about summary refs log tree commit diff
path: root/sysdeps/powerpc/powerpc64/power7/fpu
diff options
context:
space:
mode:
authorAdhemerval Zanella <adhemerval.zanella@linaro.org>2019-03-11 03:29:37 +0000
committerAdhemerval Zanella <adhemerval.zanella@linaro.org>2019-06-12 14:32:39 -0300
commit119269606996258ee0a53ff9e5828d474ee0f698 (patch)
treea8b24ad011f24ce14d08aea07f6eeab4735dcbbf /sysdeps/powerpc/powerpc64/power7/fpu
parenta72186761b2ed913fd5ea3df2de2aa446456aa98 (diff)
downloadglibc-119269606996258ee0a53ff9e5828d474ee0f698.tar.gz
glibc-119269606996258ee0a53ff9e5828d474ee0f698.tar.xz
glibc-119269606996258ee0a53ff9e5828d474ee0f698.zip
powerpc: Remove optimized finite
The powerpc finite optimization do not show much gain:

  - GCC will call libm iff -fsignaling-nans is used. This usage pattern
    is usually not performance oriented and for such calls PLT overhead
    should dominate execution time.

  - The power7 uses ftdiv to optimize for some input patterns, but at
    cost of others.  Comparing against generic C implementation built
    for powerpc64-linux-gnu-power7 (--with-cpu=power7):

    - Generic sysdeps/ieee754 implementation:
       "isfinite": {
        "": {
         "duration": 5.0082e+09,
         "iterations": 2.45299e+09,
         "max": 43.824,
         "min": 2.008,
         "mean": 2.04167
        },
        "INF": {
         "duration": 4.66554e+09,
         "iterations": 2.28288e+09,
         "max": 35.73,
         "min": 2.008,
         "mean": 2.04371
        },
        "NAN": {
         "duration": 4.66274e+09,
         "iterations": 2.28716e+09,
         "max": 34.161,
         "min": 2.009,
         "mean": 2.03866
        }
       }

    - power7 optimized one:
       "isfinite": {
        "": {
         "duration": 4.99111e+09,
         "iterations": 2.65566e+09,
         "max": 25.015,
         "min": 1.716,
         "mean": 1.87942
        },
        "INF": {
         "duration": 4.6783e+09,
         "iterations": 2.0999e+09,
         "max": 35.264,
         "min": 1.868,
         "mean": 2.22787
        },
        "NAN": {
         "duration": 4.67915e+09,
         "iterations": 2.08678e+09,
         "max": 38.099,
         "min": 1.869,
         "mean": 2.24228
        }
       }

     So it basically optimizes marginally for normal numbers while
     increasing the latency for other kind of FP.

  - The power8 implementation is just the generic implementation using
    ISA 2.07 mfvsrd instruction (which GCC uses for generic implementation).
    So generic implementation is the best option for powerpc64le.

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/powerpc32/power4/fpu/multiarch/Makefile
	(sysdeps_routines, libm-sysdep_routines): Remove s_finite*
	objects.
	* sysdeps/powerpc/powerpc32/power4/fpu/multiarch/s_finite-power7.S:
	Remove file.
	* sysdeps/powerpc/powerpc32/power4/fpu/multiarch/s_finite-ppc32.c:
	Likewise.
	* sysdeps/powerpc/powerpc32/power4/fpu/multiarch/s_finite.c: Likewise.
	* sysdeps/powerpc/powerpc32/power4/fpu/multiarch/s_finitef-ppc32.c:
	Likewise.
	* sysdeps/powerpc/powerpc32/power4/fpu/multiarch/s_finitef.c: Likewise.
	* sysdeps/powerpc/powerpc32/power7/fpu/s_finite.S: Likewise.
	* sysdeps/powerpc/powerpc32/power7/fpu/s_finitef.S: Likewise.
	* sysdeps/powerpc/powerpc64/fpu/multiarch/Makefile (sysdep_call):
	Remove s_finite* objects.
	* sysdeps/powerpc/powerpc64/fpu/multiarch/s_finite-power7.S: Remove file.
	* sysdeps/powerpc/powerpc64/fpu/multiarch/s_finite-power8.S: Likewise.
	* sysdeps/powerpc/powerpc64/fpu/multiarch/s_finite-ppc64.c: Likewise.
	* sysdeps/powerpc/powerpc64/fpu/multiarch/s_finite.c: Likewise.
	* sysdeps/powerpc/powerpc64/fpu/multiarch/s_finitef-ppc64.c: Likewise.
	* sysdeps/powerpc/powerpc64/fpu/multiarch/s_finitef.c: Likewise.
	* sysdeps/powerpc/powerpc64/power7/fpu/s_finite.S: Likewise.
	* sysdeps/powerpc/powerpc64/power7/fpu/s_finitef.S: Likewise.
	* sysdeps/powerpc/powerpc64/power8/fpu/s_finite.S: Likewise.
	* sysdeps/powerpc/powerpc64/power8/fpu/s_finitef.S: Likewise.

