I6631d317ba7aea3f25f714f31cd2aeef0f4d4e3e
Cq-Include-Trybots: luci.golang.try:gotip-linux-amd64-longtest,gotip-windows-amd64-longtest
Reviewed-on: https://go-review.googlesource.com/c/go/+/640516
Reviewed-by: Austin Clements <austin@google.com>
Reviewed-by: Sam Thanawalla <samthanawalla@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Wingrez
Robert Griesemer
Tobias Klauser
Cherry Mui
Gopher Robot
Filippo Valsorda
Sam Thanawalla
Damien Neil
Roland Shoemaker
Joe Tsai
Russ Cox
Keith Randall
cuishuang
Ian Lance Taylor
Sean Liao
Oleksandr Redko
Daniel McCarney
Michael Pratt
Michael Matloob
Michael Anthony Knyszek
Koichi Shiraishi