| Age | Commit message (Collapse) | Author |
|---|
|
This prevents the driver from discarding a buffer when the whole buffer
is mapped for writing, but only a portion is effectively written.
This is a temporary fix, because WRITE shouldn't imply DISCARD.
The full fix implies using PIPE_BUFFER_USAGE_DISCARD, throughout
the code, and will go only into master.
|
|
buffer_flush_mapped_range.
When a buffer was mapped for write and no explicit flush range was provided
the existing semantics were that the whole buffer would be flushed, mostly
for backwards compatability with non map-buffer-range aware code.
However if the buffer was mapped/unmapped with nothing really written --
something that often happens with the vbo -- we were unnecessarily assuming
that the whole buffer was written.
The new PIPE_BUFFER_USAGE_FLUSH_EXPLICIT flag (based from ARB_map_buffer_range
's GL_MAP_FLUSH_EXPLICIT_BIT flag) allows to clearly distinguish the
legacy usage from the nothing written usage.
|
|
Which are slightly different from ARB_map_buffer_range semantics, since
gallium still supports more than one mapping of the same buffer.
|
|
|
|
|
|
The core reference counting code is centralized in p_refcnt.h.
This has some consequences related to struct pipe_buffer:
* The screen member of struct pipe_buffer must be initialized, or
pipe_buffer_reference() will crash trying to destroy a buffer with reference
count 0. u_simple_screen takes care of this, but I may have missed some of
the drivers not using it.
* Except for rare exceptions deep in winsys code, buffers must always be
allocated via pipe_buffer_create() or via screen->*buffer_create() rather
than via winsys->*buffer_create().
|
|
|
|
map/unmap.
This allows the pipe driver to do more efficient DMA upload/downloads, by
avoiding to read/write unneeded/unmodified data.
|
|
Saves code, and will simplify future interface changes.
|
|
|
|
Instead, a new pipe_transfer object has to be created and mapped for
transferring data between the CPU and a texture. This gives the driver more
flexibility for textures in address spaces that aren't CPU accessible.
This is a first pass; softpipe/xlib builds and runs glxgears, but it only shows
a black window. Looks like something's off related to the Z buffer, so the
depth test always fails.
|
|
move it to pipe/internal/p_winsys_screen.h and start converting
the state trackers to the screen usage
|
|
allows the driver to overwrite buffer allocation, first step on the way
to making winsys interface internal to the drivers. state trackers and
the code above it will go through the screen
|
|
This commit is mostly just a cosmetic change that cleans-up the interfaces,
replacing pipe_winsys::surface_* calls by
/**
* Allocate storage for a display target surface.
*
* Often surfaces which are meant to be blitted to the front screen (i.e.,
* display targets) must be allocated with special characteristics, memory
* pools, or obtained directly from the windowing system.
*
* This callback is invoked by the pipe_screenwhen creating a texture marked
* with the PIPE_TEXTURE_USAGE_DISPLAY_TARGET flag to get the underlying
* buffer storage.
*/
struct pipe_buffer *(*surface_buffer_create)(struct pipe_winsys *ws,
unsigned width, unsigned height,
enum pipe_format format,
unsigned usage,
unsigned *stride);
Most drivers were updated but not all were tested. Use the softpipe pipe
driver and the xlib winsys changes as a reference when fixing other drivers.
|
|
|
|
We want to use the pipe_buffer_* inlines everywhere, but a pipe context
is not always available nor is it needed.
|
|
|
|
|
|
|
|
These functions are now per-screen, not per-context.
|
|
This will allow creating textures before a rendering context exists, for example.
Only implemented in i915 driver for now. i915pipe->texture_create() just
dispatches through to the i915screen->texture_create() to avoid state tracker
changes for now.
|
|
Added pipe field to pipe_texture (temporary, see comments).
First step toward context-less texture creation...
|
|
|
|
This is in a separate commit to ensure renames are properly preserved.
|
