> While debugging __divdi3 calls in drivers/crypto/hifn_795x.c (due to
> the
> ndelay() delay call with a s64), I found even more breakage of that
> sort. This is after a allnoconfig with ARCH=powerpc in 2.6.25-rc3,
> plus CONFIG_MODULES=y and CONFIG_CRYPTO_DEV_HIFN_795X=y:
I cannot reproduce this, but my tree has
time: prevent the loop in timespec_add_ns() from being optimised away
which is in -mm now. Could you try that patch?
Segher
_______________________________________________
Linuxppc-dev mailing list
Linuxppc-dev@ozlabs.org
https://ozlabs.org/mailman/listinfo/linuxppc-dev
