Ben Skeggs [Wed, 26 Sep 2012 22:56:24 +0000 (08:56 +1000)]
drm/nv41/vm: fix and enable use of "real" pciegart
Hopefully fixed the tlb flush timeout issue. Was able to observe this
condition occur occasionally, and it appears the binary driver doesn't
wait on the old condition either..
Should give 39-bit DMA addressing on the relevant chipsets.
Ben Skeggs [Wed, 26 Sep 2012 22:55:53 +0000 (08:55 +1000)]
drm/nv44/vm: fix and enable use of "real" pciegart
Something seems to be missing in regards to flushing specific ranges of
the TLB. For the moment, flushing the entire thing seems to make it
work alright.
Should give 39-bit DMA addressing on the relevant chipsets.
v2: allocate contig 16KiB for dummy pages, reported by mwk on irc
Martin Peres [Tue, 4 Sep 2012 11:52:00 +0000 (13:52 +0200)]
drm/nouveau/therm: add support for fan-control modes
For now, only 2 control modes are available:
- NONE: The fan is never touched (default)
- MANUAL: The fan is set to the user-defined fan speed (pwm1)
This patch introduces a distinction between ptherm internal fan management
and external fan management. The latter is bound to respect the fan mode
while the first can still select the speed it wants unless the NONE mode
is selected. This is important for automatic fan management.
Signed-off-by: Martin Peres <martin.peres@labri.fr> Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Martin Peres [Sun, 2 Sep 2012 00:55:58 +0000 (02:55 +0200)]
drm/nouveau/therm: move thermal-related functions to the therm subdev
It looks scary because of the size, but I tried to keep the differences minimal.
Further patches will fix the actual "driver" code and add new features.
v2: change filenames, split to submodules
v3: add a missing include
v4: Ben Skeggs <bskeggs@redhat.com>
- fixed set_defaults() to allow min_duty < 30 (thermal table will
override this if it's actually necessary)
- fixed set_defaults() to not provide pwm_freq so nv4x (which only has
pwm_div) can actually work. the boards using pwm_freq will have a
thermal table entry to provide us the value.
- removed unused files
Signed-off-by: Martin Peres <martin.peres@labri.fr> Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Ben Skeggs [Fri, 10 Aug 2012 05:10:34 +0000 (15:10 +1000)]
drm/nouveau/core: have fifo store a unique context identifier at attach time
This value will match something that's easily available from the engine IRQ
handlers, and used to lookup the relevant context.
Since the changes in how this is done on each generation match when the
major PFIFO changes happened, fifo is responsible for calculating the
correct value to avoid duplicating the same code among many engine modules.
Ben Skeggs [Thu, 19 Jul 2012 22:17:34 +0000 (08:17 +1000)]
drm/nouveau: port all engines to new engine module format
This is a HUGE commit, but it's not nearly as bad as it looks - any problems
can be isolated to a particular chipset and engine combination. It was
simply too difficult to port each one at a time, the compat layers are
*already* ridiculous.
Most of the changes here are simply to the glue, the process for each of the
engine modules was to start with a standard skeleton and copy+paste the old
code into the appropriate places, fixing up variable names etc as needed.
v2: Marcin Slusarz <marcin.slusarz@gmail.com>
- fix find/replace bug in license header
v3: Ben Skeggs <bskeggs@redhat.com>
- bump indirect pushbuf size to 8KiB, 4KiB barely enough for userspace and
left no space for kernel's requirements during GEM pushbuf submission.
- fix duplicate assignments noticed by clang
v4: Marcin Slusarz <marcin.slusarz@gmail.com>
- add sparse annotations to nv04_fifo_pause/nv04_fifo_start
- use ioread32_native/iowrite32_native for fifo control registers
v5: Ben Skeggs <bskeggs@redhat.com>
- rebase on v3.6-rc4, modified to keep copy engine fix intact
- nv10/fence: unmap fence bo before destroying
- fixed fermi regression when using nvidia gr fuc
- fixed typo in supported dma_mask checking
Ben Skeggs [Sat, 4 Aug 2012 08:26:50 +0000 (18:26 +1000)]
drm/nvc0-/gr: share headers between fermi and kepler graphics code
v2: Ben Skeggs <bskeggs@redhat.com>
- de-inline nv_icmd, triggers some gcc issue causing ctxnv[ce]0.c to
take a *very* *very* long time to build on some configs.
