They are dummy objects but the spec requires layout to not be
NULL, this just makes sure we are creating valid pipeline layout
objects. This will allow us to remove some useless checks.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
It uses slightly more memory (though still bounded by the number
of mapped ranges), but gives less quadratic behavior.
Cuts 4 minutes from the runtime of the CTS *.sparse.* tests.
Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com>
This reverts commit 2294d35b24.
We can't do this without adjusting the input SGPRs/VGPRs logic.
For now, just revert it. I will send a proper solution later.
It fixes a rendering issue in F1 2017 that CTS didn't catch up.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Tested-by: Alex Smith <asmith@feralinteractive.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
anv merges the tess info correctly, but radv wasn't doing this.
This fixes hangs in
dEQP-VK.tessellation.winding.default_domain.hlsl_triangles_ccw
Fixes: 60fc0544e0 (radv/pipeline: handle tessellation shader compilation)
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Signed-off-by: Dave Airlie <airlied@redhat.com>
We did not set the layer correctly for the dst, as we would keep
using the base layer. Same for the source image.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=102710
CC: <mesa-stable@lists.freedesktop.org>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Use 16_ABGR instead of 32_ABGR if Z isn't written.
Ported from RadeonSI.
No CTS regressions on Polaris.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
We should also not load the input SGPRs and VGPRS, but
let's start with this for now.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
The number of grid components is always 3 when gl_NumWorkGroups
is declared, because it relies on the number of components of
nir_instrinsic_load_num_work_groups.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
We never supported it. Missed during copy and pasting.
Fixes: 17201a2eb0 "radv: port to using updated anv entrypoint/extension generator."
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
If the app does not plan to put a buffer or image in it
(why? But it is allowed and CTS does it), they do not need to
allocate it with the deciate allocation struct.
Fixes: a639d40f13 "radv: add support for local bos. (v3)"
Reviewed-by: Dave Airlie <airlied@redhat.com>
Might be useful to know the VRAM/GTT usage, the number of VRAM
CPU page faults, etc. Nothing is currently using that new
interface, but it's a first step.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
dota2 binds a ton of index buffers but the type is always 16-bit.
Note that we have to invalidate the type when switching from
indexed draws to normal draws.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
dota2 always calls vkResetCommandBuffer() before
vkBeginCommandBuffer() which is quite useless.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
RADV_CMD_BUFFER_STATUS_INVALID is not used for now, but I think
it makes sense to declare it. Could be used later with better
command buffer error handling.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Copied from RadeonSI.
This fixes all CTS
dEQP-VK.renderpass.dedicated_allocation.formats.d32_sfloat_s8_uint.clear.*
And some other ones which use the same format.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Allows apps to determine the LLVM version so that they can decide
whether or not to enable workarounds for LLVM issues.
Signed-off-by: Alex Smith <asmith@feralinteractive.com>
Cc: "17.2 17.3" <mesa-stable@lists.freedesktop.org>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Tested-by: Dieter Nützel <Dieter@nuetzel-hh.de>
The handle type in the case statement is supposed to be VK_EXTERNAL_-
MEMORY_HANDLE_TYPE_DMA_BUF_BIT_EXT.
Fixes: 546e747867 ("radv: Implement VK_EXT_external_memory_dma_buf")
Signed-off-by: Fredrik Höglund <fredrik@kde.org>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
The WSI core code does all the hard work. Just add the wrappers and
turn it on.
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Chad Versace <chadversary@chromium.org>
Now that we have anv_device_init/finish functions, there's no reason to
have the individual driver do any more work than that.
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Chad Versace <chadversary@chromium.org>
This drops the unneeded callbacks struct as well as the queue_get_family
callback we were using before we'd pulled QueuePresent inside.
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Chad Versace <chadversary@chromium.org>
This lets us move wsi_interface to wsi_common_private.h
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Chad Versace <chadversary@chromium.org>
Unfortunately, due to the fact that AcquireNextImage does not take a
queue, the ANV trick for triggering the fence won't work in general. We
leave dealing with the fence up to the caller for now.
