-
Notifications
You must be signed in to change notification settings - Fork 1k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
jws: improve fix for CVE-2025-22868 #774
Conversation
This PR (HEAD: 3fbd01c) has been imported to Gerrit for code review. Please visit Gerrit at https://go-review.googlesource.com/c/oauth2/+/655455. Important tips:
|
Message from Gopher Robot: Patch Set 1: (1 comment) Please don’t reply on this GitHub thread. Visit golang.org/cl/655455. |
Message from jub0bs: Patch Set 1: (1 comment) Please don’t reply on this GitHub thread. Visit golang.org/cl/655455. |
3fbd01c
to
4323f8e
Compare
This PR (HEAD: 4323f8e) has been imported to Gerrit for code review. Please visit Gerrit at https://go-review.googlesource.com/c/oauth2/+/655455. Important tips:
|
Message from Neal Patel: Patch Set 3: (3 comments) Please don’t reply on this GitHub thread. Visit golang.org/cl/655455. |
4323f8e
to
d5b383c
Compare
This PR (HEAD: d5b383c) has been imported to Gerrit for code review. Please visit Gerrit at https://go-review.googlesource.com/c/oauth2/+/655455. Important tips:
|
Message from jub0bs: Patch Set 5: (3 comments) Please don’t reply on this GitHub thread. Visit golang.org/cl/655455. |
Message from Neal Patel: Patch Set 5: Code-Review+2 Commit-Queue+1 Please don’t reply on this GitHub thread. Visit golang.org/cl/655455. |
Message from Go LUCI: Patch Set 5: Dry run: CV is trying the patch. Bot data: {"action":"start","triggered_at":"2025-03-10T17:36:50Z","revision":"a1b1696d0b147aaf36dc61ffd9bc3cc6d750543e"} Please don’t reply on this GitHub thread. Visit golang.org/cl/655455. |
Message from Neal Patel: Patch Set 5: -Commit-Queue Please don’t reply on this GitHub thread. Visit golang.org/cl/655455. |
Message from Go LUCI: Patch Set 5: This CL has passed the run Please don’t reply on this GitHub thread. Visit golang.org/cl/655455. |
Message from Go LUCI: Patch Set 5: LUCI-TryBot-Result+1 Please don’t reply on this GitHub thread. Visit golang.org/cl/655455. |
Message from Neal Patel: Patch Set 5: (1 comment) Please don’t reply on this GitHub thread. Visit golang.org/cl/655455. |
The fix for CVE-2025-22868 relies on strings.Count, which isn't ideal because it precludes failing fast when the token contains an unexpected number of periods. Moreover, Verify still allocates more than necessary. Eschew strings.Count in favor of strings.Cut. Some benchmark results: goos: darwin goarch: amd64 pkg: golang.org/x/oauth2/jws cpu: Intel(R) Core(TM) i7-6700HQ CPU @ 2.60GHz │ old │ new │ │ sec/op │ sec/op vs base │ Verify/full_of_periods-8 24862.50n ± 1% 57.87n ± 0% -99.77% (p=0.000 n=20) Verify/two_trailing_periods-8 3.485m ± 1% 3.445m ± 1% -1.13% (p=0.003 n=20) geomean 294.3µ 14.12µ -95.20% │ old │ new │ │ B/op │ B/op vs base │ Verify/full_of_periods-8 16.00 ± 0% 16.00 ± 0% ~ (p=1.000 n=20) ¹ Verify/two_trailing_periods-8 2.001Mi ± 0% 1.001Mi ± 0% -49.98% (p=0.000 n=20) geomean 5.658Ki 4.002Ki -29.27% ¹ all samples are equal │ old │ new │ │ allocs/op │ allocs/op vs base │ Verify/full_of_periods-8 1.000 ± 0% 1.000 ± 0% ~ (p=1.000 n=20) ¹ Verify/two_trailing_periods-8 12.000 ± 0% 9.000 ± 0% -25.00% (p=0.000 n=20) geomean 3.464 3.000 -13.40% ¹ all samples are equal Also, remove all remaining calls to strings.Split.
