Re: mpv 0.39.0, question about "vulkan-interop"

On 2025-01-30 9:16 a.m., Stuart Henderson wrote: here's an update to mpv 0.39.0, which is working ok so far for me. the "vulkan-interop" build option (which we were setting to "disabled" was removed, and code relating to it moved under plain "vulkan" (which is "enabled"). I don't know how to te

Re: [llvm 19] boost: fix quat_traits.hpp

OK. On 2025-01-30 8:03 p.m., Theo Buehler wrote: Another obvious fix. This makes cad/pcb2gcode build with llvm 19. https://github.com/boostorg/qvm/commit/c4b9c063a076644575fe9a685c02cd4c10d24405 The pcb2gcode configure script would error out and contain: /usr/local/include/boost/qvm/quat_trai

Update: JRuby 9.4.11.0

Simple update to the latest release of JRuby, released yesterday: https://www.jruby.org/2025/01/29/jruby-9-4-11-0.html Tested on amd64, the only supported arch. Will be committing in a couple days unless I hear objections. Thanks, Jeremy Index: Makefile =

sparc64 bulk build report

Bulk build on sparc64-0a.ports.openbsd.org Started : Tue Jan 28 13:37:15 MST 2025 Finished: Thu Jan 30 17:12:51 MST 2025 Duration: 2 Days 3 hours 36 minutes Built using OpenBSD 7.6-current (GENERIC.MP) #2383: Thu Jan 23 02:29:47 MST 2025 Built 8434 packages Number of packages built each day: Ja

[update] graphics/nsxiv v32 -> v33

Hi ports@ Here is a small patch for updating nsxiv to the latest and greatest. Tested briefly on amd64 and everything seems to be working just fine as usual. Changes: - Moved key-handler prompts to the right side bar - Rename --class to --name - Print usage output to stderr on usage error Added

[llvm 19] boost: fix quat_traits.hpp

Another obvious fix. This makes cad/pcb2gcode build with llvm 19. https://github.com/boostorg/qvm/commit/c4b9c063a076644575fe9a685c02cd4c10d24405 The pcb2gcode configure script would error out and contain: /usr/local/include/boost/qvm/quat_traits.hpp:82:30: error: a template argument list is ex

Re: [llvm 19] games/allegro build hack

Le Thu, Jan 30, 2025 at 11:01:18AM +0100, Theo Buehler a écrit : > This is absolutely gross, but it makes allegro and all its consumers > build with llvm 19. I don't want to commit this, but perhaps it helps > someone to figure out the correct fix. i think this is fine, given that allegro was impo

Re: NEW emulators/llama.cpp b4589

Stuart Henderson writes: > On 2025/01/29 21:12, Chris Cappuccio wrote: >> This is a simple port that works well out-of-the-box. It's kind of an >> emulator, >> I think. >> >> It has some optional python scripts that will need numpy and torch. >> >> I'm not sure the right way to handle the versio

Re: [new] audio/spotify-player / cargo.port.mk

On Thu, Jan 30, 2025 at 07:14:11AM +0100, Sebastien Marie wrote: > Theo Buehler writes: > > > I think sthen is right that a variable for this in cargo.port.mk would > > make sense to obviate the need to keep these in sync. On rust updates, > > all rust ports are rebulit anyway, so if rust switche

Re: [new] audio/spotify-player / cargo.port.mk

On 2025/01/30 15:26, Theo Buehler wrote: > On Thu, Jan 30, 2025 at 07:14:11AM +0100, Sebastien Marie wrote: > > Theo Buehler writes: > > > > > I think sthen is right that a variable for this in cargo.port.mk would > > > make sense to obviate the need to keep these in sync. On rust updates, > > >

Re: NEW emulators/llama.cpp b4589

On 2025/01/30 08:15, Dave Voutila wrote: > > FWIW we should be able to include Vulkan support as its in ports. I've > played with llama.cpp locally with it, but I don't have a GPU that's > worth a damn top see if it's an improvement over pure CPU-based > inferencing. Makes sense, though I think i

mpv 0.39.0, question about "vulkan-interop"

here's an update to mpv 0.39.0, which is working ok so far for me. the "vulkan-interop" build option (which we were setting to "disabled" was removed, and code relating to it moved under plain "vulkan" (which is "enabled"). I don't know how to test that, any suggestions? Index: Makefile ===

Re: NEW emulators/llama.cpp b4589

Stuart Henderson [s...@spacehopper.org] wrote: > > I don't understand why it's in emulators. Perhaps misc would make sense? > I guess either misc or even a new category, like ml. Torch wuold come next, and there are plenty of other pieces that really don't fit in any other category except misc.

