Re: [BUG] More issues with arm/aes-neonbs

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Thu, 8 Aug 2024 at 10:14, Linus Torvalds
<torvalds@xxxxxxxxxxxxxxxxxxxx> wrote:
>
> (Please note: ENTIRELY UNTESTED! It compiles for me, but I might have
> done something incredibly stupid and maybe there's some silly and
> fatal bug in what _appears_ trivially correct).

It's like I have a sixth sense.

The wait_for_completion_timeout() test was entirely wrong. It returns
the time remaining if it timed out, not an error like some of the
other ones.

So t needs to be

                if (wait_for_completion_timeout(&idem.complete, 10*HZ))
                        return idem.ret;

instead.

Of course, it's only going to cause issues if it actually times out,
so that patch "works" in normal situations where module loading takes
less than 10s. But still - it was completely buggered.

               Linus




[Index of Archives]     [Kernel]     [Gnu Classpath]     [Gnu Crypto]     [DM Crypt]     [Netfilter]     [Bugtraq]
  Powered by Linux