[opensuse] benchmarking my computer?
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi, I would like to benchmark my computer before applying patches for the recent Intel fiasco. Suggestions? - -- Cheers Carlos E. R. (from 42.2 x86_64 "Malachite" at Telcontar) -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iEYEARECAAYFAlpQy00ACgkQtTMYHG2NR9W1/gCfZcen2dr52SXSmx9vB33koYnB LokAnRGU9+cgQE8DoXCRFoOglfZXSe5X =Yssr -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 06/01/2018 15:12, Carlos E. R. wrote:
Hi,
I would like to benchmark my computer before applying patches for the recent Intel fiasco.
Suggestions? I can't recommend a benchmark but the patch description gives two kernel command line options for turning off: - CVE-2017-5715 / "SpectreAttack": Local attackers on systems with modern CPUs featuring branch prediction could use mispredicted branches to speculatively execute code patterns that in turn could be made to leak other non-readable content in the same address space, an attack similar to CVE-2017-5753.
This problem is mitigated by disabling predictive branches, depending on CPU architecture either by firmware updates and/or fixes in the user-kernel privilege boundaries. Please also check with your CPU / Hardware vendor on updated firmware or BIOS images regarding this issue. As this feature can have a performance impact, it can be disabled using the "nospec" kernel commandline option. - CVE-2017-5754 / "MeltdownAttack": Local attackers on systems with modern CPUs featuring deep instruction pipelining could use code patterns in userspace to speculative executive code that would read otherwise read protected memory, an attack similar to CVE-2017-5753. This problem is mitigated by unmapping the Linux Kernel from the user address space during user code execution, following a approach called "KAISER". The terms used here are "KAISER" / "Kernel Address Isolation" and "PTI" / "Page Table Isolation". Note that this is only done on affected platforms. This feature can be enabled / disabled by the "pti=[on|off|auto]" or "nopti" commandline options. Dave P -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
Carlos E. R. composed on 2018-01-06 14:12 (UTC+0100):
I would like to benchmark my computer before applying patches for the recent Intel fiasco.
Suggestions?
http://www.stresslinux.org/sl/ -- "Wisdom is supreme; therefore get wisdom. Whatever else you get, get wisdom." Proverbs 4:7 (New Living Translation) Team OS/2 ** Reg. Linux User #211409 ** a11y rocks! Felix Miata *** http://fm.no-ip.com/ -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
Am 06.01.2018 um 15:45 schrieb Felix Miata:
Carlos E. R. composed on 2018-01-06 14:12 (UTC+0100):
I would like to benchmark my computer before applying patches for the recent Intel fiasco.
Suggestions?
If I see it correct, this comes with it's own kernel to boot, so it could give information on the bare computers capacities but not on how it runs with a given configuration/kernel? Interesting would be something to run on the current running system as is, and, for example, before/after updating the kernel or using the "switch on/off commands" of Dave Platter's post. This would then give hints about the performance differences. Or did I get it wrong? -- Daniel Bauer photographer Basel Barcelona https://www.patreon.com/danielbauer http://www.daniel-bauer.com -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On Sat, 6 Jan 2018 16:08:37 +0100 Daniel Bauer wrote:
Am 06.01.2018 um 15:45 schrieb Felix Miata:
Carlos E. R. composed on 2018-01-06 14:12 (UTC+0100):
I would like to benchmark my computer before applying patches for the recent Intel fiasco.
Suggestions?
If I see it correct, this comes with it's own kernel to boot, so it could give information on the bare computers capacities but not on how it runs with a given configuration/kernel?
Interesting would be something to run on the current running system as is, and, for example, before/after updating the kernel or using the "switch on/off commands" of Dave Platter's post. This would then give hints about the performance differences.
Or did I get it wrong?
https://software.opensuse.org/package/hardinfo e.g., after install: # hardinfo -r -f html >hardinfo.html -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 06/01/2018 17:08, Daniel Bauer wrote:
Am 06.01.2018 um 15:45 schrieb Felix Miata:
Carlos E. R. composed on 2018-01-06 14:12 (UTC+0100):
I would like to benchmark my computer before applying patches for the recent Intel fiasco.
Suggestions?
If I see it correct, this comes with it's own kernel to boot, so it could give information on the bare computers capacities but not on how it runs with a given configuration/kernel?
