Package: src:dbcsr
Version: 2.8.0-2
Severity: important
Tags: ftbfs trixie sid
Dear team:
This package used to FTBFS on single-cpu systems, so I did a team upload
and added this line to debian/rules:
export PRTE_MCA_rmaps_default_mapping_policy=:oversubscribe
which is completely standard among scientific packages dealing with OpenMPI.
Now the package finally builds ok on the AWS instances with a single-CPU
on which I tried, but paradoxically, now it fails with timeout on
AWS instances with 2 CPUs, as in the log below.
So, I request the help from other team members to find a solution which works
everywhere.
Note: Before the change, the package used to take 3 minutes and a half to build.
Now it takes 40 minutes to build. Should I just extend the timeout?
--------------------------------------------------------------------------------
[...]
debian/rules clean
dh clean --buildsystem=cmake
dh_auto_clean -O--buildsystem=cmake
dh_autoreconf_clean -O--buildsystem=cmake
dh_clean -O--buildsystem=cmake
debian/rules binary
dh binary --buildsystem=cmake
dh_update_autotools_config -O--buildsystem=cmake
dh_autoreconf -O--buildsystem=cmake
dh_auto_configure -O--buildsystem=cmake
cd obj-x86_64-linux-gnu && DEB_PYTHON_INSTALL_LAYOUT=deb
PKG_CONFIG=/usr/bin/pkg-config cmake -DCMAKE_INSTALL_PREFIX=/usr
-DCMAKE_BUILD_TYPE=None -DCMAKE_INSTALL_SYSCONFDIR=/etc
-DCMAKE_INSTALL_LOCALSTATEDIR=/var -DCMAKE_EXPORT_NO_PACKAGE_REGISTRY=ON
-DCMAKE_FIND_USE_PACKAGE_REGISTRY=OFF
-DCMAKE_FIND_PACKAGE_NO_PACKAGE_REGISTRY=ON
-DFETCHCONTENT_FULLY_DISCONNECTED=ON "-GUnix Makefiles"
-DCMAKE_VERBOSE_MAKEFILE=ON -DCMAKE_INSTALL_LIBDIR=lib/x86_64-linux-gnu ..
-- The C compiler identification is GNU 14.2.0
-- The CXX compiler identification is GNU 14.2.0
-- Detecting C compiler ABI info
[... snipped ...]
18:
-------------------------------------------------------------------------------
18: # multiplications 1
18: max memory usage/rank 278.441984E+06
18: # max total images/rank 2
18: # max 3D layers 1
18: # MPI messages exchanged 2
18: MPI messages size (bytes):
18: total size 1.808000E+03
18: min size 840.000000E+00
18: max size 968.000000E+00
18: average size 904.000000E+00
18: MPI breakdown and total messages size (bytes):
18: size <= 128 0 0
18: 128 < size <= 8192 2 1808
18: 8192 < size <= 32768 0 0
18: 32768 < size <= 131072 0 0
18: 131072 < size <= 4194304 0 0
18: 4194304 < size <= 16777216 0 0
18: 16777216 < size 0 0
18:
-------------------------------------------------------------------------------
18:
18: *** WARNING in dbcsr_mm.F:301 :: Using a non-square number of MPI ranks ***
18: *** might lead to poor performance. Used ranks: 2 Suggested: 1 4 ***
18:
18:
-------------------------------------------------------------------------------
18: -
-
18: - DBCSR MESSAGE PASSING PERFORMANCE
-
18: -
-
18:
-------------------------------------------------------------------------------
18: ROUTINE CALLS AVE VOLUME [Bytes]
18: MP_Bcast 19 12.
18: MP_Allreduce 59 26.
18: MP_Alltoall 2 48.
18: MP_ISend 6 315.
18: MP_IRecv 6 287.
18:
-------------------------------------------------------------------------------
18/19 Test #18: dbcsr_test ............................................
