On Fri, Feb 25, 2022 at 10:33 AM Jakub Jelinek <ja...@redhat.com> wrote:
>
> Hi!
>
> As mentioned in the PR, the following testcase is miscompiled for similar
> reasons as the already fixed PR78791 - we use SLOT_TEMP slots in various
> places during expansion and during expansion we can guarantee that the
> lifetime of those temporary slot doesn't overlap. But the following
> splitter uses SLOT_TEMP too and in between expansion and split1 there is
> a possibility that something extends the lifetime of SLOT_TEMP created
> slots across an instruction that will be split by this splitter.
>
> The following patch fixes it by using a new temp slot kind to make sure
> it doesn't reuse a SLOT_TEMP that could be live across the instruction.
>
> Bootstrapped/regtested on x86_64-linux and i686-linux, ok for trunk?
>
> 2022-02-24 Jakub Jelinek <ja...@redhat.com>
>
> PR target/104674
> * config/i386/i386.h (enum ix86_stack_slot): Add SLOT_FLOATxFDI_387.
> * config/i386/i386.md (splitter to floatdi<mode>2_i387_with_xmm): Use
> SLOT_FLOATxFDI_387 rather than SLOT_TEMP.
>
> * gcc.target/i386/pr104674.c: New test.
OK.
Thanks,
Uros.
>
> --- gcc/config/i386/i386.h.jj 2022-01-18 11:58:59.118988685 +0100
> +++ gcc/config/i386/i386.h 2022-02-24 13:47:02.809289843 +0100
> @@ -2414,6 +2414,7 @@ enum ix86_stack_slot
> SLOT_CW_FLOOR,
> SLOT_CW_CEIL,
> SLOT_STV_TEMP,
> + SLOT_FLOATxFDI_387,
> MAX_386_STACK_LOCALS
> };
>
> --- gcc/config/i386/i386.md.jj 2022-02-12 11:17:35.149999347 +0100
> +++ gcc/config/i386/i386.md 2022-02-24 13:35:57.832561968 +0100
> @@ -5412,9 +5412,8 @@ (define_split
> && can_create_pseudo_p ()"
> [(const_int 0)]
> {
> - emit_insn (gen_floatdi<mode>2_i387_with_xmm
> - (operands[0], operands[1],
> - assign_386_stack_local (DImode, SLOT_TEMP)));
> + rtx s = assign_386_stack_local (DImode, SLOT_FLOATxFDI_387);
> + emit_insn (gen_floatdi<mode>2_i387_with_xmm (operands[0], operands[1], s));
> DONE;
> })
>
> --- gcc/testsuite/gcc.target/i386/pr104674.c.jj 2022-02-24 13:45:31.630561330
> +0100
> +++ gcc/testsuite/gcc.target/i386/pr104674.c 2022-02-24 13:45:21.278705687
> +0100
> @@ -0,0 +1,31 @@
> +/* PR target/104674 */
> +/* { dg-do run { target sse2_runtime } } */
> +/* { dg-options "-O2 -msse2 -mfpmath=sse" } */
> +
> +__attribute__((noipa)) double
> +bar (double x, double y)
> +{
> + return x + y;
> +}
> +
> +__attribute__((noipa)) double
> +foo (long long x)
> +{
> + long long a = x / 10000000;
> + int b = x % 10000000;
> + double s = (double) a;
> + double n = (double) b / 1e7;
> + double t = s + n;
> + if (t == s + 1.0)
> + t = bar (t, s);
> + return t;
> +}
> +
> +int
> +main ()
> +{
> + long long n = 888888;
> + n = n * 10000000;
> + if (foo (n) != 888888.0)
> + __builtin_abort ();
> +}
>
> Jakub
>
