| 17 Dec 2025 |
ghpzin | Yes | 18:15:26 |
9lore |
I wonder if the queries could be optimized.
unfortunately my project list is too long to a unhealthy degree :/
| 18:16:04 |
vcunat | Or I can recommend using https://search.nixos.org/packages and after you find a package, you can click links to jobs for each platform. | 18:16:26 |
vcunat | * Or I can recommend using https://search.nixos.org/packages and after you find a package, you can click links to Hydra jobs for each platform. | 18:16:39 |
K900 | Oh great | 18:17:07 |
K900 | Node is also busted | 18:17:10 |
K900 | (possibly only i686, possibly only some versions?) | 18:17:18 |
vcunat | * Or I can recommend using https://search.nixos.org/packages and after you find a package, you can click links to Hydra jobs for each platform. And this search is actually performant, unlike Hydra searches. | 18:17:22 |
vcunat | * Or I can recommend using https://search.nixos.org/packages and after you find a package, you can click links to Hydra jobs for each platform. And this search is actually performant and has better UX overall, unlike Hydra searches. | 18:17:44 |
K900 | > In file included from ../../deps/v8/src/compiler/turboshaft/int64-lowering-phase.cc:9:
> ../../deps/v8/src/compiler/turboshaft/int64-lowering-reducer.h: In member function 'v8::internal::compiler::turboshaft::V<v8::internal::compiler::turboshaft::Tuple<v8::internal::compiler::turboshaft::WordWithBits<32>, v8::internal::compiler::turboshaft::WordWithBits<32> > > v8::internal::compiler::turboshaft::Int64LoweringReducer<Next>::LowerClz(v8::internal::compiler::turboshaft::V<v8::internal::compiler::turboshaft::Tuple<v8::internal::compiler::turboshaft::WordWithBits<32>, v8::internal::compiler::turboshaft::WordWithBits<32> > >)':
> ../../deps/v8/src/compiler/turboshaft/int64-lowering-reducer.h:640:15: error: invalid use of 'struct v8::internal::compiler::turboshaft::Tuple<v8::internal::compiler::turboshaft::WordWithBits<32>, v8::internal::compiler::turboshaft::WordWithBits<32> >' [-Wtemplate-body]
> 640 | return __ Tuple<Word32, Word32>(result, __ Word32Constant(0));
> | ^~~~~~~~~~~~~~~~~~~~~
> ../../deps/v8/src/compiler/turboshaft/int64-lowering-reducer.h: In member function 'v8::internal::compiler::turboshaft::V<v8::internal::compiler::turboshaft::Tuple<v8::internal::compiler::turboshaft::WordWithBits<32>, v8::internal::compiler::turboshaft::WordWithBits<32> > > v8::internal::compiler::turboshaft::Int64LoweringReducer<Next>::LowerCtz(v8::internal::compiler::turboshaft::V<v8::internal::compiler::turboshaft::Tuple<v8::internal::compiler::turboshaft::WordWithBits<32>, v8::internal::compiler::turboshaft::WordWithBits<32> > >)':
> ../../deps/v8/src/compiler/turboshaft/int64-lowering-reducer.h:653:15: error: invalid use of 'struct v8::internal::compiler::turboshaft::Tuple<v8::internal::compiler::turboshaft::WordWithBits<32>, v8::internal::compiler::turboshaft::WordWithBits<32> >' [-Wtemplate-body]
> 653 | return __ Tuple<Word32, Word32>(result, __ Word32Constant(0));
| 18:17:45 |
K900 | Whatever this even means | 18:17:48 |
K900 | https://chromium-review.googlesource.com/c/v8/v8/+/6830052 | 18:24:04 |
K900 | Looks like it's this | 18:24:06 |
K900 | And indeed 32-bit only | 18:24:12 |
9lore | i wonder if upstream tests 32-bit | 18:25:30 |
K900 | Evidently they do not | 18:26:02 |
vcunat | 32-bit support is dying generally. Maybe 32-bit ARM will last a bit longer, but x86... | 18:27:05 |
9lore | thought as much, had to deal with google-caused regressions in another project before | 18:27:13 |
9lore | is this v24 or earlier as well? | 18:31:09 |
9lore | * is this node 24 or earlier as well? | 18:31:19 |
9lore | armv7 is still being used alot in new embedded / iot. I doubt its going anywhere anytime soon | 18:32:26 |
vcunat | Sure, but you probably won't be running, say, Chrome on these. | 18:33:02 |
vcunat | (i.e. only a small subset of packages will be "kept alive" by such use cases) | 18:33:28 |
9lore | PoS systems is an example that comes to mind | 18:34:01 |
9lore | but yea, i686 is dead imo | 18:34:23 |
leona | valve didn't get this memo :D | 18:34:41 |
K900 | https://github.com/heylogin/nodejs-mobile/commit/79cb3ea1f8fb2ca2b1d4cbd719d5e6c61c499cff should we just steal that | 18:34:42 |
K900 | I think this is a less cursed fix than backporting that big rename | 18:34:55 |
9lore | 64-bit has hit the windows client beta as of nov 25 FYI | 18:36:59 |
9lore | cant be too far off overall | 18:37:28 |