Previously, we were crawling through the anv_cmd_buffer datastructure to
pull out batch buffers and things. This meant that every time something in
anv_cmd_buffer changed, we broke aub dumping. However, aub dumping should
just dump the stuff the kernel knows about so we really don't need to be
crawling driver internals.
This used to happen magically in cmd_buffer_new_surface_state_bo. However,
according to Ken, STATE_BASE_ADDRESS is very gen-specific so we really
shouldn't have it in the generic data-structure code.
The CTS passes in NULL names right now. It's not too hard to support that
as just "main". With this, and a patch to vulkancts, we now pass all 6
tests.
Jason started the task by creating anv_cmd_buffer.c and anv_cmd_emit.c.
This patch finishes the task by renaming all other files except
gen*_pack.h and glsl_scraper.py.
This completes the FINISHME to trim unneeded data from anv_buffer_view.
A VkExtent3D doesn't make sense for a VkBufferView.
So remove the member anv_surface_view::extent, and push it up to the two
objects that actually need it, anv_image_view and anv_attachment_view.
This removes nearly all the remaining raw Anvil<->Vulkan casts from the
C source files. (File compiler.cpp still contains many raw casts, and
I plan on ignoring that).
As far as I can tell, the only remaining raw casts are:
anv_attachment_view -> anv_depth_stencil_view
anv_attachment_view -> anv_color_attachment_view
The GLSL layer above is still hacky, so we're really just moving the
hack into GLSL-to-NIR. I'd rather not go all the way and make GLSL
support the Vulkan binding model too, since presumably we'll be
switching to SPIR-V exclusively, and so working on proper GLSL support
will be a waste of time. For now, doing this keeps it working as we add
SPIR-V->NIR support though.
The Vulkan spec does not specify that the free function provided to
CreateInstance must handle NULL properly so we do it in the wrapper. If
this ever changes in the spec, we can delete the extra 2 lines.
While updating vkDestroyObject, I discovered that vkDestroyPass reliably
crashes. That hasn't been an issue yet, though, because it is never
called.
In vkCreateRenderPass:
- Don't allocate empty attachment arrays.
- Ensure that pointers to empty attachment arrays are NULL.
- Store VkRenderPassCreateInfo::subpassCount as
anv_render_pass::subpass_count.
In vkDestroyRenderPass:
- Fix loop bounds: s/attachment_count/subpass_count/
- Don't call anv_device_free on null pointers.
