leandrof/linux
1
0
Fork 0
forked from Minki/linux
Code Issues Pull Requests Packages Projects Releases Wiki Activity

drm: Optimise power-of-two alignments in drm_mm_scan_add_block()

Browse Source 
For power-of-two alignments, we can avoid the 64bit divide and do a
simple bitwise add instead.

v2: s/alignment_mask/remainder_mask/

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-32-chris@chris-wilson.co.uk
This commit is contained in:
Chris Wilson 2016-12-22 08:36:34 +00:00 committed by Daniel Vetter
parent 0b04d474a6
commit 9a956b1548
2 changed files with 9 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
drivers/gpu/drm/drm_mm.c
View File

@ -742,8 +742,12 @@ void drm_mm_scan_init_with_range(struct drm_mm_scan *scan,
scan->mm = mm;
if (alignment <= 1)
alignment = 0;
scan->color = color;
scan->alignment = alignment;
scan->remainder_mask = is_power_of_2(alignment) ? alignment - 1 : 0;
scan->size = size;
scan->flags = flags;
@ -811,7 +815,10 @@ bool drm_mm_scan_add_block(struct drm_mm_scan *scan,
if (scan->alignment) {
u64 rem;
div64_u64_rem(adj_start, scan->alignment, &rem);
if (likely(scan->remainder_mask))
rem = adj_start & scan->remainder_mask;
else
div64_u64_rem(adj_start, scan->alignment, &rem);
if (rem) {
adj_start -= rem;
if (scan->flags != DRM_MM_CREATE_TOP)

1
include/drm/drm_mm.h
View File

@ -110,6 +110,7 @@ struct drm_mm_scan {
u64 size;
u64 alignment;
u64 remainder_mask;
u64 range_start;
u64 range_end;