[Touch-packages] [Bug 2030784] Re: Backport Intel's AVX512 patches on openssl 3.0
Thanks a lot for the tests, that's very appreciated. I ran that on my laptop (11th Gen Intel(R) Core(TM) i5-1135G7 @ 2.40GHz) which quite surprisingly has all these CPU features. Mostly idle, dynamic CPU governor but no thermal throttling at all (and if there were, it would probably slow down the AVX-512 code anyway), and tests are long enough for CPU governors to not matter much. * AES-128-GCM | AES-256-GCM - Baseline - Requires VAES and VPCMULQDQ features present on ICX or newer platform. This should be the most performant flow. AES-128-GCM 855360.29k 3158479.88k 6093932.91k 8905067.37k 13336828.91k 13788498.58k - Individual VAES Disabled and VPCLMULQDQ Disabled should fallback to AVX AESNI flow and should have equivalent performance AES-128-GCM 785422.85k 1936140.78k 4404423.77k 6481577.18k 7732716.48k 7873213.39k AES-128-GCM 790775.41k 1942054.64k 4404868.20k 6484287.87k 7711803.10k 7778795.52k - AESNI and VAESNI Disabled should fallback to 'C code' performance AES-128-GCM 150183.11k 167807.25k 598198.71k 662922.19k 681574.40k 678182.91k * RSA 2K/3K/4K Sign Performance - Baseline - Requires AVX512F, AVX512VL, AVX512DQ, and AVX512IFMA features on ICX or newer platform. This should be the most performant flow. rsa 2048 bits 0.000246s 0.15s 4057.2 65278.3 rsa 3072 bits 0.000701s 0.32s 1426.4 31247.7 rsa 4096 bits 0.001434s 0.55s697.4 18052.7 - Individual AVX512F, AVX512VL, and AVX512IFMA features should yield equivalent performance. This flow will use the ADOX/ADCX/MULX RSA flow. rsa 2048 bits 0.000523s 0.15s 1910.4 65748.2 rsa 3072 bits 0.001579s 0.32s633.3 31158.1 rsa 4096 bits 0.003529s 0.55s283.4 18093.6 rsa 2048 bits 0.000524s 0.15s 1909.0 66310.8 rsa 3072 bits 0.001577s 0.32s634.1 31309.7 rsa 4096 bits 0.003568s 0.55s280.2 18120.4 rsa 2048 bits 0.000523s 0.15s 1913.3 65234.3 rsa 3072 bits 0.001583s 0.32s631.7 31094.6 rsa 4096 bits 0.003607s 0.55s277.3 18076.8 rsa 2048 bits 0.000524s 0.15s 1907.6 66299.6 rsa 3072 bits 0.001577s 0.32s634.1 31214.4 rsa 4096 bits 0.003586s 0.55s278.9 18096.1 We see the expected behavior (AFAIU, all features must be available at the same time for the changes to have effect). I'm not comparing everything number by number because I don't think we're looking for specific percentages of improvements. Overall we see up to ~2.4 performance improvement and we always see large improvements (double digit percentages). As a control I also ran that on lunar, therefore without the patches (I acknowledge this is not the same openssl version and there are also other changes but I do not think this matters here). # AES-128-GCM | AES-256-GCM - Baseline - Requires VAES and VPCMULQDQ features present on ICX or newer platform. This should be the most performant flow. AES-128-GCM 782474.44k 1938211.66k 4430867.84k 6402298.54k 7685819.33k 7840186.37k - Individual VAES Disabled and VPCLMULQDQ Disabled should fallback to AVX AESNI flow and should have equivalent performance AES-128-GCM 750028.44k 1926234.78k 4365867.67k 6383893.16k 7742842.78k 7843146.41k AES-128-GCM 786910.34k 1934779.33k 4421411.45k 6389114.88k 7650086.87k 7797479.86k - AESNI and VAESNI Disabled should fallback to 'C code' performance AES-128-GCM 147889.72k 167843.85k 599710.04k 663642.45k 679072.96k 680631.91k # RSA 2K/3K/4K Sign Performance - Baseline - Requires AVX512F, AVX512VL, AVX512DQ, and AVX512IFMA features on ICX or newer platform. This should be the most performant flow. rsa 2048 bits 0.000247s 0.15s 4050.8 66072.6 rsa 3072 bits 0.001596s 0.32s626.5 31144.2 rsa 4096 bits 0.003534s 0.56s282.9 18003.6 - Individual AVX512F, AVX512VL, and AVX512IFMA features should yield equivalent performance. This flow will use the ADOX/ADCX/MULX RSA flow. rsa 2048 bits 0.000528s 0.15s 1892.3 66008.3 rsa 3072 bits 0.001573s 0.32s635.6 31094.2 rsa 4096 bits 0.003534s 0.55s282.9 18073.8 rsa 2048 bits 0.000522s 0.15s 1914.7 65763.4 rsa 3072 bits 0.001575s 0.32s635.0 31237.8 rsa 4096 bits 0.003530s 0.55s283.2 18093.1 rsa 2048 bits 0.000522s 0.15s 1917.4 65826.2 rsa 3072 bits 0.001575s 0.32s635.0 31177.2 rsa 4096 bits 0.003549s 0.55s281.8 18109.9 rsa 2048 bits 0.000522s 0.15s 1915.1 65760.4 rsa 3072 bits 0.001575s 0.32s635.0 31180.2 rsa 4096 bits 0.003538s 0.55s282.6 18109.9 We can see there are no change with the CPU feature flags, except for the test that disables AESNI, in which case the performance is the same in lunar and mantic. That the CPU feature flags don't change the performance except i the one aforementioned case, indicate that these patches are responsible for the large performance increase we have seen. We can also see that they don't otherwise de
