Fedora eln compose report: 20250113.n.0 changes

2025-01-12 Thread Fedora ELN Report
OLD: Fedora-eln-20250112.n.0 NEW: Fedora-eln-20250113.n.0 = SUMMARY = Added images:0 Dropped images: 0 Added packages: 0 Dropped packages:0 Upgraded packages: 11 Downgraded packages: 0 Size of added packages: 0 B Size of dropped packages:0 B Size of

Re: zswap: should we use zsmalloc as default allocator?

2025-01-12 Thread Chris Murphy
On Sun, Jan 12, 2025, at 7:39 PM, Chris Murphy wrote: > The > default configuration file uses Oops - *my* configuration file uses: > max_pool_percent=15 > compressor=zstd > zpool=zsmalloc > accept_threshold_percent=90 > shrinker_enabled=Y Default config file uses lzo/zbud, and a higher max p

Re: zswap: should we use zsmalloc as default allocator?

2025-01-12 Thread Chris Murphy
On Fri, Jan 10, 2025, at 3:28 AM, Vitaly Zaitsev via devel wrote: > On 09/01/2025 23:42, Chris Murphy wrote: >> Do I still need to enable zswap with the kernel param? Or can it all be done >> by the service unit? > > No. It will be enabled automatically since the zswap-cli.conf > configuration

Re: F42 Change Proposal: Optimized Binaries for the AMD64 / x86_64 Architecture (v2) (self-contained)

2025-01-12 Thread Neal Gompa
On Sun, Jan 12, 2025 at 4:36 PM Chris Adams wrote: > > Once upon a time, Neal Gompa said: > > For stuff installed into /usr, we > > should just allow packages to be optionally built with the higher > > microarchitecture level in addition to the base one and allow DNF to > > sort and prefer packag

Re: F42 Change Proposal: Optimized Binaries for the AMD64 / x86_64 Architecture (v2) (self-contained)

2025-01-12 Thread Carlos Rodriguez-Fernandez
Well, I take that back, there is no second fork, ... just an exec of the target binary. So, when you launch a program, you would see two exec, one for the loader itself, and one for the final binary, for what I can understand. On 1/12/25 3:54 PM, Carlos Rodriguez-Fernandez wrote: It looks into

Re: F42 Change Proposal: Optimized Binaries for the AMD64 / x86_64 Architecture (v2) (self-contained)

2025-01-12 Thread Carlos Rodriguez-Fernandez
It looks into the cmd line it was invoked with, then generates the path based on the u-arch, and then exec the actual binary. There are total two fork/exec: one to launch the hwcaps-loader (what /usr/bin/foo points to), and the fork-exec that hwcaps-loader does to run the actual binary. I'm to

Re: F42 Change Proposal: Optimized Binaries for the AMD64 / x86_64 Architecture (v2) (self-contained)

2025-01-12 Thread Christoph Erhardt
On Sunday 12 January 2025 21:25:17 Central European Standard Time Zbigniew Jędrzejewski-Szmek wrote: > > My concern is that this defeats most methods of observability. It's > > essentially the same as the double-fork problem we used to have with > > system services. The stub will fork out and exec

Re: F42 Change Proposal: Optimized Binaries for the AMD64 / x86_64 Architecture (v2) (self-contained)

2025-01-12 Thread Chris Adams
Once upon a time, Neal Gompa said: > For stuff installed into /usr, we > should just allow packages to be optionally built with the higher > microarchitecture level in addition to the base one and allow DNF to > sort and prefer packages accordingly. Is this not just resurrecting the old i386/i586

Re: Need assistance to build Blender

2025-01-12 Thread Luya Tshimbalanga
On 2025-01-12 10:58, Sérgio Basto wrote: On Sun, 2025-01-12 at 10:15 -0800, Luya Tshimbalanga wrote: Hello team, Blender failed to build due to an issue related to Python rules changes and an additional  failure from this line: /builddir/build/BUILD/blender-4.3.2-build/blender- 4.3.2

Re: F42 Change Proposal: Optimized Binaries for the AMD64 / x86_64 Architecture (v2) (self-contained)

2025-01-12 Thread Zbigniew Jędrzejewski-Szmek
On Sun, Jan 12, 2025 at 03:16:39PM -0500, Neal Gompa wrote: > On Sun, Jan 12, 2025 at 2:36 PM Zbigniew Jędrzejewski-Szmek > wrote: > > > > On Sun, Jan 12, 2025 at 05:21:18PM +0100, Vitaly Zaitsev via devel wrote: > > > On 12/01/2025 13:02, Christoph Erhardt wrote: > > > > 1. Not all packages but o

Re: F42 Change Proposal: Unification of boot loader updates, phase 1 (system-wide)

2025-01-12 Thread Zbigniew Jędrzejewski-Szmek
On Fri, Jan 10, 2025 at 07:44:29PM +, Aoife Moloney via devel-announce wrote: > Wiki - https://fedoraproject.org/wiki/Changes/BootLoaderUpdatesPhase1 > Discussion thread - > https://discussion.fedoraproject.org/t/f42-change-proposal-unification-of-boot-loader-updates-phase-1-system-wide/142035

Re: F42 Change Proposal: Optimized Binaries for the AMD64 / x86_64 Architecture (v2) (self-contained)

