Age | Commit message (Collapse) | Author |
|
r600g is up to a point where all small CPU cycle matter and pb* turn
high on profile. It's mostly because pb try to be generic and thus
trigger unecessary check for r600g driver. To avoid having too much
abstraction & too much depth in the call embedded everythings into
r600_bo. Make code simpler & faster. The performance win highly depend
on the CPU & application considered being more important on slower CPU
and marginal/unoticeable on faster one.
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
|
|
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
|
|
|
|
This opens the question of what interface the winsys layer should
really have for talking about these concepts.
For now I'm using the existing gallium resource usage concept, but
there is no reason not use terms closer to what the hardware
understands - eg. the domains themselves.
|
|
That way assert(map_count >= 0) can actually fail when we screwed up.
Signed-off-by: Tilman Sauerbeck <tilman@code-monkey.de>
|
|
Signed-off-by: Tilman Sauerbeck <tilman@code-monkey.de>
|
|
we need to know if the back is tiled so we can blit from it properly.
|
|
|
|
Avoid having object ending up in dead list of dirty object.
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
|
|
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
|
|
Since flush rework there could be only one relocation per
register in a block.
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
|
|
Got a speed up by tracking the dirty blocks in a seperate list instead of looping through all blocks. This version should work with block that get their dirty state disabled again and I added a dirty check during the flush as some blocks were already dirty.
|
|
Flush read cache before writting register. Track flushing inside
of a same cs and avoid reflushing same bo if not necessary. Allmost
properly force flush if bo rendered too and then use as a texture
in same cs (missing pipeline flush dunno if it's needed or not).
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
|
|
Allow fast lookup of relocation information & id which
was a CPU time consumming operation.
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
|
|
since we plan on using dx10 constant buffers everywhere.
|
|
we don't use this since constant buffers are now being used on all gpus.
|
|
When we go to do a lot of bos in one draw like constant bufs we need
to avoid bouncing off the busy ioctl, this mitigates by backing off
on busy bos for a short amount of times.
|
|
this just keeps a list of bos submitted together, and uses them to decide
bo busy state for the whole group.
|
|
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
|
|
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
|
|
|
|
Save a lot of call into the kernel and thus improve performances.
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
|
|
This avoid to overcount the number of dwords we need and
thus avoid maximazation of cs buffer use.
Signed-off-by: Jerome Glisse <jglisse@redhat.com
|
|
mainly remove 2 suffix from function names
|
|
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
|
|
Instead of creating group of register use a hash table
to lookup into which block each register belongs. This
simplify code a bit.
Signed-off-by: Jerome Glisse <jglisse@redhat.com
|
|
|
|
block->reg point to register value not block->pm4 which point
to packet.
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
|
|
When occlusion query are running we want to have accurate
fragment count thus disable any early culling optimization
GPU has.
Based on work from Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
|
|
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
|
|
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
|
|
So texture code can be shared btw new state design
& old one.
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
|
|
Winsys context build a list of register block a register block is
a set of consecutive register that will be emited together in the
same pm4 packet (the various r600_block* are there to provide basic
grouping that try to take advantage of states that are linked together)
Some consecutive register are emited each in a different block,
for instance the various cb[0-7]_base. At winsys context creation,
the list of block is created & an index into the list of block. So
to find into which block a register is in you simply use the register
offset and lookup the block index. Block are grouped together into
group which are the various pkt3 group of config, context, resource,
Pipe state build a list of register each state want to modify,
beside register value it also give a register mask so only subpart
of a register can be updated by a given pipe state (the oring is
in the winsys) There is no prebuild register list or define for
each pipe state. Once pipe state are built they are bound to
the winsys context.
Each of this functions will go through the list of register and
will find into which block each reg falls and will update the
value of the block with proper masking (vs/ps resource/constant
are specialized variant with somewhat limited capabilities).
Each block modified by r600_context_pipe_state_set* is marked as
dirty and we update a count of dwords needed to emit all dirty
state so far.
r600_context_pipe_state_set* should be call only when pipe context
change some of the state (thus when pipe bind state or set state)
Then to draw primitive you make a call to r600_context_draw
void r600_context_draw(struct r600_context *ctx, struct r600_draw *draw)
It will check if there is enough dwords in current cs buffer and
if not will flush. Once there is enough room it will copy packet
from dirty block and then add the draw packet3 to initiate the draw.
The flush will send the current cs, reset the count of dwords to
0 and remark all states that are enabled as dirty and recompute
the number of dwords needed to send the current context.
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
|