nv-tegra-mirror/linux-nvgpu
1
0
Fork 0
mirror of git://nv-tegra.nvidia.com/linux-nvgpu.git synced 2025-12-22 17:36:20 +03:00
Code Issues Packages Projects Releases Wiki Activity

gpu: nvgpu: Remove usage of KEPLER_C syncpt incr

Browse Source 
Using KEPLER_C for doing sync point increment has side effects.
It adds a SetObject method, which changes channel state that not all
user space accounts for.

Bug 1462255
Bug 1497928
Bug 1559462

Change-Id: I5c422ad8ca94fba15cad9bd232f7a10d94aa0973
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/554478
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
This commit is contained in:
Terje Bergstrom
2014-10-08 08:38:59 +03:00
committed by Dan Willemsen
parent 3e11a4fbb2
commit cb54368670
1 changed files with 20 additions and 44 deletions
Download Patch File Download Diff File Expand all files Collapse all files

64
drivers/gpu/nvgpu/gk20a/channel_sync_gk20a.c
View File

@@ -155,7 +155,7 @@ static void gk20a_channel_syncpt_update(void *priv, int nr_completed)
}
static int __gk20a_channel_syncpt_incr(struct gk20a_channel_sync *s,
bool gfx_class, bool wfi_cmd,
bool wfi_cmd,
bool register_irq,
struct priv_cmd_entry **entry,
struct gk20a_fence **fence)
@@ -182,39 +182,24 @@ static int __gk20a_channel_syncpt_incr(struct gk20a_channel_sync *s,
/* WAR for hw bug 1491360: syncpt needs to be incremented twice */
if (gfx_class) {
WARN_ON(wfi_cmd); /* No sense to use gfx class + wfi. */
/* setobject KEPLER_C */
incr_cmd->ptr[j++] = 0x20010000;
incr_cmd->ptr[j++] = KEPLER_C;
/* syncpt incr */
incr_cmd->ptr[j++] = 0x200100B2;
incr_cmd->ptr[j++] = sp->id |
(0x1 << 20) | (0x1 << 16);
/* syncpt incr */
incr_cmd->ptr[j++] = 0x200100B2;
incr_cmd->ptr[j++] = sp->id |
(0x1 << 20) | (0x1 << 16);
} else {
if (wfi_cmd) {
/* wfi */
incr_cmd->ptr[j++] = 0x2001001E;
/* handle, ignored */
incr_cmd->ptr[j++] = 0x00000000;
}
/* syncpoint_a */
incr_cmd->ptr[j++] = 0x2001001C;
/* payload, ignored */
incr_cmd->ptr[j++] = 0;
/* syncpoint_b */
incr_cmd->ptr[j++] = 0x2001001D;
/* syncpt_id, incr */
incr_cmd->ptr[j++] = (sp->id << 8) | 0x1;
/* syncpoint_b */
incr_cmd->ptr[j++] = 0x2001001D;
/* syncpt_id, incr */
incr_cmd->ptr[j++] = (sp->id << 8) | 0x1;
if (wfi_cmd) {
/* wfi */
incr_cmd->ptr[j++] = 0x2001001E;
/* handle, ignored */
incr_cmd->ptr[j++] = 0x00000000;
}
/* syncpoint_a */
incr_cmd->ptr[j++] = 0x2001001C;
/* payload, ignored */
incr_cmd->ptr[j++] = 0;
/* syncpoint_b */
incr_cmd->ptr[j++] = 0x2001001D;
/* syncpt_id, incr */
incr_cmd->ptr[j++] = (sp->id << 8) | 0x1;
/* syncpoint_b */
incr_cmd->ptr[j++] = 0x2001001D;
/* syncpt_id, incr */
incr_cmd->ptr[j++] = (sp->id << 8) | 0x1;
WARN_ON(j != incr_cmd_size);
thresh = nvhost_syncpt_incr_max_ext(sp->host1x_pdev, sp->id, 2);
@@ -241,7 +226,6 @@ int gk20a_channel_syncpt_incr_wfi(struct gk20a_channel_sync *s,
struct gk20a_fence **fence)
{
return __gk20a_channel_syncpt_incr(s,
false /* use host class */,
true /* wfi */,
false /* no irq handler */,
entry, fence);
@@ -251,12 +235,9 @@ int gk20a_channel_syncpt_incr(struct gk20a_channel_sync *s,
struct priv_cmd_entry **entry,
struct gk20a_fence **fence)
{
struct gk20a_channel_syncpt *sp =
container_of(s, struct gk20a_channel_syncpt, ops);
/* Don't put wfi cmd to this one since we're not returning
* a fence to user space. */
return __gk20a_channel_syncpt_incr(s,
sp->c->obj_class == KEPLER_C /* may use gfx class */,
false /* no wfi */,
true /* register irq */,
entry, fence);
@@ -268,15 +249,10 @@ int gk20a_channel_syncpt_incr_user(struct gk20a_channel_sync *s,
struct gk20a_fence **fence,
bool wfi)
{
struct gk20a_channel_syncpt *sp =
container_of(s, struct gk20a_channel_syncpt, ops);
/* Need to do 'host incr + wfi' or 'gfx incr' since we return the fence
/* Need to do 'wfi + host incr' since we return the fence
* to user space. */
return __gk20a_channel_syncpt_incr(s,
wfi &&
sp->c->obj_class == KEPLER_C /* use gfx class? */,
wfi &&
sp->c->obj_class != KEPLER_C /* wfi if host class */,
wfi,
true /* register irq */,
entry, fence);
}