Ben Skeggs [Thu, 19 Jul 2012 07:32:01 +0000 (17:32 +1000)]
drm/nvc0-nve0/graph: rename dev to priv, no code changes
There's a *lot* of code in here, and it's all going to use the PGRAPH priv
pointer rather than drm_device after the engine rework. This is handling
all the rename-only parts of the change.
Ben Skeggs [Thu, 19 Jul 2012 00:51:42 +0000 (10:51 +1000)]
drm/nouveau/fence: un-port from nouveau_exec_engine interfaces
Still the same code, but not an "engine" anymore. The fence code is more of
a policy decision rather than exposing mechanisms, so it's not appropriate
to port it to the new engine subsystem.
Ben Skeggs [Fri, 13 Jul 2012 07:21:22 +0000 (17:21 +1000)]
drm/nouveau/gpuobj: remove flags for vm-mappings
Having GPUOBJ and VM intertwined like this makes it *really* hard to
continue porting to the new driver architecture, split it out in
favour of requiring explit maps be the caller.
Ben Skeggs [Fri, 6 Jul 2012 02:14:00 +0000 (12:14 +1000)]
drm/nouveau: implement module init functions in nouveau_drm.c
These currently just call the existing ones in nouveau_drv.c, but will be
extended in upcoming commits. This needed to be separated from the current
code as there will be some header clashes until things are ported.
Ben Skeggs [Thu, 5 Jul 2012 21:36:43 +0000 (07:36 +1000)]
drm/nouveau/core: pull in most of the new core infrastructure
This commit provides most of the infrastructure to support a major overhaul
of Nouveau's internals coming in the following commits. This work aims to
take all the things we've learned over the last several years, and turn that
into a cleaner architecture that's more maintainable going forward.
RAMHT and MM bits of the new core have been left out for the moment, and
will be pulled in as I go through the process of porting the code to
become either subdev or engine modules.
There are several main goals I wanted to achieve through this work:
-- Reduce complexity
The goal here was to make each component of the driver as independent as
possible, which will ease maintainability and readability, and provide a
good base for resetting locked up GPU units in the future.
-- Better tracking of GPU units that are required at any given time
This is for future PM work, we'll be able to tell exactly what parts of the
GPU we need powered at any given point (etc).
-- Expose all available NVIDIA GPUs to the client
In order to support things such as multi-GPU channels, we want to be able
to expose all the NVIDIA GPUs to the client over a single file descriptor
so it can send a single push buffer to multiple GPUs.
-- Untangle the core hardware support code from the DRM implementation
This happened initially as an unexpected side-effect of developing the
initial core infrastructure in userspace, but it turned into a goal of
the whole project. Initial benefits will be the availablility of a
number of userspace tools and tests using the same code as the driver
itself, but will also be important as I look into some virtualisation
ideas.
v2: Ben Skeggs <bskeggs@redhat.com>
- fix duplicate assignments noticed by clang
- implement some forgotten yelling in error path
- ensure 64-bit engine mask is used everywhere
v3: Marcin Slusarz <marcin.slusarz@gmail.com>
- sparse fixes
- inline nv_printk into nv_assert to prevent recursive inlining issues
v4: Ben Skeggs <bskeggs@redhat.com>
- fixed minor memory leak on gpuobj destruction
Ben Skeggs [Wed, 4 Jul 2012 13:44:54 +0000 (23:44 +1000)]
drm/nouveau: restructure source tree, split core from drm implementation
Future work will be headed in the way of separating the policy supplied by
the nouveau drm module from the mechanisms provided by the driver core.
