[PATCH] drm/i915/kvmgt: Simplify function intel_vgpu_read/write
Zhang, Xiong Y
xiong.y.zhang at intel.com
Mon Jan 29 05:44:17 UTC 2018
> On Sun, Jan 28, 2018 at 06:05:37PM -0800, Zhang, Xiong Y wrote:
> > > From: Changbin Du <changbin.du at intel.com>
> > >
> > > The code needn't so complex, simplify it.
> > >
> > > Signed-off-by: Changbin Du <changbin.du at intel.com>
> > > ---
> > > drivers/gpu/drm/i915/gvt/kvmgt.c | 106
> > > +++++++++------------------------------
> > > 1 file changed, 24 insertions(+), 82 deletions(-)
> > >
> > > diff --git a/drivers/gpu/drm/i915/gvt/kvmgt.c
> > > b/drivers/gpu/drm/i915/gvt/kvmgt.c
> > > index 3c68727..8a4ad26 100644
> > > --- a/drivers/gpu/drm/i915/gvt/kvmgt.c
> > > +++ b/drivers/gpu/drm/i915/gvt/kvmgt.c
> > > @@ -740,50 +740,21 @@ static ssize_t intel_vgpu_read(struct
> > > mdev_device *mdev, char __user *buf,
> > > int ret;
> > >
> > > while (count) {
> > > - size_t filled;
> > > + size_t rsz = *ppos % 2 ? 1 : (*ppos % 4 ? 2 : 4);
> > [Zhang, Xiong Y] for odd *ppos is always 1 byte, which is different from the
> original without your patch.
>
> Can you give a example?
[Zhang, Xiong Y] if *ppos=3, count=3.
The original r/w would be: len =1, then len =2
With your patch: it always len =1, len =1, len =1.
>
> > >
> > > _______________________________________________
> > > intel-gvt-dev mailing list
> > > intel-gvt-dev at lists.freedesktop.org
> > > https://lists.freedesktop.org/mailman/listinfo/intel-gvt-dev
>
> --
> Thanks,
> Changbin Du
More information about the intel-gvt-dev
mailing list