CRIS atomics revisited 3/4: pattern improvements
Hans-Peter Nilsson
hans-peter.nilsson@axis.com
Mon Jul 16 03:44:00 GMT 2012
Microoptimizations for the atomic patterns themselves. Constant
operands are so common that it seems wasteful not to handle the
most common cases and avoid wasting a register.
gcc/testsuite:
* gcc.target/cris/20011127-1.c: Adjust to %P being a
valid register operand output modifier.
gcc:
* config/cris/cris.c (cris_print_operand) <case 'P', 'q'>: New cases.
* config/cris/sync.md (atomic_op_op_cnstr): New code_attr.
(atomic_op_op_pred): Ditto.
(atomic_op_mnem_pre_op2): Renamed from atomic_op_mnem_pre; to
reflect the change to include %2 in expansion. All callers changed.
(qm3): New mode_attr.
("atomic_fetch_<atomic_op_name><mode>"): Use <atomic_op_op_pred>
as predicate for operand 2.
("cris_atomic_fetch_<atomic_op_name><mode>_1"): Update FIXME. Use
"<atomic_op_op_pred>" "<atomic_op_op_cnstr>" for predicate and
constraint for operand 2.
("atomic_compare_and_swap<mode>"): Add FIXME. Change predicate to
nonmemory_operand for operand 3.
("cris_atomic_compare_and_swap<mode>_1"): Change operand 3 to
exclude memory. Improve emitted sync code for v10 and v32. Use
<qm3> instead of <m> for size designator for cmp.
Index: config/cris/cris.c
===================================================================
--- config/cris/cris.c (revision 189499)
+++ config/cris/cris.c (working copy)
@@ -981,6 +981,53 @@ cris_print_operand (FILE *file, rtx x, i
fprintf (file, INTVAL (operand) < 0 ? "adds.w" : "addq");
return;
+ case 'P':
+ /* For const_int operands, print the additive mnemonic and the
+ modified operand (byte-sized operands don't save anything):
+ N=MIN_INT..-65536: add.d N
+ -65535..-64: subu.w -N
+ -63..-1: subq -N
+ 0..63: addq N
+ 64..65535: addu.w N
+ 65536..MAX_INT: add.d N.
+ (Emitted mnemonics are capitalized to simplify testing.)
+ For anything else (N.B: only register is valid), print "add.d". */
+ if (REG_P (operand))
+ {
+ fprintf (file, "Add.d ");
+
+ /* Deal with printing the operand by dropping through to the
+ normal path. */
+ break;
+ }
+ else
+ {
+ int val;
+ gcc_assert (CONST_INT_P (operand));
+
+ val = INTVAL (operand);
+ if (!IN_RANGE (val, -65535, 65535))
+ fprintf (file, "Add.d %d", val);
+ else if (val <= -64)
+ fprintf (file, "Subu.w %d", -val);
+ else if (val <= -1)
+ fprintf (file, "Subq %d", -val);
+ else if (val <= 63)
+ fprintf (file, "Addq %d", val);
+ else if (val <= 65535)
+ fprintf (file, "Addu.w %d", val);
+ return;
+ }
+ break;
+
+ case 'q':
+ /* If the operand is an integer -31..31, print "q" else ".d". */
+ if (CONST_INT_P (operand) && IN_RANGE (INTVAL (operand), -31, 31))
+ fprintf (file, "q");
+ else
+ fprintf (file, ".d");
+ return;
+
case 'd':
/* If this is a GOT symbol, force it to be emitted as :GOT and
:GOTPLT regardless of -fpic (i.e. not as :GOT16, :GOTPLT16).
Index: config/cris/sync.md
===================================================================
--- config/cris/sync.md (revision 189501)
+++ config/cris/sync.md (working copy)
@@ -73,17 +73,32 @@ (define_code_iterator atomic_op [plus mi
(define_code_attr atomic_op_name
[(plus "add") (minus "sub") (and "and") (ior "or") (xor "xor") (mult "nand")])
+;; The operator nonatomic-operand can be memory, constant or register
+;; for all but xor. We can't use memory or addressing modes with
+;; side-effects though, so just use registers and literal constants.
