[Bug 206342] math/atlas: failed to build
bugzilla-noreply at freebsd.org
bugzilla-noreply at freebsd.org
Sun Jan 17 20:59:33 UTC 2016
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=206342
Bug ID: 206342
Summary: math/atlas: failed to build
Product: Ports & Packages
Version: Latest
Hardware: amd64
OS: Any
Status: New
Severity: Affects Only Me
Priority: ---
Component: Individual Port(s)
Assignee: freebsd-ports-bugs at FreeBSD.org
Reporter: arabesc at bk.ru
Created attachment 165723
--> https://bugs.freebsd.org/bugzilla/attachment.cgi?id=165723&action=edit
Build log
Build is failed on FreeBSD 10.2:
...
rm -f *core* *.o config?.out
mkdir src bin tune interfaces
mkdir: src: File exists
mkdir: bin: File exists
mkdir: tune: File exists
mkdir: interfaces: File exists
*** Error code 1
Stop.
make[2]: stopped in /usr/ports/math/atlas/work/ATLAS/shared
make -f Make.top startup
make -f Make.top leafstart leaf=interfaces/lapack/F77/src
cd interfaces/lapack/F77/src ; ln -s
/usr/ports/math/atlas/work/ATLAS/shared/Make.inc Make.inc
ln: Make.inc: File exists
*** Error code 1
Stop.
make[4]: stopped in /usr/ports/math/atlas/work/ATLAS/shared
*** Error code 1
Stop.
make[3]: stopped in /usr/ports/math/atlas/work/ATLAS/shared
*** Error code 1
Stop.
make[2]: stopped in /usr/ports/math/atlas/work/ATLAS/shared
ar: warning: csrot.o: not found in archive
ar: warning: zdrot.o: not found in archive
DONE configure
Building shared ATLAS:
/usr/bin/make -f Make.top build
cd bin/ ; /usr/bin/make xatlas_build
`xatlas_build' is up to date.
cd bin/ ; ./xatlas_build -1 0 -a 1
make[4]: don't know how to make
/usr/ports/math/atlas/work/ATLAS/shared/..//CONFIG/ARCHS/UNKNOWNx8664SSE3.tgz.
Stop
make[4]: stopped in /usr/ports/math/atlas/work/ATLAS/shared/bin
IN STAGE 1 INSTALL: SYSTEM PROBE/AUX COMPILE
cd /usr/ports/math/atlas/work/ATLAS/shared/src/auxil ; make -s ATL_buildinfo.o
`ATL_buildinfo.o' is up to date.
cd /usr/ports/math/atlas/work/ATLAS/shared/src/auxil ; make lib
cd /usr/ports/math/atlas/work/ATLAS/shared/src/testing ; make lib
make -j8 dlib.grd
`dlib.grd' is up to date.
make -j8 slib.grd
`slib.grd' is up to date.
make -j8 zlib.grd
`zlib.grd' is up to date.
make -j8 clib.grd
`clib.grd' is up to date.
make INSTALL_LOG/L1CacheSize
`INSTALL_LOG/L1CacheSize' is up to date.
make INSTALL_LOG/sMULADD pre=s
`INSTALL_LOG/sMULADD' is up to date.
make INSTALL_LOG/dMULADD pre=d
`INSTALL_LOG/dMULADD' is up to date.
Level 1 cache size calculated as 32KB
dFPU: Separate multiply and add instructions with 5 cycle pipeline.
Apparent number of registers : 17
Register-register performance=5657.18MFLOPS
sFPU: Separate multiply and add instructions with 5 cycle pipeline.
Apparent number of registers : 17
Register-register performance=5469.57MFLOPS
IN STAGE 2 INSTALL: TYPE-DEPENDENT TUNING
STAGE 2-1: TUNING PREC='d' (precision 1 of 4)
STAGE 2-1-1 : BUILDING BLOCK MATMUL TUNE
make -f Makefile INSTALL_LOG/dMMRES pre=d 2>&1 | ./xatlas_tee
INSTALL_LOG/dMMSEARCH.LOG
cp /usr/ports/math/atlas/work/ATLAS/shared/tune/blas/gemm/res/dMMRES
INSTALL_LOG/.
The best matmul kernel was ATL_dmm2x2x128_sse2.c, NB=40, written by R.
