David Zbarsky

Thanks for submitting! I'm about to go on vacation for a week, I will do my best to find time to look before that

We still need this PR and/or a 3.7.4 with these changes applied on top. From my POV, it's ready to go in, but @zaucy can make the final call on...

See https://github.com/localstack/localstack/issues/7924#issuecomment-1572135824 for a workaround, you should be able to add an extra RUN directive to the Dockerfile if you are willing to build/host your own image

My understanding is that the computation to go from Cargo.lock to Cargo.bazel.lock is pure/reproducible, right? If so, this works because the Bazel lockfile was essentially a performance optimization, and with...

+1 to using LLD and avoiding the host dependency. I tried it out but saw issues with clang not recognizing the fuse-ld flag. I'm sure I was holding it wrong...

I am currently poking at runfiles library, I can add this

@meteorcloudy can we get some sort of linux arm platform that is supported? A lot of us are deploying software to Graviton these days for cost efficiency reasons and it's...

> @DavidZbarsky-at the prebuilt release for this failed: > > ``` > In file included from external/libarchive~/libarchive/archive_hmac.c:32: > external/libarchive~/libarchive/archive_hmac_private.h:51:10: fatal error: 'CommonCrypto/CommonHMAC.h' file not found > 51 | #include >...

> @DavidZbarsky-at Could you also test the performance with 7.0.0, which doesn't include https://github.com/bazelbuild/bazel/commit/e9022f6731b4f62d3a08bdc4eacce70ad28e3c78? On 7.0.0 I was seeing around 600-900ms sandbox setup overhead. Note that we are using tmpfs...

Check out https://github.com/DavidZbarsky-at/nodejs-repro Repro with `CC=false bazel test -k :all --cache_test_results=no` It looks like on linux both 7.0.0 and 7.1.0 take around 700-900ms in sandbox setup with the tmpfs sandbox...