+(define_code_attr atomic_op_op_cnstr
+ [(plus "ri") (minus "ri") (and "ri") (ior "ri") (xor "r") (mult "ri")])
+
+(define_code_attr atomic_op_op_pred
+ [(plus "nonmemory_operand") (minus "nonmemory_operand")
+ (and "nonmemory_operand") (ior "nonmemory_operand")
+ (xor "register_operand") (mult "nonmemory_operand")])
+
;; Pairs of these are used to insert the "not" after the "and" for nand.
-(define_code_attr atomic_op_mnem_pre ;; Upper-case only to sinplify testing.
- [(plus "Add.d") (minus "Sub.d") (and "And.d") (ior "Or.d") (xor "Xor")
- (mult "aNd.d")])
+(define_code_attr atomic_op_mnem_pre_op2 ;; Upper-case only to simplify testing.
+ [(plus "%P2") (minus "Sub.d %2") (and "And%q2 %2") (ior "Or%q2 %2") (xor "Xor %2")
+ (mult "aNd%q2 %2")])
+
(define_code_attr atomic_op_mnem_post_op3
[(plus "") (minus "") (and "") (ior "") (xor "") (mult "not %3\;")])
+;; For SImode, emit "q" for operands -31..31.
+(define_mode_attr qm3 [(SI "%q3") (HI ".w") (QI ".b")])
+
(define_expand "atomic_fetch_<atomic_op_name><mode>"
[(match_operand:BWD 0 "register_operand")
(match_operand:BWD 1 "memory_operand")
- (match_operand:BWD 2 "register_operand")
+ (match_operand:BWD 2 "<atomic_op_op_pred>")
(match_operand 3)
(atomic_op:BWD (match_dup 0) (match_dup 1))]
""
@@ -109,8 +124,9 @@ (define_insn "cris_atomic_fetch_<atomic_
[(set (match_operand:BWD 1 "memory_operand" "+Q")
(atomic_op:BWD
(unspec_volatile:BWD [(match_dup 1)] CRIS_UNSPEC_ATOMIC_OP)
- ;; FIXME: relax this for plus, minus, and, ior.
- (match_operand:BWD 2 "register_operand" "r")))
+ ;; FIXME: improve constants more for plus, minus, and, ior.
+ ;; FIXME: handle memory operands without side-effects.
+ (match_operand:BWD 2 "<atomic_op_op_pred>" "<atomic_op_op_cnstr>")))
(set (match_operand:BWD 0 "register_operand" "=&r")
(match_dup 1))
(clobber (match_scratch:SI 3 "=&r"))]
@@ -125,7 +141,7 @@ (define_insn "cris_atomic_fetch_<atomic_
".Lsync.%=:\;"
"move<m> %1,%0\;"
"move.d %0,%3\;"
- "<atomic_op_mnem_pre> %2,%3\;<atomic_op_mnem_post_op3>"
+ "<atomic_op_mnem_pre_op2>,%3\;<atomic_op_mnem_post_op3>"
"ax\;"
"move<m> %3,%1\;"
"bcs .Lsync.%=\;"
@@ -136,7 +152,7 @@ (define_insn "cris_atomic_fetch_<atomic_
".Lsync.%=:\;"
"move<m> %1,%0\;"
"move.d %0,%3\;"
- "<atomic_op_mnem_pre> %2,%3\;<atomic_op_mnem_post_op3>"
+ "<atomic_op_mnem_pre_op2>,%3\;<atomic_op_mnem_post_op3>"
"ax\;"