Interesting would be something to run on the current running system as is, and, for example, before/after updating the kernel or using the "switch on/off commands" of Dave Platter's post. This would then give hints about the performance differences.
Or did I get it wrong?
I had a look at the link and found that one of the programs mentioned stress is available as stress-ng but all it really does is stress the cpu so I'm trying two tests myself, an rpmbuild operation and a large file transfer between two physical disks using "time" then recording the results. I'm now going to boot into the updated kernel. Dave P -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 06/01/2018 17:37, Dave Plater wrote:
On 06/01/2018 17:08, Daniel Bauer wrote:
Am 06.01.2018 um 15:45 schrieb Felix Miata:
Carlos E. R. composed on 2018-01-06 14:12 (UTC+0100):
I would like to benchmark my computer before applying patches for the recent Intel fiasco.
Suggestions?
If I see it correct, this comes with it's own kernel to boot, so it could give information on the bare computers capacities but not on how it runs with a given configuration/kernel?
Interesting would be something to run on the current running system as is, and, for example, before/after updating the kernel or using the "switch on/off commands" of Dave Platter's post. This would then give hints about the performance differences.
Or did I get it wrong?
I had a look at the link and found that one of the programs mentioned stress is available as stress-ng but all it really does is stress the cpu so I'm trying two tests myself, an rpmbuild operation and a large file transfer between two physical disks using "time" then recording the results. I'm now going to boot into the updated kernel. Dave P
Well the timings for the rpmbuild operation which I think stresses the most didn't show any discernible differences and the file copy gained 2 seconds somehow, most probably due to the reboot. Be interesting to see if someone else can find a difference. I've got a humble i3 running Leap:42.3 Dave P -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Saturday, 2018-01-06 at 16:08 +0100, Daniel Bauer wrote:
Am 06.01.2018 um 15:45 schrieb Felix Miata:
Carlos E. R. composed on 2018-01-06 14:12 (UTC+0100):
I would like to benchmark my computer before applying patches for the recent Intel fiasco.
Suggestions?
If I see it correct, this comes with it's own kernel to boot, so it could give information on the bare computers capacities but not on how it runs with a given configuration/kernel?
Interesting would be something to run on the current running system as is, and, for example, before/after updating the kernel or using the "switch on/off commands" of Dave Platter's post. This would then give hints about the performance differences.
Or did I get it wrong?
No, you are correct. - -- Cheers, Carlos E. R. (from openSUSE 42.2 x86_64 "Malachite" at Telcontar) -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iEYEARECAAYFAlpSKzAACgkQtTMYHG2NR9U69QCePrwzoB9Why9+3coxEteJCwFI M10An1PiMhF8y472VsliPxw/CJPwyzB9 =tJtD -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Saturday, 2018-01-06 at 14:12 +0100, Carlos E. R. wrote:
Hi,
I would like to benchmark my computer before applying patches for the recent Intel fiasco.
Suggestions?
I have installed some. First I try with "interbench". Runs as root, in a terminal. Another I was going to try is this: hpcc-openmpi - A high performance computing benchmark tool The HPC Challenge benchmark consists of basically 7 tests: • HPL - the Linpack TPP benchmark which measures the floating point rate of execution for solving a linear system of equations. • DGEMM - measures the floating point rate of execution of double precision real matrix-matrix multiplication. • STREAM - a simple synthetic benchmark program that measures sustainable memory bandwidth (in GB/s) and the corresponding computation rate for simple vector kernel. • PTRANS (parallel matrix transpose) - exercises the communications where pairs of processors communicate with each other simultaneously. It is a useful test of the total communications capacity of the network. • RandomAccess - measures the rate of integer random updates of memory (GUPS). • FFT - measures the floating point rate of execution of double precision complex one-dimensional Discrete Fourier Transform (DFT). • Communication bandwidth and latency - a set of tests to measure latency and bandwidth of a number of simultaneous communication patterns; based on b_eff (effective bandwidth benchmark). But fails to run: Telcontar:~ # /usr/lib64/mpi/gcc/openmpi/bin/hpcc --help /usr/lib64/mpi/gcc/openmpi/bin/hpcc: error while loading shared libraries: libmpi.so.12: cannot open shared object file: No such file or directory But the lib is installed: Telcontar:~ # locate libmpi.so /usr/lib64/mpi/gcc/mvapich2/lib64/libmpi.so.12 /usr/lib64/mpi/gcc/mvapich2/lib64/libmpi.so.12.0.5 Telcontar:~ # l /usr/lib64/mpi/gcc/mvapich2/lib64/libmpi.so.12 /usr/lib64/mpi/gcc/mvapich2/lib64/libmpi.so.12.0.5 lrwxrwxrwx 1 root root 16 Jan 9 21:10 /usr/lib64/mpi/gcc/mvapich2/lib64/libmpi.so.12 -> libmpi.so.12.0.5* - -rwxr-xr-x 1 root root 4550096 Oct 18 2016 /usr/lib64/mpi/gcc/mvapich2/lib64/libmpi.so.12.0.5* Telcontar:~ # What's the problem? - -- Cheers, Carlos E. R. (from openSUSE 42.2 x86_64 "Malachite" at Telcontar) -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iEYEARECAAYFAlpWkbwACgkQtTMYHG2NR9VcPQCfePoVxgmTHhm7HfdUGK8+mlsr YR4AoIPXHwUEL9G+0b4YMYEAJTtJ9/Kj =Z6iE -----END PGP SIGNATURE-----
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Wednesday, 2018-01-10 at 23:20 +0100, Carlos E. R. wrote:
On Saturday, 2018-01-06 at 14:12 +0100, Carlos E. R. wrote:
I have installed some. First I try with "interbench". Runs as root, in a terminal.
Another I was going to try is this:
hpcc-openmpi - A high performance computing benchmark tool
I also installed "mpitests-3.2-2.5.x86_64", but there is no binary in the package: Telcontar:~ # rpm -ql mpitests-3.2-2.5.x86_64 /usr/share/doc/packages/mpitests /usr/share/doc/packages/mpitests/IMB_Users_Guide.pdf /usr/share/doc/packages/mpitests/README /usr/share/doc/packages/mpitests/ReadMe_IMB.txt Telcontar:~ # The readme mentions some .c files for building, but they are not present. The ReadMe_IMB.txt file mentions program "IMB-MPI1", but as you can see, it is absent, the package only contains documentation, not the test. Is this a bug, or am I missing something? - -- Cheers, Carlos E. R. (from openSUSE 42.2 x86_64 "Malachite" at Telcontar) -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iEYEARECAAYFAlpWk7MACgkQtTMYHG2NR9VrygCcD+6b8xDnCfIOdYNeXgJCthBC 1MUAn3rGtG/Lz4AetfU1y031ehPozjBF =1lnZ -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 01/10/2018 04:29 PM, Carlos E. R. wrote:
On Wednesday, 2018-01-10 at 23:20 +0100, Carlos E. R. wrote:
On Saturday, 2018-01-06 at 14:12 +0100, Carlos E. R. wrote:
I have installed some. First I try with "interbench". Runs as root, in a terminal.
Another I was going to try is this:
hpcc-openmpi - A high performance computing benchmark tool
I also installed "mpitests-3.2-2.5.x86_64", but there is no binary in the package:
Telcontar:~ # rpm -ql mpitests-3.2-2.5.x86_64 /usr/share/doc/packages/mpitests /usr/share/doc/packages/mpitests/IMB_Users_Guide.pdf /usr/share/doc/packages/mpitests/README /usr/share/doc/packages/mpitests/ReadMe_IMB.txt Telcontar:~ #
The readme mentions some .c files for building, but they are not present. The ReadMe_IMB.txt file mentions program "IMB-MPI1", but as you can see, it is absent, the package only contains documentation, not the test.
Is this a bug, or am I missing something?
-- Cheers, Carlos E. R. (from openSUSE 42.2 x86_64 "Malachite" at Telcontar)
Carlos, Take a look at the benchmarks on https://www.phoronix.com/scan.php?page=home -- David C. Rankin, J.D.,P.E.
On 2018-01-16 08:41, David C. Rankin wrote:
Carlos,
Take a look at the benchmarks on https://www.phoronix.com/scan.php?page=home
I don't want benchmarks done by others, I want to do tests myself on my current operating system. I don't see there a list of tests I can download and run myself. -- Cheers / Saludos, Carlos E. R. (from 42.2 x86_64 "Malachite" at Telcontar)
participants (6)
-
Carl Hartung -
Carlos E. R. -
Daniel Bauer -
Dave Plater -
David C. Rankin -
Felix Miata