[Touch-packages] [Bug 2040484] Re: ubuntu_seccomp pseudo-syscall fails on s390 / PowerPC
** Tags added: 4.4 xenial -- You received this bug notification because you are a member of Ubuntu Touch seeded packages, which is subscribed to libseccomp in Ubuntu. https://bugs.launchpad.net/bugs/2040484 Title: ubuntu_seccomp pseudo-syscall fails on s390 / PowerPC Status in ubuntu-kernel-tests: New Status in libseccomp package in Ubuntu: New Bug description: libseccomp upstream has changed the test code for 29-sim- pseudo_syscall.c, which has broken it for s390. Perhaps s390 has been broken since forever and the test change is just uncovering it. We need to investigate if the fix would be needed in the test, libseccomp or the kernel. This seems to affect at least 4.4 and 5.4 kernels, but may affect everything. To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu-kernel-tests/+bug/2040484/+subscriptions -- Mailing list: https://launchpad.net/~touch-packages Post to : touch-packages@lists.launchpad.net Unsubscribe : https://launchpad.net/~touch-packages More help : https://help.launchpad.net/ListHelp
[Touch-packages] [Bug 2038834] Re: GPU acceleration via VirGL is broken in qemu
The autopkgtest regression above are due to armhf autopkgtest runner faults. The tests have been re-triggered and should succeed. Confirmed to fix VirGL graphics acceleration on my previously affected machine. ** Tags removed: verification-needed verification-needed-mantic ** Tags added: verification-done verification-done-mantic -- You received this bug notification because you are a member of Ubuntu Touch seeded packages, which is subscribed to mesa in Ubuntu. https://bugs.launchpad.net/bugs/2038834 Title: GPU acceleration via VirGL is broken in qemu Status in Release Notes for Ubuntu: New Status in mesa package in Ubuntu: Fix Released Status in mesa source package in Mantic: Fix Committed Status in mesa source package in Noble: Fix Released Bug description: [ Impact ] * Enabling GPU acceleration can cause host-side crashes on mantic/noble VMs * This was reported by someone else upstream and is already fixed by https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25580. [ Test Plan ] * I've tested the patch on an affected macOS host running Ubuntu in UTM with OpenGL enabled on both Mantic and Noble VMs. * Anyone else can do the same on an affected host by simply installing the patched package and booting to the desktop. [ Where problems could occur ] * This patch fixes an upstream mesa regression which caused libvirglrendrer to crash on the host side. * This makes a non-working use case work, VirGL on affected hosts cannot regress as it simply didn't work before. * Risk of breakage is mainly from other packages possible affected by a mesa rebuild. To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu-release-notes/+bug/2038834/+subscriptions -- Mailing list: https://launchpad.net/~touch-packages Post to : touch-packages@lists.launchpad.net Unsubscribe : https://launchpad.net/~touch-packages More help : https://help.launchpad.net/ListHelp
[Touch-packages] [Bug 1946513] Re: package initramfs-tools 0.140ubuntu6 failed to install/upgrade: installed initramfs-tools package post-installation script subprocess returned error exit status 1
I have what looks to be a similar issue. I suspect the boot partition may be the problem: user@host:~$ df -h /boot FilesystemSize Used Avail Use% Mounted on bpool/BOOT/ubuntu_ykzcyk 295M 289M 5,5M 99% /boot user@host:~$ mount | grep -w / rpool/ROOT/ubuntu_ykzcyk on / type zfs (rw,relatime,xattr,posixacl) This is with a ZFS root and boot partition. user@host:~$ zpool list NAME SIZE ALLOC FREE CKPOINT EXPANDSZ FRAGCAP DEDUPHEALTH ALTROOT bpool1.88G 1.74G 133M- -56%93% 1.00xONLINE - rpool 920G 38.7G 881G- - 8% 4% 1.00xONLINE - user@host:~$ df -h | grep boot bpool/BOOT/ubuntu_ykzcyk 295M 289M 5,5M 99% /boot /dev/nvme0n1p1511M 15M 497M 3% /boot/efi user@host:~$ I found a snapshot in the boot dataset and removed it but free space hasn't dropped much. Perhaps this extra information helps? -- You received this bug