On 06/11/2023 11:20, Stamatis Markianos-Wright wrote:
Patch has already been approved at:
https://gcc.gnu.org/pipermail/gcc-patches/2023-September/630326.html
... But I'm sending this again for archiving on the list after rebasing
A couple of minor nits:
1)
+#define MVE_VPT_PREDICABLE_INSN_P(INSN)
\
+ (recog_memoized (INSN) >= 0 \
+ && get_attr_mve_unpredicated_insn (INSN) != 0) \
I think it's better to write "!= CODE_FOR_nothing".
+(define_attr "mve_unpredicated_insn" "" (const_int 0))
+
And the default value here should similarly be 'symbol_ref
"CODE_FOR_nothing"'.
So that the style matches the symbol refs elsewhere.
2)
+(define_insn "*predicated_doloop_end_internal"
+ [(set (pc)
+ (if_then_else
+ (ge (plus:SI (reg:SI LR_REGNUM)
+ (match_operand:SI 0 "const_int_operand" ""))
+ (const_int 0))
+ (label_ref (match_operand 1 "" ""))
+ (pc)))
+ (set (reg:SI LR_REGNUM)
+ (plus:SI (reg:SI LR_REGNUM) (match_dup 0)))
+ (clobber (reg:CC CC_REGNUM))]
+ "TARGET_32BIT && TARGET_HAVE_LOB && TARGET_HAVE_MVE && TARGET_THUMB2"
TARGET_THUMB2 => TARGET_32BIT, so the first test is redundant. In fact,
given that TARGET_HAVE_LOB => armv8.1-m.main => thumb2, why do we need
either?
So
TARGET_HAVE_LOB && TARGET_HAVE_MVE
should be sufficient.
+(define_insn "dlstp<mode1>_insn"
+ [
+ (set (reg:SI LR_REGNUM)
+ (unspec:SI [(match_operand:SI 0 "s_register_operand" "r")]
+ DLSTP))
+ ]
+ "TARGET_32BIT && TARGET_HAVE_LOB && TARGET_HAVE_MVE && TARGET_THUMB2"
Same here.
Otherwise, OK.
R.
