search

sebhtml / ray

Ray -- Parallel genome assemblies for parallel DNA sequencing
http://denovoassembler.sf.net
Other
65 stars 12 forks source link

gcc 4.7.0 + open-mpi 1.6.2 breaks Ray #89

Closed sebhtml closed 12 years ago

sebhtml commented 12 years ago

gcc 4.7.0 + open-mpi 1.6.2 breaks Ray

the resulting coverage distribution is not coherent.

sebhtml commented 12 years ago

gcc 4.7.0 + open-mpi 1.5.4 cause the same

sebhtml commented 12 years ago

Compiler: gcc 4.7.2 MPI: Open-MPI 1.6.2 Status: FAIL

Compiler: gcc (GCC) 4.6.3 20120306 (Red Hat 4.6.3-2) MPI: mpiexec (OpenRTE) 1.5.4 Status: PASS

sebhtml commented 12 years ago

Compiler: gcc 4.7.2 MPI: OMPI 1.5.4 Status: FAIL

sebhtml commented 12 years ago

correct gcc 4.6.3

[DEBUG_GCC_4_7_2] kmer is GGGGTTGCAGTTGGTTCAGGTGGTCCAGGTGCGACTCACTTGATTAACGG [DEBUG_GCC_4_7_2] kmer is CATAATCGAAAATCCGCACCAAAGACTTATCCTTGTAGTTTCTATGAATC [DEBUG_GCC_4_7_2] kmer is GGGATAGGAACATCTTAGAACCATTTAGATGATTTTTAAAGGTTACTCCT [DEBUG_GCC_4_7_2] kmer is TGATGCAAGAAAAAATCAACATGAAGGTGGATACCTCTGAAATAGAAAAA [DEBUG_GCC_4_7_2] kmer is AGGAATTTAGAAATTTAAATTCTAAGGATGCAAGGGAATATGACTTGGCT [DEBUG_GCC_4_7_2] kmer is TGCTGACCCTAGATCAACAATTATATGTTCTGCTACATTAAGTTTTATTG [DEBUG_GCC_4_7_2] kmer is GCAACACTTCCATTTTCATCTCCCCAAGGGAAGACTGTCCATGTATCCAA

bad gcc 4.7.2

[DEBUG_GCC_4_7_2] kmer is AAAAAAAAAAAAAGTTCAGGTGGTCCAGGTGCGACTCACTTGATTAACGG [DEBUG_GCC_4_7_2] kmer is AAAAAAAAAAAAACCGCACCAAAGACTTATCCTTGTAGTTTCTATGAATC [DEBUG_GCC_4_7_2] kmer is AAAAAAAAAAAAACTTAGAACCATTTAGATGATTTTTAAAGGTTACTCCT [DEBUG_GCC_4_7_2] kmer is AAAAAAAAAAAAAAATCAACATGAAGGTGGATACCTCTGAAATAGAAAAA [DEBUG_GCC_4_7_2] kmer is AAAAAAAAAAAAATTTAAATTCTAAGGATGCAAGGGAATATGACTTGGCT [DEBUG_GCC_4_7_2] kmer is AAAAAAAAAAAAATCAACAATTATATGTTCTGCTACATTAAGTTTTATTG [DEBUG_GCC_4_7_2] kmer is AAAAAAAAAAAAATTTCATCTCCCCAAGGGAAGACTGTCCATGTATCCAA [DEBUG_GCC_4_7_2] kmer is AAAAAAAAAAAAACTGGTTTCATATGACGATGAATGATAATGGTGTCGTA

correct gcc 4.6.3

[boiseb01@ls30 RayKmerSearchDevel]$ grep GGGGTTGCAGTTGGTTCAGGTGGTCCAGGTGCGACTCACTTGATTAACGG 200xStreptococcus-pneumoniae-R6.fastafragments*.fasta 200xStreptococcus-pneumoniae-R6.fasta_fragments_1.fasta:GGGGTTGCAGTTGGTTCAGGTGGTCCAGGTGCGACTCACTTGATTAACGG

bad gcc 4.7.2

[boiseb01@ls30 RayKmerSearchDevel]$ grep TTCAGGTGGTCCAGGTGCGACTCACTTGATTAACGG$ 200xStreptococcus-pneumoniae-R6.fastafragments*.fasta 200xStreptococcus-pneumoniae-R6.fasta_fragments_1.fasta:GGGGTTGCAGTTGGTTCAGGTGGTCCAGGTGCGACTCACTTGATTAACGG

gcc 4.7.2 -O0

[Read::constructor] sequence is GGGGTTGCAGTTGGTTCAGGTGGTCCAGGTGCGACTCACTTGATTAACGG