Passed 1.00 sec
test 19
Start 19: dbcsr_tensor_test
19: Test command: /usr/bin/mpiexec "-n" "2" "./dbcsr_tensor_test"
19: Working Directory: /<<PKGBUILDDIR>>/obj-x86_64-linux-gnu/tests
19: Environment variables:
19: OMP_NUM_THREADS=2
19: Test timeout computed to be: 1500
19:
[c7a-large-1742981929][[11101,1],0][../../../../../../opal/mca/btl/tcp/btl_tcp_proc.c:266:mca_btl_tcp_proc_create_interface_graph]
Unable to find reachable pairing between local and remote interfaces
19:
[c7a-large-1742981929][[11101,1],1][../../../../../../opal/mca/btl/tcp/btl_tcp_proc.c:266:mca_btl_tcp_proc_create_interface_graph]
Unable to find reachable pairing between local and remote interfaces
19: pgrid3-dimensions:
19: 2 1 1
19:
19: pgrid4-dimensions:
19: 2 1 1 1
19:
19: dist11:
19: 0 1 0 1
19:
19: dist12:
19: 0 0 0 0 0 0 0 0 0 0 0
19:
19: dist13:
19: 0 0 0 0 0 0 0 0 0
19:
19: dist21:
19: 0 1 0 1
19:
19: dist22:
19: 0 0 0 0 0 0 0 0 0 0 0
19:
19: dist23:
19: 0 0 0 0 0
19:
19: dist24:
19: 0 0 0
19:
19: dist31:
19: 0 1 0 1 0 1 0 1 0
19:
19: dist32:
19: 0 0 0 0 0
19:
19: dist33:
19: 0 0 0
19:
19: Creating dist objects...
19:
19: Creating tensors...
19: Tensor 1
19:
19: Filling Tensor...
19: Dimension: 3
19: Reserving blocks...
19: Tensor 2
19:
19: Filling Tensor...
19: Dimension: 4
19: Reserving blocks...
19: Tensor 3
19:
19: Filling Tensor...
19: Dimension: 3
19: Reserving blocks...
19: Contracting...
19:
--------------------------------------------------------------------------------
19: DBCSR TENSOR CONTRACTION: (13|2) x (54|21) = (3|45)
19:
--------------------------------------------------------------------------------
19: GLOBAL INFO OF (13|2)
19: block dimensions: 4 11 9
19: full dimensions: 25 83 74
19: process grid dimensions: 2 1 1
19: Block sizes:
19: Dim 1: 3 9 12 1
19: Dim 2: 4 2 3 1 9 2 32 10 5 8 7
19: Dim 3: 7 3 8 7 9 5 10 23 2
19: Block distribution:
19: Dim 1: 0 1 0 1
19: Dim 2: 0 0 0 0 0 0 0 0 0 0 0
19: Dim 3: 0 0 0 0 0 0 0 0 0
19: DISTRIBUTION OF (13|2)
19: Number of non-zero blocks:
32
19: Percentage of non-zero blocks:
8.08
19: Average number of blocks per CPU:
16
19: Maximum number of blocks per CPU:
20
19: Average number of matrix elements per CPU:
8155
19: Maximum number of matrix elements per CPU:
9756
19: GLOBAL INFO OF (54|21)
19: block dimensions: 4 11 5 3
19: full dimensions: 25 83 32 28
19: process grid dimensions: 2 1 1 1
19: Block sizes:
19: Dim 1: 3 9 12 1
19: Dim 2: 4 2 3 1 9 2 32 10 5 8 7
19: Dim 3: 8 1 4 13 6
19: Dim 4: 4 2 22
19: Block distribution:
19: Dim 1: 0 1 0 1
19: Dim 2: 0 0 0 0 0 0 0 0 0 0 0
19: Dim 3: 0 0 0 0 0
19: Dim 4: 0 0 0
19: DISTRIBUTION OF (54|21)
19: Number of non-zero blocks:
36
19: Percentage of non-zero blocks:
5.45
19: Average number of blocks per CPU:
18
19: Maximum number of blocks per CPU:
21
19: Average number of matrix elements per CPU:
74745
19: Maximum number of matrix elements per CPU:
132456
19: INDEX INFO
19: tensor index: (bca) x (bcde) = (ade)
19: matrix index: (ba|c) x (ed|cb) = (a|de)
19: aligning tensor index with data
19: INDEX INFO
19: tensor index: (bac) x (edcb) = (ade)
19: matrix index: (ba|c) x (ed|cb) = (a|de)
19: large tensors: 1, 2; small tensor: 3
19: sorting contraction indices
19: compatibility of (13|2): Not compatible
