[med-svn] [Git][med-team/raxml][master] 3 commits: Update patches
Andreas Tille (@tille)
gitlab at salsa.debian.org
Tue Nov 28 10:47:56 GMT 2023
Andreas Tille pushed to branch master at Debian Med / raxml
Commits:
7c97e8df by Andreas Tille at 2023-11-28T11:41:53+01:00
Update patches
- - - - -
1b482dc8 by Andreas Tille at 2023-11-28T11:43:02+01:00
README -> README.md
- - - - -
c5f574ac by Andreas Tille at 2023-11-28T11:46:23+01:00
Upload to unstable
- - - - -
4 changed files:
- debian/changelog
- debian/docs
- debian/patches/hardening
- debian/patches/simde
Changes:
=====================================
debian/changelog
=====================================
@@ -1,11 +1,12 @@
-raxml (8.2.13+dfsg-1) UNRELEASED; urgency=medium
+raxml (8.2.13+dfsg-1) unstable; urgency=medium
* New upstream version
* Revert: Add missing build dependency on dh addon.
* Standards-Version: 4.6.2 (routine-update)
* d/watch: Fix download name
+ * README -> README.md
- -- Andreas Tille <tille at debian.org> Wed, 08 Nov 2023 21:31:57 +0100
+ -- Andreas Tille <tille at debian.org> Tue, 28 Nov 2023 11:44:47 +0100
raxml (8.2.12+dfsg-8) unstable; urgency=medium
=====================================
debian/docs
=====================================
@@ -1,5 +1,5 @@
manual/NewManual.pdf
-README
+README*
debian/README.test
debian/tests/run-unit-test
debian/tests/test_data
=====================================
debian/patches/hardening
=====================================
@@ -1,8 +1,8 @@
From: Michael R. Crusoe <crusoe at debian.org>
Subject: improve the build hardening
Forwarded: https://github.com/stamatak/standard-RAxML/pull/50
---- raxml.orig/Makefile.AVX.PTHREADS.gcc
-+++ raxml/Makefile.AVX.PTHREADS.gcc
+--- a/Makefile.AVX.PTHREADS.gcc
++++ b/Makefile.AVX.PTHREADS.gcc
@@ -3,8 +3,11 @@
CC = gcc
@@ -16,7 +16,7 @@ Forwarded: https://github.com/stamatak/standard-RAxML/pull/50
LIBRARIES = -lm -pthread
-@@ -20,10 +23,10 @@
+@@ -20,10 +23,10 @@ raxmlHPC-PTHREADS-AVX : $(objs)
$(CC) -o raxmlHPC-PTHREADS-AVX $(objs) $(LIBRARIES) $(LDFLAGS)
avxLikelihood.o : avxLikelihood.c $(GLOBAL_DEPS)
@@ -29,7 +29,7 @@ Forwarded: https://github.com/stamatak/standard-RAxML/pull/50
#rev_functions.o : rev_functions.c $(GLOBAL_DEPS)
rmqs.o : rmqs.c $(GLOBAL_DEPS)
-@@ -49,7 +52,7 @@
+@@ -49,7 +52,7 @@ ancestralStates.o : ancestralStates.c $(
mem_alloc.o eigen.o : mem_alloc.c $(GLOBAL_DEPS)
eigen.o : eigen.c $(GLOBAL_DEPS)
@@ -38,18 +38,18 @@ Forwarded: https://github.com/stamatak/standard-RAxML/pull/50
clean :
$(RM) *.o raxmlHPC-PTHREADS-AVX
---- raxml.orig/Makefile.PTHREADS.gcc
-+++ raxml/Makefile.PTHREADS.gcc
-@@ -4,7 +4,7 @@
- CC = gcc
-
+--- a/Makefile.PTHREADS.gcc
++++ b/Makefile.PTHREADS.gcc
+@@ -10,7 +10,7 @@ else ifeq ($(ARCH), aarch64)
+ ARCH_CFLAGS=
+ endif
--CFLAGS = -D_GNU_SOURCE -fomit-frame-pointer -funroll-loops -O2 -D_USE_PTHREADS -msse #-Wall -Wunused-parameter -Wredundant-decls -Wreturn-type -Wswitch-default -Wunused-value -Wimplicit -Wimplicit-function-declaration -Wimplicit-int -Wimport -Wunused -Wunused-function -Wunused-label -Wno-int-to-pointer-cast -Wbad-function-cast -Wmissing-declarations -Wmissing-prototypes -Wnested-externs -Wold-style-definition -Wstrict-prototypes -Wpointer-sign -Wextra -Wredundant-decls -Wunused -Wunused-function -Wunused-parameter -Wunused-value -Wunused-variable -Wformat -Wformat-nonliteral -Wparentheses -Wsequence-point -Wuninitialized -Wundef -Wbad-function-cast
-+CFLAGS += -D_GNU_SOURCE -fomit-frame-pointer -funroll-loops -O2 -D_USE_PTHREADS -msse #-Wall -Wunused-parameter -Wredundant-decls -Wreturn-type -Wswitch-default -Wunused-value -Wimplicit -Wimplicit-function-declaration -Wimplicit-int -Wimport -Wunused -Wunused-function -Wunused-label -Wno-int-to-pointer-cast -Wbad-function-cast -Wmissing-declarations -Wmissing-prototypes -Wnested-externs -Wold-style-definition -Wstrict-prototypes -Wpointer-sign -Wextra -Wredundant-decls -Wunused -Wunused-function -Wunused-parameter -Wunused-value -Wunused-variable -Wformat -Wformat-nonliteral -Wparentheses -Wsequence-point -Wuninitialized -Wundef -Wbad-function-cast
+-CFLAGS = -D_GNU_SOURCE -fomit-frame-pointer -funroll-loops -O2 -D_USE_PTHREADS $(ARCH_CFLAGS) #-Wall -Wunused-parameter -Wredundant-decls -Wreturn-type -Wswitch-default -Wunused-value -Wimplicit -Wimplicit-function-declaration -Wimplicit-int -Wimport -Wunused -Wunused-function -Wunused-label -Wno-int-to-pointer-cast -Wbad-function-cast -Wmissing-declarations -Wmissing-prototypes -Wnested-externs -Wold-style-definition -Wstrict-prototypes -Wpointer-sign -Wextra -Wredundant-decls -Wunused -Wunused-function -Wunused-parameter -Wunused-value -Wunused-variable -Wformat -Wformat-nonliteral -Wparentheses -Wsequence-point -Wuninitialized -Wundef -Wbad-function-cast
++CFLAGS += -D_GNU_SOURCE -fomit-frame-pointer -funroll-loops -O2 -D_USE_PTHREADS $(ARCH_CFLAGS) #-Wall -Wunused-parameter -Wredundant-decls -Wreturn-type -Wswitch-default -Wunused-value -Wimplicit -Wimplicit-function-declaration -Wimplicit-int -Wimport -Wunused -Wunused-function -Wunused-label -Wno-int-to-pointer-cast -Wbad-function-cast -Wmissing-declarations -Wmissing-prototypes -Wnested-externs -Wold-style-definition -Wstrict-prototypes -Wpointer-sign -Wextra -Wredundant-decls -Wunused -Wunused-function -Wunused-parameter -Wunused-value -Wunused-variable -Wformat -Wformat-nonliteral -Wparentheses -Wsequence-point -Wuninitialized -Wundef -Wbad-function-cast
LIBRARIES = -lm -pthread
-@@ -45,8 +45,8 @@
+@@ -51,8 +51,8 @@ ancestralStates.o : ancestralStates.c $(
mem_alloc.o eigen.o : mem_alloc.c $(GLOBAL_DEPS)
eigen.o : eigen.c $(GLOBAL_DEPS)
@@ -61,18 +61,18 @@ Forwarded: https://github.com/stamatak/standard-RAxML/pull/50
-dev : raxmlHPC-PTHREADS
\ No newline at end of file
+dev : raxmlHPC-PTHREADS
---- raxml.orig/Makefile.SSE3.PTHREADS.gcc
-+++ raxml/Makefile.SSE3.PTHREADS.gcc
-@@ -3,7 +3,7 @@
-
- CC = gcc
+--- a/Makefile.SSE3.PTHREADS.gcc
++++ b/Makefile.SSE3.PTHREADS.gcc
+@@ -10,7 +10,7 @@ else ifeq ($(ARCH), aarch64)
+ ARCH_CFLAGS=
+ endif
--CFLAGS = -D_USE_PTHREADS -D__SIM_SSE3 -D_GNU_SOURCE -msse3 -O2 -fomit-frame-pointer -funroll-loops #-Wall -pedantic -Wunused-parameter -Wredundant-decls -Wreturn-type -Wswitch-default -Wunused-value -Wimplicit -Wimplicit-function-declaration -Wimplicit-int -Wimport -Wunused -Wunused-function -Wunused-label -Wno-int-to-pointer-cast -Wbad-function-cast -Wmissing-declarations -Wmissing-prototypes -Wnested-externs -Wold-style-definition -Wstrict-prototypes -Wdeclaration-after-statement -Wpointer-sign -Wextra -Wredundant-decls -Wunused -Wunused-function -Wunused-parameter -Wunused-value -Wunused-variable -Wformat -Wformat-nonliteral -Wparentheses -Wsequence-point -Wuninitialized -Wundef -Wbad-function-cast
-+CFLAGS += -D_USE_PTHREADS -D__SIM_SSE3 -D_GNU_SOURCE -msse3 -O2 -fomit-frame-pointer -funroll-loops #-Wall -pedantic -Wunused-parameter -Wredundant-decls -Wreturn-type -Wswitch-default -Wunused-value -Wimplicit -Wimplicit-function-declaration -Wimplicit-int -Wimport -Wunused -Wunused-function -Wunused-label -Wno-int-to-pointer-cast -Wbad-function-cast -Wmissing-declarations -Wmissing-prototypes -Wnested-externs -Wold-style-definition -Wstrict-prototypes -Wdeclaration-after-statement -Wpointer-sign -Wextra -Wredundant-decls -Wunused -Wunused-function -Wunused-parameter -Wunused-value -Wunused-variable -Wformat -Wformat-nonliteral -Wparentheses -Wsequence-point -Wuninitialized -Wundef -Wbad-function-cast
