https://gcc.gnu.org/g:61bb3adee51d749d131af5582c79d1c5a432eec9
commit 61bb3adee51d749d131af5582c79d1c5a432eec9
Author: Michael Meissner <meiss...@linux.ibm.com>
Date: Thu May 22 15:49:49 2025 -0400
Update ChangeLog.*
Diff:
---
gcc/ChangeLog.bugs | 89 +++++++++++-------------------------------------------
1 file changed, 18 insertions(+), 71 deletions(-)
diff --git a/gcc/ChangeLog.bugs b/gcc/ChangeLog.bugs
index 0134634d76f9..f9eeb7439c96 100644
--- a/gcc/ChangeLog.bugs
+++ b/gcc/ChangeLog.bugs
@@ -1,4 +1,4 @@
-==================== Branch work206-bugs, patch #112 ====================
+==================== Branch work206-bugs, patch #113 ====================
Fix PR 118541, do not generate unordered fp cmoves for IEEE compares.
@@ -35,87 +35,34 @@ GCC currently generates the following code:
This is because ifcvt.c optimizes the conditional floating point move to use
the
XSCMPGTDP instruction.
-However, the XSCMPGTDP instruction will generate an interrupt if one of the
-arguments is a signalling NaN and signalling NaNs can generate an interrupt.
-The IEEE comparison functions (isgreater, etc.) require that the comparison not
-raise an interrupt.
+However, the XSCMPGTDP instruction traps if one of the arguments is a signaling
+NaN. This patch disables generating XSCMP{EQ,GT,GE}{DP,QP} instructions unless
+-ffinite-math-only is in effect so that we do not get a trap.
-The root cause of this is we allow floating point comparisons to be reversed
-(i.e. LT will be reversed to UNGE). Before power9, this was ok because we only
-generated the FCMPU or XSCMPUDP instructions.
-
-But with power9, we can generate the XSCMPEQDP, XSCMPGTDP, or XSCMPGEDP
-instructions. This code now does not convert an unordered compare into an
-ordered compare. Instead, it does the opposite comparison and swaps the
-arguments. I.e. it converts:
-
- r = (a < b) ? c : d;
-
-into:
-
- r = (b >= a) ? c : d;
-
-For the following code:
-
- double
- ordered_compare (double a, double b, double c, double d)
- {
- return __builtin_isgreater (a, b) ? c : d;
- }
-
- /* Verify normal > does generate xscmpgtdp. */
-
- double
- normal_compare (double a, double b, double c, double d)
- {
- return a > b ? c : d;
- }
-
-with the following patch, GCC generates the following for power9, power10, and
-power11:
-
- ordered_compare:
- fcmpu 0,1,2
- fmr 1,4
- bnglr 0
- fmr 1,3
- blr
-
- normal_compare:
- xscmpgtdp 1,1,2
- xxsel 1,4,3,1
- blr
-
-I have built bootstrap compilers on big endian power9 systems and little endian
-power9/power10 systems and there were no regressions. Can I check this patch
-into the GCC trunk, and after a waiting period, can I check this into the
active
-older branches?
-
-2025-05-21 Michael Meissner <meiss...@linux.ibm.com>
+2025-05-22 Michael Meissner <meiss...@linux.ibm.com>
gcc/
PR target/118541
- * config/rs6000/predicates.md (invert_fpmask_comparison_operator):
- Delete.
- (fpmask_reverse_args_comparison_operator): New predicate.
- * config/rs6000/rs6000-proto.h (rs6000_fpmask_reverse_args): New
- declaration.
- * config/rs6000/rs6000.cc (rs6000_fpmask_reverse_args): New function.
- * config/rs6000/rs6000.h (REVERSIBLE_CC_MODE): Do not allow floating
- point comparisons to be reversed unless -ffinite-math-only is used.
- * config/rs6000/rs6000.md (mov<SFDF:mode><SFDF2:mode>cc_p9): Add
- comment.
- (mov<SFDF:mode><SFDF2:mode>cc_invert_p9): Reverse the argument order for
- the comparison, and use an unordered comparison, instead of ordered
- comparison.
+ * config/rs6000/rs6000.md (mov<SFDF:mode><SFDF2:mode>cc_p9): Disable
+ generating XSCMP{EQ,GT,GE}{DP,QP} unless -ffinite-math-only is in
+ effect.
+ (mov<SFDF:mode><SFDF2:mode>cc_invert_p9): Likewise.
+ (fpmask<mode>, SFDF iterator): Likewise.
+ (xxsel<mode>, SFDF iterator): Likewise.
+ (mov<mode>cc, IEEE128 iterator): Likewise.
+ (mov<mode>cc_p10): Likewise.
(mov<mode>cc_invert_p10): Likewise.
+ (fpmask<mode>, IEEE128 iterator): Likewise.
+ (xxsel<mode>, IEEE128 iterator): Likewise.
gcc/testsuite/
PR target/118541
- * gcc.target/powerpc/pr118541.c: New test.
+ * gcc.target/powerpc/float128-cmove.c: Change optimization flag to
+ -Ofast instead of -O2.
+==================== Branch work206-bugs, patch #112 was reverted
====================
==================== Branch work206-bugs, patch #111 was reverted
====================
==================== Branch work206-bugs, patch #110 was reverted
====================
