Re: [Qemu-devel] [PATCH 1/1] tcg: Fix typos in helper_gvec_sar{8, 32, 64}v

[Aleksandar Markovic]

On Jun 14, 2019 8:03 AM, "Aleksandar Markovic" <address@hidden>
wrote:
>
>
> On Jun 14, 2019 7:22 AM, "Aleksandar Markovic" <
address@hidden> wrote:
> >
> >
> > On Jun 13, 2019 6:16 AM, "Richard Henderson" <
address@hidden> wrote:
> > >
>
> I just noticed that this patch is already applied to the upstream - no
pull request, no waiting time for community review -
>

Peter, would you please explain to me what was going on with this patch? I
am truly confused, since all rules and procedures seem to be somehow
circumvented, don't they?

Yours,
Aleksandar

> > > The loop is written with scalars, not vectors.
> > > Use the correct type when incrementing.
> > >
> > > Fixes: 5ee5c14cacd
> > > Reported-by: Laurent Vivier <address@hidden>
> > > Tested-by: Laurent Vivier <address@hidden>
> > > Reviewed-by: Laurent Vivier <address@hidden>
> > > Signed-off-by: Richard Henderson <address@hidden>
> > > ---
> >
> > This is certainly not a case of “typo” (which would mean that you
accidentally typed “vec8” instead od “int8_t”). So, change the title to
“tcg: Fix loop step in
> > helper_gvec_sar{8, 32, 64}v” or similar.
> >
> > >  accel/tcg/tcg-runtime-gvec.c | 6 +++---
> > >  1 file changed, 3 insertions(+), 3 deletions(-)
> > >
> > > diff --git a/accel/tcg/tcg-runtime-gvec.c
b/accel/tcg/tcg-runtime-gvec.c
> > > index 3b6052fe97..51cb29ca79 100644
> > > --- a/accel/tcg/tcg-runtime-gvec.c
> > > +++ b/accel/tcg/tcg-runtime-gvec.c
> > > @@ -874,7 +874,7 @@ void HELPER(gvec_sar8v)(void *d, void *a, void
*b, uint32_t desc)
> > >      intptr_t oprsz = simd_oprsz(desc);
> > >      intptr_t i;
> > >
> > > -    for (i = 0; i < oprsz; i += sizeof(vec8)) {
> > > +    for (i = 0; i < oprsz; i += sizeof(int8_t)) {
> > >          uint8_t sh = *(uint8_t *)(b + i) & 7;
> > >          *(int8_t *)(d + i) = *(int8_t *)(a + i) >> sh;
> > >      }
> > > @@ -898,7 +898,7 @@ void HELPER(gvec_sar32v)(void *d, void *a, void
*b, uint32_t desc)
> > >      intptr_t oprsz = simd_oprsz(desc);
> > >      intptr_t i;
> > >
> > > -    for (i = 0; i < oprsz; i += sizeof(vec32)) {
> > > +    for (i = 0; i < oprsz; i += sizeof(int32_t)) {
> > >          uint8_t sh = *(uint32_t *)(b + i) & 31;
> > >          *(int32_t *)(d + i) = *(int32_t *)(a + i) >> sh;
> > >      }
> > > @@ -910,7 +910,7 @@ void HELPER(gvec_sar64v)(void *d, void *a, void
*b, uint32_t desc)
> > >      intptr_t oprsz = simd_oprsz(desc);
> > >      intptr_t i;
> > >
> > > -    for (i = 0; i < oprsz; i += sizeof(vec64)) {
> > > +    for (i = 0; i < oprsz; i += sizeof(int64_t)) {
> > >          uint8_t sh = *(uint64_t *)(b + i) & 63;
> > >          *(int64_t *)(d + i) = *(int64_t *)(a + i) >> sh;
> > >      }
> > > --
> > > 2.17.1
> > >
> > >