"move<m> %3,%1\;"
"bwf .Lsync.%=\;"
@@ -167,12 +183,12 @@ (define_insn "cris_atomic_fetch_<atomic_
"bmi .Lsync.irqon.%=\;"
"move.d %0,%3\;"
- "<atomic_op_mnem_pre> %2,%3\;<atomic_op_mnem_post_op3>"
+ "<atomic_op_mnem_pre_op2>,%3\;<atomic_op_mnem_post_op3>"
"ba .Lsync.irqoff.%=\;"
"move<m> %3,%1\n"
".Lsync.irqon.%=:\;"
- "<atomic_op_mnem_pre> %2,%3\;<atomic_op_mnem_post_op3>"
+ "<atomic_op_mnem_pre_op2>,%3\;<atomic_op_mnem_post_op3>"
"move<m> %3,%1\;"
"ei\n"
".Lsync.irqoff.%=:";
@@ -232,32 +248,30 @@ (define_insn "cris_atomic_compare_and_sw
{
if (TARGET_V32)
return
- "clearf p\n"
- ".Lsync.repeat.%=:\;"
+ "\n.Lsync.repeat.%=:\;"
+ "clearf p\;"
"move<m> %2,%1\;"
- "cmp<m> %3,%1\;"
+ "cmp<qm3> %3,%1\;"
"bne .Lsync.after.%=\;"
- "seq %0\;"
-
"ax\;"
+
"move<m> %4,%2\;"
- "bcs .Lsync.repeat.%=\;"
- "clearf p\n"
- ".Lsync.after.%=:";
+ "bcs .Lsync.repeat.%=\n"
+ ".Lsync.after.%=:\;"
+ "seq %0";
else if (cris_cpu_version == 10)
return
- "clearf\n"
- ".Lsync.repeat.%=:\;"
+ "\n.Lsync.repeat.%=:\;"
+ "clearf\;"
"move<m> %2,%1\;"
- "cmp<m> %3,%1\;"
+ "cmp<qm3> %3,%1\;"
"bne .Lsync.after.%=\;"
- "seq %0\;"
-
"ax\;"
+
"move<m> %4,%2\;"
- "bwf .Lsync.repeat.%=\;"
- "clearf\n"
- ".Lsync.after.%=:";
+ "bwf .Lsync.repeat.%=\n"
+ ".Lsync.after.%=:\;"
+ "seq %0";
else
{
/* This one is for CRIS versions without load-locked-store-conditional
@@ -284,14 +298,14 @@ (define_insn "cris_atomic_compare_and_sw
"bmi .Lsync.irqon.%=\;"
"nop\;"
- "cmp<m> %3,%1\;"
+ "cmp<qm3> %3,%1\;"
"bne .Lsync.after.%=\;"
"seq %0\;"
"ba .Lsync.after.%=\;"
"move<m> %4,%2\n"
".Lsync.irqon.%=:\;"
- "cmp<m> %3,%1\;"
+ "cmp<qm3> %3,%1\;"
"bne .Lsync.after.%=\;"
"seq %0\;"
"move<m> %4,%2\;"
Index: testsuite/gcc.target/cris/20011127-1.c
===================================================================
--- testsuite/gcc.target/cris/20011127-1.c (revision 189499)
+++ testsuite/gcc.target/cris/20011127-1.c (working copy)
@@ -17,7 +17,6 @@ foo (void)
them. */
asm ("\n;# %b0" : : "r" (0)); /* { dg-error "modifier" } */
asm ("\n;# %v0" : : "r" (0)); /* { dg-error "modifier" } */
- asm ("\n;# %P0" : : "r" (0)); /* { dg-error "modifier" } */
asm ("\n;# %p0" : : "r" (0)); /* { dg-error "modifier" } */
asm ("\n;# %z0" : : "r" (0)); /* { dg-error "modifier" } */
asm ("\n;# %H0" : : "F" (0.5)); /* { dg-error "modifier" } */
brgds, H-P
More information about the Gcc-patches
mailing list