Hi! New realtime trees should be available at kernel.org. Trees are available at https://git.kernel.org/pub/scm/linux/kernel/git/cip/linux-cip.git/log/?h=linux-5.10.y-cip-rt https://git.kernel.org/pub

Hi! Interesting. So Intel hardware is buggy, and it is unsuitable in certain virtualization configurations. It is a hardware problem, and there's little we could do about it. Best regards, Pavel

Hi! I pushed a test commit to ci/pavel/linux-test. This is not first commit in the 14 days. Where can I get the results? I don't see them at https://linux.kernelci.org/test/ . pavel@amd:~/cip/4$ ./tes

Hi! I'm trying to understand how to work with kernelci. I thought I'd start with understanding 4.4.302 results, so I went to examine some. a) qemu failing I got to https://storage.kernelci.org/cip/lin

My understanding is that we are maintaining both 4.4-cip and 4.4-cip-rt going forward. I hope it is achievable. Best regards, Pavel

Hi! New realtime trees should be available at kernel.org. This is last release based on Greg Kroah-Hartman's -stable tree and Daniel Wagner's -stable-rt release. We are self-maintaining from now on. T

Hi! We did commit to maintaining 4.4-cip-rt, but upstream 4.4.X is going to be discontinued, and so is 4.4.X-stable. Best regards, Pavel

Hi! I'm sorry I missed the meeting. I was mostly reviewing 5.10.96. Aha, and useful trick, directory listings are enabled, so going to https://irclogs.baserock.org/meetings/cip/2022/02/ allows me to a

Hi! Greg says: # Message-Id: <20220201180822.148370751@...> # Subject: [PATCH 4.4 00/25] 4.4.302-rc1 review # ... # NOTE! This is the proposed LAST 4.4.y kernel release to happen under

Hi! Testing did not uncover any kernel problems. I have applied the patches and pushed the tree. Best regards, Pavel

Hi! I found some minor details, but overall series looks okay. I'm currently running tests, I can apply it if it passes and if there are no other comments. Best regards, Pavel

Hi! This needs to rpcif_disable_rpm() before returning, AFAICT. Same here. Best regards, Pavel

Hi! AFAICT this needs to be set to 0, not commented out. Best regards, Pavel

Hi! This confused me for a while. Some of hex numbers are 7 digits, some are 8. Making it consistent at 8 digits would be good. (Or maybe using 0x0 so it is clear we are not aiming for 8 digits). Best

Hi! Should pm_runtime_put() be called before returning failure to the caller? Best regards, Pavel

Hi! Along with refactoring code, this also removes locking from the reads. Which is okay AFAICT, but note in changelog would not hurt. Best regards, Pavel

Hi! Yes, sorry, I was looking at wrong tree. Best regards, Pavel

Hi! Wow. This must have been important. It looks like 5.10.95 (+4.4 and 4.19) was released just to get this fixed. Fix is "interesting" but... it should be fixed. Best regards, Pavel

Hi! Are they? I see fail label being unused in our 5.10 tree (but mainline uses it and I don't think we need it removed). But more importantly... staring at the code some more: err = register_candev(n

Hi! Looks good to me, go ahead :-). Best regards, Pavel