[DEBUG_GCC_4_7_2] read is AAAAAAAAAAAAAGTTCAGGTGGTCCAGGTGCGACTCACTTGATTAACGG [DEBUG_GCC_4_7_2] read is AAAAAAAAAAAAACCGCACCAAAGACTTATCCTTGTAGTTTCTATGAATC [DEBUG_GCC_4_7_2] read is AAAAAAAAAAAAACTTAGAACCATTTAGATGATTTTTAAAGGTTACTCCT [DEBUG_GCC_4_7_2] read is AAAAAAAAAAAAAAATCAACATGAAGGTGGATACCTCTGAAATAGAAAAA [DEBUG_GCC_4_7_2] read is AAAAAAAAAAAAATTTAAATTCTAAGGATGCAAGGGAATATGACTTGGCT [DEBUG_GCC_4_7_2] read is AAAAAAAAAAAAATCAACAATTATATGTTCTGCTACATTAAGTTTTATTG [DEBUG_GCC_4_7_2] read is AAAAAAAAAAAAATTTCATCTCCCCAAGGGAAGACTGTCCATGTATCCAA [DEBUG_GCC_4_7_2] read is AAAAAAAAAAAAACTGGTTTCATATGACGATGAATGATAATGGTGTCGTA

full stack for 0 with gcc 4.7.2

[Read::constructor] sequence is GGGGTTGCAGTTGGTTCAGGTGGTCCAGGTGCGACTCACTTGATTAACGG

| 0 0 0 0| 0 0 0 0| 0 0 0 0| 0 2 3 3| 1 0 2 2| 3 2 2 3| 1 1 0 2| 2 3 2 1| 2 0 1 3| 1 0 1 3| 3 2 0 3| 3 0 0 1| 2 2 0 0 0 248 161 235 133 110 210 209 203 67 10

with gcc 4.6.3:

[Read::constructor] sequence is GGGGTTGCAGTTGGTTCAGGTGGTCCAGGTGCGACTCACTTGATTAACGG

| 2 2 2 2| 3 3 2 1| 0 2 3 3| 2 2 3 3| 1 0 2 2| 3 2 2 3| 1 1 0 2| 2 3 2 1| 2 0 1 3| 1 0 1 3| 3 2 0 3| 3 0 0 1| 2 2 170 111 248 250 161 235 133 110 210 209 203 67 10

with 4.7.2:

[Read::constructor] sequence is GGGGTTGCAGTTGGTTCAGGTGGTCCAGGTGCGACTCACTTGATTAACGG

[DEBUG_GCC_4_7_2] after trim GGGGTTGCAGTTGGTTCAGGTGGTCCAGGTGCGACTCACTTGATTAACGG | 0(A) 0(A) 0(A) 0(A)| 0(A) 0(A) 0(A) 0(A)| 0(A) 0(A) 0(A) 0(A)| 0(A) 2(G) 3(T) 3(T)| 1(C) 0(A) 2(G) 2(G)| 3(T) 2(G) 2(G) 3(T)| 1(C) 1(C) 0(A) 2(G)| 2(G) 3(T) 2(G) 1(C)| 2(G) 0(A) 1(C) 3(T)| 1(C) 0(A) 1(C) 3(T)| 3(T) 2(G) 0(A) 3(T)| 3( ) 0 0 0 248 161 235 133 110 210 209 203 67 10

[Read::constructor] sequence is GGGGTTGCAGTTGGTTCAGGTGGTCCAGGTGCGACTCACTTGATTAACGG

[DEBUG_GCC_4_7_2] after trim GGGGTTGCAGTTGGTTCAGGTGGTCCAGGTGCGACTCACTTGATTAACGG [DEBUG_GCC_4_7_2] nucleotide ^@ is not in {A,T,C,G}, position 0 in , length is 50 | 0(A)[DEBUG_GCC_4_7_2] nucleotide ^@ is not in {A,T,C,G}, position 1 in , length is 50 0(A)[DEBUG_GCC_4_7_2] nucleotide ^@ is not in {A,T,C,G}, position 2 in , length is 50 0(A)[DEBUG_GCC_4_7_2] nucleotide ^@ is not in {A,T,C,G}, position 3 in , length is 50 0(A)[DEBUG_GCC_4_7_2] nucleotide ^@ is not in {A,T,C,G}, position 4 in , length is 50 | 0(A)[DEBUG_GCC_4_7_2] nucleotide ^@ is not in {A,T,C,G}, position 5 in , length is 50 0(A)[DEBUG_GCC_4_7_2] nucleotide ^@ is not in {A,T,C,G}, position 6 in , length is 50 0(A)[DEBUG_GCC_4_7_2] nucleotide ^@ is not in {A,T,C,G}, position 7 in , length is 50 0(A)[DEBUG_GCC_4_7_2] nucleotide ^@ is not in {A,T,C,G}, position 8 in , length is 50 | 0(A)[DEBUG_GCC_4_7_2] nucleotide ^@ is not in {A,T,C,G}, position 9 in , length is 50 0(A)[DEBUG_GCC_4_7_2] nucleotide ^@ is not in {A,T,C,G}, position 10 in , length is 50 0(A)[DEBUG_GCC_4_7_2] nucleotide ^@ is not in {A,T,C,G}, position 11 in , length is 50 0(A)[DEBUG_GCC_4_7_2] nucleotide ^@ is not in {A,T,C,G}, position 12 in , length is 50 | 0(A) 2(G) 3(T) 3(T)| 1(C) 0(A) 2(G) 2(G)| 3(T) 2(G) 2(G) 3(T)| 1(C) 1(C) 0(A) 2(G)| 2(G) 3(T) 2(G) 1(C)| 2(G) 0(A) 1(C) 3(T)| 1(C) 0(A) 1(C) 3(T)| 3(T) 2(G) 0(A) 3(T)| 3(T) 0(A) 0(A) 1(C)| 2(G) 2(G) 0 0 0 248 161 235 133 110 210 209 203 67 10

sebhtml commented 12 years ago

Patch fcd8150b7837daaac61bdad8e528ac9e46be3c69 solves the problem.

Ray now works properly with gcc 4.7.2 and Open-MPI 1.6.2.