[libc-commits] [PATCH] D159110: [libc] Fix set_thread_ptr call in rv32 start up code

Jessica Clarke via Phabricator via libc-commits libc-commits at lists.llvm.org
Tue Aug 29 17:03:42 PDT 2023


jrtc27 added a comment.

In D159110#4626688 <https://reviews.llvm.org/D159110#4626688>, @mikhail.ramalho wrote:

> In D159110#4626641 <https://reviews.llvm.org/D159110#4626641>, @jrtc27 wrote:
>
>> In D159110#4626619 <https://reviews.llvm.org/D159110#4626619>, @mikhail.ramalho wrote:
>>
>>> - Updated commit message to reflect new changes
>>
>> You need to do that manually in Phabricator, arc won't sync your local changes to the commit message.
>
> Ok.
>
>> Change looks fine, but message is not, and I'd like to know how you're testing this. Presumably it's currently completely untested given it couldn't possibly have worked before.
>
> For rv64: I test it locally on my VisionFive V2 board with `ninja check-libc`, which covers all the tests in libc, even more than what's being tested on the buildbots.
> For rv32: I test it locally with qemu-system, with `ninja libc-unit-tests`, as our buildbots do for rv64. I do get some random crashes with  `ninja check-libc` with or without this patch. I'll only investigate these crashes once I get `ninja libc-unit-tests` passing.

Does `ninja check-libc` on RV64 pass //without// this change? If so then there's no test coverage for this. If not then it's concerning that broken code was committed.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D159110/new/

https://reviews.llvm.org/D159110



More information about the libc-commits mailing list