public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
* [Bug target/52479] New: SH Target: SH4A DFmode fsca tests failing
@ 2012-03-04 13:20 olegendo at gcc dot gnu.org
2012-03-04 13:47 ` [Bug target/52479] " kkojima at gcc dot gnu.org
` (4 more replies)
0 siblings, 5 replies; 6+ messages in thread
From: olegendo at gcc dot gnu.org @ 2012-03-04 13:20 UTC (permalink / raw)
To: gcc-bugs
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52479
Bug #: 52479
Summary: SH Target: SH4A DFmode fsca tests failing
Classification: Unclassified
Product: gcc
Version: 4.8.0
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: target
AssignedTo: unassigned@gcc.gnu.org
ReportedBy: olegendo@gcc.gnu.org
CC: kkojima@gcc.gnu.org
Target: sh4a-*-*
As of rev 184877 the following SH4A target tests are failing:
gcc.target/sh/sh4a-cos.c scan-assembler fsca
gcc.target/sh/sh4a-sin.c scan-assembler fsca
gcc.target/sh/sh4a-sincos.c scan-assembler-times fsca 1
These tests check whether the fsca instruction is used for DFmode standard sin
and cos functions when -ffast-math is specified.
While using fsca for SFmode and -ffast-math makes sense, for DFmode I think the
results might be too imprecise. I would like to propose to remove the DFmode
fsca code altogether.
Using built-in specs.
COLLECT_GCC=sh-elf-gcc
COLLECT_LTO_WRAPPER=/usr/local/libexec/gcc/sh-elf/4.8.0/lto-wrapper
Target: sh-elf
Configured with: ../gcc-trunk/configure --target=sh-elf --prefix=/usr/local
--enable-languages=c,c++ --enable-multilib --disable-libssp --disable-nls
--disable-werror --enable-lto --with-newlib --with-gnu-as --with-gnu-ld
--with-system-zlib
Thread model: single
gcc version 4.8.0 20120304 (experimental) (GCC)
^ permalink raw reply [flat|nested] 6+ messages in thread* [Bug target/52479] SH Target: SH4A DFmode fsca tests failing
2012-03-04 13:20 [Bug target/52479] New: SH Target: SH4A DFmode fsca tests failing olegendo at gcc dot gnu.org
@ 2012-03-04 13:47 ` kkojima at gcc dot gnu.org
2012-03-15 5:28 ` aoliva at gcc dot gnu.org
` (3 subsequent siblings)
4 siblings, 0 replies; 6+ messages in thread
From: kkojima at gcc dot gnu.org @ 2012-03-04 13:47 UTC (permalink / raw)
To: gcc-bugs
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52479
Kazumoto Kojima <kkojima at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |aoliva at gcc dot gnu.org
--- Comment #1 from Kazumoto Kojima <kkojima at gcc dot gnu.org> 2012-03-04 13:46:58 UTC ---
I'd like to add Alex to the CC list. Alex, what do you think?
^ permalink raw reply [flat|nested] 6+ messages in thread* [Bug target/52479] SH Target: SH4A DFmode fsca tests failing
2012-03-04 13:20 [Bug target/52479] New: SH Target: SH4A DFmode fsca tests failing olegendo at gcc dot gnu.org
2012-03-04 13:47 ` [Bug target/52479] " kkojima at gcc dot gnu.org
@ 2012-03-15 5:28 ` aoliva at gcc dot gnu.org
2012-03-16 5:32 ` kkojima at gcc dot gnu.org
` (2 subsequent siblings)
4 siblings, 0 replies; 6+ messages in thread
From: aoliva at gcc dot gnu.org @ 2012-03-15 5:28 UTC (permalink / raw)
To: gcc-bugs
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52479
--- Comment #2 from Alexandre Oliva <aoliva at gcc dot gnu.org> 2012-03-15 04:47:07 UTC ---
IMHO, the whole point of -ffast-math is to trade precision for speed, so the
use of fsca makes sense to me, but I wouldn't mind if we no longer used it for
DFmode.
