dev-python/diff-gaussian-rasterization
Differentiable antialiased Gaussian rasterization (mip-splatting) for TRELLIS
Screenshots
View
Download
BrowseChangeLog
commit 5bab2a452e8f701397a6ea07d92ec513c0b7de2f
Author: Ivan S. Titov <iohann.s.titov@gmail.com>
Date: Sun Oct 4 13:12:22 2026 +0200
dev-python/diff-gaussian-rasterization: drop the redundant S
S="$/$" repeats the default. No behaviour change.
commit c080ae13b8cf5fa53e776385047786defb0b4c53
Author: Ivan S. Titov <iohann.s.titov@gmail.com>
Date: Wed Sep 16 18:23:29 2026 +0200
dev-python/diff-gaussian-rasterization: add a CUDA arch fallback
The ebuild probed the visible GPU and otherwise left
TORCH_CUDA_ARCH_LIST unset, on the assumption that cpp_extension would
pick a fallback list. It does not: with no device to query it assembles
an empty architecture list and then indexes it, so the build ends in
IndexError. That hits any build on a machine with no GPU, or where the
devices are absent, such as a container or a build host.
Fall back to compute capability 7.5 with PTX and say so. PTX is what
makes the fallback useful: such a build is by definition running
somewhere without the target GPU, and a bare 7.5 cubin would not load
on a newer card. An explicit list is still honoured.
commit ecc73e2034735298a1c6e379d4ee7f6b219cb13b
Author: Ivan S. Titov <iohann.s.titov@gmail.com>
Date: Sat Sep 12 16:13:08 2026 +0200
dev-python/diff-gaussian-rasterization: enable Python 3.14
The extension uses Torch C++ bindings without a Python upper bound, and
its dependency stack supports Python 3.14.
commit 07b16ecc1244a03d07ad673f5c74c17643001e64
Author: Ivan S. Titov <iohann.s.titov@gmail.com>
Date: Thu Sep 10 04:54:22 2026 +0200
dev-python/diff-gaussian-rasterization: trim snapshot comment prose
commit 976697f1e71d7be42efd6416cf1f2ac5e1196383
Author: Ivan S. Titov <iohann.s.titov@gmail.com>
Date: Mon Aug 17 17:13:56 2026 +0200
dev-python/diff-gaussian-rasterization: build for the native GPU arch
The cuda compile forced TORCH_CUDA_ARCH_LIST to a fixed 8.0 8.6 8.9 9.0
list, so every host built four Ampere-through-Hopper targets regardless
of its GPU, and a machine outside that window got no runnable kernel by
default. Probe the native compute capability with nvcc's device query
and build only that; an explicit TORCH_CUDA_ARCH_LIST still wins, and
when no GPU is visible at build time the value is left unset so torch
falls back to its full arch list.
commit bc28583b1be0a4ecb128e7cb29eac7a318aacfef
Author: Ivan S. Titov <iohann.s.titov@gmail.com>
Date: Tue Jul 7 23:27:44 2026 +0200
dev-python/diff-gaussian-rasterization: reword vendoring comment
commit c674d3e095434edf93086598e27a8c24a79ef8de
Author: Ivan S. Titov <iohann.s.titov@gmail.com>
Date: Wed Jun 17 10:16:56 2026 +0200
dev-python/diff-gaussian-rasterization: new package, add 0_p20241201
Author: Ivan S. Titov <iohann.s.titov@gmail.com>
Date: Sun Oct 4 13:12:22 2026 +0200
dev-python/diff-gaussian-rasterization: drop the redundant S
S="$/$" repeats the default. No behaviour change.
commit c080ae13b8cf5fa53e776385047786defb0b4c53
Author: Ivan S. Titov <iohann.s.titov@gmail.com>
Date: Wed Sep 16 18:23:29 2026 +0200
dev-python/diff-gaussian-rasterization: add a CUDA arch fallback
The ebuild probed the visible GPU and otherwise left
TORCH_CUDA_ARCH_LIST unset, on the assumption that cpp_extension would
pick a fallback list. It does not: with no device to query it assembles
an empty architecture list and then indexes it, so the build ends in
IndexError. That hits any build on a machine with no GPU, or where the
devices are absent, such as a container or a build host.
Fall back to compute capability 7.5 with PTX and say so. PTX is what
makes the fallback useful: such a build is by definition running
somewhere without the target GPU, and a bare 7.5 cubin would not load
on a newer card. An explicit list is still honoured.
commit ecc73e2034735298a1c6e379d4ee7f6b219cb13b
Author: Ivan S. Titov <iohann.s.titov@gmail.com>
Date: Sat Sep 12 16:13:08 2026 +0200
dev-python/diff-gaussian-rasterization: enable Python 3.14
The extension uses Torch C++ bindings without a Python upper bound, and
its dependency stack supports Python 3.14.
commit 07b16ecc1244a03d07ad673f5c74c17643001e64
Author: Ivan S. Titov <iohann.s.titov@gmail.com>
Date: Thu Sep 10 04:54:22 2026 +0200
dev-python/diff-gaussian-rasterization: trim snapshot comment prose
commit 976697f1e71d7be42efd6416cf1f2ac5e1196383
Author: Ivan S. Titov <iohann.s.titov@gmail.com>
Date: Mon Aug 17 17:13:56 2026 +0200
dev-python/diff-gaussian-rasterization: build for the native GPU arch
The cuda compile forced TORCH_CUDA_ARCH_LIST to a fixed 8.0 8.6 8.9 9.0
list, so every host built four Ampere-through-Hopper targets regardless
of its GPU, and a machine outside that window got no runnable kernel by
default. Probe the native compute capability with nvcc's device query
and build only that; an explicit TORCH_CUDA_ARCH_LIST still wins, and
when no GPU is visible at build time the value is left unset so torch
falls back to its full arch list.
commit bc28583b1be0a4ecb128e7cb29eac7a318aacfef
Author: Ivan S. Titov <iohann.s.titov@gmail.com>
Date: Tue Jul 7 23:27:44 2026 +0200
dev-python/diff-gaussian-rasterization: reword vendoring comment
commit c674d3e095434edf93086598e27a8c24a79ef8de
Author: Ivan S. Titov <iohann.s.titov@gmail.com>
Date: Wed Jun 17 10:16:56 2026 +0200
dev-python/diff-gaussian-rasterization: new package, add 0_p20241201