git: Apply patch to fix uninitialised memory & t8020 test on big-endian

[OPNA2608]

A failure of the t8020-last-modified test on s390x was reported on the mailing list, which uncovered an issue with a memset call in 2.52.0.

Apply the patch that resulted from this report. Most immediate result from that is that on big-endian platforms, git-last-modified now behaves correctly again and the t8020-last-modified test suite passes.

Things done

• Built on platform:
  • x86_64-linux
  • powerpc64-linux (native)
  • aarch64-linux
  • x86_64-darwin
  • aarch64-darwin
• Tested, as applicable:
  • NixOS tests in nixos/tests.
  • Package tests at passthru.tests.
  • Tests in lib/tests or pkgs/test for functions and "core" functionality.
• Ran nixpkgs-review on this PR. See nixpkgs-review usage.
• Tested basic functionality of all binary files, usually in ./result/bin/.
• Nixpkgs Release Notes
  • Package update: when the change is major or breaking.
• NixOS Release Notes
  • Module addition: when adding a new NixOS module.
  • Module update: when the change is significant.
• Fits CONTRIBUTING.md, pkgs/README.md, maintainers/README.md and other READMEs.

---

Add a 👍 reaction to pull requests you find important.

[me-and]

I'm a bit wary about just suppressing this test: if it's failing, it's failing for a reason. I don't object to having this in place as a temporary workaround, but I'd like someone who has a platform to reproduce the problem to be raising this with the upstream Git community too.

Having tests disabled indefinitely makes me sad; having tests disabled with a reference to wherever is tracking/discussing fixing things properly is far preferable from my perspective.

[OPNA2608]

Was already reported & discussed on the Git ML. Agh… _{grumble grumble mailing lists grumble grumble}

Will switch to fetching the resulting patch, and send this to staging.

[OPNA2608]

Verified that the fix from the ML fixes the test on ppc64.

[philiptaron]

I'll do a build and tests then merge. Thanks for running this upstream.