Wilco Dijkstra 656b84c2ef 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.

2014-08-07 16:29:55 +00:00

..

fclrexcpt.c

[AArch64] Suppress unnecessary FPSR and FPCR writes.

2014-04-24 07:15:33 +01:00

fedisblxcpt.c

[AArch64] Suppress unnecessary FPSR and FPCR writes.

2014-04-24 07:15:33 +01:00

feenablxcpt.c

[AArch64] Suppress unnecessary FPSR and FPCR writes.

2014-04-24 07:15:33 +01:00

fegetenv.c

…

fegetexcept.c

…

fegetround.c

…

feholdexcpt.c

[AArch64] Suppress unnecessary FPSR and FPCR writes.

2014-04-24 07:15:33 +01:00

fesetenv.c

[AArch64] Suppress unnecessary FPSR and FPCR writes.

2014-04-24 07:15:33 +01:00

fesetround.c

[AArch64] Suppress unnecessary FPSR and FPCR writes.

2014-04-24 07:15:33 +01:00

feupdateenv.c

[AArch64] Rewrite feupdateenv (BZ 17009).

2014-06-02 12:36:34 +01:00

fgetexcptflg.c

…

fpu_control.h

[AArch64] Remove ISB after FPCR write.

2014-06-02 12:44:21 +01:00

fraiseexcpt.c

…

fsetexcptflg.c

[AArch64] Suppress unnecessary FPSR and FPCR writes.

2014-04-24 07:15:33 +01:00

ftestexcept.c

…

get-rounding-mode.h

[AArch64] Switch from FE_TOWARDZERO to _FPU_FPCR_RM_MASK

2014-06-03 12:44:50 +00:00

math_private.h

This patch adds new function libc_feholdsetround_noex_aarch64_ctx, enabling

2014-08-07 16:29:55 +00:00

s_ceil.c

…

s_ceilf.c

…

s_floor.c

…

s_floorf.c

…

s_fma.c

…

s_fmaf.c

…

s_fmax.c

…

s_fmaxf.c

…

s_fmin.c

…

s_fminf.c

…

s_frint.c

…

s_frintf.c

…

s_llrint.c

…

s_llrintf.c

…

s_llround.c

…

s_llroundf.c

…

s_lrint.c

…

s_lrintf.c

…

s_lround.c

…

s_lroundf.c

…

s_nearbyint.c

…

s_nearbyintf.c

…

s_rint.c

…

s_rintf.c

…

s_round.c

…

s_roundf.c

…

s_trunc.c

…

s_truncf.c

…