d9793a8a60
zink: reuse descriptor barriers across draws
...
if we aren't invalidating the descriptor set then we can safely reuse its
barriers to avoid doing any sort of hashing during descriptor updating
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9348 >
2021-03-12 16:32:53 +00:00
c55e2fb59c
zink: move descriptor binding out of the update codepath
...
we won't always need to update sets that we bind
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9348 >
2021-03-12 16:32:53 +00:00
dc020db723
zink: pre-size descriptor transition hash table
...
this avoids costly rehashing
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9348 >
2021-03-12 16:32:53 +00:00
80d9f3aa34
zink: simplify check for knowing whether descriptor updating is needed
...
if a program has at least one pool object set then it will have descriptors
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9348 >
2021-03-12 16:32:53 +00:00
fe31566fbb
zink: remove intermediate func for descriptor set getting
...
we can simplify this a bit by just getting the objects in the function
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9348 >
2021-03-12 16:32:53 +00:00
141ff0e091
zink: allow reuse of zink_descriptor_pools between programs
...
now that we've split these into their own objects, we can reuse them across
programs with matching binding layouts for a given type, giving even more value to
the descriptor set cache by increasing reuse
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9348 >
2021-03-12 16:32:53 +00:00
ee40550b87
zink: move descriptor type to pool object from set
...
this is a little less memory usage
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9348 >
2021-03-12 16:32:53 +00:00
58500979a8
zink: track number of sets currently allocated per descriptor pool
...
if we're going to be sharing pools across programs, it's no longer sufficient
to check the hash populations, so we need a separate counter to know when we have
to flush
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9348 >
2021-03-12 16:32:53 +00:00
871022dc84
zink: change program pointer on struct zink_descriptor_set to pool pointer
...
the program using the set isn't very relevant now that there's the intermediate
pool struct which stores the sets
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9348 >
2021-03-12 16:32:53 +00:00
817de3c776
zink: break out all the descriptor pool/layout stuff into a new struct
...
by pushing these into a separate struct, we can eventually reuse them between
programs to make the descriptor set caches even more valuable
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9348 >
2021-03-12 16:32:53 +00:00
3d4e6c641b
zink: break out descriptor stuff into new files
...
this is getting to be enough code that it's getting to be a hassle to
keep with the program stuff
also rename a couple of the moved functions
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9348 >
2021-03-12 16:32:53 +00:00
eee488abad
zink: deduplicate VkWriteDescriptorSet setup
...
no functional changes
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9348 >
2021-03-12 16:32:53 +00:00
ce0524f009
zink: break out image descriptor updating
...
all the descriptor updates are now broken out, so update_descriptors is now just
a dispatch for calling the other update functions
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9348 >
2021-03-12 16:32:53 +00:00
eb375dc939
zink: break out sampler descriptor updating
...
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9348 >
2021-03-12 16:32:53 +00:00
e94c2d84b3
zink: break out ssbo descriptor updating
...
by the power of grayskull, this is now readable
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9348 >
2021-03-12 16:32:53 +00:00
e4076d557f
zink: split out ubo descriptor updating
...
now that we're doing incremental updates, we can refactor this to be much
more readable by having separate functions for each type of descriptor update
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9348 >
2021-03-12 16:32:53 +00:00
2c83ea13d2
zink: add VkPipelineLayout to zink_program meta struct
...
this lets us simplify a little more code
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9348 >
2021-03-12 16:32:53 +00:00
0248e6b969
zink: don't double iterate all the per-batch sets on reset
...
we're already iterating once here, so we can definitely skip iterating again
and just do the removals inlined
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9348 >
2021-03-12 16:32:53 +00:00
74a0badb91
zink: add flag for recycled descriptor sets
...
this lets us skip a hash lookup in the last_set fastpath
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9348 >
2021-03-12 16:32:53 +00:00
3f561bbc83
zink: improve descriptor cache invalidation
...