Clint Whaley
Performance: 9248.64MFLOPS (369.80 percent of of detected clock rate)
(Gen case got 6081.21MFLOPS)
cp /usr/ports/math/atlas/work/ATLAS/shared/tune/blas/gemm/res/dNCNB
INSTALL_LOG/.
cp /usr/ports/math/atlas/work/ATLAS/shared/tune/blas/gemm/res/dbestNN_40x40x40
INSTALL_LOG/.
make -f Makefile INSTALL_LOG/dNCNB pre=d 2>&1 | ./xatlas_tee
INSTALL_LOG/dMMSEARCH.LOGmake -f Makefile INSTALL_LOG/dbestNN_40x40x40 pre=d
nb=40 2>&1 | ./xatlas_tee INSTALL_LOG/dMMSEARCH.LOG NCgemmNN : muladd=1,
lat=5, pf=513, nb=40, mu=5, nu=1 ku=40,
ForceFetch=0, ifetch=6 nfetch=1
Performance = 6004.15 (64.92 of copy matmul, 240.07 of clock)
cp /usr/ports/math/atlas/work/ATLAS/shared/tune/blas/gemm/res/dbestNT_40x40x40
INSTALL_LOG/.
make -f Makefile INSTALL_LOG/dbestNT_40x40x40 pre=d nb=40 2>&1 | ./xatlas_tee
INSTALL_LOG/dMMSEARCH.LOG NCgemmNT : muladd=1, lat=5, pf=513, nb=40, mu=5,
nu=1 ku=1,
ForceFetch=0, ifetch=6 nfetch=1
Performance = 3875.97 (41.91 of copy matmul, 154.98 of clock)
cd /usr/ports/math/atlas/work/ATLAS/shared/tune/blas/gemm ; make
res/dbestTN_40x40x40 pre=d bnb=40
./xmmsearch -p d
Precision='d', FORCE=0, LAT=-1, nreg=-1, MaxL1=128
Read in L1 Cache size as = 32KB.
L1Size=4096, pre=d, Smallnb=35
Probing for a good latency value:
lat = 1, mf=4289.88
lat = 2, mf=4289.88
lat=1 selected!
L1Size=4096, pre=d, Smallnb=36
MATMUL FPU PROBE RESULTS: muladd=1, lat=5 (4306.80) selected over (4370.17)!!
Assertion failed: (fscanf(fp, "%lf", &mflop[i]) == 1), function mmcase0, file
/usr/ports/math/atlas/work/ATLAS/shared/..//tune/blas/gemm/mmsearch.c, line
515.
mmnreg = 23
NB's to try: 40
pre=d, muladd=0, lat=1, pf=0, nb=35, mu=5, nu=1, ku=35, mflop=4289.88
pre=d, muladd=0, lat=1, pf=0, nb=35, mu=5, nu=1, ku=35, mflop=4289.88
pre=d, muladd=0, lat=1, pf=0, nb=36, mu=4, nu=1, ku=36, mflop=4370.17
pre=d, muladd=1, lat=1, pf=0, nb=36, mu=4, nu=1, ku=36, mflop=4306.80
BEST USER CASE: NB=40, MFLOP=9248.64
BEST USER CASE: NB=40, MFLOP=9248.64
pre=d, loopO=JIK, ta=N tb=N, mb=40, nb=40, kb=40, lda=0, ldb=40, ldc=40
mu=5, nu=1, ku=40, muladd=1, lat=5 ====> mflop=5811.673450
pre=d, loopO=JIK, ta=N tb=N, mb=40, nb=40, kb=40, lda=0, ldb=40, ldc=40
mu=5, nu=1, ku=40, muladd=1, lat=5 ====> mflop=5811.673450
dNB = 40 (-1.00), No copy dNB = 40 (5811.67)
*** Signal 6
Stop.
make[5]: stopped in /usr/ports/math/atlas/work/ATLAS/shared/tune/blas/gemm
*** Error code 1
Stop.
make[4]: stopped in /usr/ports/math/atlas/work/ATLAS/shared/bin
file INSTALL_LOG/dbestTN_40x40x40 not found!!
Assertion failed: (fp), function GetInstLogFile, file
/usr/ports/math/atlas/work/ATLAS/shared/..//bin/atlas_install.c, line 337.
make -f Makefile INSTALL_LOG/dbestTN_40x40x40 pre=d nb=40 2>&1 | ./xatlas_tee
INSTALL_LOG/dMMSEARCH.LOG*** Signal 6
Stop.
make[3]: stopped in /usr/ports/math/atlas/work/ATLAS/shared
*** Error code 1
Stop.
make[2]: stopped in /usr/ports/math/atlas/work/ATLAS/shared
*** Error code 1
Stop.
make[1]: stopped in /usr/ports/math/atlas
*** Error code 1
Stop.
make: stopped in /usr/ports/math/atlas
--
You are receiving this mail because:
You are the assignee for the bug.
More information about the freebsd-ports-bugs
mailing list