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Chad Versace <chadversary@chromium.org>
v2 (Jason Ekstrand):
- Rebase
- Alter the names of the helpers to better match the vulkan entrypoints
- Use the helpers in anv
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Chad Versace <chadversary@chromium.org>
This moves bits out of all four corners (anv, radv, x11, wayland) and
into the wsi common code. We also switch to using an outarray to ensure
we get our return code right.
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Chad Versace <chadversary@chromium.org>
v2 (Jason Ekstrand):
- Better comit message
- Rebase
- Re-indent to follow wsi_common style
- Drop the unneeded _swapchain from the newly added helper
- Make the clone more true to the original (as per the rebase)
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Chad Versace <chadversary@chromium.org>
Just check if image has scanout flag set
v2 (Jason Ekstrand):
- Rebase
- Also drop the now unused radv_mem_flag_bits enum
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Chad Versace <chadversary@chromium.org>
This uses the mock extension created in a previous commit to tell the
driver that the image it's just been asked to create is, in fact, a
window system image with whatever assumptions that implies. There was a
lot of redundant code between the two drivers to do basically exactly
the same thing.
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Chad Versace <chadversary@chromium.org>
We need it to happen after memory type setup so that we can query memory
types in wsi_device_init.
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Chad Versace <chadversary@chromium.org>
This gives the opportunity to collect some function pointers if we'd
like which will be very useful in future.
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Chad Versace <chadversary@chromium.org>
This is used to hold information about the allocated image, rather than
an ever-growing function argument list.
v2 (Jason Ekstrand):
- Rename wsi_image_base to wsi_image
Signed-off-by: Daniel Stone <daniels@collabora.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Chad Versace <chadversary@chromium.org>
This just seems cleaner, and we may expand this in future.
Signed-off-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
This is a bit more general and lets us pass additional options into the
spirv_to_nir pass beyond what capabilities we support.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
"vkAllocateCommandBuffers can be used to create multiple command
buffers. If the creation of any of those command buffers fails, the
implementation must destroy all successfully created command buffer
objects from this command, set all entries of the pCommandBuffers
array to NULL and return the error."
This has been suggested by gabriel@system.is.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
It's really annoying and this pollutes the output especially
when a bunch of non-meta shaders are compiled.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
The idea is ported from RadeonSI, but using 512x512 instead of
256x256 seems slightly better. This improves dota2 performance
by +2%.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Tested-by: Dieter Nützel <Dieter@nuetzel-hh.de>
The state no longer exists on GFX9.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Fix a bunch of labels indicating when registers were added/removed
and normalize the SI-class GRBM_GFX_INDEX.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
The next commit will reduce the size even more.
v2: typecast to uint64_t manually
v3: add more typecasts, add asserts
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
This fixes two CTS regressions:
- dEQP-VK.api.object_management.alloc_callback_fail_multiple.command_buffer_primary
- dEQP-VK.api.object_management.alloc_callback_fail_multiple.command_buffer_secondary
These two tests are part the mustpass lists, so presumably they
are correct and my change was wrong.
This reverts commit 0f68208f1d.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
This happens when all BOs have the RADEON_FLAG_NO_INTERPROCESS_SHARING
(DRM version >= 3.23) flag set. This flag is mainly used for reducing
overhead on the userspace side because we don't have to put those BOs
inside the list.
Though, if the driver tries to create a list with 0 buffers inside it,
libdrm returns -EINVAL and the app just crashes.
This fixes a bunch of CTS dEQP-VK.sparse_resources.* fails (~100).
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
This allows to update them with only one memcpy().