19: compatibility of (54|21): Transposed
19: No redistribution of (54|21)
19: Redistribution of (13|2) compatible with (54|21)
19: compatibility of (13|2): Transposed
19: compatibility of (3|45): Not compatible
19: Redistribution of (3|45)
19: compatibility of (3|45): Normal
19: INDEX INFO
19: tensor index: (bac) x (edcb) = (ade)
19: matrix index: (a|cb) x (ed|cb) = (a|ed)
19: GLOBAL INFO OF (13|2)
19: block dimensions: 4 9 11
19: full dimensions: 25 74 83
19: process grid dimensions: 2 1 1
19: Block sizes:
19: Dim 1: 3 9 12 1
19: Dim 2: 7 3 8 7 9 5 10 23 2
19: Dim 3: 4 2 3 1 9 2 32 10 5 8 7
19: Block distribution:
19: Dim 1: 0 1 0 1
19: Dim 2: 0 0 0 0 0 0 0 0 0
19: Dim 3: 0 0 0 0 0 0 0 0 0 0 0
19: DISTRIBUTION OF (13|2)
19: Number of non-zero blocks:
32
19: Percentage of non-zero blocks:
8.08
19: Average number of blocks per CPU:
16
19: Maximum number of blocks per CPU:
20
19: Average number of matrix elements per CPU:
8155
19: Maximum number of matrix elements per CPU:
9756
19:
--------------------------------------------------------------------------------
19: DBCSR TAS MATRIX MULTIPLICATION: (13|2) matrix x (54|21) matrix = (3|45)
matrix
19:
--------------------------------------------------------------------------------
19: mm dims: 9 44 15
19: MM PARAMETERS
19: Est. number of matrix elements per CPU of result matrix:
5269
19: Est. optimal split factor:
2
19: No redistribution of (13|2) matrix and (54|21) matrix
19: Change split factor of (13|2) matrix : No
19: Change split factor of (54|21) matrix : No
19: mm case: -- x --T = +
19: SPLIT / PARALLELIZATION INFO
19: splitting columns by factor 2
19: global grid sizes: 1x 2
19: grid sizes on subgroups: 1x 1
19: GLOBAL INFO OF (13|2) matrix
19: block dimensions: 9 44
19: full dimensions: 74 2075
19: process grid dimensions: 1 2
19: Block sizes:
19: Row: 7 3 8 7 9 5 10 23 2
19: Column: 12 6 9 3 27 6 96 30 15 24 21 36
18 27 9 81 18 288 90 45 72 63 48 24 36 12 108 24
384 120 60 96 84 4 2 3 1 9 2 32 10 5 8 7
19: Block distribution:
19: Row: 0 0 0 0 0 0 0 0 0
19: Column: 0 0 0 0 0 0 0 0 0 0 0 1
1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0
0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1
19: GLOBAL INFO OF (54|21) matrix
19: block dimensions: 15 44
19: full dimensions: 896 2075
19: process grid dimensions: 1 2
19: Block sizes:
19: Row: 32 16 176 4 2 22 16 8 88 52 26 286 24
12 132
19: Column: 12 6 9 3 27 6 96 30 15 24 21 36
18 27 9 81 18 288 90 45 72 63 48 24 36 12 108 24
384 120 60 96 84 4 2 3 1 9 2 32 10 5 8 7
19: Block distribution:
19: Row: 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0
19: Column: 0 0 0 0 0 0 0 0 0 0 0 1
1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0
0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1
19: GLOBAL INFO OF (3|45) matrix
19: block dimensions: 9 15
19: full dimensions: 74 896
19: process grid dimensions: 1 2
19: Block sizes:
19: Row: 7 3 8 7 9 5 10 23 2
19: Column: 32 16 176 4 2 22 16 8 88 52 26 286
24 12 132
19: Block distribution:
19: Row: 0 0 0 0 0 0 0 0 0
19: Column: 0 1 0 0 1 1 0 1 1 0 1 1
0 0 0
19: Change process grid: No
19: DISTRIBUTION OF (13|2) matrix
19: Number of non-zero blocks:
32
19: Percentage of non-zero blocks:
8.08
