[PATCH] drm/radeon: set VM base addr using the PFP v2

[Christian König]

Am 30.07.2014 um 19:24 schrieb Jerome Glisse:
> On Wed, Jul 30, 2014 at 11:54 AM, Alex Deucher <alexdeucher at gmail.com 
> <mailto:alexdeucher at gmail.com>> wrote:
>
>     On Wed, Jul 30, 2014 at 11:18 AM, Christian König
>     <deathsimple at vodafone.de <mailto:deathsimple at vodafone.de>> wrote:
>     > From: Christian König <christian.koenig at amd.com
>     <mailto:christian.koenig at amd.com>>
>     >
>     > Seems to make VM flushes more stable on SI and CIK.
>     >
>     > v2: only use the PFP on the GFX ring on CIK
>     >
>     > Signed-off-by: Christian König <christian.koenig at amd.com
>     <mailto:christian.koenig at amd.com>>
>     > Cc: stable at vger.kernel.org <mailto:stable at vger.kernel.org>
>
>     Applied to my 3.17 tree.  We should use PFP whenever possible for
>     all packets.
>
>
> Can all this be explained with more words ? What are the choice ? What 
> are the difference ? What we gain ?

The PFP is responsible of fetching the data the CP executes from memory, 
classify them and if necessary forward them to the ME. The ME is the 
part where normally all the juicy stuff happens, e.g. register writes, 
most type3 packet execution etc...

What we did before was setting the VM base address with the ME instead 
of the PFP, but since the PFP needs to fetch the IB data we inserted an 
"Wait for ME" packet into the command stream to let the ME catch up to 
the PFP before proceeding.

The problem is that this is sometimes not enough, e.g. on CIK it could 
happen that the IB read goes out while the the page table base address 
wasn't updated yet with the obviously bad consequences.

Compute rings don't have a PFP on CIK so this won't work there.

Regards,
Christian.

>
>     Alex
>
>     > ---
>     >  drivers/gpu/drm/radeon/cik.c | 5 +++--
>     >  drivers/gpu/drm/radeon/si.c  | 2 +-
>     >  2 files changed, 4 insertions(+), 3 deletions(-)
>     >
>     > diff --git a/drivers/gpu/drm/radeon/cik.c
>     b/drivers/gpu/drm/radeon/cik.c
>     > index cc1f02f..db53616 100644
>     > --- a/drivers/gpu/drm/radeon/cik.c
>     > +++ b/drivers/gpu/drm/radeon/cik.c
>     > @@ -5641,12 +5641,13 @@ static void cik_vm_decode_fault(struct
>     radeon_device *rdev,
>     >  void cik_vm_flush(struct radeon_device *rdev, int ridx, struct
>     radeon_vm *vm)
>     >  {
>     >         struct radeon_ring *ring = &rdev->ring[ridx];
>     > +       int usepfp = (ridx == RADEON_RING_TYPE_GFX_INDEX);
>     >
>     >         if (vm == NULL)
>     >                 return;
>     >
>     >         radeon_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3));
>     > -       radeon_ring_write(ring, (WRITE_DATA_ENGINE_SEL(0) |
>     > +       radeon_ring_write(ring, (WRITE_DATA_ENGINE_SEL(usepfp) |
>     >  WRITE_DATA_DST_SEL(0)));
>     >         if (vm->id < 8) {
>     >                 radeon_ring_write(ring,
>     > @@ -5696,7 +5697,7 @@ void cik_vm_flush(struct radeon_device
>     *rdev, int ridx, struct radeon_vm *vm)
>     >         radeon_ring_write(ring, 1 << vm->id);
>     >
>     >         /* compute doesn't have PFP */
>     > -       if (ridx == RADEON_RING_TYPE_GFX_INDEX) {
>     > +       if (usepfp) {
>     >                 /* sync PFP to ME, otherwise we might get
>     invalid PFP reads */
>     >                 radeon_ring_write(ring,
>     PACKET3(PACKET3_PFP_SYNC_ME, 0));
>     >                 radeon_ring_write(ring, 0x0);
>     > diff --git a/drivers/gpu/drm/radeon/si.c
>     b/drivers/gpu/drm/radeon/si.c
>     > index 9e854fd..f87d82a 100644
>     > --- a/drivers/gpu/drm/radeon/si.c
>     > +++ b/drivers/gpu/drm/radeon/si.c
>     > @@ -4815,7 +4815,7 @@ void si_vm_flush(struct radeon_device
>     *rdev, int ridx, struct radeon_vm *vm)
>     >
>     >         /* write new base address */
>     >         radeon_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3));
>     > -       radeon_ring_write(ring, (WRITE_DATA_ENGINE_SEL(0) |
>     > +       radeon_ring_write(ring, (WRITE_DATA_ENGINE_SEL(1) |
>     >  WRITE_DATA_DST_SEL(0)));
>     >
>     >         if (vm->id < 8) {
>     > --
>     > 1.9.1
>     >
>     _______________________________________________
>     dri-devel mailing list
>     dri-devel at lists.freedesktop.org
>     <mailto:dri-devel at lists.freedesktop.org>
>     http://lists.freedesktop.org/mailman/listinfo/dri-devel
>
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freedesktop.org/archives/dri-devel/attachments/20140730/1cfc9138/attachment.html>