^ permalink raw reply [flat|nested] 6+ messages in thread* [Bug target/52479] SH Target: SH4A DFmode fsca tests failing
2012-03-04 13:20 [Bug target/52479] New: SH Target: SH4A DFmode fsca tests failing olegendo at gcc dot gnu.org
2012-03-04 13:47 ` [Bug target/52479] " kkojima at gcc dot gnu.org
2012-03-15 5:28 ` aoliva at gcc dot gnu.org
@ 2012-03-16 5:32 ` kkojima at gcc dot gnu.org
2012-03-21 20:28 ` olegendo at gcc dot gnu.org
2012-04-11 23:05 ` olegendo at gcc dot gnu.org
4 siblings, 0 replies; 6+ messages in thread
From: kkojima at gcc dot gnu.org @ 2012-03-16 5:32 UTC (permalink / raw)
To: gcc-bugs
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52479
--- Comment #3 from Kazumoto Kojima <kkojima at gcc dot gnu.org> 2012-03-16 03:21:08 UTC ---
There is no concrete definition of -ffast-math and users will have
different expectations. Numerical programs for astrodynamics may
expect precisions even for -ffast-math and OTOH there are many
programs which don't take care of precisions at all and prefer
sin&cos instead of sinf&cosf. I guess that the former doesn't make
much sense for SH4A in the first place and for users of the latter
with -ffast-math, the proposed change may be a bit surprising.
I have no strong opinion for this, though. I also won't object to
the suggested patch.
^ permalink raw reply [flat|nested] 6+ messages in thread
* [Bug target/52479] SH Target: SH4A DFmode fsca tests failing
2012-03-04 13:20 [Bug target/52479] New: SH Target: SH4A DFmode fsca tests failing olegendo at gcc dot gnu.org
` (2 preceding siblings ...)
2012-03-16 5:32 ` kkojima at gcc dot gnu.org
@ 2012-03-21 20:28 ` olegendo at gcc dot gnu.org
2012-04-11 23:05 ` olegendo at gcc dot gnu.org
4 siblings, 0 replies; 6+ messages in thread
From: olegendo at gcc dot gnu.org @ 2012-03-21 20:28 UTC (permalink / raw)
To: gcc-bugs
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52479
--- Comment #4 from Oleg Endo <olegendo at gcc dot gnu.org> 2012-03-21 20:24:59 UTC ---
Author: olegendo
Date: Wed Mar 21 20:24:41 2012
New Revision: 185617
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=185617
Log:
PR target/52479
* config/sh/sh-protos.h (sh_fsca_df2int): Remove.
* config/sh/sh.c (sh_fsca_df2int_rtx, sh_fsca_df2int): Remove.
* config/sh/sh.md (sindf2, cosdf2): Remove.
* gcc.target/sh/sh4a-cos.c: Remove.
* gcc.target/sh/sh4a-sin.c: Remove.
* gcc.target/sh/sh4a-sincos.c: Remove.
Removed:
trunk/gcc/testsuite/gcc.target/sh/sh4a-cos.c
trunk/gcc/testsuite/gcc.target/sh/sh4a-sin.c
trunk/gcc/testsuite/gcc.target/sh/sh4a-sincos.c
Modified:
trunk/gcc/ChangeLog
trunk/gcc/config/sh/sh-protos.h
trunk/gcc/config/sh/sh.c
trunk/gcc/config/sh/sh.md
trunk/gcc/testsuite/ChangeLog
^ permalink raw reply [flat|nested] 6+ messages in thread* [Bug target/52479] SH Target: SH4A DFmode fsca tests failing
2012-03-04 13:20 [Bug target/52479] New: SH Target: SH4A DFmode fsca tests failing olegendo at gcc dot gnu.org
` (3 preceding siblings ...)
2012-03-21 20:28 ` olegendo at gcc dot gnu.org
@ 2012-04-11 23:05 ` olegendo at gcc dot gnu.org
4 siblings, 0 replies; 6+ messages in thread
From: olegendo at gcc dot gnu.org @ 2012-04-11 23:05 UTC (permalink / raw)
To: gcc-bugs
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52479
Oleg Endo <olegendo at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|UNCONFIRMED |RESOLVED
Resolution| |FIXED
--- Comment #5 from Oleg Endo <olegendo at gcc dot gnu.org> 2012-04-11 23:04:53 UTC ---
Removed support for DFmode fsca insn in rev 185617.
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2012-04-11 23:05 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2012-03-04 13:20 [Bug target/52479] New: SH Target: SH4A DFmode fsca tests failing olegendo at gcc dot gnu.org
2012-03-04 13:47 ` [Bug target/52479] " kkojima at gcc dot gnu.org
2012-03-15 5:28 ` aoliva at gcc dot gnu.org
2012-03-16 5:32 ` kkojima at gcc dot gnu.org
2012-03-21 20:28 ` olegendo at gcc dot gnu.org
2012-04-11 23:05 ` olegendo at gcc dot gnu.org
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).