19: Average number of blocks per group:
16
19: Maximum number of blocks per group:
20
19: Average number of matrix elements per group:
8155
19: Maximum number of matrix elements per group:
9756
19: Average number of blocks per CPU:
16
19: Maximum number of blocks per CPU:
20
19: Average number of matrix elements per CPU:
8155
19: Maximum number of matrix elements per CPU:
9756
19: DISTRIBUTION OF (54|21) matrix
19: Number of non-zero blocks:
36
19: Percentage of non-zero blocks:
5.45
19: Average number of blocks per group:
18
19: Maximum number of blocks per group:
21
19: Average number of matrix elements per group:
74745
19: Maximum number of matrix elements per group:
132456
19: Average number of blocks per CPU:
18
19: Maximum number of blocks per CPU:
21
19: Average number of matrix elements per CPU:
74745
19: Maximum number of matrix elements per CPU:
132456
19: DISTRIBUTION OF (3|45) matrix replicated
19: Number of non-zero blocks:
23
19: Percentage of non-zero blocks:
8.52
19: Average number of blocks per group:
12
19: Maximum number of blocks per group:
13
19: Average number of matrix elements per group:
5351
19: Maximum number of matrix elements per group:
7306
19: Average number of blocks per CPU:
12
19: Maximum number of blocks per CPU:
13
19: Average number of matrix elements per CPU:
5351
19: Maximum number of matrix elements per CPU:
7306
19: Row distribution on subgroup:
19: 0 0 0 0 0 0 0 0 0
19: Column distribution on subgroup:
19: 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
19: MM PARAMETERS
19: Number of matrix elements per CPU of result matrix:
3653
19: Optimal split factor:
2
19:
--------------------------------------------------------------------------------
19: TAS MATRIX MULTIPLICATION DONE
19:
--------------------------------------------------------------------------------
19: GLOBAL INFO OF (3|45)
19: block dimensions: 9 5 3
19: full dimensions: 74 32 28
19: process grid dimensions: 2 1 1
19: Block sizes:
19: Dim 1: 7 3 8 7 9 5 10 23 2
19: Dim 2: 8 1 4 13 6
19: Dim 3: 4 2 22
19: Block distribution:
19: Dim 1: 0 0 0 1 0 1 0 1 1
19: Dim 2: 0 0 0 0 0
19: Dim 3: 0 0 0
19: DISTRIBUTION OF (3|45)
19: Number of non-zero blocks:
21
19: Percentage of non-zero blocks:
15.56
19: Average number of blocks per CPU:
11
19: Maximum number of blocks per CPU:
13
19: Average number of matrix elements per CPU:
5269
19: Maximum number of matrix elements per CPU:
9052
19: GLOBAL INFO OF (3|45)
19: block dimensions: 9 5 3
19: full dimensions: 74 32 28
19: process grid dimensions: 2 1 1
19: Block sizes:
19: Dim 1: 7 3 8 7 9 5 10 23 2
19: Dim 2: 8 1 4 13 6
19: Dim 3: 4 2 22
19: Block distribution:
19: Dim 1: 0 1 0 1 0 1 0 1 0
19: Dim 2: 0 0 0 0 0
19: Dim 3: 0 0 0
19: DISTRIBUTION OF (3|45)
19: Number of non-zero blocks:
29
19: Percentage of non-zero blocks:
21.48
19: Average number of blocks per CPU:
15
19: Maximum number of blocks per CPU:
15
19: Average number of matrix elements per CPU:
6431
19: Maximum number of matrix elements per CPU:
9322
19:
--------------------------------------------------------------------------------
19: TENSOR CONTRACTION DONE
19:
--------------------------------------------------------------------------------
19: Testing get_info for Tensor 1...
