| commit | 656b84c2ef525e3b69802c9057c5897e327b0332 | |
| author | Wilco Dijkstra <wdijkstr@arm.com> | |
| Thu, 7 Aug 2014 16:29:55 +0000 (7 16:29 +0000) | ||
| committer | Wilco Dijkstra <wdijkstr@arm.com> | |
| Thu, 7 Aug 2014 16:29:55 +0000 (7 16:29 +0000) | ||
| tree | b6cf594f11d613c778e7c4cbd063c316a42ce5b2 | treesnapshot (tar.gz zip) |
| parent | 538e9e454de75c5528d1a7c8f57ede9ccad39d00 | commitdiff |
This patch adds new function libc_feholdsetround_noex_aarch64_ctx, enabling
further optimization. libc_feholdsetround_aarch64_ctx now only needs to
read the FPCR in the typical case, avoiding a redundant FPSR read.
Performance results show a good improvement (5-10% on sin()) on cores with
expensive FPCR/FPSR instructions.
further optimization. libc_feholdsetround_aarch64_ctx now only needs to
read the FPCR in the typical case, avoiding a redundant FPSR read.
Performance results show a good improvement (5-10% on sin()) on cores with
expensive FPCR/FPSR instructions.
| ChangeLog | diffblobblamehistory | |
| sysdeps/aarch64/fpu/math_private.h | diffblobblamehistory |