diff options
| author | Evandro Menezes <e.menezes@samsung.com> | 2016-02-26 23:41:53 +0000 |
|---|---|---|
| committer | Evandro Menezes <evandro@gcc.gnu.org> | 2016-02-26 23:41:53 +0000 |
| commit | 0c30e0f3160716430de36f359cd120aa875dcb82 (patch) | |
| tree | fd190d707411a2a565e5dc493446708ec6eccf62 /gcc/doc | |
| parent | 20ba5f3318378eb29b15674e2ef37c6c26a1d465 (diff) | |
| download | gcc-0c30e0f3160716430de36f359cd120aa875dcb82.zip gcc-0c30e0f3160716430de36f359cd120aa875dcb82.tar.gz gcc-0c30e0f3160716430de36f359cd120aa875dcb82.tar.bz2 | |
Rename the tuning option and related functions to enable the Newton series for the reciprocal square root to reflect its approximative characteristic.
gcc/
* config/aarch64/aarch64-protos.h (aarch64_emit_swrsqrt): Rename
function to "aarch64_emit_approx_rsqrt".
* config/aarch64/aarch64-tuning-flags.def: Rename tuning flag to
AARCH64_EXTRA_TUNE_APPROX_RSQRT.
* config/aarch64/aarch64.c (exynosm1_tunigs): Use new flag name.
(xgene1_tunings): Likewise.
(use_rsqrt_p): Likewise.
(aarch64_emit_swrsqrt): Use new function name.
* config/aarch64/aarch64-simd.md (aarch64_rsqrts_*): Likewise.
* config/aarch64/aarch64.opt (mlow-precision-recip-sqrt): Reword the
text explaining this option.
* doc/invoke.texi (-mlow-precision-recip-sqrt): Likewise.
From-SVN: r233772
Diffstat (limited to 'gcc/doc')
| -rw-r--r-- | gcc/doc/invoke.texi | 10 |
1 files changed, 4 insertions, 6 deletions
diff --git a/gcc/doc/invoke.texi b/gcc/doc/invoke.texi index 18b2b8f..4b5df0b 100644 --- a/gcc/doc/invoke.texi +++ b/gcc/doc/invoke.texi @@ -12884,12 +12884,10 @@ corresponding flag to the linker. @item -mno-low-precision-recip-sqrt @opindex -mlow-precision-recip-sqrt @opindex -mno-low-precision-recip-sqrt -The square root estimate uses two steps instead of three for double-precision, -and one step instead of two for single-precision. -Thus reducing latency and precision. -This is only relevant if @option{-ffast-math} activates -reciprocal square root estimate instructions. -Which in turn depends on the target processor. +When calculating the reciprocal square root approximation, +uses one less step than otherwise, thus reducing latency and precision. +This is only relevant if @option{-ffast-math} enables the reciprocal square root +approximation, which in turn depends on the target processor. @item -march=@var{name} @opindex march |