d5b383c
to
3a82900
Compare
This PR (HEAD: 3a82900) has been imported to Gerrit for code review. Please visit Gerrit at https://go-review.googlesource.com/c/oauth2/+/655455. Important tips:
|
Message from jub0bs: Patch Set 6: (1 comment) Please don’t reply on this GitHub thread. Visit golang.org/cl/655455. |
Message from Neal Patel: Patch Set 7: Code-Review+2 Commit-Queue+1 Please don’t reply on this GitHub thread. Visit golang.org/cl/655455. |
Message from Go LUCI: Patch Set 7: Dry run: CV is trying the patch. Bot data: {"action":"start","triggered_at":"2025-03-11T18:44:59Z","revision":"a9c20d14a9b323542435f1f8a017c879fe331163"} Please don’t reply on this GitHub thread. Visit golang.org/cl/655455. |
Message from Neal Patel: Patch Set 7: -Commit-Queue Please don’t reply on this GitHub thread. Visit golang.org/cl/655455. |
Message from Go LUCI: Patch Set 7: This CL has passed the run Please don’t reply on this GitHub thread. Visit golang.org/cl/655455. |
Message from Go LUCI: Patch Set 7: LUCI-TryBot-Result+1 Please don’t reply on this GitHub thread. Visit golang.org/cl/655455. |
Message from Roland Shoemaker: Patch Set 7: Code-Review+1 Please don’t reply on this GitHub thread. Visit golang.org/cl/655455. |
The fix for CVE-2025-22868 relies on strings.Count, which isn't ideal because it precludes failing fast when the token contains an unexpected number of periods. Moreover, Verify still allocates more than necessary. Eschew strings.Count in favor of strings.Cut. Some benchmark results: goos: darwin goarch: amd64 pkg: golang.org/x/oauth2/jws cpu: Intel(R) Core(TM) i7-6700HQ CPU @ 2.60GHz │ old │ new │ │ sec/op │ sec/op vs base │ Verify/full_of_periods-8 24862.50n ± 1% 57.87n ± 0% -99.77% (p=0.000 n=20) Verify/two_trailing_periods-8 3.485m ± 1% 3.445m ± 1% -1.13% (p=0.003 n=20) geomean 294.3µ 14.12µ -95.20% │ old │ new │ │ B/op │ B/op vs base │ Verify/full_of_periods-8 16.00 ± 0% 16.00 ± 0% ~ (p=1.000 n=20) ¹ Verify/two_trailing_periods-8 2.001Mi ± 0% 1.001Mi ± 0% -49.98% (p=0.000 n=20) geomean 5.658Ki 4.002Ki -29.27% ¹ all samples are equal │ old │ new │ │ allocs/op │ allocs/op vs base │ Verify/full_of_periods-8 1.000 ± 0% 1.000 ± 0% ~ (p=1.000 n=20) ¹ Verify/two_trailing_periods-8 12.000 ± 0% 9.000 ± 0% -25.00% (p=0.000 n=20) geomean 3.464 3.000 -13.40% ¹ all samples are equal Also, remove all remaining calls to strings.Split. Updates golang/go#71490 Change-Id: Icac3c7a81562161ab6533d892ba19247d6d5b943 GitHub-Last-Rev: 3a82900 GitHub-Pull-Request: #774 Reviewed-on: https://go-review.googlesource.com/c/oauth2/+/655455 Commit-Queue: Neal Patel <nealpatel@google.com> Reviewed-by: Roland Shoemaker <roland@golang.org> LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Neal Patel <nealpatel@google.com> Auto-Submit: Neal Patel <nealpatel@google.com>
This PR is being closed because golang.org/cl/655455 has been merged. |
The fix for CVE-2025-22868 relies on strings.Count, which isn't ideal
because it precludes failing fast when the token contains an unexpected
number of periods. Moreover, Verify still allocates more than necessary.
Eschew strings.Count in favor of strings.Cut. Some benchmark results:
goos: darwin
goarch: amd64
pkg: golang.org/x/oauth2/jws
cpu: Intel(R) Core(TM) i7-6700HQ CPU @ 2.60GHz
│ old │ new │
│ sec/op │ sec/op vs base │
Verify/full_of_periods-8 24862.50n ± 1% 57.87n ± 0% -99.77% (p=0.000 n=20)
Verify/two_trailing_periods-8 3.485m ± 1% 3.445m ± 1% -1.13% (p=0.003 n=20)
geomean 294.3µ 14.12µ -95.20%
Verify/full_of_periods-8 16.00 ± 0% 16.00 ± 0% ~ (p=1.000 n=20) ¹
Verify/two_trailing_periods-8 2.001Mi ± 0% 1.001Mi ± 0% -49.98% (p=0.000 n=20)
geomean 5.658Ki 4.002Ki -29.27%
¹ all samples are equal
Verify/full_of_periods-8 1.000 ± 0% 1.000 ± 0% ~ (p=1.000 n=20) ¹
Verify/two_trailing_periods-8 12.000 ± 0% 9.000 ± 0% -25.00% (p=0.000 n=20)
geomean 3.464 3.000 -13.40%
¹ all samples are equal
Also, remove all remaining calls to strings.Split.
Updates golang/go#71490