+-CFLAGS = -D_USE_PTHREADS -D__SIM_SSE3 -D_GNU_SOURCE $(ARCH_CFLAGS) -O2 -fomit-frame-pointer -funroll-loops #-Wall -pedantic -Wunused-parameter -Wredundant-decls -Wreturn-type -Wswitch-default -Wunused-value -Wimplicit -Wimplicit-function-declaration -Wimplicit-int -Wimport -Wunused -Wunused-function -Wunused-label -Wno-int-to-pointer-cast -Wbad-function-cast -Wmissing-declarations -Wmissing-prototypes -Wnested-externs -Wold-style-definition -Wstrict-prototypes -Wdeclaration-after-statement -Wpointer-sign -Wextra -Wredundant-decls -Wunused -Wunused-function -Wunused-parameter -Wunused-value -Wunused-variable -Wformat -Wformat-nonliteral -Wparentheses -Wsequence-point -Wuninitialized -Wundef -Wbad-function-cast
++CFLAGS += -D_USE_PTHREADS -D__SIM_SSE3 -D_GNU_SOURCE $(ARCH_CFLAGS) -O2 -fomit-frame-pointer -funroll-loops #-Wall -pedantic -Wunused-parameter -Wredundant-decls -Wreturn-type -Wswitch-default -Wunused-value -Wimplicit -Wimplicit-function-declaration -Wimplicit-int -Wimport -Wunused -Wunused-function -Wunused-label -Wno-int-to-pointer-cast -Wbad-function-cast -Wmissing-declarations -Wmissing-prototypes -Wnested-externs -Wold-style-definition -Wstrict-prototypes -Wdeclaration-after-statement -Wpointer-sign -Wextra -Wredundant-decls -Wunused -Wunused-function -Wunused-parameter -Wunused-value -Wunused-variable -Wformat -Wformat-nonliteral -Wparentheses -Wsequence-point -Wuninitialized -Wundef -Wbad-function-cast
LIBRARIES = -lm -pthread
-@@ -41,11 +41,11 @@
+@@ -48,11 +48,11 @@ fastDNAparsimony.o : fastDNAparsimony.c
fastSearch.o : fastSearch.c $(GLOBAL_DEPS)
leaveDropping.o : leaveDropping.c $(GLOBAL_DEPS)
ancestralStates.o : ancestralStates.c $(GLOBAL_DEPS)
=====================================
debian/patches/simde
=====================================
@@ -1,8 +1,8 @@
From: Michael R. Crusoe <crusoe at debian.org>
Subject: enable building on non-x86 via the SIMDE library
https://github.com/stamatak/standard-RAxML/pull/50
---- raxml.orig/avxLikelihood.c
-+++ raxml/avxLikelihood.c
+--- a/avxLikelihood.c
++++ b/avxLikelihood.c
@@ -10,12 +10,10 @@
#include <limits.h>
#include "axml.h"
@@ -19,7 +19,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
#define FMAMACC(a,b,c) _mm256_fmadd_pd(b,c,a)
#endif
-@@ -154,7 +152,7 @@
+@@ -154,7 +152,7 @@ void newviewGTRGAMMA_AVX(int tipCase,
__m256d
evv = _mm256_load_pd(&extEV[l * 4]);
@@ -28,7 +28,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
xv = FMAMACC(xv,x1v,evv);
#else
xv = _mm256_add_pd(xv, _mm256_mul_pd(x1v, evv));
-@@ -226,7 +224,7 @@
+@@ -226,7 +224,7 @@ void newviewGTRGAMMA_AVX(int tipCase,
__m256d
evv = _mm256_load_pd(&extEV[l * 4]);
@@ -37,7 +37,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
xv[k] = FMAMACC(xv[k],x1v,evv);
#else
xv[k] = _mm256_add_pd(xv[k], _mm256_mul_pd(x1v, evv));
-@@ -443,7 +441,7 @@
+@@ -443,7 +441,7 @@ void newviewGTRGAMMA_AVX_GAPPED_SAVE(in
__m256d
evv = _mm256_load_pd(&extEV[l * 4]);
@@ -46,7 +46,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
xv = FMAMACC(xv,x1v,evv);
#else
xv = _mm256_add_pd(xv, _mm256_mul_pd(x1v, evv));
-@@ -478,7 +476,7 @@
+@@ -478,7 +476,7 @@ void newviewGTRGAMMA_AVX_GAPPED_SAVE(in
__m256d
evv = _mm256_load_pd(&extEV[l * 4]);
@@ -55,7 +55,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
xv = FMAMACC(xv,x1v,evv);
#else
xv = _mm256_add_pd(xv, _mm256_mul_pd(x1v, evv));
-@@ -555,7 +553,7 @@
+@@ -555,7 +553,7 @@ void newviewGTRGAMMA_AVX_GAPPED_SAVE(in
__m256d
evv = _mm256_load_pd(&extEV[l * 4]);
@@ -64,7 +64,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
xv[k] = FMAMACC(xv[k],x1v,evv);
#else
xv[k] = _mm256_add_pd(xv[k], _mm256_mul_pd(x1v, evv));
-@@ -640,7 +638,7 @@
+@@ -640,7 +638,7 @@ void newviewGTRGAMMA_AVX_GAPPED_SAVE(in
__m256d
evv = _mm256_load_pd(&extEV[l * 4]);
@@ -73,7 +73,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
xv[k] = FMAMACC(xv[k],x1v,evv);
#else
xv[k] = _mm256_add_pd(xv[k], _mm256_mul_pd(x1v, evv));
-@@ -902,7 +900,7 @@
+@@ -902,7 +900,7 @@ void newviewGTRCAT_AVX(int tipCase, dou
__m256d
evv = _mm256_load_pd(&EV[l * 4]);
@@ -82,7 +82,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
vv = FMAMACC(vv,x1v,evv);
#else
vv = _mm256_add_pd(vv, _mm256_mul_pd(x1v, evv));
-@@ -938,7 +936,7 @@
+@@ -938,7 +936,7 @@ void newviewGTRCAT_AVX(int tipCase, dou
__m256d
evv = _mm256_load_pd(&EV[l * 4]);
@@ -91,7 +91,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
vv = FMAMACC(vv,x1v,evv);
#else
vv = _mm256_add_pd(vv, _mm256_mul_pd(x1v, evv));
-@@ -990,7 +988,7 @@
+@@ -990,7 +988,7 @@ void newviewGTRCAT_AVX(int tipCase, dou
__m256d
evv = _mm256_load_pd(&EV[l * 4]);
@@ -100,7 +100,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
vv = FMAMACC(vv,x1v,evv);
#else
vv = _mm256_add_pd(vv, _mm256_mul_pd(x1v, evv));
-@@ -1077,7 +1075,7 @@
+@@ -1077,7 +1075,7 @@ void newviewGTRCAT_AVX_GAPPED_SAVE(int t
__m256d
evv = _mm256_load_pd(&EV[l * 4]);
@@ -109,7 +109,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
vv = FMAMACC(vv,x1v,evv);
#else
vv = _mm256_add_pd(vv, _mm256_mul_pd(x1v, evv));
-@@ -1139,7 +1137,7 @@
+@@ -1139,7 +1137,7 @@ void newviewGTRCAT_AVX_GAPPED_SAVE(int t
__m256d
evv = _mm256_load_pd(&EV[l * 4]);
@@ -118,7 +118,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
vv = FMAMACC(vv,x1v,evv);
#else
vv = _mm256_add_pd(vv, _mm256_mul_pd(x1v, evv));
-@@ -1204,7 +1202,7 @@
+@@ -1204,7 +1202,7 @@ void newviewGTRCAT_AVX_GAPPED_SAVE(int t
__m256d
evv = _mm256_load_pd(&EV[l * 4]);
@@ -127,7 +127,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
vv = FMAMACC(vv,x1v,evv);
#else
vv = _mm256_add_pd(vv, _mm256_mul_pd(x1v, evv));
-@@ -1290,7 +1288,7 @@
+@@ -1290,7 +1288,7 @@ void newviewGTRCAT_AVX_GAPPED_SAVE(int t
__m256d
evv = _mm256_load_pd(&EV[l * 4]);
@@ -136,7 +136,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
vv = FMAMACC(vv,x1v,evv);
#else
vv = _mm256_add_pd(vv, _mm256_mul_pd(x1v, evv));
-@@ -1338,7 +1336,7 @@
+@@ -1338,7 +1336,7 @@ void newviewGTRCATPROT_AVX(int tipCase,
int i, l, scale, addScale = 0;
@@ -145,7 +145,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
int k;
#endif
-@@ -1374,7 +1372,7 @@
+@@ -1374,7 +1372,7 @@ void newviewGTRCATPROT_AVX(int tipCase,
*lv = &le[l * 20],
*rv = &ri[l * 20];
@@ -154,7 +154,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
for(k = 0; k < 20; k += 4)
{
__m256d vlv = _mm256_load_pd(&vl[k]);
-@@ -1399,7 +1397,7 @@
+@@ -1399,7 +1397,7 @@ void newviewGTRCATPROT_AVX(int tipCase,
#endif
x1v = hadd4(x1v, x2v);
@@ -163,7 +163,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
for(k = 0; k < 5; k++)
{
__m256d evv = _mm256_load_pd(&ev[k*4]);
-@@ -1460,7 +1458,7 @@
+@@ -1460,7 +1458,7 @@ void newviewGTRCATPROT_AVX(int tipCase,
*ev = &extEV[l * 20],
*lv = &le[l * 20],
*rv = &ri[l * 20];
@@ -172,7 +172,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
for(k = 0; k < 20; k += 4)
{
__m256d vlv = _mm256_load_pd(&vl[k]);
-@@ -1495,7 +1493,7 @@
+@@ -1495,7 +1493,7 @@ void newviewGTRCATPROT_AVX(int tipCase,
evv[3] = _mm256_load_pd(&ev[12]);
