r300g: rework command submission and resource space checking

The motivation behind this rework is to get some speed by reducing
CPU overhead. The performance increase depends on many factors,
but it's measurable (I think it's about 10% increase in Torcs).

This commit replaces libdrm's radeon_cs_gem with our own implemention.
It's optimized specifically for r300g, but r600g could use it as well.
Reloc writes and space checking are faster and simpler than their
counterparts in libdrm (the time complexity of all the functions
is O(1) in nearly all scenarios, thanks to hashing).
(libdrm's radeon_bo_gem is still being used in the driver.)

It works like this:

cs_add_reloc(cs, buf, read_domain, write_domain) adds a new relocation and
also adds the size of 'buf' to the used_gart and used_vram winsys variables
based on the domains, which are simply or'd for the accounting purposes.
The adding is skipped if the reloc is already present in the list, but it
accounts any newly-referenced domains.

cs_validate is then called, which just checks:
    used_vram/gart < vram/gart_size * 0.8
The 0.8 number allows for some memory fragmentation. If the validation
fails, the pipe driver flushes CS and tries do the validation again,
i.e. it validates only that one operation. If it fails again, it drops
the operation on the floor and prints some nasty message to stderr.

cs_write_reloc(cs, buf) just writes a reloc that has been added using
cs_add_reloc. The read_domain and write_domain parameters have been removed,
because we already specify them in cs_add_reloc.

The space checking has been tested by putting small values in vram/gart_size
variables.

1 files changed, 0 insertions, 99 deletions

diff --git a/src/gallium/winsys/radeon/drm/radeon_r300.c b/src/gallium/winsys/radeon/drm/radeon_r300.c
index ae7020a063..bacf181b47 100644
--- a/src/gallium/winsys/radeon/drm/radeon_r300.c
+++ b/src/gallium/winsys/radeon/drm/radeon_r300.c
@@ -25,7 +25,6 @@
 #include "util/u_memory.h"
 #include "pipebuffer/pb_bufmgr.h"
 
-#include "radeon_cs_gem.h"
 #include "state_tracker/drm_driver.h"
 
 static unsigned get_pb_usage_from_create_flags(unsigned bind, unsigned usage,
@@ -134,65 +133,6 @@ static boolean radeon_r300_winsys_buffer_get_handle(struct r300_winsys_screen *r
     return radeon_drm_bufmgr_get_handle(_buf, whandle);
 }
 
-static void radeon_r300_winsys_cs_set_flush(struct r300_winsys_cs *rcs,
-                                            void (*flush)(void *),
-                                            void *user)
-{
-    struct radeon_drm_cs *cs = radeon_drm_cs(rcs);
-    cs->flush_cs = flush;
-    cs->flush_data = user;
-    radeon_cs_space_set_flush(cs->cs, flush, user);
-}
-
-static boolean radeon_r300_winsys_cs_validate(struct r300_winsys_cs *rcs)
-{
-    struct radeon_drm_cs *cs = radeon_drm_cs(rcs);
-
-    return radeon_cs_space_check(cs->cs) >= 0;
-}
-
-static void radeon_r300_winsys_cs_reset_buffers(struct r300_winsys_cs *rcs)
-{
-    struct radeon_drm_cs *cs = radeon_drm_cs(rcs);
-    radeon_cs_space_reset_bos(cs->cs);
-}
-
-static void radeon_r300_winsys_cs_flush(struct r300_winsys_cs *rcs)
-{
-    struct radeon_drm_cs *cs = radeon_drm_cs(rcs);
-    int retval;
-
-    /* Don't flush a zero-sized CS. */
-    if (!cs->base.cdw) {
-        return;
-    }
-
-    cs->cs->cdw = cs->base.cdw;
-
-    radeon_drm_bufmgr_flush_maps(cs->ws->kman);
-
-    /* Emit the CS. */
-    retval = radeon_cs_emit(cs->cs);
-    if (retval) {
-        if (debug_get_bool_option("RADEON_DUMP_CS", FALSE)) {
-            fprintf(stderr, "radeon: The kernel rejected CS, dumping...\n");
-            radeon_cs_print(cs->cs, stderr);
-        } else {
-            fprintf(stderr, "radeon: The kernel rejected CS, "
-                            "see dmesg for more information.\n");
-        }
-    }
-
-    /* Reset CS.
-     * Someday, when we care about performance, we should really find a way
-     * to rotate between two or three CS objects so that the GPU can be
-     * spinning through one CS while another one is being filled. */
-    radeon_cs_erase(cs->cs);
-
-    cs->base.buf = cs->cs->packets;
-    cs->base.cdw = cs->cs->cdw;
-}
-
 static uint32_t radeon_get_value(struct r300_winsys_screen *rws,
                                  enum r300_value_id id)
 {
@@ -219,39 +159,6 @@ static uint32_t radeon_get_value(struct r300_winsys_screen *rws,
     return 0;
 }
 
-static struct r300_winsys_cs *radeon_r300_winsys_cs_create(struct r300_winsys_screen *rws)
-{
-    struct radeon_drm_winsys *ws = radeon_drm_winsys(rws);
-    struct radeon_drm_cs *cs = CALLOC_STRUCT(radeon_drm_cs);
-
-    if (!cs)
-        return NULL;
-
-    /* Size limit on IBs is 64 kibibytes. */
-    cs->cs = radeon_cs_create(ws->csm, 1024 * 64 / 4);
-    if (!cs->cs) {
-        FREE(cs);
-        return NULL;
-    }
-
-    radeon_cs_set_limit(cs->cs,
-            RADEON_GEM_DOMAIN_GTT, ws->gart_size);
-    radeon_cs_set_limit(cs->cs,
-            RADEON_GEM_DOMAIN_VRAM, ws->vram_size);
-
-    cs->ws = ws;
-    cs->base.buf = cs->cs->packets;
-    cs->base.cdw = cs->cs->cdw;
-    return &cs->base;
-}
-
-static void radeon_r300_winsys_cs_destroy(struct r300_winsys_cs *rcs)
-{
-    struct radeon_drm_cs *cs = radeon_drm_cs(rcs);
-    radeon_cs_destroy(cs->cs);
-    FREE(cs);
-}
-
 void radeon_winsys_init_functions(struct radeon_drm_winsys *ws)
 {
     ws->base.get_value = radeon_get_value;
@@ -259,10 +166,4 @@ void radeon_winsys_init_functions(struct radeon_drm_winsys *ws)
     ws->base.buffer_reference = radeon_r300_winsys_buffer_reference;
     ws->base.buffer_from_handle = radeon_r300_winsys_buffer_from_handle;
     ws->base.buffer_get_handle = radeon_r300_winsys_buffer_get_handle;
-    ws->base.cs_create = radeon_r300_winsys_cs_create;
-    ws->base.cs_destroy = radeon_r300_winsys_cs_destroy;
-    ws->base.cs_validate = radeon_r300_winsys_cs_validate;
-    ws->base.cs_flush = radeon_r300_winsys_cs_flush;
-    ws->base.cs_reset_buffers = radeon_r300_winsys_cs_reset_buffers;
-    ws->base.cs_set_flush = radeon_r300_winsys_cs_set_flush;
 }