ttm tt rework modified the way we allocate and populate the
ttm_tt structure, the AGP side was missing some bit to properly
work. Fix those and fix radeon and nouveau AGP support.
Tested on radeon only so far.
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
Reviewed-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Provide helper function to compute the kernel memory size needed
for each buffer object. Move all the accounting inside ttm, simplifying
driver and avoiding code duplication accross them.
v2 fix accounting of ghost object, one would have thought that i
would have run into the issue since a longtime but it seems
ghost object are rare when you have plenty of vram ;)
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
Reviewed-by: Thomas Hellstrom <thellstrom@vmware.com>
Move dma data to a superset ttm_dma_tt structure which herit
from ttm_tt. This allow driver that don't use dma functionalities
to not have to waste memory for it.
V2 Rebase on top of no memory account changes (where/when is my
delorean when i need it ?)
V3 Make sure page list is initialized empty
V4 typo/syntax fixes
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
Reviewed-by: Thomas Hellstrom <thellstrom@vmware.com>
In TTM world the pages for the graphic drivers are kept in three different
pools: write combined, uncached, and cached (write-back). When the pages
are used by the graphic driver the graphic adapter via its built in MMU
(or AGP) programs these pages in. The programming requires the virtual address
(from the graphic adapter perspective) and the physical address (either System RAM
or the memory on the card) which is obtained using the pci_map_* calls (which does the
virtual to physical - or bus address translation). During the graphic application's
"life" those pages can be shuffled around, swapped out to disk, moved from the
VRAM to System RAM or vice-versa. This all works with the existing TTM pool code
- except when we want to use the software IOTLB (SWIOTLB) code to "map" the physical
addresses to the graphic adapter MMU. We end up programming the bounce buffer's
physical address instead of the TTM pool memory's and get a non-worky driver.
There are two solutions:
1) using the DMA API to allocate pages that are screened by the DMA API, or
2) using the pci_sync_* calls to copy the pages from the bounce-buffer and back.
This patch fixes the issue by allocating pages using the DMA API. The second
is a viable option - but it has performance drawbacks and potential correctness
issues - think of the write cache page being bounced (SWIOTLB->TTM), the
WC is set on the TTM page and the copy from SWIOTLB not making it to the TTM
page until the page has been recycled in the pool (and used by another application).
The bounce buffer does not get activated often - only in cases where we have
a 32-bit capable card and we want to use a page that is allocated above the
4GB limit. The bounce buffer offers the solution of copying the contents
of that 4GB page to an location below 4GB and then back when the operation has been
completed (or vice-versa). This is done by using the 'pci_sync_*' calls.
Note: If you look carefully enough in the existing TTM page pool code you will
notice the GFP_DMA32 flag is used - which should guarantee that the provided page
is under 4GB. It certainly is the case, except this gets ignored in two cases:
- If user specifies 'swiotlb=force' which bounces _every_ page.
- If user is using a Xen's PV Linux guest (which uses the SWIOTLB and the
underlaying PFN's aren't necessarily under 4GB).
To not have this extra copying done the other option is to allocate the pages
using the DMA API so that there is not need to map the page and perform the
expensive 'pci_sync_*' calls.
This DMA API capable TTM pool requires for this the 'struct device' to
properly call the DMA API. It also has to track the virtual and bus address of
the page being handed out in case it ends up being swapped out or de-allocated -
to make sure it is de-allocated using the proper's 'struct device'.
Implementation wise the code keeps two lists: one that is attached to the
'struct device' (via the dev->dma_pools list) and a global one to be used when
the 'struct device' is unavailable (think shrinker code). The global list can
iterate over all of the 'struct device' and its associated dma_pool. The list
in dev->dma_pools can only iterate the device's dma_pool.
/[struct device_pool]\
/---------------------------------------------------| dev |
/ +-------| dma_pool |
/-----+------\ / \--------------------/
|struct device| /-->[struct dma_pool for WC]</ /[struct device_pool]\
| dma_pools +----+ /-| dev |
| ... | \--->[struct dma_pool for uncached]<-/--| dma_pool |
\-----+------/ / \--------------------/
\----------------------------------------------/
[Two pools associated with the device (WC and UC), and the parallel list
containing the 'struct dev' and 'struct dma_pool' entries]
The maximum amount of dma pools a device can have is six: write-combined,
uncached, and cached; then there are the DMA32 variants which are:
write-combined dma32, uncached dma32, and cached dma32.
Currently this code only gets activated when any variant of the SWIOTLB IOMMU
code is running (Intel without VT-d, AMD without GART, IBM Calgary and Xen PV
with PCI devices).
Tested-by: Michel Dänzer <michel@daenzer.net>
[v1: Using swiotlb_nr_tbl instead of swiotlb_enabled]
[v2: Major overhaul - added 'inuse_list' to seperate used from inuse and reorder
the order of lists to get better performance.]
[v3: Added comments/and some logic based on review, Added Jerome tag]
[v4: rebase on top of ttm_tt & ttm_backend merge]
[v5: rebase on top of ttm memory accounting overhaul]
[v6: New rebase on top of more memory accouting changes]
[v7: well rebase on top of no memory accounting changes]
[v8: make sure pages list is initialized empty]
[v9: calll ttm_mem_global_free_page in unpopulate for accurate accountg]
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Reviewed-by: Jerome Glisse <jglisse@redhat.com>
Acked-by: Thomas Hellstrom <thellstrom@vmware.com>
Move the page allocation and freeing to driver callback and
provide ttm code helper function for those.
Most intrusive change, is the fact that we now only fully
populate an object this simplify some of code designed around
the page fault design.