2025-01-12 Thread Neal Gompa
On Sun, Jan 12, 2025 at 2:36 PM Zbigniew Jędrzejewski-Szmek wrote: > > On Sun, Jan 12, 2025 at 05:21:18PM +0100, Vitaly Zaitsev via devel wrote: > > On 12/01/2025 13:02, Christoph Erhardt wrote: > > > 1. Not all packages but only those whose maintainers explicitly opt in. > > > > Yes, at the first

Re: SPDX Statistics - 161 packages remaining

2025-01-12 Thread Miroslav Suchý
Dne 12. 01. 25 v 6:50 odp. Fabio Valentini napsal(a): Thanks for the reminder - I backported fixes from upstream for all four of my packages that are in this list, so there will be 51 later today. 🙂 Thank you. ("neither in SPDX nor in Callaway format" is also a bit of a lie - these four pack

Re: F42 Change Proposal: Optimized Binaries for the AMD64 / x86_64 Architecture (v2) (self-contained)

2025-01-12 Thread Zbigniew Jędrzejewski-Szmek
On Sun, Jan 12, 2025 at 05:21:18PM +0100, Vitaly Zaitsev via devel wrote: > On 12/01/2025 13:02, Christoph Erhardt wrote: > > 1. Not all packages but only those whose maintainers explicitly opt in. > > Yes, at the first stage. But over time the number of packages can increase > significantly. I t

Re: Need assistance to build Blender

2025-01-12 Thread Sérgio Basto
On Sun, 2025-01-12 at 10:15 -0800, Luya Tshimbalanga wrote: >   > Hello team, >   > Blender failed to build due to an issue related to Python rules > changes and an additional >  failure from this line: >   > /builddir/build/BUILD/blender-4.3.2-build/blender- > 4.3.2/source/blender/python/gpu/gpu_p

Need assistance to build Blender

2025-01-12 Thread Luya Tshimbalanga
Hello team, Blender failed to build due to an issue related to Python rules changes and an additional failure from this line: /builddir/build/BUILD/blender-4.3.2-build/blender-4.3.2/source/blender/python/gpu/gpu_py_batch.cc:341:7: error: cannot convert ‘const char*’ to ‘const char* const*’ in

Re: SPDX Statistics - 161 packages remaining

2025-01-12 Thread Fabio Valentini
On Fri, Jan 10, 2025 at 7:08 AM Miroslav Suchý wrote: > Packages that are neither in SPDX nor in Callaway format (highest priority > for now) - 55 packages: > > https://pagure.io/copr/license-validate/blob/main/f/neither-nor-remaining-packagers.txt Thanks for the reminder - I backported fixes fr

Re: F42 Change Proposal: Optimized Binaries for the AMD64 / x86_64 Architecture (v2) (self-contained)

2025-01-12 Thread Vitaly Zaitsev via devel
On 12/01/2025 13:02, Christoph Erhardt wrote: 1. Not all packages but only those whose maintainers explicitly opt in. Yes, at the first stage. But over time the number of packages can increase significantly. 2. The duplication will affect only the executable ELFs - i.e. the contents of `/us

Re: Retiring python-nose from Fedora 43+

2025-01-12 Thread Miro Hrončok
On 11. 01. 25 23:02, Nico Kadel-Garcia wrote: The dependency chains still using python-nose are pretty popular, and include awscli and Samba. This may not be as easy as it seems at first glance. Could you please point out that dependency trees? I see neither package as affected by nose retire

Re: F42 Change Proposal: Optimized Binaries for the AMD64 / x86_64 Architecture (v2) (self-contained)

2025-01-12 Thread Christoph Erhardt
Hi Vitaly, if I understand the proposal correctly, the consequences will be far less dramatic than you believe. > So all Fedora packages will take 2x (x86_64-v1, x86_64-v2) disk space? > Am I right? 1. Not all packages but only those whose maintainers explicitly opt in. 2. The duplication will

Re: F42 Change Proposal: Optimized Binaries for the AMD64 / x86_64 Architecture (v2) (self-contained)

2025-01-12 Thread drago01
On Sunday, January 12, 2025, Vitaly Zaitsev via devel < devel@lists.fedoraproject.org> wrote: > On 10/01/2025 20:28, Aoife Moloney via devel-announce wrote: > >> Individual packages can provide already optimized libraries via the >> glibc-hwcaps mechanism. This approach will be extended to executa

Re: F42 Change Proposal: Optimized Binaries for the AMD64 / x86_64 Architecture (v2) (self-contained)

2025-01-12 Thread Vitaly Zaitsev via devel
On 10/01/2025 20:28, Aoife Moloney via devel-announce wrote: Individual packages can provide already optimized libraries via the glibc-hwcaps mechanism. This approach will be extended to executables. The package provides an optimized variant of a binary in a different directory. So all Fedora p

Fedora rawhide compose report: 20250112.n.0 changes

2025-01-12 Thread Fedora Rawhide Report
OLD: Fedora-Rawhide-20250111.n.0 NEW: Fedora-Rawhide-20250112.n.0 = SUMMARY = Added images:4 Dropped images: 2 Added packages: 2 Dropped packages:0 Upgraded packages: 44 Downgraded packages: 0 Size of added packages: 371.39 KiB Size of dropped packages:0