Hi,
This has been tested with the following configurations:
rv64gcv_zvl128b
rv64gcv_zvl256b
rv32imafd_zve32x1p0
rv32gc_zve64f_zvl128b
I'll drop the constraints and add the testing info to the cover email in
v3. I'll hold off submitting v3 until gcc-15 as requested.
On Wed, Dec 20, 2023 at 5:38 AM Jeff Law <jeffreya...@gmail.com> wrote:
>
>
> On 12/19/23 02:53, Sergei Lewis wrote:
> > gcc/ChangeLog
> >
> > * config/riscv/riscv-protos.h (riscv_vector::expand_vec_setmem):
> New function
> > declaration.
> >
> > * config/riscv/riscv-string.cc (riscv_vector::expand_vec_setmem):
> New
> > function: this generates an inline vectorised memory set, if and
> only if we
> > know the entire operation can be performed in a single vector store
> >
> > * config/riscv/riscv.md (setmem<mode>): Try
> riscv_vector::expand_vec_setmem
> > for constant lengths
> >
> > gcc/testsuite/ChangeLog
> > * gcc.target/riscv/rvv/base/setmem-1.c: New tests
> > * gcc.target/riscv/rvv/base/setmem-2.c: New tests
> > * gcc.target/riscv/rvv/base/setmem-3.c: New tests
> As with patch 1/3 this needs to be regression tested. The other
> concern, which I should have voiced with patch 1/3 is that this was
> submitted after the gcc-14 development window closed. While we do have
> some degrees of freedom to accept backend specific new features, we
> really shouldn't be adding new features/optimizations at this point. We
> really should just be fixing bugs and new features should be queued for
> gcc-15.
>
>
>
>
> > diff --git a/gcc/config/riscv/riscv.md b/gcc/config/riscv/riscv.md
> > index 1b3f66fd15c..dd34211ca80 100644
> > --- a/gcc/config/riscv/riscv.md
> > +++ b/gcc/config/riscv/riscv.md
> > @@ -2387,6 +2387,20 @@
> > FAIL;
> > })
> >
> > +(define_expand "setmemsi"
> > + [(set (match_operand:BLK 0 "memory_operand") ;; Dest
> > + (match_operand:QI 2 "nonmemory_operand")) ;; Value
> > + (use (match_operand:SI 1 "const_int_operand")) ;; Length
> > + (match_operand:SI 3 "const_int_operand")] ;; Align
> > + "TARGET_VECTOR"
> > +{
> > + if (riscv_vector::expand_vec_setmem (operands[0], operands[1],
> operands[2],
> > + operands[3]))
> > + DONE;
> > + else
> > + FAIL;
> > +})
> Is the :SI really needed for operands1 and operands3? a CONST_INT node
> never has a mode. Or is the existence of the mode just to keep the
> gen* programs from generating a warning? And if we're going to keep a
> mode, particularly on the length, shouldn't the length be in mode P?
>
>
> Jeff
>