evv[4] = _mm256_load_pd(&ev[16]);
@@ -181,7 +181,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
for(k = 0; k < 5; k++)
vv[k] = FMAMACC(vv[k],x1v,evv[k]);
#else
-@@ -1586,7 +1584,7 @@
+@@ -1586,7 +1584,7 @@ void newviewGTRCATPROT_AVX(int tipCase,
x2v = _mm256_add_pd(x2v, _mm256_mul_pd(_mm256_load_pd(&vr[16]), _mm256_load_pd(&rv[16])));
x1v = hadd4(x1v, x2v);
@@ -190,7 +190,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
for(k = 0; k < 5; k++)
{
__m256d evv = _mm256_load_pd(&ev[k*4]);
-@@ -1680,7 +1678,7 @@
+@@ -1680,7 +1678,7 @@ void newviewGTRCATPROT_AVX_GAPPED_SAVE(i
addScale = 0,
scaleGap = 0;
@@ -199,7 +199,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
int k;
#endif
-@@ -1724,7 +1722,7 @@
+@@ -1724,7 +1722,7 @@ void newviewGTRCATPROT_AVX_GAPPED_SAVE(i
x2v = _mm256_add_pd(x2v, _mm256_mul_pd(_mm256_load_pd(&vr[16]), _mm256_load_pd(&rv[16])));
x1v = hadd4(x1v, x2v);
@@ -208,7 +208,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
for(k = 0; k < 5; k++)
{
__m256d evv = _mm256_load_pd(&ev[k*4]);
-@@ -1827,7 +1825,7 @@
+@@ -1827,7 +1825,7 @@ void newviewGTRCATPROT_AVX_GAPPED_SAVE(i
*lv = &le[l * 20],
*rv = &ri[l * 20];
@@ -217,7 +217,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
for(k = 0; k < 20; k += 4)
{
__m256d vlv = _mm256_load_pd(&vl[k]);
-@@ -1852,7 +1850,7 @@
+@@ -1852,7 +1850,7 @@ void newviewGTRCATPROT_AVX_GAPPED_SAVE(i
#endif
x1v = hadd4(x1v, x2v);
@@ -226,7 +226,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
for(k = 0; k < 5; k++)
{
__m256d evv = _mm256_load_pd(&ev[k*4]);
-@@ -1942,7 +1940,7 @@
+@@ -1942,7 +1940,7 @@ void newviewGTRCATPROT_AVX_GAPPED_SAVE(i
*ev = &extEV[l * 20],
*lv = &le[l * 20],
*rv = &ri[l * 20];
@@ -235,7 +235,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
for(k = 0; k < 20; k += 4)
{
__m256d vlv = _mm256_load_pd(&vl[k]);
-@@ -1977,7 +1975,7 @@
+@@ -1977,7 +1975,7 @@ void newviewGTRCATPROT_AVX_GAPPED_SAVE(i
evv[3] = _mm256_load_pd(&ev[12]);
evv[4] = _mm256_load_pd(&ev[16]);
@@ -244,7 +244,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
for(k = 0; k < 5; k++)
vv[k] = FMAMACC(vv[k],x1v,evv[k]);
#else
-@@ -2102,7 +2100,7 @@
+@@ -2102,7 +2100,7 @@ void newviewGTRCATPROT_AVX_GAPPED_SAVE(i
x2v = _mm256_add_pd(x2v, _mm256_mul_pd(_mm256_load_pd(&vr[16]), _mm256_load_pd(&rv[16])));
x1v = hadd4(x1v, x2v);
@@ -253,7 +253,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
for(k = 0; k < 5; k++)
{
__m256d evv = _mm256_load_pd(&ev[k*4]);
-@@ -2237,7 +2235,7 @@
+@@ -2237,7 +2235,7 @@ void newviewGTRGAMMAPROT_AVX(int tipCase
for(l = 0; l < 20; l+=4)
{
__m256d vv = _mm256_load_pd(&v[l]);
@@ -262,7 +262,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
__m256d llv = _mm256_load_pd(&ll[l]);
umpX1v = FMAMACC(umpX1v,vv,llv);
__m256d rrv = _mm256_load_pd(&rr[l]);
-@@ -2279,7 +2277,7 @@
+@@ -2279,7 +2277,7 @@ void newviewGTRGAMMAPROT_AVX(int tipCase
__m256d x1px2v = _mm256_set1_pd(x1px2);
__m256d extEvv = _mm256_load_pd(&extEV[20 * k]);
@@ -271,7 +271,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
vv[0] = FMAMACC(vv[0],x1px2v,extEvv);
#else
vv[0] = _mm256_add_pd(vv[0],_mm256_mul_pd(x1px2v,extEvv));
-@@ -2287,7 +2285,7 @@
+@@ -2287,7 +2285,7 @@ void newviewGTRGAMMAPROT_AVX(int tipCase
_mm256_store_pd(&v[0],vv[0]);
extEvv = _mm256_load_pd(&extEV[20 * k + 4]);
@@ -280,7 +280,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
vv[1] = FMAMACC(vv[1],x1px2v,extEvv);
#else
vv[1] = _mm256_add_pd(vv[1],_mm256_mul_pd(x1px2v,extEvv));
-@@ -2295,7 +2293,7 @@
+@@ -2295,7 +2293,7 @@ void newviewGTRGAMMAPROT_AVX(int tipCase
_mm256_store_pd(&v[4],vv[1]);
extEvv = _mm256_load_pd(&extEV[20 * k + 8]);
@@ -289,7 +289,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
vv[2] = FMAMACC(vv[2],x1px2v,extEvv);
#else
vv[2] = _mm256_add_pd(vv[2],_mm256_mul_pd(x1px2v,extEvv));
-@@ -2303,7 +2301,7 @@
+@@ -2303,7 +2301,7 @@ void newviewGTRGAMMAPROT_AVX(int tipCase
_mm256_store_pd(&v[8],vv[2]);
extEvv = _mm256_load_pd(&extEV[20 * k + 12]);
@@ -298,7 +298,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
vv[3] = FMAMACC(vv[3],x1px2v,extEvv);
#else
vv[3] = _mm256_add_pd(vv[3],_mm256_mul_pd(x1px2v,extEvv));
-@@ -2311,7 +2309,7 @@
+@@ -2311,7 +2309,7 @@ void newviewGTRGAMMAPROT_AVX(int tipCase
_mm256_store_pd(&v[12],vv[3]);
extEvv = _mm256_load_pd(&extEV[20 * k + 16]);
@@ -307,7 +307,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
vv[4] = FMAMACC(vv[4],x1px2v,extEvv);
#else
vv[4] = _mm256_add_pd(vv[4],_mm256_mul_pd(x1px2v,extEvv));
-@@ -2340,7 +2338,7 @@
+@@ -2340,7 +2338,7 @@ void newviewGTRGAMMAPROT_AVX(int tipCase
{
__m256d vv = _mm256_load_pd(&v[l]);
__m256d leftv = _mm256_load_pd(&left[k * 20 + l]);
@@ -316,7 +316,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
umpX1v = FMAMACC(umpX1v, vv, leftv);
#else
-@@ -2366,7 +2364,7 @@
+@@ -2366,7 +2364,7 @@ void newviewGTRGAMMAPROT_AVX(int tipCase
__m256d vv = _mm256_load_pd(&v[0]);
__m256d rightv = _mm256_load_pd(&right[k*400+l*20+0]);
@@ -325,7 +325,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ump_x2v = FMAMACC(ump_x2v,vv,rightv);
#else
ump_x2v = _mm256_add_pd(ump_x2v, _mm256_mul_pd(vv, rightv));
-@@ -2374,7 +2372,7 @@
+@@ -2374,7 +2372,7 @@ void newviewGTRGAMMAPROT_AVX(int tipCase
vv = _mm256_load_pd(&v[4]);
rightv = _mm256_load_pd(&right[k*400+l*20+4]);
@@ -334,7 +334,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ump_x2v = FMAMACC(ump_x2v,vv,rightv);
#else
ump_x2v = _mm256_add_pd(ump_x2v, _mm256_mul_pd(vv, rightv));
-@@ -2382,7 +2380,7 @@
+@@ -2382,7 +2380,7 @@ void newviewGTRGAMMAPROT_AVX(int tipCase
vv = _mm256_load_pd(&v[8]);
rightv = _mm256_load_pd(&right[k*400+l*20+8]);
@@ -343,7 +343,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ump_x2v = FMAMACC(ump_x2v,vv,rightv);
#else
ump_x2v = _mm256_add_pd(ump_x2v, _mm256_mul_pd(vv, rightv));
-@@ -2390,7 +2388,7 @@
+@@ -2390,7 +2388,7 @@ void newviewGTRGAMMAPROT_AVX(int tipCase
vv = _mm256_load_pd(&v[12]);
rightv = _mm256_load_pd(&right[k*400+l*20+12]);
@@ -352,7 +352,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ump_x2v = FMAMACC(ump_x2v,vv,rightv);
#else
ump_x2v = _mm256_add_pd(ump_x2v, _mm256_mul_pd(vv, rightv));
-@@ -2398,7 +2396,7 @@
+@@ -2398,7 +2396,7 @@ void newviewGTRGAMMAPROT_AVX(int tipCase
vv = _mm256_load_pd(&v[16]);
rightv = _mm256_load_pd(&right[k*400+l*20+16]);
@@ -361,7 +361,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ump_x2v = FMAMACC(ump_x2v,vv,rightv);
#else
ump_x2v = _mm256_add_pd(ump_x2v, _mm256_mul_pd(vv, rightv));
-@@ -2424,7 +2422,7 @@
+@@ -2424,7 +2422,7 @@ void newviewGTRGAMMAPROT_AVX(int tipCase
x1px2 = uX1[k * 20 + l] * ump_x2[l];
__m256d x1px2v = _mm256_set1_pd(x1px2);
@@ -370,7 +370,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
__m256d ev = _mm256_load_pd(&extEV[l * 20 + 0]);
vv[0] = FMAMACC(vv[0],x1px2v, ev);
#else
-@@ -2432,7 +2430,7 @@
+@@ -2432,7 +2430,7 @@ void newviewGTRGAMMAPROT_AVX(int tipCase
#endif
_mm256_store_pd(&v[0],vv[0]);
@@ -379,7 +379,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ev = _mm256_load_pd(&extEV[l * 20 + 4]);
vv[1] = FMAMACC(vv[1],x1px2v, ev);
#else
-@@ -2440,7 +2438,7 @@
+@@ -2440,7 +2438,7 @@ void newviewGTRGAMMAPROT_AVX(int tipCase
#endif
_mm256_store_pd(&v[4],vv[1]);
@@ -388,7 +388,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ev = _mm256_load_pd(&extEV[l * 20 + 8]);
vv[2] = FMAMACC(vv[2],x1px2v, ev);
#else
-@@ -2448,7 +2446,7 @@
+@@ -2448,7 +2446,7 @@ void newviewGTRGAMMAPROT_AVX(int tipCase
#endif
_mm256_store_pd(&v[8],vv[2]);
@@ -397,7 +397,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ev = _mm256_load_pd(&extEV[l * 20 + 12]);
vv[3] = FMAMACC(vv[3],x1px2v, ev);
#else
-@@ -2457,7 +2455,7 @@
+@@ -2457,7 +2455,7 @@ void newviewGTRGAMMAPROT_AVX(int tipCase
_mm256_store_pd(&v[12],vv[3]);
@@ -406,7 +406,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ev = _mm256_load_pd(&extEV[l * 20 + 16]);
vv[4] = FMAMACC(vv[4],x1px2v, ev);
#else
-@@ -2525,7 +2523,7 @@
+@@ -2525,7 +2523,7 @@ void newviewGTRGAMMAPROT_AVX(int tipCase
__m256d vlv = _mm256_load_pd(&vl[0]);
__m256d vrv = _mm256_load_pd(&vr[0]);
@@ -415,7 +415,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
al = FMAMACC(al, vlv, leftv);
ar = FMAMACC(ar, vrv, rightv);
-@@ -2538,7 +2536,7 @@
+@@ -2538,7 +2536,7 @@ void newviewGTRGAMMAPROT_AVX(int tipCase
rightv = _mm256_load_pd(&right[k * 400 + l * 20 + 4]);
vlv = _mm256_load_pd(&vl[4]);
vrv = _mm256_load_pd(&vr[4]);
@@ -424,7 +424,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
al = FMAMACC(al, vlv, leftv);
ar = FMAMACC(ar, vrv, rightv);
-@@ -2551,7 +2549,7 @@
+@@ -2551,7 +2549,7 @@ void newviewGTRGAMMAPROT_AVX(int tipCase
rightv = _mm256_load_pd(&right[k * 400 + l * 20 + 8]);
vlv = _mm256_load_pd(&vl[8]);
vrv = _mm256_load_pd(&vr[8]);
@@ -433,7 +433,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
al = FMAMACC(al, vlv, leftv);
ar = FMAMACC(ar, vrv, rightv);
-@@ -2564,7 +2562,7 @@
+@@ -2564,7 +2562,7 @@ void newviewGTRGAMMAPROT_AVX(int tipCase
rightv = _mm256_load_pd(&right[k * 400 + l * 20 + 12]);
vlv = _mm256_load_pd(&vl[12]);
vrv = _mm256_load_pd(&vr[12]);
@@ -442,7 +442,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
al = FMAMACC(al, vlv, leftv);
ar = FMAMACC(ar, vrv, rightv);
-@@ -2578,7 +2576,7 @@
+@@ -2578,7 +2576,7 @@ void newviewGTRGAMMAPROT_AVX(int tipCase
vlv = _mm256_load_pd(&vl[16]);
vrv = _mm256_load_pd(&vr[16]);
@@ -451,7 +451,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
al = FMAMACC(al, vlv, leftv);
ar = FMAMACC(ar, vrv, rightv);
#else
-@@ -2593,7 +2591,7 @@
+@@ -2593,7 +2591,7 @@ void newviewGTRGAMMAPROT_AVX(int tipCase
al = _mm256_mul_pd(ar,al);
/************************************************************************************************************/
@@ -460,7 +460,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
__m256d ev = _mm256_load_pd(&extEV[20 * l + 0]);
vv[0] = FMAMACC(vv[0], al, ev);
#else
-@@ -2601,7 +2599,7 @@
+@@ -2601,7 +2599,7 @@ void newviewGTRGAMMAPROT_AVX(int tipCase
#endif
_mm256_store_pd(&v[0],vv[0]);
@@ -469,7 +469,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ev = _mm256_load_pd(&extEV[20 * l + 4]);
vv[1] = FMAMACC(vv[1], al, ev);
#else
-@@ -2609,7 +2607,7 @@
+@@ -2609,7 +2607,7 @@ void newviewGTRGAMMAPROT_AVX(int tipCase
#endif
_mm256_store_pd(&v[4],vv[1]);
@@ -478,7 +478,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ev = _mm256_load_pd(&extEV[20 * l + 8]);
vv[2] = FMAMACC(vv[2], al, ev);
#else
-@@ -2617,7 +2615,7 @@
+@@ -2617,7 +2615,7 @@ void newviewGTRGAMMAPROT_AVX(int tipCase
#endif
_mm256_store_pd(&v[8],vv[2]);
@@ -487,7 +487,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ev = _mm256_load_pd(&extEV[20 * l + 12]);
vv[3] = FMAMACC(vv[3], al, ev);
#else
-@@ -2625,7 +2623,7 @@
+@@ -2625,7 +2623,7 @@ void newviewGTRGAMMAPROT_AVX(int tipCase
#endif
_mm256_store_pd(&v[12],vv[3]);
@@ -496,7 +496,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ev = _mm256_load_pd(&extEV[20 * l + 16]);
vv[4] = FMAMACC(vv[4], al, ev);
#else
-@@ -2735,7 +2733,7 @@
+@@ -2735,7 +2733,7 @@ void newviewGTRGAMMAPROT_AVX_LG4(int tip
for(l = 0; l < 20; l+=4)
{
__m256d vv = _mm256_load_pd(&v[l]);
@@ -505,7 +505,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
__m256d llv = _mm256_load_pd(&ll[l]);
umpX1v = FMAMACC(umpX1v,vv,llv);
__m256d rrv = _mm256_load_pd(&rr[l]);
-@@ -2777,7 +2775,7 @@
+@@ -2777,7 +2775,7 @@ void newviewGTRGAMMAPROT_AVX_LG4(int tip
__m256d x1px2v = _mm256_set1_pd(x1px2);
__m256d extEvv = _mm256_load_pd(&extEV[j][20 * k]);
@@ -514,7 +514,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
vv[0] = FMAMACC(vv[0],x1px2v,extEvv);
#else
vv[0] = _mm256_add_pd(vv[0],_mm256_mul_pd(x1px2v,extEvv));
-@@ -2785,7 +2783,7 @@
+@@ -2785,7 +2783,7 @@ void newviewGTRGAMMAPROT_AVX_LG4(int tip
_mm256_store_pd(&v[0],vv[0]);
extEvv = _mm256_load_pd(&extEV[j][20 * k + 4]);
@@ -523,7 +523,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
vv[1] = FMAMACC(vv[1],x1px2v,extEvv);
#else
vv[1] = _mm256_add_pd(vv[1],_mm256_mul_pd(x1px2v,extEvv));
-@@ -2793,7 +2791,7 @@
+@@ -2793,7 +2791,7 @@ void newviewGTRGAMMAPROT_AVX_LG4(int tip
_mm256_store_pd(&v[4],vv[1]);
extEvv = _mm256_load_pd(&extEV[j][20 * k + 8]);
@@ -532,7 +532,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
vv[2] = FMAMACC(vv[2],x1px2v,extEvv);
#else
vv[2] = _mm256_add_pd(vv[2],_mm256_mul_pd(x1px2v,extEvv));
-@@ -2801,7 +2799,7 @@
+@@ -2801,7 +2799,7 @@ void newviewGTRGAMMAPROT_AVX_LG4(int tip
_mm256_store_pd(&v[8],vv[2]);
extEvv = _mm256_load_pd(&extEV[j][20 * k + 12]);
@@ -541,7 +541,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
vv[3] = FMAMACC(vv[3],x1px2v,extEvv);
#else
vv[3] = _mm256_add_pd(vv[3],_mm256_mul_pd(x1px2v,extEvv));
-@@ -2809,7 +2807,7 @@
+@@ -2809,7 +2807,7 @@ void newviewGTRGAMMAPROT_AVX_LG4(int tip
_mm256_store_pd(&v[12],vv[3]);
extEvv = _mm256_load_pd(&extEV[j][20 * k + 16]);
@@ -550,7 +550,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
vv[4] = FMAMACC(vv[4],x1px2v,extEvv);
#else
vv[4] = _mm256_add_pd(vv[4],_mm256_mul_pd(x1px2v,extEvv));
-@@ -2839,7 +2837,7 @@
+@@ -2839,7 +2837,7 @@ void newviewGTRGAMMAPROT_AVX_LG4(int tip
{
__m256d vv = _mm256_load_pd(&v[l]);
__m256d leftv = _mm256_load_pd(&left[k * 20 + l]);
@@ -559,7 +559,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
umpX1v = FMAMACC(umpX1v, vv, leftv);
#else
-@@ -2865,7 +2863,7 @@
+@@ -2865,7 +2863,7 @@ void newviewGTRGAMMAPROT_AVX_LG4(int tip
__m256d vv = _mm256_load_pd(&v[0]);
__m256d rightv = _mm256_load_pd(&right[k*400+l*20+0]);
@@ -568,7 +568,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ump_x2v = FMAMACC(ump_x2v,vv,rightv);
#else
ump_x2v = _mm256_add_pd(ump_x2v, _mm256_mul_pd(vv, rightv));
-@@ -2873,7 +2871,7 @@
+@@ -2873,7 +2871,7 @@ void newviewGTRGAMMAPROT_AVX_LG4(int tip
vv = _mm256_load_pd(&v[4]);
rightv = _mm256_load_pd(&right[k*400+l*20+4]);
@@ -577,7 +577,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ump_x2v = FMAMACC(ump_x2v,vv,rightv);
#else
ump_x2v = _mm256_add_pd(ump_x2v, _mm256_mul_pd(vv, rightv));
-@@ -2881,7 +2879,7 @@
+@@ -2881,7 +2879,7 @@ void newviewGTRGAMMAPROT_AVX_LG4(int tip
vv = _mm256_load_pd(&v[8]);
rightv = _mm256_load_pd(&right[k*400+l*20+8]);
@@ -586,7 +586,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ump_x2v = FMAMACC(ump_x2v,vv,rightv);
#else
ump_x2v = _mm256_add_pd(ump_x2v, _mm256_mul_pd(vv, rightv));
-@@ -2889,7 +2887,7 @@
+@@ -2889,7 +2887,7 @@ void newviewGTRGAMMAPROT_AVX_LG4(int tip
vv = _mm256_load_pd(&v[12]);
rightv = _mm256_load_pd(&right[k*400+l*20+12]);
@@ -595,7 +595,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ump_x2v = FMAMACC(ump_x2v,vv,rightv);
#else
ump_x2v = _mm256_add_pd(ump_x2v, _mm256_mul_pd(vv, rightv));
-@@ -2897,7 +2895,7 @@
+@@ -2897,7 +2895,7 @@ void newviewGTRGAMMAPROT_AVX_LG4(int tip
vv = _mm256_load_pd(&v[16]);
rightv = _mm256_load_pd(&right[k*400+l*20+16]);
@@ -604,7 +604,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ump_x2v = FMAMACC(ump_x2v,vv,rightv);
#else
ump_x2v = _mm256_add_pd(ump_x2v, _mm256_mul_pd(vv, rightv));
-@@ -2923,7 +2921,7 @@
+@@ -2923,7 +2921,7 @@ void newviewGTRGAMMAPROT_AVX_LG4(int tip
x1px2 = uX1[k * 20 + l] * ump_x2[l];
__m256d x1px2v = _mm256_set1_pd(x1px2);
@@ -613,7 +613,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
__m256d ev = _mm256_load_pd(&extEV[k][l * 20 + 0]);
vv[0] = FMAMACC(vv[0],x1px2v, ev);
#else
-@@ -2931,7 +2929,7 @@
+@@ -2931,7 +2929,7 @@ void newviewGTRGAMMAPROT_AVX_LG4(int tip
#endif
_mm256_store_pd(&v[0],vv[0]);
@@ -622,7 +622,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ev = _mm256_load_pd(&extEV[k][l * 20 + 4]);
vv[1] = FMAMACC(vv[1],x1px2v, ev);
#else
-@@ -2939,7 +2937,7 @@
+@@ -2939,7 +2937,7 @@ void newviewGTRGAMMAPROT_AVX_LG4(int tip
#endif
_mm256_store_pd(&v[4],vv[1]);
@@ -631,7 +631,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ev = _mm256_load_pd(&extEV[k][l * 20 + 8]);
vv[2] = FMAMACC(vv[2],x1px2v, ev);
#else
-@@ -2947,7 +2945,7 @@
+@@ -2947,7 +2945,7 @@ void newviewGTRGAMMAPROT_AVX_LG4(int tip
#endif
_mm256_store_pd(&v[8],vv[2]);
@@ -640,7 +640,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ev = _mm256_load_pd(&extEV[k][l * 20 + 12]);
vv[3] = FMAMACC(vv[3],x1px2v, ev);
#else
-@@ -2956,7 +2954,7 @@
+@@ -2956,7 +2954,7 @@ void newviewGTRGAMMAPROT_AVX_LG4(int tip
_mm256_store_pd(&v[12],vv[3]);
@@ -649,7 +649,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ev = _mm256_load_pd(&extEV[k][l * 20 + 16]);
vv[4] = FMAMACC(vv[4],x1px2v, ev);
#else
-@@ -3024,7 +3022,7 @@
+@@ -3024,7 +3022,7 @@ void newviewGTRGAMMAPROT_AVX_LG4(int tip
__m256d vlv = _mm256_load_pd(&vl[0]);
__m256d vrv = _mm256_load_pd(&vr[0]);
@@ -658,7 +658,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
al = FMAMACC(al, vlv, leftv);
ar = FMAMACC(ar, vrv, rightv);
-@@ -3037,7 +3035,7 @@
+@@ -3037,7 +3035,7 @@ void newviewGTRGAMMAPROT_AVX_LG4(int tip
rightv = _mm256_load_pd(&right[k * 400 + l * 20 + 4]);
vlv = _mm256_load_pd(&vl[4]);
vrv = _mm256_load_pd(&vr[4]);
@@ -667,7 +667,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
al = FMAMACC(al, vlv, leftv);
ar = FMAMACC(ar, vrv, rightv);
-@@ -3050,7 +3048,7 @@
+@@ -3050,7 +3048,7 @@ void newviewGTRGAMMAPROT_AVX_LG4(int tip
rightv = _mm256_load_pd(&right[k * 400 + l * 20 + 8]);
vlv = _mm256_load_pd(&vl[8]);
vrv = _mm256_load_pd(&vr[8]);
@@ -676,7 +676,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
al = FMAMACC(al, vlv, leftv);
ar = FMAMACC(ar, vrv, rightv);
-@@ -3063,7 +3061,7 @@
+@@ -3063,7 +3061,7 @@ void newviewGTRGAMMAPROT_AVX_LG4(int tip
rightv = _mm256_load_pd(&right[k * 400 + l * 20 + 12]);
vlv = _mm256_load_pd(&vl[12]);
vrv = _mm256_load_pd(&vr[12]);
@@ -685,7 +685,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
al = FMAMACC(al, vlv, leftv);
ar = FMAMACC(ar, vrv, rightv);
-@@ -3077,7 +3075,7 @@
+@@ -3077,7 +3075,7 @@ void newviewGTRGAMMAPROT_AVX_LG4(int tip
vlv = _mm256_load_pd(&vl[16]);
vrv = _mm256_load_pd(&vr[16]);
@@ -694,7 +694,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
al = FMAMACC(al, vlv, leftv);
ar = FMAMACC(ar, vrv, rightv);
#else
-@@ -3092,7 +3090,7 @@
+@@ -3092,7 +3090,7 @@ void newviewGTRGAMMAPROT_AVX_LG4(int tip
al = _mm256_mul_pd(ar,al);
/************************************************************************************************************/
@@ -703,7 +703,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
__m256d ev = _mm256_load_pd(&extEV[k][20 * l + 0]);
vv[0] = FMAMACC(vv[0], al, ev);
#else
-@@ -3100,7 +3098,7 @@
+@@ -3100,7 +3098,7 @@ void newviewGTRGAMMAPROT_AVX_LG4(int tip
#endif
_mm256_store_pd(&v[0],vv[0]);
@@ -712,7 +712,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ev = _mm256_load_pd(&extEV[k][20 * l + 4]);
vv[1] = FMAMACC(vv[1], al, ev);
#else
-@@ -3108,7 +3106,7 @@
+@@ -3108,7 +3106,7 @@ void newviewGTRGAMMAPROT_AVX_LG4(int tip
#endif
_mm256_store_pd(&v[4],vv[1]);
@@ -721,7 +721,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ev = _mm256_load_pd(&extEV[k][20 * l + 8]);
vv[2] = FMAMACC(vv[2], al, ev);
#else
-@@ -3116,7 +3114,7 @@
+@@ -3116,7 +3114,7 @@ void newviewGTRGAMMAPROT_AVX_LG4(int tip
#endif
_mm256_store_pd(&v[8],vv[2]);
@@ -730,7 +730,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ev = _mm256_load_pd(&extEV[k][20 * l + 12]);
vv[3] = FMAMACC(vv[3], al, ev);
#else
-@@ -3124,7 +3122,7 @@
+@@ -3124,7 +3122,7 @@ void newviewGTRGAMMAPROT_AVX_LG4(int tip
#endif
_mm256_store_pd(&v[12],vv[3]);
@@ -739,7 +739,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ev = _mm256_load_pd(&extEV[k][20 * l + 16]);
vv[4] = FMAMACC(vv[4], al, ev);
#else
-@@ -3240,7 +3238,7 @@
+@@ -3240,7 +3238,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
for(l = 0; l < 20; l+=4)
{
__m256d vv = _mm256_load_pd(&v[l]);
@@ -748,7 +748,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
__m256d llv = _mm256_load_pd(&ll[l]);
umpX1v = FMAMACC(umpX1v,vv,llv);
__m256d rrv = _mm256_load_pd(&rr[l]);
-@@ -3282,7 +3280,7 @@
+@@ -3282,7 +3280,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
__m256d x1px2v = _mm256_set1_pd(x1px2);
__m256d extEvv = _mm256_load_pd(&extEV[20 * k]);
@@ -757,7 +757,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
vv[0] = FMAMACC(vv[0],x1px2v,extEvv);
#else
vv[0] = _mm256_add_pd(vv[0],_mm256_mul_pd(x1px2v,extEvv));
-@@ -3290,7 +3288,7 @@
+@@ -3290,7 +3288,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
_mm256_store_pd(&v[0],vv[0]);
extEvv = _mm256_load_pd(&extEV[20 * k + 4]);
@@ -766,7 +766,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
vv[1] = FMAMACC(vv[1],x1px2v,extEvv);
#else
vv[1] = _mm256_add_pd(vv[1],_mm256_mul_pd(x1px2v,extEvv));
-@@ -3298,7 +3296,7 @@
+@@ -3298,7 +3296,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
_mm256_store_pd(&v[4],vv[1]);
extEvv = _mm256_load_pd(&extEV[20 * k + 8]);
@@ -775,7 +775,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
vv[2] = FMAMACC(vv[2],x1px2v,extEvv);
#else
vv[2] = _mm256_add_pd(vv[2],_mm256_mul_pd(x1px2v,extEvv));
-@@ -3306,7 +3304,7 @@
+@@ -3306,7 +3304,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
_mm256_store_pd(&v[8],vv[2]);
extEvv = _mm256_load_pd(&extEV[20 * k + 12]);
@@ -784,7 +784,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
vv[3] = FMAMACC(vv[3],x1px2v,extEvv);
#else
vv[3] = _mm256_add_pd(vv[3],_mm256_mul_pd(x1px2v,extEvv));
-@@ -3314,7 +3312,7 @@
+@@ -3314,7 +3312,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
_mm256_store_pd(&v[12],vv[3]);
extEvv = _mm256_load_pd(&extEV[20 * k + 16]);
@@ -793,7 +793,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
vv[4] = FMAMACC(vv[4],x1px2v,extEvv);
#else
vv[4] = _mm256_add_pd(vv[4],_mm256_mul_pd(x1px2v,extEvv));
-@@ -3351,7 +3349,7 @@
+@@ -3351,7 +3349,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
__m256d x1px2v = _mm256_set1_pd(x1px2);
__m256d extEvv = _mm256_load_pd(&extEV[20 * k]);
@@ -802,7 +802,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
vv[0] = FMAMACC(vv[0],x1px2v,extEvv);
#else
vv[0] = _mm256_add_pd(vv[0],_mm256_mul_pd(x1px2v,extEvv));
-@@ -3359,7 +3357,7 @@
+@@ -3359,7 +3357,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
_mm256_store_pd(&v[0],vv[0]);
extEvv = _mm256_load_pd(&extEV[20 * k + 4]);
@@ -811,7 +811,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
vv[1] = FMAMACC(vv[1],x1px2v,extEvv);
#else
vv[1] = _mm256_add_pd(vv[1],_mm256_mul_pd(x1px2v,extEvv));
-@@ -3367,7 +3365,7 @@
+@@ -3367,7 +3365,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
_mm256_store_pd(&v[4],vv[1]);
extEvv = _mm256_load_pd(&extEV[20 * k + 8]);
@@ -820,7 +820,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
vv[2] = FMAMACC(vv[2],x1px2v,extEvv);
#else
vv[2] = _mm256_add_pd(vv[2],_mm256_mul_pd(x1px2v,extEvv));
-@@ -3375,7 +3373,7 @@
+@@ -3375,7 +3373,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
_mm256_store_pd(&v[8],vv[2]);
extEvv = _mm256_load_pd(&extEV[20 * k + 12]);
@@ -829,7 +829,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
vv[3] = FMAMACC(vv[3],x1px2v,extEvv);
#else
vv[3] = _mm256_add_pd(vv[3],_mm256_mul_pd(x1px2v,extEvv));
-@@ -3383,7 +3381,7 @@
+@@ -3383,7 +3381,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
_mm256_store_pd(&v[12],vv[3]);
extEvv = _mm256_load_pd(&extEV[20 * k + 16]);
@@ -838,7 +838,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
vv[4] = FMAMACC(vv[4],x1px2v,extEvv);
#else
vv[4] = _mm256_add_pd(vv[4],_mm256_mul_pd(x1px2v,extEvv));
-@@ -3415,7 +3413,7 @@
+@@ -3415,7 +3413,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
{
__m256d vv = _mm256_load_pd(&v[l]);
__m256d leftv = _mm256_load_pd(&left[k * 20 + l]);
@@ -847,7 +847,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
umpX1v = FMAMACC(umpX1v, vv, leftv);
#else
-@@ -3440,7 +3438,7 @@
+@@ -3440,7 +3438,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
__m256d vv = _mm256_load_pd(&v[0]);
__m256d rightv = _mm256_load_pd(&right[k*400+l*20+0]);
@@ -856,7 +856,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ump_x2v = FMAMACC(ump_x2v,vv,rightv);
#else
ump_x2v = _mm256_add_pd(ump_x2v, _mm256_mul_pd(vv, rightv));
-@@ -3448,7 +3446,7 @@
+@@ -3448,7 +3446,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
vv = _mm256_load_pd(&v[4]);
rightv = _mm256_load_pd(&right[k*400+l*20+4]);
@@ -865,7 +865,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ump_x2v = FMAMACC(ump_x2v,vv,rightv);
#else
ump_x2v = _mm256_add_pd(ump_x2v, _mm256_mul_pd(vv, rightv));
-@@ -3456,7 +3454,7 @@
+@@ -3456,7 +3454,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
vv = _mm256_load_pd(&v[8]);
rightv = _mm256_load_pd(&right[k*400+l*20+8]);
@@ -874,7 +874,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ump_x2v = FMAMACC(ump_x2v,vv,rightv);
#else
ump_x2v = _mm256_add_pd(ump_x2v, _mm256_mul_pd(vv, rightv));
-@@ -3464,7 +3462,7 @@
+@@ -3464,7 +3462,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
vv = _mm256_load_pd(&v[12]);
rightv = _mm256_load_pd(&right[k*400+l*20+12]);
@@ -883,7 +883,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ump_x2v = FMAMACC(ump_x2v,vv,rightv);
#else
ump_x2v = _mm256_add_pd(ump_x2v, _mm256_mul_pd(vv, rightv));
-@@ -3472,7 +3470,7 @@
+@@ -3472,7 +3470,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
vv = _mm256_load_pd(&v[16]);
rightv = _mm256_load_pd(&right[k*400+l*20+16]);
@@ -892,7 +892,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ump_x2v = FMAMACC(ump_x2v,vv,rightv);
#else
ump_x2v = _mm256_add_pd(ump_x2v, _mm256_mul_pd(vv, rightv));
-@@ -3497,7 +3495,7 @@
+@@ -3497,7 +3495,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
x1px2 = uX1[k * 20 + l] * ump_x2[l];
__m256d x1px2v = _mm256_set1_pd(x1px2);
@@ -901,7 +901,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
__m256d ev = _mm256_load_pd(&extEV[l * 20 + 0]);
vv[0] = FMAMACC(vv[0],x1px2v, ev);
#else
-@@ -3505,7 +3503,7 @@
+@@ -3505,7 +3503,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
#endif
_mm256_store_pd(&v[0],vv[0]);
@@ -910,7 +910,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ev = _mm256_load_pd(&extEV[l * 20 + 4]);
vv[1] = FMAMACC(vv[1],x1px2v, ev);
#else
-@@ -3513,7 +3511,7 @@
+@@ -3513,7 +3511,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
#endif
_mm256_store_pd(&v[4],vv[1]);
@@ -919,7 +919,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ev = _mm256_load_pd(&extEV[l * 20 + 8]);
vv[2] = FMAMACC(vv[2],x1px2v, ev);
#else
-@@ -3521,7 +3519,7 @@
+@@ -3521,7 +3519,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
#endif
_mm256_store_pd(&v[8],vv[2]);
@@ -928,7 +928,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ev = _mm256_load_pd(&extEV[l * 20 + 12]);
vv[3] = FMAMACC(vv[3],x1px2v, ev);
#else
-@@ -3530,7 +3528,7 @@
+@@ -3530,7 +3528,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
_mm256_store_pd(&v[12],vv[3]);
@@ -937,7 +937,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ev = _mm256_load_pd(&extEV[l * 20 + 16]);
vv[4] = FMAMACC(vv[4],x1px2v, ev);
#else
-@@ -3600,7 +3598,7 @@
+@@ -3600,7 +3598,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
__m256d vv = _mm256_load_pd(&v[0]);
__m256d rightv = _mm256_load_pd(&right[k*400+l*20+0]);
@@ -946,7 +946,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ump_x2v = FMAMACC(ump_x2v,vv,rightv);
#else
ump_x2v = _mm256_add_pd(ump_x2v, _mm256_mul_pd(vv, rightv));
-@@ -3608,7 +3606,7 @@
+@@ -3608,7 +3606,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
vv = _mm256_load_pd(&v[4]);
rightv = _mm256_load_pd(&right[k*400+l*20+4]);
@@ -955,7 +955,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ump_x2v = FMAMACC(ump_x2v,vv,rightv);
#else
ump_x2v = _mm256_add_pd(ump_x2v, _mm256_mul_pd(vv, rightv));
-@@ -3616,7 +3614,7 @@
+@@ -3616,7 +3614,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
vv = _mm256_load_pd(&v[8]);
rightv = _mm256_load_pd(&right[k*400+l*20+8]);
@@ -964,7 +964,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ump_x2v = FMAMACC(ump_x2v,vv,rightv);
#else
ump_x2v = _mm256_add_pd(ump_x2v, _mm256_mul_pd(vv, rightv));
-@@ -3624,7 +3622,7 @@
+@@ -3624,7 +3622,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
vv = _mm256_load_pd(&v[12]);
rightv = _mm256_load_pd(&right[k*400+l*20+12]);
@@ -973,7 +973,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ump_x2v = FMAMACC(ump_x2v,vv,rightv);
#else
ump_x2v = _mm256_add_pd(ump_x2v, _mm256_mul_pd(vv, rightv));
-@@ -3632,7 +3630,7 @@
+@@ -3632,7 +3630,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
vv = _mm256_load_pd(&v[16]);
rightv = _mm256_load_pd(&right[k*400+l*20+16]);
@@ -982,7 +982,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ump_x2v = FMAMACC(ump_x2v,vv,rightv);
#else
ump_x2v = _mm256_add_pd(ump_x2v, _mm256_mul_pd(vv, rightv));
-@@ -3658,7 +3656,7 @@
+@@ -3658,7 +3656,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
x1px2 = uX1[k * 20 + l] * ump_x2[l];
__m256d x1px2v = _mm256_set1_pd(x1px2);
@@ -991,7 +991,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
__m256d ev = _mm256_load_pd(&extEV[l * 20 + 0]);
vv[0] = FMAMACC(vv[0],x1px2v, ev);
#else
-@@ -3666,7 +3664,7 @@
+@@ -3666,7 +3664,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
#endif
_mm256_store_pd(&v[0],vv[0]);
@@ -1000,7 +1000,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ev = _mm256_load_pd(&extEV[l * 20 + 4]);
vv[1] = FMAMACC(vv[1],x1px2v, ev);
#else
-@@ -3674,7 +3672,7 @@
+@@ -3674,7 +3672,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
#endif
_mm256_store_pd(&v[4],vv[1]);
@@ -1009,7 +1009,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ev = _mm256_load_pd(&extEV[l * 20 + 8]);
vv[2] = FMAMACC(vv[2],x1px2v, ev);
#else
-@@ -3682,7 +3680,7 @@
+@@ -3682,7 +3680,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
#endif
_mm256_store_pd(&v[8],vv[2]);
@@ -1018,7 +1018,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ev = _mm256_load_pd(&extEV[l * 20 + 12]);
vv[3] = FMAMACC(vv[3],x1px2v, ev);
#else
-@@ -3691,7 +3689,7 @@
+@@ -3691,7 +3689,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
_mm256_store_pd(&v[12],vv[3]);
@@ -1027,7 +1027,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ev = _mm256_load_pd(&extEV[l * 20 + 16]);
vv[4] = FMAMACC(vv[4],x1px2v, ev);
#else
-@@ -3757,7 +3755,7 @@
+@@ -3757,7 +3755,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
__m256d vlv = _mm256_load_pd(&vl[0]);
__m256d vrv = _mm256_load_pd(&vr[0]);
@@ -1036,7 +1036,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
al = FMAMACC(al, vlv, leftv);
ar = FMAMACC(ar, vrv, rightv);
-@@ -3770,7 +3768,7 @@
+@@ -3770,7 +3768,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
rightv = _mm256_load_pd(&right[k * 400 + l * 20 + 4]);
vlv = _mm256_load_pd(&vl[4]);
vrv = _mm256_load_pd(&vr[4]);
@@ -1045,7 +1045,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
al = FMAMACC(al, vlv, leftv);
ar = FMAMACC(ar, vrv, rightv);
-@@ -3783,7 +3781,7 @@
+@@ -3783,7 +3781,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
rightv = _mm256_load_pd(&right[k * 400 + l * 20 + 8]);
vlv = _mm256_load_pd(&vl[8]);
vrv = _mm256_load_pd(&vr[8]);
@@ -1054,7 +1054,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
al = FMAMACC(al, vlv, leftv);
ar = FMAMACC(ar, vrv, rightv);
-@@ -3796,7 +3794,7 @@
+@@ -3796,7 +3794,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
rightv = _mm256_load_pd(&right[k * 400 + l * 20 + 12]);
vlv = _mm256_load_pd(&vl[12]);
vrv = _mm256_load_pd(&vr[12]);
@@ -1063,7 +1063,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
al = FMAMACC(al, vlv, leftv);
ar = FMAMACC(ar, vrv, rightv);
-@@ -3810,7 +3808,7 @@
+@@ -3810,7 +3808,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
vlv = _mm256_load_pd(&vl[16]);
vrv = _mm256_load_pd(&vr[16]);
@@ -1072,7 +1072,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
al = FMAMACC(al, vlv, leftv);
ar = FMAMACC(ar, vrv, rightv);
#else
-@@ -3825,7 +3823,7 @@
+@@ -3825,7 +3823,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
al = _mm256_mul_pd(ar,al);
/************************************************************************************************************/
@@ -1081,7 +1081,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
__m256d ev = _mm256_load_pd(&extEV[20 * l + 0]);
vv[0] = FMAMACC(vv[0], al, ev);
#else
-@@ -3833,7 +3831,7 @@
+@@ -3833,7 +3831,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
#endif
_mm256_store_pd(&v[0],vv[0]);
@@ -1090,7 +1090,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ev = _mm256_load_pd(&extEV[20 * l + 4]);
vv[1] = FMAMACC(vv[1], al, ev);
#else
-@@ -3841,7 +3839,7 @@
+@@ -3841,7 +3839,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
#endif
_mm256_store_pd(&v[4],vv[1]);
@@ -1099,7 +1099,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ev = _mm256_load_pd(&extEV[20 * l + 8]);
vv[2] = FMAMACC(vv[2], al, ev);
#else
-@@ -3849,7 +3847,7 @@
+@@ -3849,7 +3847,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
#endif
_mm256_store_pd(&v[8],vv[2]);
@@ -1108,7 +1108,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ev = _mm256_load_pd(&extEV[20 * l + 12]);
vv[3] = FMAMACC(vv[3], al, ev);
#else
-@@ -3857,7 +3855,7 @@
+@@ -3857,7 +3855,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
#endif
_mm256_store_pd(&v[12],vv[3]);
@@ -1117,7 +1117,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ev = _mm256_load_pd(&extEV[20 * l + 16]);
vv[4] = FMAMACC(vv[4], al, ev);
#else
-@@ -3950,7 +3948,7 @@
+@@ -3950,7 +3948,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
__m256d vlv = _mm256_load_pd(&vl[0]);
__m256d vrv = _mm256_load_pd(&vr[0]);
@@ -1126,7 +1126,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
al = FMAMACC(al, vlv, leftv);
ar = FMAMACC(ar, vrv, rightv);
-@@ -3963,7 +3961,7 @@
+@@ -3963,7 +3961,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
rightv = _mm256_load_pd(&right[k * 400 + l * 20 + 4]);
vlv = _mm256_load_pd(&vl[4]);
vrv = _mm256_load_pd(&vr[4]);
@@ -1135,7 +1135,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
al = FMAMACC(al, vlv, leftv);
ar = FMAMACC(ar, vrv, rightv);
-@@ -3976,7 +3974,7 @@
+@@ -3976,7 +3974,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
rightv = _mm256_load_pd(&right[k * 400 + l * 20 + 8]);
vlv = _mm256_load_pd(&vl[8]);
vrv = _mm256_load_pd(&vr[8]);
@@ -1144,7 +1144,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
al = FMAMACC(al, vlv, leftv);
ar = FMAMACC(ar, vrv, rightv);
-@@ -3989,7 +3987,7 @@
+@@ -3989,7 +3987,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
rightv = _mm256_load_pd(&right[k * 400 + l * 20 + 12]);
vlv = _mm256_load_pd(&vl[12]);
vrv = _mm256_load_pd(&vr[12]);
@@ -1153,7 +1153,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
al = FMAMACC(al, vlv, leftv);
ar = FMAMACC(ar, vrv, rightv);
-@@ -4003,7 +4001,7 @@
+@@ -4003,7 +4001,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
vlv = _mm256_load_pd(&vl[16]);
vrv = _mm256_load_pd(&vr[16]);
@@ -1162,7 +1162,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
al = FMAMACC(al, vlv, leftv);
ar = FMAMACC(ar, vrv, rightv);
#else
-@@ -4018,7 +4016,7 @@
+@@ -4018,7 +4016,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
al = _mm256_mul_pd(ar,al);
/************************************************************************************************************/
@@ -1171,7 +1171,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
__m256d ev = _mm256_load_pd(&extEV[20 * l + 0]);
vv[0] = FMAMACC(vv[0], al, ev);
#else
-@@ -4026,7 +4024,7 @@
+@@ -4026,7 +4024,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
#endif
_mm256_store_pd(&v[0],vv[0]);
@@ -1180,7 +1180,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ev = _mm256_load_pd(&extEV[20 * l + 4]);
vv[1] = FMAMACC(vv[1], al, ev);
#else
-@@ -4034,7 +4032,7 @@
+@@ -4034,7 +4032,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
#endif
_mm256_store_pd(&v[4],vv[1]);
@@ -1189,7 +1189,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ev = _mm256_load_pd(&extEV[20 * l + 8]);
vv[2] = FMAMACC(vv[2], al, ev);
#else
-@@ -4042,7 +4040,7 @@
+@@ -4042,7 +4040,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
#endif
_mm256_store_pd(&v[8],vv[2]);
@@ -1198,7 +1198,7 @@ https://github.com/stamatak/standard-RAxML/pull/50
ev = _mm256_load_pd(&extEV[20 * l + 12]);
vv[3] = FMAMACC(vv[3], al, ev);
#else
-@@ -4050,7 +4048,7 @@
+@@ -4050,7 +4048,7 @@ void newviewGTRGAMMAPROT_AVX_GAPPED_SAVE
#endif
_mm256_store_pd(&v[12],vv[3]);
@@ -1207,13 +1207,13 @@ https://github.com/stamatak/standard-RAxML/pull/50
ev = _mm256_load_pd(&extEV[20 * l + 16]);
vv[4] = FMAMACC(vv[4], al, ev);
#else
---- raxml.orig/axml.c
-+++ raxml/axml.c
-@@ -66,17 +66,8 @@
+--- a/axml.c
++++ b/axml.c
+@@ -66,16 +66,9 @@
#endif
--#if ! (defined(__ppc) || defined(__powerpc__) || defined(PPC))
+-#if ! (defined(__ppc) || defined(__powerpc__) || defined(PPC) || defined(__aarch64__))
-#include <xmmintrin.h>
-/*
- special bug fix, enforces denormalized numbers to be flushed to zero,
@@ -1223,13 +1223,13 @@ https://github.com/stamatak/standard-RAxML/pull/50
- #define MM_DAZ_ON 0x0040
- #define MM_DAZ_OFF 0x0000
-*/
--#endif
++#ifndef __aarch64__
+#define SIMDE_ENABLE_NATIVE_ALIASES
+#include <simde/x86/sse.h>
-
- #include "axml.h"
- #include "globalVariables.h"
-@@ -13707,7 +13698,7 @@
+ #elif __aarch64__
+ #include "sse2neon.h"
+ #endif
+@@ -13709,7 +13702,7 @@ int main (int argc, char *argv[])
@@ -1238,68 +1238,24 @@ https://github.com/stamatak/standard-RAxML/pull/50
/*
David Defour's command
---- raxml.orig/bipartitionList.c
-+++ raxml/bipartitionList.c
-@@ -45,12 +45,8 @@
- #include "axml.h"
- #include "rmq.h" //include range minimum queries for fast plausibility checker
-
--#ifdef __SIM_SSE3
--
--#include <xmmintrin.h>
--#include <pmmintrin.h>
--
--#endif
-+#define SIMDE_ENABLE_NATIVE_ALIASES
-+#include <simde/x86/sse3.h>
-
- #ifdef _USE_PTHREADS
- #include <pthread.h>
---- raxml.orig/evaluateGenericSpecial.c
-+++ raxml/evaluateGenericSpecial.c
-@@ -39,12 +39,8 @@
- #include <string.h>
- #include "axml.h"
-
--
--#ifdef __SIM_SSE3
--#include <xmmintrin.h>
--#include <pmmintrin.h>
--/*#include <tmmintrin.h>*/
--#endif
-+#define SIMDE_ENABLE_NATIVE_ALIASES
-+#include <simde/x86/sse3.h>
+--- a/fastDNAparsimony.c
++++ b/fastDNAparsimony.c
+@@ -59,23 +59,13 @@
+ #endif
- #ifdef _USE_PTHREADS
- extern volatile double *reductionBuffer;
---- raxml.orig/evaluatePartialGenericSpecial.c
-+++ raxml/evaluatePartialGenericSpecial.c
-@@ -40,11 +40,8 @@
- #include <string.h>
- #include "axml.h"
-#ifdef __SIM_SSE3
+-
+-#ifdef __x86_64__
-#include <xmmintrin.h>
-#include <pmmintrin.h>
--#endif
--
++#ifndef __aarch64__
+#define SIMDE_ENABLE_NATIVE_ALIASES
-+#include <simde/x86/sse3.h>
-
- /********************** GTRCAT ***************************************/
-
---- raxml.orig/fastDNAparsimony.c
-+++ raxml/fastDNAparsimony.c
-@@ -58,21 +58,8 @@
-
++#include <simde/x86/avx.h>
+ #elif __aarch64__
+ #include "sse2neon.h"
#endif
--
--#ifdef __SIM_SSE3
--
--#include <xmmintrin.h>
--#include <pmmintrin.h>
--
-#endif
-
-#ifdef __AVX
@@ -1308,13 +1264,10 @@ https://github.com/stamatak/standard-RAxML/pull/50
-#include <immintrin.h>
-
-#endif
--
-+#define SIMDE_ENABLE_NATIVE_ALIASES
-+#include <simde/x86/avx.h>
- #include "axml.h"
-@@ -102,8 +89,8 @@
+ #include "axml.h"
+@@ -106,8 +96,8 @@ extern int globalArgc;
#define INTS_PER_VECTOR 8
#define INT_TYPE __m256d
#define CAST double*
@@ -1325,8 +1278,8 @@ https://github.com/stamatak/standard-RAxML/pull/50
#define VECTOR_LOAD _mm256_load_pd
#define VECTOR_BIT_AND _mm256_and_pd
#define VECTOR_BIT_OR _mm256_or_pd
---- raxml.orig/makenewzGenericSpecial.c
-+++ raxml/makenewzGenericSpecial.c
+--- a/makenewzGenericSpecial.c
++++ b/makenewzGenericSpecial.c
@@ -33,8 +33,6 @@
#include <unistd.h>
#endif
@@ -1336,37 +1289,45 @@ https://github.com/stamatak/standard-RAxML/pull/50
#include <math.h>
#include <time.h>
#include <stdlib.h>
-@@ -43,11 +41,8 @@
+@@ -43,15 +41,12 @@
#include <string.h>
#include "axml.h"
-#ifdef __SIM_SSE3
+-#ifdef __x86_64__
-#include <xmmintrin.h>
-#include <pmmintrin.h>
-/*#include <tmmintrin.h>*/
--#endif
++#ifndef __aarch64__
+#define SIMDE_ENABLE_NATIVE_ALIASES
+#include <simde/x86/sse3.h>
+ #elif __aarch64__
+ #include "sse2neon.h"
+ #endif
+-#endif
#ifdef _USE_PTHREADS
extern volatile double *reductionBuffer;
---- raxml.orig/newviewGenericSpecial.c
-+++ raxml/newviewGenericSpecial.c
-@@ -41,11 +41,10 @@
+--- a/newviewGenericSpecial.c
++++ b/newviewGenericSpecial.c
+@@ -41,13 +41,12 @@
#include <limits.h>
#include "axml.h"
-#ifdef __SIM_SSE3
#include <stdint.h>
+
+-#ifdef __x86_64__
-#include <xmmintrin.h>
-#include <pmmintrin.h>
++#ifndef __aarch64__
+#define SIMDE_ENABLE_NATIVE_ALIASES
+#include <simde/x86/avx.h>
-
- const union __attribute__ ((aligned (BYTE_ALIGNMENT)))
- {
-@@ -53,12 +52,6 @@
+ #elif __aarch64__
+ #include "sse2neon.h"
+ #endif
+@@ -58,12 +57,6 @@ const union __attribute__ ((aligned (BYT
__m128d m;
} absMask = {{0x7fffffffffffffffULL , 0x7fffffffffffffffULL }};
@@ -1379,8 +1340,8 @@ https://github.com/stamatak/standard-RAxML/pull/50
#ifdef _USE_PTHREADS
#include <pthread.h>
extern volatile int NumberOfThreads;
---- raxml.orig/Makefile.AVX.PTHREADS.gcc
-+++ raxml/Makefile.AVX.PTHREADS.gcc
+--- a/Makefile.AVX.PTHREADS.gcc
++++ b/Makefile.AVX.PTHREADS.gcc
@@ -3,7 +3,7 @@
CC = gcc
@@ -1390,10 +1351,71 @@ https://github.com/stamatak/standard-RAxML/pull/50
LIBRARIES = -lm -pthread
-@@ -53,4 +53,4 @@
+@@ -53,4 +53,4 @@ eigen.o : eigen.c $(GLOBAL_DEPS)
clean :
$(RM) *.o raxmlHPC-PTHREADS-AVX
-dev: raxmlHPC-PTHREADS-AVX
\ No newline at end of file
+dev: raxmlHPC-PTHREADS-AVX
+--- a/bipartitionList.c
++++ b/bipartitionList.c
+@@ -45,14 +45,12 @@
+ #include "axml.h"
+ #include "rmq.h" //include range minimum queries for fast plausibility checker
+
+-#ifdef __SIM_SSE3
+-#ifdef __x86_64__
+-#include <xmmintrin.h>
+-#include <pmmintrin.h>
++#ifndef __aarch64__
++#define SIMDE_ENABLE_NATIVE_ALIASES
++#include <simde/x86/sse3.h>
+ #elif __aarch64__
+ #include "sse2neon.h"
+ #endif
+-#endif
+
+ #ifdef _USE_PTHREADS
+ #include <pthread.h>
+--- a/evaluateGenericSpecial.c
++++ b/evaluateGenericSpecial.c
+@@ -39,15 +39,12 @@
+ #include <string.h>
+ #include "axml.h"
+
+-#ifdef __SIM_SSE3
+-#ifdef __x86_64__
+-#include <xmmintrin.h>
+-#include <pmmintrin.h>
+-/*#include <tmmintrin.h>*/
++#ifndef __aarch64__
++#define SIMDE_ENABLE_NATIVE_ALIASES
++#include <simde/x86/sse3.h>
+ #elif __aarch64__
+ #include "sse2neon.h"
+ #endif
+-#endif
+
+ #ifdef _USE_PTHREADS
+ extern volatile double *reductionBuffer;
+--- a/evaluatePartialGenericSpecial.c
++++ b/evaluatePartialGenericSpecial.c
+@@ -40,14 +40,12 @@
+ #include <string.h>
+ #include "axml.h"
+
+-#ifdef __SIM_SSE3
+-#ifdef __x86_64__
+-#include <xmmintrin.h>
+-#include <pmmintrin.h>
++#ifndef __aarch64__
++#define SIMDE_ENABLE_NATIVE_ALIASES
++#include <simde/x86/sse3.h>
+ #elif __aarch64__
+ #include "sse2neon.h"
+ #endif
+-#endif
+
+ /********************** GTRCAT ***************************************/
+
View it on GitLab: https://salsa.debian.org/med-team/raxml/-/compare/4f21c1bc37e330b06c140522fac77207a90c9ae9...c5f574ac5dabffa3829cf2c9942298f81738bdde
--
View it on GitLab: https://salsa.debian.org/med-team/raxml/-/compare/4f21c1bc37e330b06c140522fac77207a90c9ae9...c5f574ac5dabffa3829cf2c9942298f81738bdde
You're receiving this email because of your account on salsa.debian.org.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://alioth-lists.debian.net/pipermail/debian-med-commit/attachments/20231128/03cdcdad/attachment-0001.htm>
More information about the debian-med-commit
mailing list