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
As per the spec, the query identified by queryPool and query
must currently be active. Applications have to call vkCmdBeginQuery()
before, and thus the query pool BO will already be in the list.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Similar to how the driver sets the depth clear regs after a
fast depth clear. Most of the time, this will copy a 32-bit reg
instead of a 64-bit reg.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
For the fast path, radv_fill_buffer() ensures that the BO is
already in the list. For the slow path, the depth surface is
part of the framebuffer which means the BO is added to the list
when the framebuffer is emitted.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
aspects can't be zero and there is an assertion that ensures
it's not in emit_clear().
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
This reverts two of the vk_error changes:
reporting unsupported format is common,
and testing non-amdgpu drivers and ignoring them is also common.
Fixes: cd64a4f70 (radv: use vk_error() everywhere an error is returned)
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
This should reduce the overhead of adding a BO to the current
list, especially when the list is huge. Also, when a new pipeline
is bound, we only need to update the descriptor, the buffer objects
should already be in the list.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
I don't think we will need a 64-bit unsigned integer for the
dirty flags in the future, and there is still 20 bits left.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
VkDeviceQueueCreateInfo::queueFamilyIndex is an unsigned 32-bit
integer.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
radv_fill_buffer() ensures that the image BO is added to the list.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
radv_fill_buffer() ensures that the image BO is added to the list.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
It seems safe and it improves performance by +4% (73->76).
A drirc based solution is not what we want for now, keep it
simple and improve later if it's really needed.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Otherwise we can leak the old syncobj.
Fixes: eaa56eab6d "radv: initial support for shared semaphores (v2)"
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
This should not be needed, if the allocation fails an error is
returned and the host should handle it.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
All radv_fence fields are initialized here.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
For consistency and it might help for debugging purposes.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Instead just dirty RADV_CMD_DIRTY_FRAMEBUFFER and it will be
re-emitted if necessary before the next draw.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Just after the vertex shader.
This seems to give a minor boost for, at least, Serious Sam
Fusion 2017 and Dawn of War 3. I don't see any real impacts
with The Talos Principle.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Will be used for VBO descriptors prefetching.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
For consistency because this function will also prefetch VBO
descriptors.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
This register is the same on all gpus so far, so emit it in one
place and also for the pre-gfx9 gpus set the value in the pipeline
creation.
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Signed-off-by: Dave Airlie <airlied@redhat.com>
This moves some calculations of register values into the pipeline
construction, it saves looking at outinfo in the cmd buffer emit.
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Signed-off-by: Dave Airlie <airlied@redhat.com>
This uses C++11 initializer lists.
I just overwrote all Mesa files with internal addrlib and discarded
hunks that we should probably keep, but I might have missed something.
The code depending on ADDR_AM_BUILD is removed. We can add it back next
time if needed.
Acked-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
We can avoid adding the buffer in the non-local case, this will
avoid all the overhead of the indirect call.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
The next patch will try and avoid calling the indirect function.
v2: add a missing conversion.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
The function that calls us has just added the buffer to the
list already, no need to try and add it again.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
There's no point recalculating these the whole time on descriptor
emission, just store them at pipeline creation.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
It appears the latest dota2 vulkan uses this,
and we get a hang in VR mode without it.
v2: remove finishme I left in after finishing.
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Andres Rodriguez <andresx7@gmail.com>
Cc: "17.2 17.3" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Instead of storing all the pointers and zeroing them all out,
just store a valid bitmask in the state. This also moves
the CmdBindPipeline path down the cpu usage path for the
multithreading demo as it no longer has to traverse MAX_SETS
to find the active descriptor sets.
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Signed-off-by: Dave Airlie <airlied@redhat.com>
This isn't required to be cleared, since buffers are only linked
by vertex elements, so if elements are clear then no buffers
should be referenced.
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Signed-off-by: Dave Airlie <airlied@redhat.com>
This just removes a hole in the cmd_state and packs some bools
together.
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Signed-off-by: Dave Airlie <airlied@redhat.com>
If we allocate attachments in the begin command buffer due to the
render pass continue bit, we were leaking them.
