diff options
author | Adhemerval Zanella <adhemerval.zanella@linaro.org> | 2019-03-18 11:28:13 +0000 |
---|---|---|
committer | Adhemerval Zanella <adhemerval.zanella@linaro.org> | 2019-06-17 09:28:21 -0300 |
commit | dee07df1a40533f1f2856111946eafbd45b81dbe (patch) | |
tree | 9febce88c8ce942f2ea7f259856caa2e58694a27 /sysdeps/powerpc/powerpc64/fpu/s_llroundf.S | |
parent | 2166283fcc8ba3b4a4505acd2a91f2e7decf689c (diff) | |
download | glibc-dee07df1a40533f1f2856111946eafbd45b81dbe.tar.gz |
powerpc: Refactor powerpc64 lround/lroundf/llround/llroundf
This patches consolidates all the powerpc {l}lround{f} implementations
on the generic sysdeps/powerpc/fpu/s_{l}lround{f}.c.
The IFUNC support is also moved only to powerpc64 only, since for
powerpc64le generic implementation resulting in optimized code.
Checked on powerpc-linux-gnu (built without --with-cpu, with
--with-cpu=power4 and with --with-cpu=power5+ and --disable-multi-arch),
powerpc64-linux-gnu (built without --with-cp and with --with-cpu=power5+
and --disable-multi-arch).
* sysdeps/powerpc/powerpc64/be/fpu/multiarch/Makefile
(libm-sysdep_routines): Add s_llround-power8, s_llround-power6x,
s_llround-power5+, s_llround-ppc64, and s_llroundf-ppc64.
(CFLAGS-s_llround-power8.c, CFLAGS-s_llround-power6x.c,
CFLAGS-s_llround-power5+.c): New rule.
* sysdeps/powerpc/powerpc64/be/fpu/multiarch/s_llround-power5+.c:
New file.
* sysdeps/powerpc/powerpc64/be/fpu/multiarch/s_llround-power6x.c:
Likewise.
* sysdeps/powerpc/powerpc64/be/fpu/multiarch/s_llround-power8.c:
Likewise.
* sysdeps/powerpc/powerpc64/be/fpu/multiarch/s_llround-ppc64.c:
Likewise.
* sysdeps/powerpc/powerpc64/be/fpu/multiarch/s_llroundf-ppc64.c:
Likewise.
* sysdeps/powerpc/powerpc64/fpu/multiarch/s_llround.c: Move to ...
* sysdeps/powerpc/powerpc64/be/fpu/multiarch/s_llround.c: ... here.
* sysdeps/powerpc/powerpc64/fpu/multiarch/s_llroundf.c: Move to ...
* sysdeps/powerpc/powerpc64/be/fpu/multiarch/s_llroundf.c: ... here.
* sysdeps/powerpc/powerpc64/fpu/multiarch/s_lround.c: Move to ...
* sysdeps/powerpc/powerpc64/be/fpu/multiarch/s_lround.c: ... here.
* sysdeps/powerpc/powerpc64/fpu/Makefile
[$(subdir) == math] (CFLAGS-s_llround.c): New rule.
* sysdeps/powerpc/powerpc64/fpu/multiarch/Makefile
(libm-sysdep_routines): Remove s_llround-* objects.
* sysdeps/powerpc/powerpc64/fpu/multiarch/s_llround-power5+.S: Remove
file.
* sysdeps/powerpc/powerpc64/fpu/multiarch/s_llround-power6x.S:
Likewise.
* sysdeps/powerpc/powerpc64/fpu/multiarch/s_llround-power8.S:
Likewise.
* sysdeps/powerpc/powerpc64/fpu/multiarch/s_llround-ppc64.S:
Likewise.
* sysdeps/powerpc/powerpc64/fpu/multiarch/s_llroundf-ppc64.S:
Likewise.
* sysdeps/powerpc/powerpc64/fpu/s_llround.S: Likewise.
* sysdeps/powerpc/powerpc64/fpu/s_llroundf.S: Likewise.
* sysdeps/powerpc/powerpc64/fpu/s_lround.S: Likewise.
* sysdeps/powerpc/powerpc64/fpu/s_lroundf.S: Likewise.
* sysdeps/powerpc/powerpc64/fpu/s_llround.c: New file.
* sysdeps/powerpc/powerpc64/fpu/s_llroundf.c: Likewise.
* sysdeps/powerpc/powerpc64/fpu/s_lround.c: Likewise.
* sysdeps/powerpc/powerpc64/fpu/s_lroundf.c: Likewise.
* sysdeps/powerpc/powerpc64/power5+/fpu/s_llround.S: Likewise.
* sysdeps/powerpc/powerpc64/power5+/fpu/s_llroundf.S: Likewise.
* sysdeps/powerpc/powerpc64/power6x/fpu/s_llround.S: Likewise.
* sysdeps/powerpc/powerpc64/power6x/fpu/s_llroundf.S: Likewise.
