[Mesa-dev] [PATCH v4 1/2] intel/ppgtt: memory address alignment

[Sergii Romantsov]

Hello, Kenneth.
Thanks for remarks.
Will update patch soon and also will try to look on 3DSTATE_SO_BUFFER.

On Tue, Aug 14, 2018 at 8:39 PM, Kenneth Graunke <kenneth at whitecape.org>
wrote:

> Hi Sergii,
>
> This patch causes 2,384 failures in CI.  The issue is that we're
> apparently trying to allocate 0 size BOs in some places, which are
> getting rounded up to 4096 with the current code...but with your patch,
> we get ALIGN(0, 4096) == 0, and assert(bo_size) triggers.
>
> We might want to continue rounding up for now.  Additionally, we
> probably ought to fix the callers to stop allocating 0 size BOs.
> It looks like most of them come from the 3DSTATE_SO_BUFFER code,
> where one stream has valid transform feedback info, and the other
> 3 are empty.  Whoops.
>
> Would you like to fix that, or should I?
>
> --Ken
>
> On Tuesday, August 14, 2018 4:28:35 AM PDT Sergii Romantsov wrote:
> > Hello,
> > seems some part of the World is still may waiting for a possibility to
> play
> > Dying Light... till pushed :)
> >
> > On Mon, Aug 6, 2018 at 4:26 PM, Lionel Landwerlin <
> > lionel.g.landwerlin at intel.com> wrote:
> >
> > > On 06/08/18 13:41, Sergii Romantsov wrote:
> > >
> > >> Kernel (for ppgtt) requires memory address to be
> > >> aligned to page size (4096).
> > >>
> > >> -v2: added marking that also fixes initial commit 01058a552294.
> > >> -v3: numbers replaced by PAGE_SIZE; buffer-object size is aligned
> > >> instead of alignment of offsets (Chris Wilson).
> > >> -v4: changes related to PAGE_SIZE moved to separate commit
> > >>
> > >> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=106997
> > >> Fixes: a363bb2cd0e2 (i965: Allocate VMA in userspace for full-PPGTT
> > >> systems.)
> > >> Fixes: 01058a552294 (i965: Add virtual memory allocator
> infrastructure to
> > >> brw_bufmgr.)
> > >> Signed-off-by: Sergii Romantsov <sergii.romantsov at globallogic.com>
> > >>
> > >
> > > Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin at intel.com>
> > >
> > > Thanks!
> > >
> > >
> > > ---
> > >>   src/mesa/drivers/dri/i965/brw_bufmgr.c | 7 +++----
> > >>   1 file changed, 3 insertions(+), 4 deletions(-)
> > >>
> > >> diff --git a/src/mesa/drivers/dri/i965/brw_bufmgr.c
> > >> b/src/mesa/drivers/dri/i965/brw_bufmgr.c
> > >> index 09d45e3..8274c2e 100644
> > >> --- a/src/mesa/drivers/dri/i965/brw_bufmgr.c
> > >> +++ b/src/mesa/drivers/dri/i965/brw_bufmgr.c
> > >> @@ -496,7 +496,6 @@ bo_alloc_internal(struct brw_bufmgr *bufmgr,
> > >>                     uint32_t stride)
> > >>   {
> > >>      struct brw_bo *bo;
> > >> -   unsigned int page_size = getpagesize();
> > >>      int ret;
> > >>      struct bo_cache_bucket *bucket;
> > >>      bool alloc_from_cache;
> > >> @@ -522,12 +521,12 @@ bo_alloc_internal(struct brw_bufmgr *bufmgr,
> > >>       * allocation up.
> > >>       */
> > >>      if (bucket == NULL) {
> > >> -      bo_size = size;
> > >> -      if (bo_size < page_size)
> > >> -         bo_size = page_size;
> > >> +      unsigned int page_size = getpagesize();
> > >> +      bo_size = ALIGN(size, page_size);
> > >>      } else {
> > >>         bo_size = bucket->size;
> > >>      }
> > >> +   assert(bo_size);
> > >>        mtx_lock(&bufmgr->lock);
> > >>      /* Get a buffer out of the cache if available */
> > >>
> > >
> > >
> > > _______________________________________________
> > > mesa-dev mailing list
> > > mesa-dev at lists.freedesktop.org
> > > https://lists.freedesktop.org/mailman/listinfo/mesa-dev
> > >
> >
> >
> >
> >
>
>


-- 
Sergii Romantsov
GlobalLogic Inc.
www.globallogic.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/mesa-dev/attachments/20180815/a629e4f6/attachment.html>