19: Name: (13|2)
19: Data_type: 3
19: ======= Process: 0 ========
19: Total number of blocks:
19: 4 11 9
19:
19: Total number of elements:
19: 25 83 74
19:
19: Total number of local blocks:
19: 2 11 9
19:
19: Total number of local elements:
19: 15 83 74
19:
19: Pgrid dimensions:
19: 2 1 1
19:
19: Process coordinates:
19: 0 0 0
19:
19: blks_local:
19: 0 2
19:
19: 0 1 2 3 4 5 6 7 8 9 10
19:
19: 0 1 2 3 4 5 6 7 8
19:
19: proc_dist:
19: 0 1 0 1
19:
19: 0 0 0 0 0 0 0 0 0 0 0
19:
19: 0 0 0 0 0 0 0 0 0
19:
19: blk_size:
19: 3 9 12 1
19:
19: 4 2 3 1 9 2 32 10 5 8 7
19:
19: 7 3 8 7 9 5 10 23 2
19:
19: blk_offset:
19: 0 3 12 24
19:
19: 0 4 6 9 10 19 21 53 63 68 76
19:
19: 0 7 10 18 25 34 39 49 72
19:
19: ======= Process: 1 ========
19: Total number of blocks:
19: 4 11 9
19:
19: Total number of elements:
19: 25 83 74
19:
19: Total number of local blocks:
19: 2 11 9
19:
19: Total number of local elements:
19: 10 83 74
19:
19: Pgrid dimensions:
19: 2 1 1
19:
19: Process coordinates:
19: 1 0 0
19:
19: blks_local:
19: 1 3
19:
19: 0 1 2 3 4 5 6 7 8 9 10
19:
19: 0 1 2 3 4 5 6 7 8
19:
19: proc_dist:
19: 0 1 0 1
19:
19: 0 0 0 0 0 0 0 0 0 0 0
19:
19: 0 0 0 0 0 0 0 0 0
19:
19: blk_size:
19: 3 9 12 1
19:
19: 4 2 3 1 9 2 32 10 5 8 7
19:
19: 7 3 8 7 9 5 10 23 2
19:
19: blk_offset:
19: 0 3 12 24
19:
19: 0 4 6 9 10 19 21 53 63 68 76
19:
19: 0 7 10 18 25 34 39 49 72
19:
19: Testing get_mapping_info for Tensor 1...
19: ndim_nd = 3
19: ndim1_2d = 2
19: ndim2_2d = 1
19: dims_2d_i8: 36 11
19:
19: dims_2d: 36 11
19:
19: dims_nd:
19: 3 10 8
19:
19: dims1_2d:
19: 3 8
19:
19: dims2_2d:
19: 10
19:
19: map1_2d:
19: 0 2
19:
19: map2_2d:
19: 1
19:
19: map_nd:
19: 0 2 1
19:
19: Base: 1
19: col_major 1
19:
19: Testing c_dbcsr_t_contract_index...
19:
19: Result Indices on Rank 0
19: 0 3 1
19: 1 0 0
19: 1 3 2
19: 2 1 0
19: 2 3 1
19: 4 1 0
19: 4 2 1
19: 4 3 1
19: 4 3 2
19: 6 0 0
19: 6 3 1
19: 6 3 2
19: 6 4 2
19: Result Indices on Rank 1
19: 3 2 1
19: 5 0 0
19: 5 2 0
19: 5 2 1
19: 7 1 0
19: 7 1 2
19: 7 2 0
19: 7 2 1
19/19 Test #19: dbcsr_tensor_test .....................................
Passed 2.55 sec
95% tests passed, 1 tests failed out of 19
Total Test time (real) = 2334.56 sec
The following tests FAILED:
11 - dbcsr_unittest1 (Timeout)
Errors while running CTest
make[2]: *** [Makefile:74: test] Error 8
make[2]: Leaving directory '/<<PKGBUILDDIR>>/obj-x86_64-linux-gnu'
dh_auto_test: error: cd obj-x86_64-linux-gnu && make -j1 test ARGS\+=--verbose
ARGS\+=-j1 returned exit code 2
make[1]: *** [debian/rules:21: override_dh_auto_test] Error 25
make[1]: Leaving directory '/<<PKGBUILDDIR>>'
make: *** [debian/rules:12: binary] Error 2
dpkg-buildpackage: error: debian/rules binary subprocess returned exit status 2
--------------------------------------------------------------------------------
The above is just how the build ends and not necessarily the most relevant part.
If required, the full build log is available here:
https://people.debian.org/~sanvila/build-logs/202503/
About the archive rebuild: The build was made on virtual machines from AWS,
using sbuild and a reduced chroot with only build-essential packages.
If you could not reproduce the bug please contact me privately, as I
am willing to provide ssh access to a virtual machine where the bug is
fully reproducible.
If this is really a bug in one of the build-depends, please use
reassign and add an affects on src:dbcsr, so that this is still
visible in the BTS web page for this package.
Thanks.
