| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| |
|
|
|
|
|
|
|
|
|
|
| |
The access functions are forced to be inline, so performance shouldn't
be impacted for stores.
WebGL performance in Firefox is more than doubled, and track loading
in STK is noticeably faster.
Reviewed-by: Alyssa Rosenzweig <[email protected]>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4317>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4317>
|
| |
|
|
|
|
|
|
|
| |
Split it into shared part since we're going to re-use it in lima.
Reviewed-by: Alyssa Rosenzweig <[email protected]>
Reviewed-by: Boris Brezillon <[email protected]>
Signed-off-by: Vasily Khoruzhick <[email protected]>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4051>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
There's a lot going on here (it's a ton of commits squashed together
since otherwise this would be impossible to review...)
1. We have a fast path for linear->tiled for whole (aligned) tiles, but we
have to use a slow path for unaligned accesses. We can get a pretty
major win for partial updates by using this slow path simply on the
borders of the update region, and then hit the fast path for the
tile-aligned interior. This does require some shuffling.
2. Mark the LUTs constant, which allows the compiler to inline them,
which pairs well with loop unrolling (eliminating the memory accesses
and just becoming some immediates.. which are not as immediate on
aarch64 as I'd like..)
3. Add fast path for bpp1/2/8/16. These use the same algorithm and we
have native types for them, so may as well get the fast path.
4. Drop generic path for bpp != 1/2/8/16, since these formats are
generally awful and there's no way to tile them efficienctly and
honestly there's not a good reason too either. Lima doesn't support any
of these formats; Panfrost can make the opinionated choice to make them
linear.
5. Specialize the unaligned routines. They don't have to be fully
generic, they just can't assume alignment. So now they should be nearly
as fast as the aligned versions (which get some extra tricks to be even
faster but the difference might be neglible on some workloads).
6. Specialize also for the size of the tile, to allow 4x4 tiling as well
as 16x16 tiling. This allows compressed textures to be efficiently tiled
with the same routines (so we add support for tiling ASTC/ETC textures
while we're at it)
Signed-off-by: Alyssa Rosenzweig <[email protected]>
Reviewed-by: Vasily Khoruzhick <[email protected]>
Tested-by: Vasily Khoruzhick <[email protected]> #lima on Mali400
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3414>
|
| |
|
|
|
|
|
|
|
|
|
|
| |
There's an implicit dependence on Gallium here that will add more
complexity than needed when testing/optimizing out of driver as well as
potentially Vulkanizing. We don't need a full pipe_box, just the x/y/w/h
properties directly.
Signed-off-by: Alyssa Rosenzweig <[email protected]>
Reviewed-by: Vasily Khoruzhick <[email protected]>
Tested-by: Vasily Khoruzhick <[email protected]> #lima on Mali400
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3414>
|
| |
|
|
|
|
|
|
|
|
| |
These are major hot spots for panfrost and lima; better let the compiler
do its thing even on debug builds.
Signed-off-by: Alyssa Rosenzweig <[email protected]>
Reviewed-by: Vasily Khoruzhick <[email protected]>
Tested-by: Vasily Khoruzhick <[email protected]> #lima on Mali400
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3414>
|
| |
|
|
|
|
| |
Should not affect lima.
Signed-off-by: Alyssa Rosenzweig <[email protected]>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Rather than using a magic lookup table with no explanations, let's add
liberal comments to the code to explain what this tiling scheme is and
how to encode/decode it efficiently.
It's not so mysterious after all -- just reordering bits with some XORs
thrown in.
v2: Correct copyright identifier. Fix spelling error. Switch space_4 to
a LUT. Fix comment typo. Use LUT instead of space_x tricks. Fallback on
generic rather than split up unaligned writes.
v3: Correct stride order (fixes crash loading). Correct coordinate
system mishap.
Signed-off-by: Alyssa Rosenzweig <[email protected]>
Reviewed-by: Vasily Khoruzhick <[email protected]>
Tested-by: Andreas Baierl <[email protected]>
|
|
|
This will allow both drivers to share this code. Both drivers
build-tested with meson. Android build not tested.
v2: Change naming from tiling->shared, in case Lima and Panfrost can
share more in the future. Fix Android build system.
Signed-off-by: Alyssa Rosenzweig <[email protected]>
Reviewed-and-tested-by: Qiang Yu <[email protected]>
|