Since renderpasses inside a cmd buffer malloc/free these properly,
and set to NULL, we just need to call free at end.
Fixes a memory leak with multithreading demo.
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Cc: "17.2 17.3" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Dave Airlie <airlied@redhat.com>
uint32_t data[MAX_SETS * 2] = {}; was getting executed before
the exit and took significant amounts of time. By having the
check outside the function, we skip the execution of the clear.
Reviewed-by: Dave Airlie <airlied@redhat.com>
The vram_list linked list resulted in lots of pointer chasing.
Replacing this with an array instead improves descriptor set
allocation CPU usage by 3x at least (when also considering the free),
because it had to iterate through 300-400 sets on average.
Not a huge improvement as the pre-improvement CPU usage was only
about 2.3% in the busiest thread.
Reviewed-by: Dave Airlie <airlied@redhat.com>
This should reduce the time where compute units are idle, mainly
for meta operations because they use a bunch of compute shaders.
This seems to have a really minor positive effect for Talos, at least.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
It confuses CTS. This pregenerates the heap info into the
physical device, so we can use it for translating contiguous
indices into our "standard" ones.
This also makes the WSI a bit smarter in case the first preferred
heap does not exist.
Reviewed-by: Dave Airlie <airlied@redhat.com>
CC: <mesa-stable@lists.freedesktop.org>
This should make sure we don't treat exports buffers as local
bos.
Fixes: a639d40f13 (radv: add support for local bos. (v3))
Tested-by: Andres Rodriguez <andresx7@gmail.com>
Reviewed-by: Andres Rodriguez <andresx7@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Fixes regression in:
dEQP-VK.api.object_management.alloc_callback_fail.graphics_pipeline
Fixes: 1e84e53712 "radv: add cache items to in memory cache when reading from disk"
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=103513
Fixes: de88979413 ("radv: Implement VK_AMD_shader_info")
Signed-off-by: Alex Smith <asmith@feralinteractive.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Otherwise we will leak them, load duplicates from disk rather
than memory and never write items loaded from disk to the apps
pipeline cache.
Fixes: fd24be134f 'radv: make use of on-disk cache'
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
This allows an app to query shader statistics and get a disassembly of
a shader. RenderDoc git has support for it, so this allows you to view
shader disassembly from a capture.
When this extension is enabled on a device (or when tracing), we now
disable pipeline caching, since we don't get the shader debug info when
we retrieve cached shaders.
v2: Improvements to resource usage reporting
v3: Disassembly string must be null terminated (string_buffer's length
does not include the terminator)
v4: Fixed LDS reporting. (Bas)
Signed-off-by: Alex Smith <asmith@feralinteractive.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Serious Sam Fusion 2017 uses a huge number of occlusion queries,
and the allocated query pool buffer is greater than 4096 bytes.
This slightly improves performance (tested in Ultra) from
117.2 FPS to 119.7 FPS (~+2%) on my RX480.
This also improves Talos, from 69 FPS to 72/73 FPS (~+5%).
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Only needed when the CS path is used.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
This uses the new kernel interfaces for reduced cs overhead,
We only set the local flag for memory allocations that don't have
a dedicated allocation and ones that aren't imports.
v2: add to all the internal buffer creation paths.
v3: missed some command submission paths, handle 0/empty bo lists.
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Signed-off-by: Dave Airlie <airlied@redhat.com>
When binding a new pipeline, we applied all dynamic states
without checking if they really need to be re-emitted. This
doesn't seem to be useful for the meta operations because only
the viewports/scissors are updated.
This should reduce the number of commands added to the IB
when a new graphics pipeline is bound.
Also, rename radv_dynamic_state_copy() to radv_bind_dynamic_state()
and set the dirty flags directly there.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
The depth bounds test values are either set at pipeline
creation or dynamically using vkCmdSetDepthBounds().
