diff options
| author | Andreas Krebbel <krebbel@linux.vnet.ibm.com> | 2016-03-11 07:24:41 +0000 |
|---|---|---|
| committer | Andreas Krebbel <krebbel@gcc.gnu.org> | 2016-03-11 07:24:41 +0000 |
| commit | ae8e301e08b084da8e2142aa195135cae718cd70 (patch) | |
| tree | ff8ea622f6b57f8864a346ec9013099a62faf4f1 /gcc/tree-ssa-loop.c | |
| parent | dd95128ba97f4b43145d76d6624de52e2cbda074 (diff) | |
| download | gcc-ae8e301e08b084da8e2142aa195135cae718cd70.zip gcc-ae8e301e08b084da8e2142aa195135cae718cd70.tar.gz gcc-ae8e301e08b084da8e2142aa195135cae718cd70.tar.bz2 | |
S/390: Define macros for rounding mode constants
This patch replaces a few magic numbers used for floating point
rounding modes with macros. This is mostly a NoOp change apart from:
fixuns_truncdddi2, fixuns_trunctddi2, fixuns_trunc<mode>si2: Replace 5
with DFP_RND_TOWARD_0 (which is 9).
5 as well as 9 represent round towards 0 with the difference that for
5 the new DFP quantum exception is enabled as well. This exception
isn't IEEE754 and we do not have an interface to enable and test it
anyway. So we do not intend to enable it. This so far should not
have any noticable effect since the quantum exception was not
observable through the Posix functions.
Some pattern ("fix_trunc<mode>di2") is already using rounding mode 9
correctly.
gcc/ChangeLog:
2016-03-11 Andreas Krebbel <krebbel@linux.vnet.ibm.com>
* config/s390/s390.md (BFP_RND_*, DFP_RND_*): Add new constant
definitions for BFP and DFP rounding modes.
("fixuns_truncdddi2", "fixuns_trunctddi2")
("fixuns_trunc<BFP:mode><GPR:mode>2", "fixuns_trunc<mode>si2")
("fix_trunc<DSF:mode><GPR:mode>2", "fix_trunc<mode>di2")
("fix_trunctf<mode>2"): Use the new constants instead of magic
numbers.
From-SVN: r234133
Diffstat (limited to 'gcc/tree-ssa-loop.c')
0 files changed, 0 insertions, 0 deletions