| Message ID | 20250312103818.370-1-philmd@linaro.org |
|---|---|
| State | Superseded |
| Headers | show |
| Series | [PATCH-for-10.1] target/i386: Replace MO_TE -> MO_LE | expand |
On 3/12/25 03:38, Philippe Mathieu-Daudé wrote: > The x86 architecture is only implemented as little-endian. > The MO_TE definition always expands to MO_LE. > > Replace: > - MO_TEUQ -> MO_LE | MO_UQ > - MO_TE -> MO_LE > > Signed-off-by: Philippe Mathieu-Daudé<philmd@linaro.org> > --- > target/i386/tcg/emit.c.inc | 7 ++++--- > 1 file changed, 4 insertions(+), 3 deletions(-) Is there a reason not to use MO_LEUQ? Anyway, Reviewed-by: Richard Henderson <richard.henderson@linaro.org> r~
On 12/3/25 13:52, Richard Henderson wrote: > On 3/12/25 03:38, Philippe Mathieu-Daudé wrote: >> The x86 architecture is only implemented as little-endian. >> The MO_TE definition always expands to MO_LE. >> >> Replace: >> - MO_TEUQ -> MO_LE | MO_UQ >> - MO_TE -> MO_LE >> >> Signed-off-by: Philippe Mathieu-Daudé<philmd@linaro.org> >> --- >> target/i386/tcg/emit.c.inc | 7 ++++--- >> 1 file changed, 4 insertions(+), 3 deletions(-) > > Is there a reason not to use MO_LEUQ? No, I'll respin using it. > Anyway, > > Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Thanks!
diff --git a/target/i386/tcg/emit.c.inc b/target/i386/tcg/emit.c.inc index 0fa1664a24f..c7a40a2f9e4 100644 --- a/target/i386/tcg/emit.c.inc +++ b/target/i386/tcg/emit.c.inc @@ -1796,7 +1796,7 @@ static void gen_CMPXCHG(DisasContext *s, X86DecodedInsn *decode) static void gen_CMPXCHG16B(DisasContext *s, X86DecodedInsn *decode) { #ifdef TARGET_X86_64 - MemOp mop = MO_TE | MO_128 | MO_ALIGN; + MemOp mop = MO_LE | MO_128 | MO_ALIGN; TCGv_i64 t0, t1; TCGv_i128 cmp, val; @@ -1853,10 +1853,11 @@ static void gen_CMPXCHG8B(DisasContext *s, X86DecodedInsn *decode) /* Only require atomic with LOCK; non-parallel handled in generator. */ if (s->prefix & PREFIX_LOCK) { - tcg_gen_atomic_cmpxchg_i64(old, s->A0, cmp, val, s->mem_index, MO_TEUQ); + tcg_gen_atomic_cmpxchg_i64(old, s->A0, cmp, val, + s->mem_index, MO_LE | MO_UQ); } else { tcg_gen_nonatomic_cmpxchg_i64(old, s->A0, cmp, val, - s->mem_index, MO_TEUQ); + s->mem_index, MO_LE | MO_UQ); } /* Set tmp0 to match the required value of Z. */
The x86 architecture is only implemented as little-endian. The MO_TE definition always expands to MO_LE. Replace: - MO_TEUQ -> MO_LE | MO_UQ - MO_TE -> MO_LE Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org> --- target/i386/tcg/emit.c.inc | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-)