we can pass the offset of the 'invalid' flag directly to the resources
to let them run through and invalidate their sets in time for us to detect
it when we recycle the set during batch reset and throw it onto our allocation array
additionally, by adding refs for the actual objects used in a descriptor set, we can
ensure that our cache is as accurate as possible
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9348 >
2021-03-12 16:32:53 +00:00
1a9f3ba2d7
zink: only batch-reference the program in use once per descriptor update
...
now that we're guaranteed to know what our batch is earlier, we can move this
referencing around to reduce the number of hash lookups we'll perform here
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9348 >
2021-03-12 16:32:53 +00:00
9df3147264
zink: move descriptor set allocation near the top of update_descriptors
...
no functional changes, but it moves the only point in the function where a
flush can occur out of the way which lets us get the batch we'll be using
immediately
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9348 >
2021-03-12 16:32:53 +00:00
514b5ced97
zink: add program pointer to desc set struct
...
we generally want to avoid this, but it lets us skip a lot of hash lookups,
which is a performance hit
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9348 >
2021-03-12 16:32:53 +00:00
b4105e8ac0
zink: actually flag all used resources as used during update_descriptors
...
num descriptors != num bound resources, so this needs to be a separate count/index
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9348 >
2021-03-12 16:32:53 +00:00
40be7f6353
zink: store last-used descriptor set for each type of set for quick reuse
...
we can avoid some hash lookups this way, and we can also avoid putting the null
descriptor sets back into the array since we know they'll always be the last-used
set
this also helps our null set reuse be more explicit since we never have to put these sets
back into an array or anything
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9348 >
2021-03-12 16:32:53 +00:00
8c908ee6a4
zink: pre-hash sampler views and states
...
this greatly reduces the amount of on-demand hashing that we have to do,
as now in worst case we'll be hashing 2x uint32_t per sampler descriptor
vs 2x vulkan object pointer (sometimes a uint64_t) and a uint32_t
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9348 >
2021-03-12 16:32:53 +00:00
fa77cc2fbb
zink: add a null sampler view descriptor hash to the screen
...
pre-hashing this saves us time later since we can just reuse it
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9348 >
2021-03-12 16:32:53 +00:00
67f9038b85
zink: introduce descriptor states
...
this aims to track the states of descriptors so that we can do more incremental
updating
it also enables the descriptor cache to be more robust by providing the incremental
data as the key
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9348 >
2021-03-12 16:32:53 +00:00
6d233e74ad
zink: use dynamic offsets for first ubo
...
this lets us avoid invalidating the ubo descriptor state, which reduces our
cache overhead
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9348 >
2021-03-12 16:32:53 +00:00
d4304ab0d2
zink: split descriptor sets based on usage
...
this uses multiple descriptor sets so that we can perform more incremental
updating and increase the value that we get from our cache by only invalidating
one state at a time
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9348 >
2021-03-12 16:32:53 +00:00
c5ce20e97a
zink: move surface refs to the end of descriptor updating
...
keeping descriptor stuff with descriptor stuff
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9348 >
2021-03-12 16:32:53 +00:00
d06c2ae21f
zink: reorder descriptor barrier applying during updating
...
just putting this with the rest of the non-descriptor stuff
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9348 >
2021-03-12 16:32:53 +00:00
761a04a464
zink: move streamout to draw_vbo
...
this isn't part of descriptor updating
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9348 >
2021-03-12 16:32:53 +00:00
b10a3be45e
zink: add second level cache for descriptor sets
...
this lets us store sets that are valid but not currently used so that we
can either prolongue a cache entry's lifetime or cannibalize a valid entry
if necessary to avoid needing to allocate more sets
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9348 >
2021-03-12 16:32:53 +00:00
6c4c995836
zink: add caching for descriptor sets
...
this is a lot of churn that more or less amounts to hashing the descriptor
state during draw and then performing lookups with this to determine whether
we can reuse an existing descriptor set instead of allocating one
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9348 >
2021-03-12 16:32:52 +00:00
840ea21fa1
zink: add scaling factor for descriptor set bucket allocations
...
