nv-tegra-mirror/linux-nvgpu
1
0
Fork 0
mirror of git://nv-tegra.nvidia.com/linux-nvgpu.git synced 2025-12-24 10:34:43 +03:00
Code Issues Packages Projects Releases Wiki Activity

gpu: nvgpu: Optimize validate_fixed_buffer

Browse Source 
Function validate_fixed_buffer used to do a linear search for
collision detection of already mapped buffers. Optimize this by doing
a nice logarithmic search instead.

Change-Id: Ifbf2ec015741d44883da27bc6f8cc090c48da145
Signed-off-by: Sami Kiminki <skiminki@nvidia.com>
Reviewed-on: http://git-master/r/739682
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
This commit is contained in:
Sami Kiminki
2015-05-06 17:50:34 +03:00
committed by Ishan Mittal
parent e7ea596363
commit 6a5cc11171
1 changed files with 29 additions and 13 deletions
Download Patch File Download Diff File Expand all files Collapse all files

42
drivers/gpu/nvgpu/gk20a/mm_gk20a.c
View File

@@ -1054,6 +1054,27 @@ static struct mapped_buffer_node *find_mapped_buffer_range_locked(
return NULL;
}
/* find the first mapped buffer with GPU VA less than addr */
static struct mapped_buffer_node *find_mapped_buffer_less_than_locked(
struct rb_root *root, u64 addr)
{
struct rb_node *node = root->rb_node;
struct mapped_buffer_node *ret = NULL;
while (node) {
struct mapped_buffer_node *mapped_buffer =
container_of(node, struct mapped_buffer_node, node);
if (mapped_buffer->addr >= addr)
node = node->rb_left;
else {
ret = mapped_buffer;
node = node->rb_right;
}
}
return ret;
}
#define BFR_ATTRS (sizeof(nvmap_bfr_param)/sizeof(nvmap_bfr_param[0]))
struct buffer_attrs {
@@ -1165,19 +1186,14 @@ static int validate_fixed_buffer(struct vm_gk20a *vm,
return -EINVAL;
}
/* check that this mappings does not collide with existing
* mappings by checking the overlapping area between the current
* buffer and all other mapped buffers */
list_for_each_entry(buffer,
&va_node->va_buffers_list, va_buffers_list) {
s64 begin = max(buffer->addr, map_offset);
s64 end = min(buffer->addr +
buffer->size, map_offset + map_size);
if (end - begin > 0) {
gk20a_warn(dev, "overlapping buffer map requested");
return -EINVAL;
}
/* check that this mapping does not collide with existing
* mappings by checking the buffer with the highest GPU VA
* that is less than our buffer end */
buffer = find_mapped_buffer_less_than_locked(
&vm->mapped_buffers, map_offset + map_size);
if (buffer && buffer->addr + buffer->size > map_offset) {
gk20a_warn(dev, "overlapping buffer map requested");
return -EINVAL;
}
return 0;