* sysdeps/powerpc/powerpc64/power8/fpu/s_llround.S: Likewise.
* sysdeps/powerpc/powerpc64/power8/fpu/s_llroundf.S: Likewise.
Reviewed-by: Gabriel F. T. Gomes <gabrielftg@linux.ibm.com>
Diffstat (limited to 'sysdeps/powerpc/powerpc64/fpu/s_llroundf.S')
-rw-r--r-- | sysdeps/powerpc/powerpc64/fpu/s_llroundf.S | 89 |
1 files changed, 0 insertions, 89 deletions
diff --git a/sysdeps/powerpc/powerpc64/fpu/s_llroundf.S b/sysdeps/powerpc/powerpc64/fpu/s_llroundf.S deleted file mode 100644 index b8afebc7dc..0000000000 --- a/sysdeps/powerpc/powerpc64/fpu/s_llroundf.S +++ /dev/null @@ -1,89 +0,0 @@ -/* llroundf function. PowerPC64 version. - Copyright (C) 2004-2019 Free Software Foundation, Inc. - This file is part of the GNU C Library. - - The GNU C Library is free software; you can redistribute it and/or - modify it under the terms of the GNU Lesser General Public - License as published by the Free Software Foundation; either - version 2.1 of the License, or (at your option) any later version. - - The GNU C Library is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - Lesser General Public License for more details. - - You should have received a copy of the GNU Lesser General Public - License along with the GNU C Library; if not, see - <http://www.gnu.org/licenses/>. */ - -#include <sysdep.h> -#include <libm-alias-float.h> - - .section ".toc","aw" -.LC0: /* 2^23 */ - .tc FD_41600000_0[TC],0x4160000000000000 -.LC1: /* 0.5 */ - .tc FD_3fe00000_0[TC],0x3fe0000000000000 -.LC2: /* 2^52 */ - .tc FD_43300000_0[TC],0x4330000000000000 - .section ".text" - -/* long long [r3] llroundf (float x [fp1]) - IEEE 1003.1 llroundf function. IEEE specifies "roundf to the nearest - integer value, rounding halfway cases away from zero, regardless of - the current rounding mode." However PowerPC Architecture defines - "roundf to Nearest" as "Choose the best approximation. In case of a - tie, choose the one that is even (least significant bit o).". - So we can't use the PowerPC "round to Nearest" mode. Instead we set - "round toward Zero" mode and round by adding +-0.5 before rounding - to the integer value. - - It is necessary to detect when x is (+-)0x1.fffffffffffffp-2 - because adding +-0.5 in this case will cause an erroneous shift, - carry and round. We simply return 0 if 0.5 > x > -0.5. Likewise - if x is and odd number between +-(2^23 and 2^24-1) a shift and - carry will erroneously round if biased with +-0.5. Therefore if x - is greater/less than +-2^23 we don't need to bias the number with - +-0.5. */ - -ENTRY (__llroundf) - CALL_MCOUNT 0 - lfd fp9,.LC0@toc(2) /* Load 2^23 into fpr9. */ - lfd fp10,.LC1@toc(2)/* Load 0.5 into fpr10. */ - lfd fp11,.LC2@toc(2) /* Load 2^52 into fpr11. */ - fabs fp2,fp1 /* Get the absolute value of x. */ - fsub fp12,fp10,fp10 /* Compute 0.0 into fp12. */ - fcmpu cr6,fp2,fp10 /* if |x| < 0.5 */ - fcmpu cr7,fp2,fp9 /* if |x| >= 2^23 */ - fcmpu cr1,fp1,fp12 /* x is negative? x < 0.0 */ - blt- cr6,.Lretzero /* 0.5 > x < -0.5 so just return 0. */ - bge- cr7,.Lnobias /* 2^23 > x < -2^23 just convert with no bias. */ - /* Test whether an integer to avoid spurious "inexact". */ - fadd fp3,fp2,fp11 - fsub fp3,fp3,fp11 - fcmpu cr5,fp2,fp3 - beq cr5,.Lnobias - fadd fp3,fp2,fp10 /* |x|+=0.5 bias to prepare to round. */ - bge cr1,.Lconvert /* x is positive so don't negate x. */ - fnabs fp3,fp3 /* -(|x|+=0.5) */ -.Lconvert: - fctidz fp4,fp3 /* Convert to Integer double word round toward 0. */ - stfd fp4,-16(r1) - nop - nop - nop - ld r3,-16(r1) /* Load return as integer. */ -.Lout: - blr -.Lretzero: /* 0.5 > x > -0.5 */ - li r3,0 /* return 0. */ - b .Lout -.Lnobias: - fmr fp3,fp1 - b .Lconvert - END (__llroundf) - -strong_alias (__llroundf, __lroundf) -libm_alias_float (__llround, llround) -libm_alias_float (__lround, lround) - |