There will be a couple of major classes (subdev, engine) of driver modules
that have clearly defined tasks, and the further directory structure change
is to reflect this.
No code changes here whatsoever, aside from fixing up a couple of include
file pathnames.
Ben Skeggs [Fri, 3 Aug 2012 19:46:01 +0000 (05:46 +1000)]
drm/nouveau/gem: use bo.offset rather than mm_node.start
Won't necessarily be a drm_mm_node in the future, and I can't think of any
good reason to not use the offset from the bo struct. There may have been
some reason once apon a time, but, separate commit just in case.
Ben Skeggs [Fri, 28 Sep 2012 01:50:29 +0000 (11:50 +1000)]
drm/nvc0/fence: restore pre-suspend fence buffer context on resume
Fixes some unfortunate races on resume. The G84 version of the code doesn't
need this as "gpuobj"s are automagically suspended/resumed by the core code
whereas pinned buffer objects are not.
Cc: stable@vger.kernel.org Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
IBM reported a deadlock in select_parent(). This was found to be caused
by taking rename_lock when already locked when restarting the tree
traversal.
There are two cases when the traversal needs to be restarted:
1) concurrent d_move(); this can only happen when not already locked,
since taking rename_lock protects against concurrent d_move().
2) racing with final d_put() on child just at the moment of ascending
to parent; rename_lock doesn't protect against this rare race, so it
can happen when already locked.
Because of case 2, we need to be able to handle restarting the traversal
when rename_lock is already held. This patch fixes all three callers of
try_to_ascend().
IBM reported that the deadlock is gone with this patch.
[ I rewrote the patch to be smaller and just do the "goto again" if the
lock was already held, but credit goes to Miklos for the real work.
- Linus ]
Merge tag 'iommu-fixes-v3.6-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu
Pull IOMMU fixes from Joerg Roedel:
"Two small patches:
* One patch to fix the function declarations for
!CONFIG_IOMMU_API. This is causing build errors
in linux-next and should be fixed for v3.6.
* Another patch to fix an IOMMU group related NULL pointer
dereference."
* tag 'iommu-fixes-v3.6-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu:
iommu/amd: Fix wrong assumption in iommu-group specific code
iommu: static inline iommu group stub functions
Pull NVMe driver fixes from Matthew Wilcox:
"Now that actual hardware has been released (don't have any yet
myself), people are starting to want some of these fixes merged."
Willy doesn't have hardware? Guys...
* git://git.infradead.org/users/willy/linux-nvme:
NVMe: Cancel outstanding IOs on queue deletion
NVMe: Free admin queue memory on initialisation failure
NVMe: Use ida for nvme device instance
NVMe: Fix whitespace damage in nvme_init
NVMe: handle allocation failure in nvme_map_user_pages()
NVMe: Fix uninitialized iod compiler warning
NVMe: Do not set IO queue depth beyond device max
NVMe: Set block queue max sectors
NVMe: use namespace id for nvme_get_features
NVMe: replace nvme_ns with nvme_dev for user admin
NVMe: Fix nvme module init when nvme_major is set
NVMe: Set request queue logical block size
Sasha Levin has been running trinity in a KVM tools guest, and was able
to trigger the BUG_ON() at arch/x86/mm/pat.c:279 (verifying the range of
the memory type). The call trace showed that it was mtdchar_mmap() that
created an invalid remap_pfn_range().
The problem is that mtdchar_mmap() does various really odd and subtle
things with the vma page offset etc, and uses the wrong types (and the
wrong overflow) detection for it.
For example, the page offset may well be 32-bit on a 32-bit
architecture, but after shifting it up by PAGE_SHIFT, we need to use a
potentially 64-bit resource_size_t to correctly hold the full value.
Also, we need to check that the vma length plus offset doesn't overflow
before we check that it is smaller than the length of the mtdmap region.
This fixes things up and tries to make the code a bit easier to read.