Avi Kivity wrote:
Fabrice Bellard wrote:
Avi Kivity wrote:
Aurelien Jarno wrote:
Well the IDE code hasn't changed a lot recently, so I checked the CVS
history and easily (first test) found the commit that causes the
problem:
Last AIO patch, by Vladimir N. Oleynik.
http://cvs.savannah.nongnu.org/viewvc/qemu/hw/ide.c?root=qemu&r1=1.64&r2=1.65
Have we learned something about this issue? Should I revert the patch
from kvm? Should qemu?
Sure. The conversion from sync to async mode done in this patch is
incomplete to say the least.
Is reverting the patch safe, or will it cause new problems?
The patch must be reverted because it implements incorrectly the AIO
write case (e.g. nsector is modified before the IO actually takes
place). The previous code was at least correct if win2k_install_hack was
not enabled and if one considered than the PIO I/Os were instantaneous.
Regards,
Fabrice.
