Skip to content

R
riscv-gcc-1
Commit 592188a5 by Richard Henderson
Options

i386.c (test splitter): Narrow tests vs paradoxical subregs. 

        * i386.c (test splitter): Narrow tests vs paradoxical subregs.
        (jcc splitters): Add two splitters to simplify compound compares
        that simplify_comparison can't handle.

From-SVN: r46178
parent 7f16eed4
Showing with 77 additions and 15 deletions
gcc/ChangeLog
2001-10-10 Richard Henderson <rth@redhat.com>
* combine.c (try_combine): Handle a SEQUENCE of one insn.
* combine.c (try_combine): Handle a SEQUENCE of one insn.
* i386.c (test splitter): Narrow tests vs paradoxical subregs.
(jcc splitters): Add two splitters to simplify compound compares
that simplify_comparison can't handle.
2001-10-10 Richard Henderson <rth@redhat.com>
......@@ -301,8 +305,8 @@ Wed Oct 10 00:41:29 EDT 2001 John Wehle (john@feith.com)
2001-10-08 Richard Henderson <rth@redhat.com>
* i386.md (movsi_xor): Export.
(setcc peep2): Use it when available; add an alternative to
* i386.md (movsi_xor): Export.
(setcc peep2): Use it when available; add an alternative to
match zero_extendhisi2_and.
2001-10-08 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
......@@ -533,7 +537,7 @@ Sat Oct 6 07:42:39 2001 Richard Kenner <kenner@vlsi1.ultra.nyu.edu>
2001-10-06 Ulrich Weigand <uweigand@de.ibm.com>
* config/s390/s390.c (s390_va_arg): Use set_mem_alias_set.
* config/s390/s390.c (s390_va_arg): Use set_mem_alias_set.
2001-10-05 Joseph S. Myers <jsm28@cam.ac.uk>
......@@ -541,15 +545,15 @@ Sat Oct 6 07:42:39 2001 Richard Kenner <kenner@vlsi1.ultra.nyu.edu>
2001-10-05 Catherine Moore <clm@redhat.com>
* config/stormy16/stormy-abi: Document abi changes.
* config/stormy16/stormy16.h (REG_ALLOC_ORDER): Redefine.
(ICALL_REGS): New register class.
(NUM_ARGUMENT_REGISTERS): Now 6.
(DEFAULT_PCC_STRUCT_RETURN): Don't try to return structs
in registers.
(REG_CLASS_FROM_LETTER): Map 'z' to ICALL_REGS.
* config/stormy16.md (call_internal, call_value_internal,
indirect_jump): Use new 'z' constraint.
* config/stormy16/stormy-abi: Document abi changes.
* config/stormy16/stormy16.h (REG_ALLOC_ORDER): Redefine.
(ICALL_REGS): New register class.
(NUM_ARGUMENT_REGISTERS): Now 6.
(DEFAULT_PCC_STRUCT_RETURN): Don't try to return structs
in registers.
(REG_CLASS_FROM_LETTER): Map 'z' to ICALL_REGS.
* config/stormy16.md (call_internal, call_value_internal,
indirect_jump): Use new 'z' constraint.
2001-10-05 Neil Booth <neil@daikokuya.demon.co.uk>
......@@ -1055,7 +1059,7 @@ Mon Oct 1 06:43:41 2001 Richard Kenner <kenner@vlsi1.ultra.nyu.edu>
storehps builtins.
* doc/extend.texi (Vector Extensions): New node.
* doc/invoke.texi (Machine Dependent Options): Add documentation for
i386 -mmmx, -msse, -m3dnow.
i386 -mmmx, -msse, -m3dnow.
Sat Sep 29 15:08:16 CEST 2001 Jan Hubicka <jh@suse.cz>
......
gcc/config/i386/i386.md
......@@ -8125,7 +8125,7 @@
HOST_WIDE_INT len = INTVAL (operands[1]);
HOST_WIDE_INT pos = INTVAL (operands[2]);
HOST_WIDE_INT mask;
enum machine_mode mode;
enum machine_mode mode, submode;
mode = GET_MODE (operands[0]);
if (GET_CODE (operands[0]) == MEM)
......@@ -8138,6 +8138,15 @@
operands[0] = adjust_address (operands[0], mode, 0);
}
}
else if (GET_CODE (operands[0]) == SUBREG
&& (submode = GET_MODE (SUBREG_REG (operands[0])),
GET_MODE_BITSIZE (mode) > GET_MODE_BITSIZE (submode))
&& pos + len <= GET_MODE_BITSIZE (submode))
{
/* Narrow a paradoxical subreg to prevent partial register stalls. */
mode = submode;
operands[0] = SUBREG_REG (operands[0]);
}
else if (mode == HImode && pos + len <= 8)
{
/* Small HImode tests can be converted to QImode. */
......@@ -12745,6 +12754,55 @@
(const_int 0)
(const_int 1)))])
;; In general it is not safe to assume too much about CCmode registers,
;; so simplify-rtx stops when it sees a second one. Under certain
;; conditions this is safe on x86, so help combine not create
;;
;; seta %al
;; testb %al, %al
;; je Lfoo
(define_split
[(set (pc)
(if_then_else (ne (match_operator 0 "ix86_comparison_operator"
[(reg 17) (const_int 0)])
(const_int 0))
(label_ref (match_operand 1 "" ""))
(pc)))]
""
[(set (pc)
(if_then_else (match_dup 0)
(label_ref (match_dup 1))
(pc)))]
{
PUT_MODE (operands[0], VOIDmode);
})
(define_split
[(set (pc)
(if_then_else (eq (match_operator 0 "ix86_comparison_operator"
[(reg 17) (const_int 0)])
(const_int 0))
(label_ref (match_operand 1 "" ""))
(pc)))]
""
[(set (pc)
(if_then_else (match_dup 0)
(label_ref (match_dup 1))
(pc)))]
{
rtx new_op0 = copy_rtx (operands[0]);
operands[0] = new_op0;
PUT_MODE (new_op0, VOIDmode);
PUT_CODE (new_op0, REVERSE_CONDITION (GET_CODE (new_op0),
GET_MODE (XEXP (new_op0, 0))));
/* Make sure that (a) the CCmode we have for the flags is strong
enough for the reversed compare or (b) we have a valid FP compare. */
if (! ix86_comparison_operator (new_op0, VOIDmode))
FAIL;
})
;; Define combination compare-and-branch fp compare instructions to use
;; during early optimization. Splitting the operation apart early makes
;; for bad code when we want to reverse the operation.
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment