[med-svn] [Git][med-team/last-align][master] 10 commits: debian/rules, simd-dispatch: Pass an environment variable to use the plain...
Andreas Tille (@tille)
gitlab at salsa.debian.org
Mon Jan 10 09:10:01 GMT 2022
Andreas Tille pushed to branch master at Debian Med / last-align
Commits:
b01cab59 by Vagrant Cascadian at 2022-01-10T09:37:58+01:00
debian/rules, simd-dispatch: Pass an environment variable to use the plain variant to when building manpages with help2man.
The relevent simd variant of the build environment is embedded in the
manpage, breaking build reproducibility depending on the cpu features
of the running build environment.
- - - - -
ad46d4b4 by Andreas Tille at 2022-01-10T09:41:28+01:00
Do not embed cpu-specific features in manpages
- - - - -
9bf099a2 by Andreas Tille at 2022-01-10T09:42:41+01:00
Re-enable reprotest in Salca-CI to see whether the means above are working properly
- - - - -
08c0f517 by Andreas Tille at 2022-01-10T09:43:12+01:00
New upstream version 1257
- - - - -
16dc9571 by Andreas Tille at 2022-01-10T09:43:12+01:00
routine-update: New upstream version
- - - - -
43cb47e8 by Andreas Tille at 2022-01-10T09:43:19+01:00
Update upstream source from tag 'upstream/1257'
Update to upstream version '1257'
with Debian dir e871743ce190974de5be73c0780988d22bb02559
- - - - -
c4da81d3 by Andreas Tille at 2022-01-10T09:43:19+01:00
routine-update: Standards-Version: 4.6.0
- - - - -
c3852a7f by Andreas Tille at 2022-01-10T09:43:31+01:00
Add missing build dependency on dh addon.
Changes-By: lintian-brush
Fixes: lintian: missing-build-dependency-for-dh_-command
See-also: https://lintian.debian.org/tags/missing-build-dependency-for-dh_-command.html
- - - - -
92585f43 by Andreas Tille at 2022-01-10T09:52:42+01:00
Refresh patches
- - - - -
aedcfb14 by Andreas Tille at 2022-01-10T10:07:04+01:00
Upload to unstable
- - - - -
11 changed files:
- README.rst
- debian/bin/simd-dispatch
- debian/changelog
- debian/control
- debian/patches/helpMakefiles.patch
- debian/patches/simde
- debian/rules
- debian/salsa-ci.yml
- doc/last-dotplot.rst
- makefile
- src/makefile
Changes:
=====================================
README.rst
=====================================
@@ -23,8 +23,9 @@ Usage
-----
Please see the cookbook_. **Warning:** this documentation may not
-apply to older versions of LAST! You can see your version with
-``lastal --version``.
+apply to older versions of LAST! You can see your version with::
+
+ lastal --version
Install
-------
@@ -38,8 +39,16 @@ into the downloaded directory and type::
This assumes you have a C++ compiler. On Linux, you might need to
install a package called "g++". On Mac, you might need to install
command-line developer tools. On Windows, you might need to install
-Cygwin. (It's possible to specify a compiler like this:
-``make CXX=MyOtherCompiler``.)
+Cygwin. You might also need to install something like "zlib-devel".
+
+For ARM CPUs, the default "make" seems to work in some cases but not
+others (sigh). This seems to be good for ARM::
+
+ make CXXFLAGS="-mcpu=native -O3 -pthread"
+
+It's possible to specify a compiler like this: ``make CXX=MyOtherCompiler``.
+If you re-run ``make`` in different ways, it may be good to do ``make clean``
+first, to remove any previously-made files.
The programs are in the ``bin`` directory. For convenient usage, set
up your computer to find them automatically. Some possible ways:
=====================================
debian/bin/simd-dispatch
=====================================
@@ -14,7 +14,8 @@ function test_and_run () {
fi
}
-for SIMD in avx2 avx sse4.1 ssse3 sse3 sse2 sse mmx ; do test_and_run ${SIMD} "$@" ; done
+SIMD_LIST=${SIMD_LIST:-"avx2 avx sse4.1 ssse3 sse3 sse2 sse mmx"}
+for SIMD in $SIMD_LIST ; do test_and_run ${SIMD} "$@" ; done
# fallback to plain option
$BASE-plain "$@"
=====================================
debian/changelog
=====================================
@@ -1,3 +1,17 @@
+last-align (1257-1) unstable; urgency=medium
+
+ [ Vagrant Cascadian ]
+ * Do not embed cpu-specific features in manpages
+ Closes: #1003379
+
+ [ Andreas Tille ]
+ * Re-enable reprotest in Salca-CI to see whether the means above are
+ working properly
+ * New upstream version
+ * Standards-Version: 4.6.0 (routine-update)
+
+ -- Andreas Tille <tille at debian.org> Mon, 10 Jan 2022 09:53:09 +0100
+
last-align (1256-1) unstable; urgency=medium
* d/watch: Fix fetch URL - last-align has moved to gitlab
=====================================
debian/control
=====================================
@@ -9,8 +9,9 @@ Build-Depends: debhelper-compat (= 13),
help2man,
python3-pil,
zlib1g-dev,
- libsimde-dev
-Standards-Version: 4.5.1
+ libsimde-dev,
+ debhelper
+Standards-Version: 4.6.0
Vcs-Browser: https://salsa.debian.org/med-team/last-align
Vcs-Git: https://salsa.debian.org/med-team/last-align.git
Homepage: https://gitlab.com/mcfrith/last
=====================================
debian/patches/helpMakefiles.patch
=====================================
@@ -6,7 +6,7 @@ Description: Propagate variables to Makefile
--- a/src/makefile
+++ b/src/makefile
@@ -1,4 +1,7 @@
--CXXFLAGS = -msse4 -O3 -Wall -g -std=c++11 -pthread -DHAS_CXX_THREADS
+-CXXFLAGS = -msse4 -O3 -Wall -g -std=c++11 -pthread
+CXXFLAGS = -O3 -Wall -Wextra -g
+CXXFLAGS += -msse4
+CXXFLAGS += -std=c++11
@@ -14,16 +14,16 @@ Description: Propagate variables to Makefile
# -fomit-frame-pointer ?
# The number of symbol types that we can keep in sequences (e.g. 20
-@@ -7,7 +10,7 @@
+@@ -7,7 +10,7 @@ CXXFLAGS = -msse4 -O3 -Wall -g -std=c++1
ALPHABET_CAPACITY = 66
- CPPF = -DALPHABET_CAPACITY=$(ALPHABET_CAPACITY) $(CPPFLAGS)
+ CPPF = -DALPHABET_CAPACITY=$(ALPHABET_CAPACITY) -DHAS_CXX_THREADS $(CPPFLAGS)
-CFLAGS = -Wall -O2
+CFLAGS ?= -Wall -O2
alpObj = alp/sls_alignment_evaluer.o alp/sls_pvalues.o \
alp/sls_alp_sim.o alp/sls_alp_regression.o alp/sls_alp_data.o \
-@@ -72,45 +75,45 @@
+@@ -72,45 +75,45 @@ last8: $(LAST8)
indexAllObj4 = $(indexObj0) $(indexObj4)
../bin/lastdb: $(indexAllObj4)
@@ -80,7 +80,7 @@ Description: Propagate variables to Makefile
.SUFFIXES:
.SUFFIXES: .o .c .cc .cpp .o5 .o8
-@@ -156,10 +159,10 @@
+@@ -156,10 +159,10 @@ fix8 = sed 's/.*\.o/& &5 &8/'
depend:
sed '/[m][v]/q' makefile > m
=====================================
debian/patches/simde
=====================================
@@ -27,7 +27,7 @@ equivalents automatically
typedef __m256i SimdInt;
typedef __m256i SimdUint1;
-@@ -146,279 +141,6 @@
+@@ -146,279 +141,6 @@ static inline SimdInt simdChoose1(SimdIn
return _mm256_shuffle_epi8(items, choices);
}
@@ -309,7 +309,7 @@ equivalents automatically
#endif
--- a/src/GappedXdropAligner.cc
+++ b/src/GappedXdropAligner.cc
-@@ -140,17 +140,13 @@
+@@ -140,17 +140,13 @@ int GappedXdropAligner::align(const ucha
if (isAffine) {
for (int i = 0; i < numCells; i += simdLen) {
SimdInt s = simdSet(
@@ -329,7 +329,7 @@ equivalents automatically
SimdInt y = simdSub(simdLoad(y1+i), mDelGrowCost);
--- a/src/GappedXdropAlignerPssm.cc
+++ b/src/GappedXdropAlignerPssm.cc
-@@ -91,17 +91,13 @@
+@@ -91,17 +91,13 @@ int GappedXdropAligner::alignPssm(const
if (isAffine) {
for (int i = 0; i < numCells; i += simdLen) {
SimdInt s = simdSet(
@@ -349,16 +349,12 @@ equivalents automatically
SimdInt y = simdSub(simdLoad(y1+i), mDelGrowCost);
--- a/makefile
+++ b/makefile
-@@ -1,7 +1,8 @@
--CXXFLAGS = -msse4 -O3 -std=c++11 -pthread -DHAS_CXX_THREADS
-+CXXFLAGS = -O3 -std=c++11 -pthread -DHAS_CXX_THREADS
+@@ -1,4 +1,4 @@
+-CXXFLAGS = -msse4 -O3 -std=c++11 -pthread
++CXXFLAGS = -O3 -std=c++11 -pthread
all:
@cd src && $(MAKE) CXXFLAGS="$(CXXFLAGS)"
-+SFX :=
- prefix = /usr/local
- exec_prefix = $(prefix)
- bindir = $(exec_prefix)/bin
--- a/src/makefile
+++ b/src/makefile
@@ -1,5 +1,4 @@
@@ -367,7 +363,7 @@ equivalents automatically
CXXFLAGS += -std=c++11
CXXFLAGS += -pthread -DHAS_CXX_THREAD
# -fomit-frame-pointer ?
-@@ -56,11 +55,12 @@
+@@ -56,11 +55,12 @@ split/last-split.o
PPOBJ = last-pair-probs.o last-pair-probs-main.o
MBOBJ = last-merge-batches.o
@@ -384,7 +380,7 @@ equivalents automatically
indexObj5 = $(indexObj4:.o=.o5)
alignObj5 = $(alignObj4:.o=.o5)
-@@ -74,45 +74,45 @@
+@@ -74,45 +74,45 @@ all: $(ALL)
last8: $(LAST8)
indexAllObj4 = $(indexObj0) $(indexObj4)
@@ -452,7 +448,7 @@ equivalents automatically
//#include <iostream> // for debugging
namespace cbrc {
-@@ -43,12 +41,10 @@
+@@ -43,12 +41,10 @@ int GappedXdropAligner::alignDna(const u
const SimdUint1 scorer4x4 =
simdSet1(
@@ -465,7 +461,7 @@ equivalents automatically
scorer[3][3], scorer[3][2], scorer[3][1], scorer[3][0],
scorer[2][3], scorer[2][2], scorer[2][1], scorer[2][0],
scorer[1][3], scorer[1][2], scorer[1][1], scorer[1][0],
-@@ -126,7 +122,6 @@
+@@ -126,7 +122,6 @@ int GappedXdropAligner::alignDna(const u
for (int i = 0; i < numCells; i += simdBytes) {
SimdUint1 s = simdSet1(
@@ -473,7 +469,7 @@ equivalents automatically
scorer[s1[31]][s2[31]],
scorer[s1[30]][s2[30]],
scorer[s1[29]][s2[29]],
-@@ -143,7 +138,6 @@
+@@ -143,7 +138,6 @@ int GappedXdropAligner::alignDna(const u
scorer[s1[18]][s2[18]],
scorer[s1[17]][s2[17]],
scorer[s1[16]][s2[16]],
@@ -481,7 +477,7 @@ equivalents automatically
scorer[s1[15]][s2[15]],
scorer[s1[14]][s2[14]],
scorer[s1[13]][s2[13]],
-@@ -275,5 +269,3 @@
+@@ -275,5 +269,3 @@ bool GappedXdropAligner::getNextChunkDna
}
}
@@ -489,7 +485,7 @@ equivalents automatically
-#endif
--- a/src/Alignment.cc
+++ b/src/Alignment.cc
-@@ -365,12 +365,10 @@
+@@ -365,12 +365,10 @@ void Alignment::extend( std::vector< Seg
del.openCost, del.growCost,
ins.openCost, ins.growCost,
gap.pairCost, gap.isAffine, maxDrop, smMax)
@@ -502,7 +498,7 @@ equivalents automatically
: aligner.align(seq1, seq2, isForward, globality, sm,
del.openCost, del.growCost,
ins.openCost, ins.growCost,
-@@ -387,14 +385,12 @@
+@@ -387,14 +385,12 @@ void Alignment::extend( std::vector< Seg
while( greedyAligner.getNextChunk( end1, end2, size ) )
chunks.push_back( SegmentPair( end1 - size, end2 - size, size ) );
}
@@ -519,7 +515,7 @@ equivalents automatically
del.openCost, del.growCost,
--- a/src/GappedXdropAligner.hh
+++ b/src/GappedXdropAligner.hh
-@@ -352,7 +352,6 @@
+@@ -352,7 +352,6 @@ class GappedXdropAligner {
void initFrame();
// Everything below here is for alignDna & getNextChunkDna
@@ -527,7 +523,7 @@ equivalents automatically
std::vector<TinyScore> xTinyScores;
std::vector<TinyScore> yTinyScores;
std::vector<TinyScore> zTinyScores;
-@@ -402,7 +401,6 @@
+@@ -402,7 +401,6 @@ class GappedXdropAligner {
while (*x2 != target) ++x2;
bestSeq1position = x2 - x2beg + seq1beg;
}
=====================================
debian/rules
=====================================
@@ -22,6 +22,10 @@ export DEB_LDFLAGS_MAINT_APPEND += -pthread
export DEB_BUILD_MAINT_OPTIONS = hardening=+all
+# Ensure the COMMAND-plain variants is used to generate the manpages
+# for reproducible builds
+export SIMD_LIST = fallback-to-plain-simd
+
BUILT_USING=$(shell dpkg-query -f '$${source:Package} (= $${source:Version}), ' -W "libsimde-dev")
%:
=====================================
debian/salsa-ci.yml
=====================================
@@ -5,4 +5,4 @@ include:
variables:
SALSA_CI_DISABLE_BLHC: 1
- SALSA_CI_DISABLE_REPROTEST: 1
+# SALSA_CI_DISABLE_REPROTEST: 1
=====================================
doc/last-dotplot.rst
=====================================
@@ -159,10 +159,10 @@ Options ``-a`` and ``-b`` can read annotations in these formats:
* AGP_, gap.txt: Unsequenced gaps are shown, but only if the gap
covers at least one whole pixel.
-You can use these options multiple times, e.g. ``-a stuff.bed -a
-more.bed -a rmsk.txt``. Annotations look good only if reasonably
-sparse, e.g. you can't sensibly view 20000 gene annotations in one
-small dotplot.
+You can use these options multiple times, e.g.
+``-a stuff.bed -a more.bed -a rmsk.txt``. Annotations look good only
+if reasonably sparse, e.g. you can't sensibly view 20000 gene
+annotations in one small dotplot.
Choosing sequences
------------------
=====================================
makefile
=====================================
@@ -1,4 +1,4 @@
-CXXFLAGS = -msse4 -O3 -std=c++11 -pthread -DHAS_CXX_THREADS
+CXXFLAGS = -msse4 -O3 -std=c++11 -pthread
all:
@cd src && $(MAKE) CXXFLAGS="$(CXXFLAGS)"
=====================================
src/makefile
=====================================
@@ -1,11 +1,11 @@
-CXXFLAGS = -msse4 -O3 -Wall -g -std=c++11 -pthread -DHAS_CXX_THREADS
+CXXFLAGS = -msse4 -O3 -Wall -g -std=c++11 -pthread
# -fomit-frame-pointer ?
# The number of symbol types that we can keep in sequences (e.g. 20
# amino acids, plus ambiguous ones, in upper & lower case, plus one
# delimiter):
ALPHABET_CAPACITY = 66
-CPPF = -DALPHABET_CAPACITY=$(ALPHABET_CAPACITY) $(CPPFLAGS)
+CPPF = -DALPHABET_CAPACITY=$(ALPHABET_CAPACITY) -DHAS_CXX_THREADS $(CPPFLAGS)
CFLAGS = -Wall -O2
@@ -143,7 +143,7 @@ ScoreMatrixData.hh: ../data/*.mat
../build/mat-inc.sh ../data/*.mat > $@
VERSION1 = git describe --dirty
-VERSION2 = echo ' (HEAD -> main, tag: 1256) ' | sed -e 's/.*tag: *//' -e 's/[,) ].*//'
+VERSION2 = echo ' (HEAD -> main, tag: 1257) ' | sed -e 's/.*tag: *//' -e 's/[,) ].*//'
VERSION = \"`test -e ../.git && $(VERSION1) || $(VERSION2)`\"
View it on GitLab: https://salsa.debian.org/med-team/last-align/-/compare/6ff84ca5c95964eec6989226c209afacdcc937c8...aedcfb14fe84a3032c230de48cb18b028a19b0bd
--
View it on GitLab: https://salsa.debian.org/med-team/last-align/-/compare/6ff84ca5c95964eec6989226c209afacdcc937c8...aedcfb14fe84a3032c230de48cb18b028a19b0bd
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/20220110/bb0cc8f7/attachment-0001.htm>
More information about the debian-med-commit
mailing list