Bug#569436: [Pkg-scicomp-devel] Bug#569436: lapack: zgesvd seems to give incorrect results
Le samedi 13 février 2010 à 08:57 -0600, Kumar Appaiah a écrit : On Sat, Feb 13, 2010 at 03:20:19PM +0100, Sylvestre Ledru wrote: severity 569436 normal thanks Hello, I installed a chroot (through debootstrap) and I don't have the problem. Did you try rebuilding libitpp on amd64? Did it FTBFS for you? I did not. I only checked lapack. With the intel code, I get: Left singular vectors (stored columnwise) ( -0.86, -0.00) ( 0.40, -0.00) ( 0.32, 0.00) ( -0.35, 0.13) ( -0.24, -0.21) ( -0.63, 0.60) ( 0.15, 0.32) ( 0.61, 0.61) ( -0.36, 0.10) The code is working fine with Octave or Scilab. Thanks for checking. Did you use amd64? Did you notice that the FTBFS is on amd64? No, I just tried with i386. I did not see any reference that it was a amd64-only bug. Sylvestre PS: I changed the severity to normal. FYI, grave means: makes the package in question unusable or mostly so, or causes data loss, or introduces a security hole allowing access to the accounts of users who use the package. Thanks for being pedantic without providing any alternate suggestions, You should relax. I didn't mean to be pedantic. I was just direct. And I am not providing suggestions, it is because I don't have any for now. but without SVD, I'd say most of the math related packages are unusable, since they give absurdly wrong results. But I won't play a severity ping-pong although at least my work and my package(s) are affected. And, I just created a fresh amd64 chroot, and Octave gives me this: You confirm that you only get this bug while you are in a chroot ? [...] Some help as to how to debug this package would be appreciated, since, even according to your definition above, several packages are unusable for me, and the bug is grave for me. I would really appreciate some further pointers in finding where the issue is. You should contact upstream mailing list here. Sylvestre -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#569436: [Pkg-scicomp-devel] Bug#569436: lapack: zgesvd seems to give incorrect results
On Sat, Feb 13, 2010 at 04:12:49PM +0100, Sylvestre Ledru wrote: Thanks for checking. Did you use amd64? Did you notice that the FTBFS is on amd64? No, I just tried with i386. I did not see any reference that it was a amd64-only bug. If you checked the reason why this bug was filed, Lucas says: During a rebuild of all packages in sid, your package failed to build on amd64. Sylvestre PS: I changed the severity to normal. FYI, grave means: makes the package in question unusable or mostly so, or causes data loss, or introduces a security hole allowing access to the accounts of users who use the package. Thanks for being pedantic without providing any alternate suggestions, You should relax. I didn't mean to be pedantic. I was just direct. And I am not providing suggestions, it is because I don't have any for now. Thanks. but without SVD, I'd say most of the math related packages are unusable, since they give absurdly wrong results. But I won't play a severity ping-pong although at least my work and my package(s) are affected. And, I just created a fresh amd64 chroot, and Octave gives me this: You confirm that you only get this bug while you are in a chroot ? Yes. My working environment is not sid, so I cannot make any comments on it. [...] Some help as to how to debug this package would be appreciated, since, even according to your definition above, several packages are unusable for me, and the bug is grave for me. I would really appreciate some further pointers in finding where the issue is. You should contact upstream mailing list here. Frankly, I think you could forward this bug to them on my behalf. But then, let me try doing so myself; at least I consider the bug serious enough to tackle now, since anyone who installs a fresh new sid installation on amd64 is likely to be bitten with incorrect SVD results unless he/she installs Atlas. Kumar -- If you shed tears when you miss the sun, you also miss the stars. - Rabindranath Tagore (Stray Birds, 1916) signature.asc Description: Digital signature
Bug#569436: [Pkg-scicomp-devel] Bug#569436: lapack: zgesvd seems to give incorrect results
Dear Sylvestre, I've zeroed down the bug to the last blas upload. Note that for the intel program, the expected Left singular vectors are: Left singular vectors (stored columnwise) ( -0.86, 0.00) ( 0.40, 0.00) ( 0.32, 0.00) ( -0.35, 0.13) ( -0.24, -0.21) ( -0.63, 0.60) ( 0.15, 0.32) ( 0.61, 0.61) ( -0.36, 0.10) Here is with the older Blas: [$ /tmp] sudo dpkg -i libblas3gf_1.2-2_amd64.deb (Reading database ... 22762 files and directories currently installed.) Preparing to replace libblas3gf 1.2-2 (using libblas3gf_1.2-2_amd64.deb) ... Unpacking replacement libblas3gf ... Setting up libblas3gf (1.2-2) ... [$ /tmp] ./a.out ZGESVD Example Program Results Singular values 17.63 11.61 6.78 Left singular vectors (stored columnwise) ( -0.86, -0.00) ( 0.40, -0.00) ( 0.32, 0.00) ( -0.35, 0.13) ( -0.24, -0.21) ( -0.63, 0.60) ( 0.15, 0.32) ( 0.61, 0.61) ( -0.36, 0.10) Right singular vectors (stored rowwise) ( -0.22, 0.51) ( -0.37, -0.32) ( -0.53, 0.11) ( 0.15, 0.38) ( 0.31, 0.31) ( 0.09, -0.57) ( 0.18, -0.39) ( 0.38, -0.39) ( 0.53, 0.24) ( 0.49, 0.28) ( -0.47, -0.25) ( -0.15, 0.19) With the newer Blas: [# /tmp] sudo dpkg -i libblas3gf_1.2-3_amd64.deb (Reading database ... 22762 files and directories currently installed.) Preparing to replace libblas3gf 1.2-2 (using libblas3gf_1.2-3_amd64.deb) ... Unpacking replacement libblas3gf ... Setting up libblas3gf (1.2-3) ... [# /tmp] ./a.out ZGESVD Example Program Results Singular values 17.63 11.61 6.78 Left singular vectors (stored columnwise) ( -0.86, -0.00) ( 0.40, -0.00) ( 0.32, 0.00) ( -0.35, -0.21) ( -0.24, -0.21) ( -0.63, 0.60) ( 0.15, 0.61) ( 0.61, 0.61) ( -0.36, 0.10) Right singular vectors (stored rowwise) ( -0.22, 0.51) ( -0.37, -0.32) ( -0.53, 0.11) ( 0.15, 0.38) ( 0.31, 0.31) ( 0.09, -0.57) ( 0.18, -0.39) ( 0.38, -0.39) ( 0.53, 0.24) ( 0.49, 0.28) ( -0.47, -0.25) ( -0.15, 0.19) Note that the second line of the Left Singular vectors doesn't match. I couldn't progress further on my probe, but hope this helps when you come back to this bug. Thanks! Kumar -- We do not stray out of all words into the ever silent; We do not raise our hands to the void for things beyond hope. - Rabindranath Tagore (The Gardener, 1915) signature.asc Description: Digital signature