notification because you are a member of Ubuntu Touch seeded packages, which is subscribed to initramfs-tools in Ubuntu. https://bugs.launchpad.net/bugs/1946513 Title: package initramfs-tools 0.140ubuntu6 failed to install/upgrade: installed initramfs-tools package post-installation script subprocess returned error exit status 1 Status in initramfs-tools package in Ubuntu: Confirmed Bug description: I don't know ProblemType: Package DistroRelease: Ubuntu 21.10 Package: initramfs-tools 0.140ubuntu6 ProcVersionSignature: Ubuntu 5.13.0-16.16-generic 5.13.13 Uname: Linux 5.13.0-16-generic x86_64 ApportVersion: 2.20.11-0ubuntu70 Architecture: amd64 CasperMD5CheckResult: pass Date: Fri Oct 8 12:36:07 2021 ErrorMessage: installed initramfs-tools package post-installation script subprocess returned error exit status 1 InstallationDate: Installed on 2021-10-08 (0 days ago) InstallationMedia: Ubuntu 21.10 "Impish Indri" - Beta amd64 (20210922) PackageArchitecture: all Python3Details: /usr/bin/python3.9, Python 3.9.7, python3-minimal, 3.9.4-1build1 PythonDetails: N/A RelatedPackageVersions: dpkg 1.20.9ubuntu2 apt 2.3.9 SourcePackage: initramfs-tools Title: package initramfs-tools 0.140ubuntu6 failed to install/upgrade: installed initramfs-tools package post-installation script subprocess returned error exit status 1 UpgradeStatus: No upgrade log present (probably fresh install) To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/initramfs-tools/+bug/1946513/+subscriptions -- Mailing list: https://launchpad.net/~touch-packages Post to : touch-packages@lists.launchpad.net Unsubscribe : https://launchpad.net/~touch-packages More help : https://help.launchpad.net/ListHelp
[Touch-packages] [Bug 2044127] Re: ubuntu-but does not accept "program" paramater as documented in --help
Thank you for taking the time to report this bug and helping to make Ubuntu better. The --help output mislead you. "symptom" refers to one short name like "audio", "display", or "installer" (look into /usr/share/apport/symptoms/ to see all), but not describing the symptom in full sentences. Suggestions how to rephrase the documentation to avoid this confusion will be appreciated. -- You received this bug notification because you are a member of Ubuntu Touch seeded packages, which is subscribed to apport in Ubuntu. https://bugs.launchpad.net/bugs/2044127 Title: ubuntu-but does not accept "program" paramater as documented in --help Status in apport package in Ubuntu: New Bug description: ubuntu-bug --help says $ ubuntu-bug --help usage: ubuntu-bug [options] [symptom|pid|package|program path|.apport/.crash file] which suggest I can type a symptom after "ubuntu-bug" that I want to report. But doing something like: $ ubuntu-bug "Ubuntu Software tells me at every boot that it installed important OS update 'UEFI dbx'" only results in the error: dpkg-query: no packages found matching Ubuntu The man page of ubuntu-bug states that the symptoms are asked for when I run "ubuntu-bug" without parameters. So the --help output is wrong or the program is lacking some fallback if the message given as sysmptom is neither a pid, nor a package name nor a program path nor a crash file, the given parameter must be accepted as a symptom description. Probably the easiest part to solve that is to remove the nonworking symptom parameter from the --help output as long as that is not working. ProblemType: Bug DistroRelease: Ubuntu 23.10 Package: apport 2.27.0-0ubuntu5 ProcVersionSignature: Ubuntu 6.2.0-1015.15-lowlatency 6.2.16 Uname: Linux 6.2.0-1015-lowlatency x86_64 ApportVersion: 2.27.0-0ubuntu5 Architecture: amd64 CasperMD5CheckResult: pass CurrentDesktop: ubuntu:GNOME Date: Tue Nov 21 14:31:49 2023 InstallationDate: Installed on 2023-08-21 (92 days ago) InstallationMedia: Ubuntu 23.04 "Lunar Lobster" - Release amd64 (20230418) PackageArchitecture: all ProcEnviron: LANG=en_US.UTF-8 PATH=(custom, no user) SHELL=/bin/bash TERM=xterm-256color XDG_RUNTIME_DIR= RebootRequiredPkgs: Error: path contained symlinks. SourcePackage: apport UpgradeStatus: Upgraded to mantic on 2023-11-05 (16 days ago) To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/apport/+bug/2044127/+subscriptions -- Mailing list: https://launchpad.net/~touch-packages Post to : touch-packages@lists.launchpad.net Unsubscribe : https://launchpad.net/~touch-packages More help : https://help.launchpad.net/ListHelp
