[Gmp-commit] /var/hg/gmp: 2 new changesets
mercurial at gmplib.org
mercurial at gmplib.org
Sun Apr 3 14:50:49 UTC 2016
details: /var/hg/gmp/rev/855294110147
changeset: 17065:855294110147
user: Torbjorn Granlund <torbjorng at google.com>
date: Sun Apr 03 16:50:07 2016 +0200
description:
Undo "Remove workaround for slow addmul_2".
details: /var/hg/gmp/rev/8ccd8e7de27b
changeset: 17066:8ccd8e7de27b
user: Torbjorn Granlund <torbjorng at google.com>
date: Sun Apr 03 16:50:36 2016 +0200
description:
Add cycle numbers.
diffstat:
mpn/x86_64/coreibwl/gmp-mparam.h | 1 +
mpn/x86_64/coreihwl/addmul_2.asm | 9 ++++++---
mpn/x86_64/coreihwl/mul_2.asm | 9 ++++++---
mpn/x86_64/skylake/gmp-mparam.h | 1 +
4 files changed, 14 insertions(+), 6 deletions(-)
diffs (76 lines):
diff -r acbe3fa79aea -r 8ccd8e7de27b mpn/x86_64/coreibwl/gmp-mparam.h
--- a/mpn/x86_64/coreibwl/gmp-mparam.h Sat Apr 02 12:42:23 2016 +0200
+++ b/mpn/x86_64/coreibwl/gmp-mparam.h Sun Apr 03 16:50:36 2016 +0200
@@ -33,6 +33,7 @@
/* Disable use of slow functions. FIXME: We should disable lib inclusion. */
#undef HAVE_NATIVE_mpn_mul_2
+#undef HAVE_NATIVE_mpn_addmul_2
/* 2100 MHz Intel i3-5100 */
/* FFT tuning limit = 200 M */
diff -r acbe3fa79aea -r 8ccd8e7de27b mpn/x86_64/coreihwl/addmul_2.asm
--- a/mpn/x86_64/coreihwl/addmul_2.asm Sat Apr 02 12:42:23 2016 +0200
+++ b/mpn/x86_64/coreihwl/addmul_2.asm Sun Apr 03 16:50:36 2016 +0200
@@ -37,17 +37,20 @@
C AMD K10 n/a
C AMD bull n/a
C AMD pile n/a
-C AMD steam ?
+C AMD steam n/a
+C AMD excavator ?
C AMD bobcat n/a
-C AMD jaguar ?
+C AMD jaguar n/a
C Intel P4 n/a
C Intel core n/a
C Intel NHM n/a
C Intel SBR n/a
C Intel IBR n/a
C Intel HWL 2.15
-C Intel BWL ?
+C Intel BWL 2.33
+C Intel SKL 2.22
C Intel atom n/a
+C Intel SLM n/a
C VIA nano n/a
C The loop of this code is the result of running a code generation and
diff -r acbe3fa79aea -r 8ccd8e7de27b mpn/x86_64/coreihwl/mul_2.asm
--- a/mpn/x86_64/coreihwl/mul_2.asm Sat Apr 02 12:42:23 2016 +0200
+++ b/mpn/x86_64/coreihwl/mul_2.asm Sun Apr 03 16:50:36 2016 +0200
@@ -37,17 +37,20 @@
C AMD K10 n/a
C AMD bull n/a
C AMD pile n/a
-C AMD steam ?
+C AMD steam n/a
+C AMD excavator ?
C AMD bobcat n/a
-C AMD jaguar ?
+C AMD jaguar n/a
C Intel P4 n/a
C Intel core n/a
C Intel NHM n/a
C Intel SBR n/a
C Intel IBR n/a
C Intel HWL 1.86
-C Intel BWL ?
+C Intel BWL 2.1
+C Intel SKL 2.08
C Intel atom n/a
+C Intel SLM n/a
C VIA nano n/a
C The loop of this code is the result of running a code generation and
diff -r acbe3fa79aea -r 8ccd8e7de27b mpn/x86_64/skylake/gmp-mparam.h
--- a/mpn/x86_64/skylake/gmp-mparam.h Sat Apr 02 12:42:23 2016 +0200
+++ b/mpn/x86_64/skylake/gmp-mparam.h Sun Apr 03 16:50:36 2016 +0200
@@ -33,6 +33,7 @@
/* Disable use of slow functions. FIXME: We should disable lib inclusion. */
#undef HAVE_NATIVE_mpn_mul_2
+#undef HAVE_NATIVE_mpn_addmul_2
/* 3500 MHz i5-6600K Skylake */
/* FFT tuning limit = 300 M */
More information about the gmp-commit
mailing list