Re: [new] audio/spotify-player / cargo.port.mk

On Thu, Jan 30, 2025 at 04:55:34PM +, Stuart Henderson wrote: > On 2025/01/30 15:26, Theo Buehler wrote: > > On Thu, Jan 30, 2025 at 07:14:11AM +0100, Sebastien Marie wrote: > > > Theo Buehler writes: > > > > > > > I think sthen is right that a variable for this in cargo.port.mk would > > > >

Re: NEW emulators/llama.cpp b4589

Stuart Henderson [s...@spacehopper.org] wrote: > > I'd be happy with misc. If we end up with dozens of related ports then > maybe a new category makes sense but misc seems to fit and is not over-full. Ok, here's a new spin for misc/llama.cpp with your patch applied. Using this model an AMD EPYC

Re: NEW emulators/llama.cpp b4589

Stuart Henderson writes: > On 2025/01/30 08:15, Dave Voutila wrote: >> >> FWIW we should be able to include Vulkan support as its in ports. I've >> played with llama.cpp locally with it, but I don't have a GPU that's >> worth a damn top see if it's an improvement over pure CPU-based >> inferencin

Re: NEW emulators/llama.cpp b4589

On 2025/01/30 13:27, Dave Voutila wrote: > Stuart Henderson writes: > > > On 2025/01/30 08:15, Dave Voutila wrote: > >> > >> FWIW we should be able to include Vulkan support as its in ports. I've > >> played with llama.cpp locally with it, but I don't have a GPU that's > >> worth a damn top see i

Re: NEW emulators/llama.cpp b4589

On 2025/01/30 10:03, Chris Cappuccio wrote: > Stuart Henderson [s...@spacehopper.org] wrote: > > > > I'd be happy with misc. If we end up with dozens of related ports then > > maybe a new category makes sense but misc seems to fit and is not over-full. > > Ok, here's a new spin for misc/llama.cpp

[llvm 19] boost: fix typo in future.hpp

This fixes a typo that breaks the build of poedit with LLVM 19. /usr/local/include/boost/thread/future.hpp:4671:19: error: no member named 'that' in 'run_it'; did you mean 'that_'? 4671 | that_=x.that; | ^~~~ | that_ /usr/local/include/bo

Re: [llvm 19] boost: fix typo in future.hpp

OK. On 2025-01-30 4:38 p.m., Theo Buehler wrote: This fixes a typo that breaks the build of poedit with LLVM 19. /usr/local/include/boost/thread/future.hpp:4671:19: error: no member named 'that' in 'run_it'; did you mean 'that_'? 4671 | that_=x.that; | ^~~~

Re: NEW emulators/llama.cpp b4589

On 2025/01/29 21:12, Chris Cappuccio wrote: > This is a simple port that works well out-of-the-box. It's kind of an > emulator, > I think. > > It has some optional python scripts that will need numpy and torch. > > I'm not sure the right way to handle the version since it's "b4589" and not > 1.

[update] editors/neovim 0.10.4

Hi ports@ [ccc edd@ as maintainer] a simple update for the latest version 0.10.4 (maintenance release). Changelog: https://github.com/neovim/neovim/releases/tag/v0.10.4 - update Makefile and distinfo - remove comment for fail test 'Test_linebreak_list' => OK with this version. Build and tests O

[llvm 19] games/allegro build hack

This is absolutely gross, but it makes allegro and all its consumers build with llvm 19. I don't want to commit this, but perhaps it helps someone to figure out the correct fix. Index: Makefile === RCS file: /cvs/ports/games/allegro/M

[llvm 19] fix build of prusaslicer

llvm 19 got a lot pickier about templates. Here's the simple adjustment needed to make prusaslicer build with it. This depends on the update to CGLA 5.6.2 that I just committed. Index: Makefile === RCS file: /cvs/ports/cad/prusaslicer