now descriptor sets allocate in increasingly large batches based on how many
sets a program has allocated, multiplying by 10 any time the sets hit a power of
10, e.g., if 100 sets are allocated, we now allocate in batches of 100
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9348 >
2021-03-12 16:32:52 +00:00
bd78710d21
zink: add bucket allocating for descriptor sets
...
now instead of allocating a single descriptorset at a time, we allocate
a defined count of descriptorsets (currently 10) at once and keep a separate
array of allocated-and-unused sets that we can pop sets off of
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9348 >
2021-03-12 16:32:52 +00:00
4461a871b8
zink: remove flushes for batch descriptor use
...
now that we do all our tracking and flushing per-program, we can throw
out the batch-based flushing and let our descriptors free-range
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9348 >
2021-03-12 16:32:52 +00:00
d21e88719a
zink: even better handling for descriptor oom
...
in addition to ensuring that all our batches stay under the max size by cycling
them whenever we get too many active descriptors going, we now do per-program
descriptor pools, so we can do some limiting there as well to ensure that we
aren't letting any one program hog all the resources
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9348 >
2021-03-12 16:32:52 +00:00
4ec1d5e70c
zink: add helper function for cycling a batch
...
this saves a bit of typing here and there
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9348 >
2021-03-12 16:32:52 +00:00
7ae8509673
zink: use more precise sizing for descriptor pools
...
now that these are on the program structs, we can size the pools optimally
based on the set layouts
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9348 >
2021-03-12 16:32:52 +00:00
9b3e2ba30c
zink: move descriptor set alloc function to zink_program.c
...
try to keep things organized...
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9348 >
2021-03-12 16:32:52 +00:00
0b1a356393
zink: store and reuse descriptorsets after batch completion
...
since we know that the layout is going to match, we can store descriptorsets
in the program and then overwrite them instead of needing to free sets or reset
the pool
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9348 >
2021-03-12 16:32:52 +00:00
0e6ef05878
zink: move descriptor sets/pools from batches to programs
...
this lets us much more accurately create descriptor sets using the exact
size required by a given program instead of creating gigantic monolithic sets
it does (temporarily) incur a perf hit since sets are now freed after each use rather
than being reset
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9348 >
2021-03-12 16:32:52 +00:00
5ddfffcca1
zink: simplify some update_descriptor code
...
by using the generic zink_program here we can shorten the code a little
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9348 >
2021-03-12 16:32:52 +00:00
15fc0c351a
radv: fix potential clears with non renderable images on GFX9+
...
Found by inspection.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com >
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9474 >
2021-03-12 16:09:42 +00:00
9863ed9bf3
radv: fix meta save/restore state with non renderable images
...
For non renderable images, the driver performs some transfer operations
with compute shaders on the gfx queue, but it was saving the gfx state
instead of the compute state.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com >
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9474 >
2021-03-12 16:09:42 +00:00
b42204e4d0
radv: disable sampling with VK_FORMAT_R64_SFLOAT
...
It's not supported.
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/4433
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com >
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9519 >
2021-03-12 16:07:54 +01:00
44c38cae9a
virgl: Support the ETC1_RGB8 format as virglrenderer supports it
...
Signed-off-by: Rohan Garg <rohan.garg@collabora.com >
Reviewed-by: Gert Wollny <gert.wollny@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9497 >
2021-03-12 15:35:47 +01:00
c453921c91
lima: run nir dce after nir_lower_vec_to_movs
...
Some of the 'vec*' nir instructions may hold references to dead code
until the nir_lower_vec_to_movs pass runs.
After nir_lower_vec_to_movs, that code can finally be cleaned by dce,
so add an additional dce pass.
This not only potentially further removes unneeded code from the nir
representation but also prevents bugs with the compiler from special
case unused code that is not expected (e.g. root undef type nodes).
Signed-off-by: Erico Nunes <nunes.erico@gmail.com >
Reviewed-by: Vasily Khoruzhick <anarsoul@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9507 >
2021-03-12 11:46:54 +00:00
Mike Blumenkrantz