Commit Graph

30 Commits

Author SHA1 Message Date
Konstantin Belousov
b11d58b63f Do not call malloc(M_WAITOK) while bodev->fence_lock mutex is
held. The ttm_buffer_object_transfer() does not need the mutex locked
at all, except for the call to the driver sync_obj_ref() method.

Reported and tested by:	dumbbell
MFC after:   2 weeks
2013-03-23 22:23:15 +00:00
Jean-Sébastien Pédron
accadf8de2 drm/ttm: Fix a typo: s/pTTM]/[TTM]/ 2013-03-23 20:46:47 +00:00
Jean-Sébastien Pédron
76c40c6986 drm/ttm: Explain why we don't need to acquire a ref in ttm_bo_vm_ctor() 2013-03-23 20:43:26 +00:00
Jean-Sébastien Pédron
a649986089 drm/ttm: Fix TTM buffer object refcount
This fixes memory leaks in the radeonkms driver.

Reviewed by:	Konstantin Belousov (kib@)
Tested by:	J.R. Oldroyd <jr@opal.com>
2013-03-23 19:19:19 +00:00
Attilio Rao
89f6b8632c Switch the vm_object mutex to be a rwlock. This will enable in the
future further optimizations where the vm_object lock will be held
in read mode most of the time the page cache resident pool of pages
are accessed for reading purposes.

The change is mostly mechanical but few notes are reported:
* The KPI changes as follow:
  - VM_OBJECT_LOCK() -> VM_OBJECT_WLOCK()
  - VM_OBJECT_TRYLOCK() -> VM_OBJECT_TRYWLOCK()
  - VM_OBJECT_UNLOCK() -> VM_OBJECT_WUNLOCK()
  - VM_OBJECT_LOCK_ASSERT(MA_OWNED) -> VM_OBJECT_ASSERT_WLOCKED()
    (in order to avoid visibility of implementation details)
  - The read-mode operations are added:
    VM_OBJECT_RLOCK(), VM_OBJECT_TRYRLOCK(), VM_OBJECT_RUNLOCK(),
    VM_OBJECT_ASSERT_RLOCKED(), VM_OBJECT_ASSERT_LOCKED()
* The vm/vm_pager.h namespace pollution avoidance (forcing requiring
  sys/mutex.h in consumers directly to cater its inlining functions
  using VM_OBJECT_LOCK()) imposes that all the vm/vm_pager.h
  consumers now must include also sys/rwlock.h.
* zfs requires a quite convoluted fix to include FreeBSD rwlocks into
  the compat layer because the name clash between FreeBSD and solaris
  versions must be avoided.
  At this purpose zfs redefines the vm_object locking functions
  directly, isolating the FreeBSD components in specific compat stubs.

The KPI results heavilly broken by this commit.  Thirdy part ports must
be updated accordingly (I can think off-hand of VirtualBox, for example).

Sponsored by:	EMC / Isilon storage division
Reviewed by:	jeff
Reviewed by:	pjd (ZFS specific review)
Discussed with:	alc
Tested by:	pho
2013-03-09 02:32:23 +00:00
Jean-Sébastien Pédron
b3469f58a7 Import Linux commit a1b7736dac5f2d5876e68c47a0fce3f423840070
Author: Sascha Hauer <s.hauer@pengutronix.de>
Date:   Wed Feb 1 11:38:22 2012 +0100

    drm drm_fb_helper: destroy modes

    drm_setup_crtcs allocated modes using drm_mode_duplicate. Free
    them in drm_fb_helper_crtc_free.

    Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
    Signed-off-by: Dave Airlie <airlied@redhat.com>

Discussed with:	Konstantin Belousov (kib@)
2013-03-08 18:15:07 +00:00
Jean-Sébastien Pédron
3f81c67746 drm: Fix a call to free(9) with an incorrect malloc type
While here, the call to free(9) is moved to drm_global.c, near the
initial malloc(9).

Reviewed by:	Konstantin Belousov (kib@)
2013-03-08 18:11:02 +00:00
Konstantin Belousov
257d427d5f Fix build with gcc, do not use unnamed union.
Reported and tested by:	gjb
MFC after:	1 month
2013-03-05 16:15:34 +00:00
Konstantin Belousov
94415dd90e Fix build with gcc, remove redundand declarations.
Reported and tested by:	gjb
MFC after:	1 month
2013-03-05 16:14:55 +00:00
Jean-Sébastien Pédron
f0916543a6 drm_global.c: Destroy sx in drm_global_release()
This fixes a build error at the same time (unused variable "item"), if
the kernel is compiled without INVARIANTS.

Reported by:	Hartmann, O. <ohartman@zedat.fu-berlin.de> (build error)
Reviewed by:	Konstantin Belousov (kib@)
2013-03-05 11:18:57 +00:00
Konstantin Belousov
9b7efb7458 Correct the r247832.
Noted by:	marius, rdivacky
MFC after:	1 month
2013-03-05 11:02:38 +00:00
Konstantin Belousov
e6cd8542ed Import the preliminary port of the TTM.
The early commit is done to facilitate the off-tree work on the
porting of the Radeon driver.