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
The beginning of the end for the shader keys. Not entirely sure
what I'm going to replace them with for the compiler though, so this
is the first step.
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
It's still printed after linking, but it makes more sense to
have SPIRV->NIR->LLVM IR->ASM.
Fixes: f0a2bbd1a4 (radv: move nir print after linking is done)
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Not sure how useful this is, but it makes it more consistent.
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Cc: "17.3" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Dave Airlie <airlied@redhat.com>
For certain buffer meta ops we can use the CP or a compute shader,
we should use a define to rather than hardcoding 4096, allows
for easier testing and more consistency.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Signed-off-by: Dave Airlie <airlied@redhat.com>
radeonsi only emits these when dfsm is enabled, so for now
just hinge them on a flag we never set.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Signed-off-by: Dave Airlie <airlied@redhat.com>
We now have linking optimisations so we want to delay dumping the
nir until after these are complete.
Fixes: 06f05040eb (radv: Link shaders)
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
The IR is reused in different pipeline combinations so we need
to clone it to avoid link time optimistaions messing up the
original copy.
Fixes: 06f05040eb (radv: Link shaders)
Reviewed-by: Dave Airlie <airlied@redhat.com>
This was the actual cause of GPU hangs fixed by 0fdd531457 ("radv:
Fix pipeline cache locking issues"), since multiple threads would end
up trying to create the variants for a single entry.
Now that we're locking around the whole of this function, this isn't
really necessary (we either create all or none of the variants), but
fix this anyway in case things change later.
Signed-off-by: Alex Smith <asmith@feralinteractive.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
CC: 17.3 <mesa-stable@lists.freedesktop.org>
Since it also uses the output vector before writing to memory.
Fixes: e38685cc62 'Revert "radv: disable support for VEGA for now."'
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Otherwise we just need to write them to the tf ring.
this seems to improve the tessellation demo on Bonarie
~2190->~2230 fps
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Due to LLVM bugs. Fixes a bunch of dEQP-VK.glsl.indexing.*
tests.
Fixes: e38685cc62 'Revert "radv: disable support for VEGA for now."'
Reviewed-by: Dave Airlie <airlied@redhat.com>
When the gs_copy_shader is NULL (due to an incomplete cache), but
the main shaders are found, we still do the nir, but we shouldn't
compile the shaders again. For merged shaders we should also account
for the missing shaders.
Fixes: ce03c119ce 'radv: Add code to compile merged shaders.'
Reviewed-by: Dave Airlie <airlied@redhat.com>
With merged shaders the vertex shader may not exist. This got in
because the offending patch was written before merged shaders were
upstream, but committed after.
Fixes: 75dfab24a2 'radv: refactor indirect draws with radv_draw_info'
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Otherwise for non-indexed draws we set and immediately unset
RADV_CMD_DIRTY_INDEX_BUFFER. As all the set functions should
clear their own bit, this is unnecessary.
Fixes: 341529dbee 'radv: use optimal packet order for draws'
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
As they were emitted after the new pipeline, the changed pipeline
detection was not working anymore.
Fixes: 341529dbee 'radv: use optimal packet order for draws'
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Need to lock around the whole process of retrieving cached shaders, and
around GetPipelineCacheData.
This fixes GPU hangs observed when creating multiple pipelines in
parallel, which appeared to be due to invalid shader code being pulled
from the cache.
Signed-off-by: Alex Smith <asmith@feralinteractive.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Implicit sync kicks in when a buffer is used by two different amdgpu
contexts simultaneously. Jobs that use explicit synchronization
mechanisms end up needlessly waiting to be scheduled for long periods
of time in order to achieve serialized execution.
This patch disables implicit synchronization for all radv allocations
except for wsi bos. The only systems that require implicit
synchronization are DRI2/3 and PRIME.
v2: mark wsi bos as RADV_MEM_IMPLICIT_SYNC
v3: Add drm version check (Bas)
Signed-off-by: Andres Rodriguez <andresx7@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
This allows us to pass extra parameters to the memory allocation
operation that are not defined in the vulkan spec. This is useful for
internal usage.
