On Fri, Feb 07, 2025 at 09:25:08PM -0800, Barry K. Nathan wrote:
On 2/6/25 08:06, Greg Kroah-Hartman wrote:
This is the start of the stable review cycle for the 6.6.76 release. There are 389 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.
Responses should be made by Sat, 08 Feb 2025 15:51:12 +0000. Anything received after that time might be too late.
[snip]
Herbert Xu herbert@gondor.apana.org.au crypto: api - Fix boot-up self-test race
[snip] I mentioned in an email to the stable mailing list earlier this week that the crypto-api-fix-boot-up-self-test-race patch, on my systems, is causing a one-minute freeze followed by a crypto self-test failure. See the log excerpt below for an example. (In that email I erroneously said the affected kernel was 6.6.76-rc1. It was 6.6.75 plus stable-queue patches, but not actually 6.6.76-rc1. Sorry for my mistake.)
I'm still experiencing this with 6.6.76-rc2. As before, reverting "crypto: api - Fix boot-up self-test race" fixes it. I also tried testing 6.6.75 plus this patch, and that reproduces the problem. In case it might have been a compiler issue, I tried upgrading my build system from Debian bookworm to trixie, but the issue reproduces with trixie's gcc 14.2 just as it does with bookworm's gcc 12.2.
I also tested 6.12.13-rc2 and 6.13.2-rc2. Those kernels work fine in my testing and do not reproduce this issue.
To be clear, it's personally not a real problem for me if 6.6.76 is released with this patch included -- if necessary, I can just keep reverting this patch until I get my systems upgraded to 6.12. However, I figure this is still worth reporting, in case it eventually turns out to be something that doesn't only affect me.
Anyway, I'll keep digging to see if I can figure out more. Since essentially the same patch that's breaking 6.6 is working just fine on 6.12 and 6.13, I feel like I should be able to find more clues. (However, I may be busy for the next several days, so I'm not sure how soon I'll be able to make progress.)
Thanks for the info, for some reason I thought this was hitting an already released kernel. I'll go drop this patch from the 6.6.y queue now, thanks for pointing it out again.
greg k-h