On Sun, Mar 28, 2021 at 04:05:54PM -0700, Linus Torvalds wrote:
> So if rc4 was perhaps a bit smaller than average, it looks like rc5 is
> a bigger than average. We're not breaking any records, but it
> certainly isn't tiny, and the rc's aren't shrinking.
>
> I'm not overly worried yet, but let's just say that the trend had
> better not continue, or I'll start feeling like we will need to make
> this one of those releases that need an rc8.
>
> Most of the changes are drivers (gpu and networking stand out, but
> there's various other smaller driver updates elsewhere too) with core
> networking (including bpf) fixes being another noticeable subsystem.
>
> Other than that, there's a smattering of noise all over: minor arch
> fixes, some filesystem fixes (btrfs, cifs, squashfs), selinux, perf
> tools, documentation.
>
> io_uring continues to have noise in it, this time mainly due to some
> signal handling fixes. That removed a fair amount of problematic
> special casing, but the timing certainly isn't great.
>
> So again, nothing really scary, just rather more than I would have
> liked to have in an rc5.
>
> Shortlog appended for people who want to delve into the details,
>
Build results:
total: 151 pass: 151 fail: 0
Qemu test results:
total: 458 pass: 457 fail: 1
Failed tests:
openrisc:or1ksim_defconfig
This is not really a new problem. I enabled devicetree unit tests
in the openrisc kernel and was rewarded with a crash.
https://lore.kernel.org/lkml/20210327224116.69309-1-li...@roeck-us.net/
has all the glorious details.
Guenter
