xiaomi-sm8450/android_kernel_xiaomi_sm8450
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

drm/nouveau/fifo/gk104-: require explicit runlist selection for channel allocation

Browse Source 
We didn't used to be aware that runlist/engine IDs weren't the same thing,
or that there was such variability in configuration between GPUs.

By exposing this information to a client, and giving it explicit control
of which runlist it's allocating a channel on, we're able to make better
choices.

The immediate effect of this is that on GPUs where CE0 is the "GRCE", we
will now be allocating a copy engine running asynchronously to GR for BO
migrations - as intended.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
This commit is contained in:
Ben Skeggs
2018-05-08 20:39:46 +10:00
parent cc36205085
commit a7cf01809b
5 changed files with 43 additions and 105 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
drivers/gpu/drm/nouveau/nouveau_chan.c
View File

@@ -214,7 +214,7 @@ nouveau_channel_prep(struct nouveau_drm *drm, struct nvif_device *device,
static int
nouveau_channel_ind(struct nouveau_drm *drm, struct nvif_device *device,
u32 engine, struct nouveau_channel **pchan)
u64 runlist, struct nouveau_channel **pchan)
{
struct nouveau_cli *cli = (void *)device->object.client;
static const u16 oclasses[] = { PASCAL_CHANNEL_GPFIFO_A,
@@ -245,9 +245,9 @@ nouveau_channel_ind(struct nouveau_drm *drm, struct nvif_device *device,
do {
if (oclass[0] >= KEPLER_CHANNEL_GPFIFO_A) {
args.kepler.version = 0;
args.kepler.engines = engine;
args.kepler.ilength = 0x02000;
args.kepler.ioffset = 0x10000 + chan->push.addr;
args.kepler.runlist = runlist;
args.kepler.vmm = nvif_handle(&cli->vmm.vmm.object);
size = sizeof(args.kepler);
} else