On 8/9/24 17:00, Peter Bergner wrote:
On 8/9/24 12:02 PM, Vladimir Makarov wrote:
I believe your should reverse the original patch and all the patches you
submitted to fix the issues with the original patch.
I agree this commit should be reverted and Kyrill has pushed that already,
so bootstrap should be restored now. Thank you Kyrill. Surya was testing
her revert commit too, but was waiting for her bootstrap to complete before
pushing (per process rules). It's just the cfarm systems she has access
to were being slow.
I do *not* agree that all the previous patches should be reverted, since they
fixed *real* bugs and really only exposed other latent issues which Surya
has been working on.
You should do a better testing you patches
It is not correct to presume she did not test her patches thoroughly.
I know she has, but with the large number of supported architectures
and ABIs on those architectures, she cannot possibly test everything.
That's why we have stage1 and our patch revert process.
Sorry, I did not finish writing my email and sent it accidentally. I
wanted to explain more in my email.
I am agree I was wrong about reverting all her previous patches.
Still, for GCC developer novice, I think it is important to test all
major targets and aarch64 (one target on which bootstrap was broken) is
the 2nd most important target. I myself always test x86-64, aarch64,
and ppc64le (except for trivial changes). For patches which probably
affect performance, I do SPEC17 on x86-64 but I don't require it from
others for RA patches because it might be not possible to do this.
Still I benchmark others patches by myself.
Reason for my email is that when any bootstrap is broken, I think it
should be fixed as soon as possible and the fastest way to do is to
revert it. At least that is what I'd do if somebody reported about this
yesterday.
Everybody can make a mistake, especially people with little experience.
I hope my email will not discourage Surya to work on GCC.