Sponsored by:	The FreeBSD Foundation
Debugged and tested by:	    dumbbell
MFC after:	1 month
2013-03-05 09:49:34 +00:00
Konstantin Belousov
9cfa0e9e3c Import the drm_global references helpers.
Sponsored by:	The FreeBSD Foundation
MFC after:	1 month
2013-03-05 09:27:21 +00:00
Konstantin Belousov
8f3993c1f1 Import the drm_mm_debug_table() function.
Sponsored by:	The FreeBSD Foundation
MFC after:	1 month
2013-03-05 09:07:58 +00:00
Konstantin Belousov
214bb83805 Import the likely() compat macro.
Sponsored by:	The FreeBSD Foundation
MFC after:	1 month
2013-03-05 09:07:01 +00:00
Gleb Smirnoff
c109de1e07 - If DRM_DEBUG_DEFAULT_ON is defined, then initialize drm_debug_flagi to
all supported debugging bits.
- If DRM_DEBUG_DEFAULT_ON isn't defined, then initialize drm_debug_flag
  to zero.

DRM_DEBUG_DEFAULT_ON is defined when module is build with -DDEBUG_DRM
or if kernel config has 'options DEBUG_DRM'.

Reviewed by:	kib
2012-11-02 05:26:33 +00:00
Kevin Lo
9823d52705 Revert previous commit...
Pointyhat to:	kevlo (myself)
2012-10-10 08:36:38 +00:00
Kevin Lo
a10cee30c9 Prefer NULL over 0 for pointers 2012-10-09 08:27:40 +00:00
Alexander Motin
aa91d1a85c Reduce delays in several wait loops from 10ms to 10us, same is it is done
in Linux. This substantially increases graphics performance on Ivy Bridge.

Submitted by:	avg@
Reviewed by:	kib@
2012-09-25 10:52:49 +00:00
Ed Schouten
7cbef24e1a Prefer __containerof() above member2struct().
The first does proper checking of the argument types, while the latter
does not.
2012-09-15 19:28:54 +00:00
Konstantin Belousov
d1c3436c1c Add drm and i915 ioctl translations for 32 bit process on 64 bit host.
Submitted by:	meowthink@gmail.com
MFC after:	2 weeks
2012-08-18 18:26:25 +00:00
Hans Petter Selasky
07da61a6cc Streamline use of cdevpriv and correct some corner cases.
1) It is not useful to call "devfs_clear_cdevpriv()" from
"d_close" callbacks, hence for example read, write, ioctl and
so on might be sleeping at the time of "d_close" being called
and then then freed private data can still be accessed.
Examples: dtrace, linux_compat, ksyms (all fixed by this patch)

2) In sys/dev/drm* there are some cases in which memory will
be freed twice, if open fails, first by code in the open
routine, secondly by the cdevpriv destructor. Move registration
of the cdevpriv to the end of the drm open routines.

3) devfs_clear_cdevpriv() is not called if the "d_open" callback
registered cdevpriv data and the "d_open" callback function
returned an error. Fix this.

Discussed with:	phk
MFC after:	2 weeks
2012-08-15 16:19:39 +00:00
Alexander Motin
d7f7144a73 Fix millisecond to ticks conversion in drm_msleep().
On systems with HZ=100 it caused Intel eDP video output initialization
(and Xorg startup) to take several minutes instead of several seconds.

Reviewed by:	kib
MFC after:	3 days
2012-06-28 17:15:16 +00:00
Ed Maste
078a150402 Add PCI IDs for Ivy Bridge 2012-06-21 22:06:57 +00:00
Konstantin Belousov
92bd74bb2f Use right size when freeing unneeded GTT mapping.
MFC after:	3 days
2012-06-15 08:50:44 +00:00
Alan Cox
2444cf8fd0 A small simplification to i915_gem_pager_fault().
Reviewed by:	kib
2012-05-28 21:15:54 +00:00
Konstantin Belousov
46932515ac Disable end of buffer fixup by default. New DDX does not need this, and
since batch_len is unused by Linux driver, it seems that it is sometimes
gets passed wrong. This causes command buffer corruption and GPU hung.

Old GEMified DDX drivers that needs this workaround are not supported.

MFC after:	1 month
2012-05-28 13:58:08 +00:00
Konstantin Belousov
886add7764 Fix calculation of the execution buffer end in the mapped pages
when it is spilled into the next page.

MFC after:	1 month
2012-05-28 13:55:49 +00:00
Dimitry Andric
648466fc8d Fix the following clang warning in drm2:
sys/dev/drm2/i915/intel_display.c:8861:3: error: expression result unused [-Werror,-Wunused-value]
                  _intel_wait_for(dev,
                  ^~~~~~~~~~~~~~~~~~~~
  @/dev/drm2/i915/intel_drv.h:55:2: note: expanded from macro '_intel_wait_for'
          ret;                                                            \
          ^~~

Reviewed by:	kib
MFC after:	1 week
2012-05-24 19:13:54 +00:00
Konstantin Belousov
e27f871969 Add the code for new Intel GPU driver, which supports GEM, KMS and
works with new generations of GPUs (IronLake, SandyBridge and
supposedly IvyBridge).

The driver is not connected to the build yet.

Sponsored by:	The FreeBSD Foundation
MFC after:	1 week
2012-05-22 11:07:44 +00:00