linux.git/drivers/gpu/drm/vmwgfx, branch v5.8

drm/vmwgfx: fix update of display surface when resolution changes

2020-07-14T02:05:52+00:00

The assignment of metadata overwrote the new display resolution values,
hence we'd miss the size actually changed and wouldn't redefine the
surface. This would then lead to command buffer error when trying to
update the screen target (due to the size mismatch), and result in a
VM with black screen.

Fixes: 504901dbb0b5 ("drm/vmwgfx: Refactor surface_define to use vmw_surface_metadata")
Reviewed-by: Charmaine Lee 
Signed-off-by: Roland Scheidegger 
Cc: stable@vger.kernel.org

drm: remove drm specific kmap_atomic code

2020-06-05T02:06:22+00:00

kmap_atomic_prot() is now exported by all architectures.  Use this
function rather than open coding a driver specific kmap_atomic.

[arnd@arndb.de: include linux/highmem.h]
  Link: http://lkml.kernel.org/r/20200508220150.649044-1-arnd@arndb.de
Signed-off-by: Ira Weiny 
Signed-off-by: Arnd Bergmann 
Signed-off-by: Andrew Morton 
Reviewed-by: Christian König 
Reviewed-by: Christoph Hellwig 
Acked-by: Daniel Vetter 
Cc: Al Viro 
Cc: Andy Lutomirski 
Cc: Benjamin Herrenschmidt 
Cc: Borislav Petkov 
Cc: Chris Zankel 
Cc: Dan Williams 
Cc: Dave Hansen 
Cc: "David S. Miller" 
Cc: Helge Deller 
Cc: "H. Peter Anvin" 
Cc: Ingo Molnar 
Cc: "James E.J. Bottomley" 
Cc: Max Filippov 
Cc: Paul Mackerras 
Cc: Peter Zijlstra 
Cc: Thomas Bogendoerfer 
Cc: Thomas Gleixner 
Link: http://lkml.kernel.org/r/20200507150004.1423069-12-ira.weiny@intel.com
Signed-off-by: Linus Torvalds

drm/vmwgfx: Return true in function vmw_fence_obj_signaled()

2020-05-16T04:50:37+00:00

Fix the following coccicheck warning:

drivers/gpu/drm/vmwgfx/vmwgfx_fence.c:518:9-10: WARNING: return of 0/1
in function 'vmw_fence_obj_signaled' with return type bool

Signed-off-by: Jason Yan 
Signed-off-by: Roland Scheidegger

drm/vmwgfx: remove redundant assignment to variable ret

2020-05-16T04:50:32+00:00

The variable ret is being initialized with a value that is never read
and it is being updated later with a new value. The initialization
is redundant and can be removed.

Addresses-Coverity: ("Unused value")
Signed-off-by: Colin Ian King 
Signed-off-by: Roland Scheidegger

drm/vmwgfx: Fix parameter name in vmw_bo_init

2020-05-16T04:50:21+00:00

The parameter name should be interruptible instead of interuptable.

Signed-off-by: Guixiong Wei 
Signed-off-by: Roland Scheidegger

Merge branch 'ttm-transhuge' of git://people.freedesktop.org/~thomash/linux into drm-next

2020-04-02T23:07:49+00:00

Huge page-table entries for TTM

In order to reduce CPU usage [1] and in theory TLB misses this patchset enables
huge- and giant page-table entries for TTM and TTM-enabled graphics drivers.

Signed-off-by: Dave Airlie 
From: Thomas Hellstrom (VMware) 
Link: https://patchwork.freedesktop.org/patch/msgid/20200325073102.6129-1-thomas_os@shipmail.org

drm/vmwgfx: Hook up the helpers to align buffer objects

2020-03-24T17:50:35+00:00

Start using the helpers that align buffer object user-space addresses and
buffer object vram addresses to huge page boundaries.
This is to improve the chances of allowing huge page-table entries.

Cc: Andrew Morton 
Cc: Michal Hocko 
Cc: "Matthew Wilcox (Oracle)" 
Cc: "Kirill A. Shutemov" 
Cc: Ralph Campbell 
Cc: "Jérôme Glisse" 
Cc: "Christian König" 
Cc: Dan Williams 
Signed-off-by: Thomas Hellstrom (VMware) 
Reviewed-by: Roland Scheidegger 
Acked-by: Christian König

drm/vmwgfx: Introduce a huge page aligning TTM range manager

2020-03-24T17:50:12+00:00

Using huge page-table entries requires that the physical address of the
start of a buffer object is huge page size aligned.
Make a special version of the TTM range manager that accomplishes this,
but falls back to a smaller page size alignment (PUD->PMD, PMD->NORMAL)
to avoid eviction.
If other drivers want to use it in the future, it can be made a
TTM generic helper. Note that drivers can force eviction for a certain
alignment by assigning the TTM GPU alignment correspondingly.

Cc: Andrew Morton 
Cc: Michal Hocko 
Cc: "Matthew Wilcox (Oracle)" 
Cc: "Kirill A. Shutemov" 
Cc: Ralph Campbell 
Cc: "Jérôme Glisse" 
Cc: "Christian König" 
Cc: Dan Williams 
Signed-off-by: Thomas Hellstrom (VMware) 
Reviewed-by: Roland Scheidegger 
Acked-by: Christian König

drm/vmwgfx: Support huge page faults

2020-03-24T17:48:55+00:00

With vmwgfx dirty-tracking we need a specialized huge_fault
callback. Implement and hook it up.

Cc: Andrew Morton 
Cc: Michal Hocko 
Cc: "Matthew Wilcox (Oracle)" 
Cc: "Kirill A. Shutemov" 
Cc: Ralph Campbell 
Cc: "Jérôme Glisse" 
Cc: "Christian König" 
Cc: Dan Williams 
Signed-off-by: Thomas Hellstrom (VMware) 
Reviewed-by: Roland Scheidegger 
Acked-by: Christian König

drm/ttm, drm/vmwgfx: Support huge TTM pagefaults

2020-03-24T17:48:33+00:00

Support huge (PMD-size and PUD-size) page-table entries by providing a
huge_fault() callback.
We still support private mappings and write-notify by splitting the huge
page-table entries on write-access.

Note that for huge page-faults to occur, either the kernel needs to be
compiled with trans-huge-pages always enabled, or the kernel needs to be
compiled with trans-huge-pages enabled using madvise, and the user-space
app needs to call madvise() to enable trans-huge pages on a per-mapping
basis.

Furthermore huge page-faults will not succeed unless buffer objects and
user-space addresses are aligned on huge page size boundaries.

Cc: Andrew Morton 
Cc: Michal Hocko 
Cc: "Matthew Wilcox (Oracle)" 
Cc: "Kirill A. Shutemov" 
Cc: Ralph Campbell 
Cc: "Jérôme Glisse" 
Cc: "Christian König" 
Cc: Dan Williams 
Signed-off-by: Thomas Hellstrom (VMware) 
Reviewed-by: Roland Scheidegger 
Reviewed-by: Christian König