Signed-off-by: Andres Rodriguez <andresx7@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
This patch helps lower high priority compute latency. Found by
bisecting a perf regression on computeparticles with high priority
compute queues enabled.
Reverting this micro-optimization doesn't seem to have any negative
effect on performance on Dota2 or ssao.
Signed-off-by: Andres Rodriguez <andresx7@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
This extension allows the caller to change a queue's system wide
priority. This is useful for applications with specific
latency constraints.
Signed-off-by: Andres Rodriguez <andresx7@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
When WAVE_LIMIT is set, a submission will opt-in for SPI based resource
scheduling. Because this mechanism is cooperative, we must ensure that
all submissions have this field set, otherwise they will bypass resource
arbitration.
We always hardcode the field to its maximum value, instead of attempting
to calculate an approximate usage. In testing, there were no benefits to
using anything other than the maximum.
Signed-off-by: Andres Rodriguez <andresx7@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Ported from RadeonSI. The time where shaders are idle should
be shorter now. This can give a little boost, like +6% with
the dynamicubo Vulkan demo.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
The Vulkan specification says:
"... an execution dependency with only VK_PIPELINE_STAGE_TOP_OF_-
PIPE_BIT in the source stage mask will effectively not wait for
any prior commands to complete."
Signed-off-by: Fredrik Höglund <fredrik@kde.org>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Indirect draws with a count buffer will be refactored in a
separate patch.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Similar to the dispatch compute logic but for draw calls. For
convenience, indirect draws will be converted in a separate
patch.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Only on CIK and later. We should only update VGT_INDEX_TYPE but
it seems easier to re-emit all the index buffer packets.
Fixes: 966d66f28f (radv: do not re-emit the index buffer for every draw call)
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
This will allow us to fix the VGT_INDEX_TYPE issue properly.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
To be consistent with the emit function name.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
CLEAR_STATE will initialize DB_COUNT_CONTROL to 0 for CIK+.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
This fixes some access to the tess eval shader when it's combined
with geometry on gfx9.
This is a review of Bas's commit:
radv: Prevent crashing by accessing TES for VGT reuse depth.
Signed-off-by: Dave Airlie <airlied@redhat.com>
This struct used to rely on being in a union, it isn't anymore,
so we have to pick the correct outinfo struct now.
This should fix a regression since the union became a struct.
dEQP-VK.tessellation.geometry_interaction.point_size.vertex_set_geometry_set
Fixes: 6078a3bd51 (ac/nir: Allow ac_shader_variant_info to contain info about multiple stages.)
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Looking at shader traces I noticed some registers were missing,
one of them was being eaten by the wrong clear state length.
Fixes: 4f42ea4dc (radv: use CLEAR_STATE for initializing some registers)
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
It looks the original indirect mask was probably copied from
ANV.
Sascha Willems demo results:
tessellation ~4000 -> ~4200 fps
V2: continue lowering local indirects due to llvm deficiencies.
Tested-by: Alex Smith <asmith@feralinteractive.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Move it to radv_cmd_buffer_flush_state() because if
rasterizerDiscardEnable is true, the flags are not cleared.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
It can only be changed when CmdBindIndexBuffer() is called
or when a secondary buffer is used. Though not always, but
let's re-emit the packets in this situation for now.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
This saves few CPU cycles when CmdDrawIndexed() is used a lot.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Fixes: 7d45d22fdd ("radv: switch to using radv_create_shaders()")
Signed-off-by: Alex Smith <asmith@feralinteractive.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Fixes a crash while generating a hang report.
Fixes: 7d45d22fdd ("radv: switch to using radv_create_shaders()")
Signed-off-by: Alex Smith <asmith@feralinteractive.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>