Reviewed-by: Gabriel F. T. Gomes <gabrielftg@linux.ibm.com>
Diffstat (limited to 'sysdeps/powerpc/powerpc64/power7/fpu')
-rw-r--r--sysdeps/powerpc/powerpc64/power7/fpu/s_finite.S70
-rw-r--r--sysdeps/powerpc/powerpc64/power7/fpu/s_finitef.S1
2 files changed, 0 insertions, 71 deletions
diff --git a/sysdeps/powerpc/powerpc64/power7/fpu/s_finite.S b/sysdeps/powerpc/powerpc64/power7/fpu/s_finite.S
deleted file mode 100644
index 5b5746c9a5..0000000000
--- a/sysdeps/powerpc/powerpc64/power7/fpu/s_finite.S
+++ /dev/null
@@ -1,70 +0,0 @@
-/* finite().  PowerPC64/POWER7 version.
-   Copyright (C) 2010-2019 Free Software Foundation, Inc.
-   Contributed by Luis Machado <luisgpm@br.ibm.com>.
-   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 <math_ldbl_opt.h>
-
-/* int __finite(x)  */
-	.section    ".toc","aw"
-.LC0:   /* 1.0 */
-	.tc	    FD_ONE[TC],0x3ff0000000000000
-	.section    ".text"
-	.type	    __finite, @function
-	.machine    power7
-ENTRY (__finite, 4)
-	CALL_MCOUNT 0
-	lfd     fp0,.LC0@toc(r2)
-	ftdiv   cr7,fp1,fp0
-	li	r3,1
-	bflr    30
-
-	/* If we are here, we either have +/-INF,
-	NaN or denormal.  */
-
-	stfd    fp1,-16(r1)   /* Transfer FP to GPR's.  */
-	ori	2,2,0	      /* Force a new dispatch group.  */
-	lhz     r4,-16+HISHORT(r1)  /* Fetch the upper 16 bits of the FP value
-				    (biased exponent and sign bit).  */
-	clrlwi  r4,r4,17      /* r4 = abs(r4).  */
-	cmpwi   cr7,r4,0x7ff0 /* r4 == 0x7ff0?  */
-	bltlr   cr7	      /* LT means finite, other non-finite.  */
-	li      r3,0
-	blr
-	END (__finite)
-
-hidden_def (__finite)
-weak_alias (__finite, finite)
-
-/* It turns out that the 'double' version will also always work for
-   single-precision.  */
-strong_alias (__finite, __finitef)
-hidden_def (__finitef)
-weak_alias (__finitef, finitef)
-
-#if IS_IN (libm)
-# if LONG_DOUBLE_COMPAT (libm, GLIBC_2_0)
-compat_symbol (libm, __finite, __finitel, GLIBC_2_0)
-compat_symbol (libm, finite, finitel, GLIBC_2_0)
-# endif
-#else
-# if LONG_DOUBLE_COMPAT (libc, GLIBC_2_0)
-compat_symbol (libc, __finite, __finitel, GLIBC_2_0);
-compat_symbol (libc, finite, finitel, GLIBC_2_0);
-# endif
-#endif
diff --git a/sysdeps/powerpc/powerpc64/power7/fpu/s_finitef.S b/sysdeps/powerpc/powerpc64/power7/fpu/s_finitef.S
deleted file mode 100644
index 54bd94176d..0000000000
--- a/sysdeps/powerpc/powerpc64/power7/fpu/s_finitef.S
+++ /dev/null
@@ -1 +0,0 @@
-/* This function uses the same code as s_finite.S.  */