V2 Rebase on top of memory accounting overhaul
V3 New rebase on top of more memory accouting changes
V4 Rebase on top of no memory account changes (where/when is my
delorean when i need it ?)
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
Reviewed-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Reviewed-by: Thomas Hellstrom <thellstrom@vmware.com>
ttm_backend will only exist with a ttm_tt, and ttm_tt
will only be of interest when bound to a backend. Merge them
to avoid code and data duplication.
V2 Rebase on top of memory accounting overhaul
V3 Rebase on top of more memory accounting changes
V4 Rebase on top of no memory account changes (where/when is my
delorean when i need it ?)
V5 make sure ttm is unbound before destroying, change commit
message on suggestion from Tormod Volden
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
Reviewed-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Reviewed-by: Thomas Hellstrom <thellstrom@vmware.com>
This field is not use by any of the driver just drop it.
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
Reviewed-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Reviewed-by: Thomas Hellstrom <thellstrom@vmware.com>
Split btw highmem and lowmem page was rendered useless by the
pool code. Remove it. Note further cleanup would change the
ttm page allocation helper to actualy take an array instead
of relying on list this could drasticly reduce the number of
function call in the common case of allocation whole buffer.
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
Reviewed-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Reviewed-by: Thomas Hellstrom <thellstrom@vmware.com>
This was never use in none of the driver, properly using userspace
page for bo would need more code (vma interaction mostly). Removing
this dead code in preparation of ttm_tt & backend merge.
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
Reviewed-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Reviewed-by: Thomas Hellstrom <thellstrom@vmware.com>
. and some comments to make it easier to understand.
Ackedby: Randy Dunlap <randy.dunlap@oracle.com>
[v2: Added some more updates from Randy Dunlap]
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
This reverts commit 5a893fc28f.
This causes a use after free in the ttm free alloc pages path,
when it tries to get the be after the be has been destroyed.
Signed-off-by: Dave Airlie <airlied@redhat.com>
* 'stable/ttm.pci-api.v5' of git://git.kernel.org/pub/scm/linux/kernel/git/konrad/xen:
ttm: Include the 'struct dev' when using the DMA API.
nouveau/ttm/PCIe: Use dma_addr if TTM has set it.
radeon/ttm/PCIe: Use dma_addr if TTM has set it.
ttm: Expand (*populate) to support an array of DMA addresses.
ttm: Utilize the DMA API for pages that have TTM_PAGE_FLAG_DMA32 set.
ttm: Introduce a placeholder for DMA (bus) addresses.
This makes the accounting when using 'debug_dma_dump_mappings()'
and CONFIG_DMA_API_DEBUG=y be assigned to the correct device
instead of 'fallback'.
No functional change - just cosmetic.
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
We pass in the array of ttm pages to be populated in the GART/MM
of the card (or AGP). Patch titled: "ttm: Utilize the DMA API for
pages that have TTM_PAGE_FLAG_DMA32 set." uses the DMA API to make
those pages have a proper DMA addresses (in the situation where
page_to_phys or virt_to_phys do not give use the DMA (bus) address).
Since we are using the DMA API on those pages, we should pass in the
DMA address to this function so it can save it in its proper fields
(later patches use it).
[v2: Added reviewed-by tag]
Reviewed-by: Thomas Hellstrom <thellstrom@shipmail.org>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Tested-by: Ian Campbell <ian.campbell@citrix.com>
This is right now limited to only non-pool constructs.
[v2: Fixed indentation issues, add review-by tag]
Reviewed-by: Thomas Hellstrom <thomas@shipmail.org>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Tested-by: Ian Campbell <ian.campbell@citrix.com>
This patch attempts to fix up shortcomings with the current calling
sequences.
1) There's a fastpath where no locking occurs and only io_mem_reserved is
called to obtain needed info for mapping. The fastpath is set per
memory type manager.
2) If the fastpath is disabled, io_mem_reserve and io_mem_free will be exactly
balanced and not called recursively for the same struct ttm_mem_reg.
3) Optionally the driver can choose to enable a per memory type manager LRU
eviction mechanism that, when io_mem_reserve returns -EAGAIN will attempt
to kill user-space mappings of memory in that manager to free up needed
resources
Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
Reviewed-by: Ben Skeggs <bskeggs@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Rather than having the driver supply the validation sequence, leave that
responsibility to TTM. This saves some confusion and a function argument.
Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Drastically reduce the number of spin lock / unlock operations by performing
unreserving and fencing under global locks.
Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
Reviewed-by: Jerome Glisse <j.glisse@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
The bo lock used only to protect the bo sync object members, and since it
is a per bo lock, fencing a buffer list will see a lot of locks and unlocks.
Replace it with a per-device lock that protects the sync object members on
*all* bos. Reading and setting these members will always be very quick, so
the risc of heavy lock contention is microscopic. Note that waiting for
sync objects will always take place outside of this lock.
The bo device fence lock will eventually be replaced with a seqlock /
rcu mechanism so we can determine that a bo is idle under a
rcu / read seqlock.
However this change will allow us to batch fencing and unreserving of
buffers with a minimal amount of locking.
Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
Reviewed-by: Jerome Glisse <j.glisse@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Add an aid for the driver to detect deadlocks on multi-bo reservations
Update documentation.
Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
Reviewed-by: Jerome Glisse <j.glisse@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Makes it possible to reserve a list of buffer objects with a single
spin lock / unlock if there is no contention.
Should improve cpu usage on SMP kernels.
v2: Initialize private list members on reserve and don't call
ttm_bo_list_ref_sub() with zero put_count.
Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Remove an obsolete comment about mm nodes.
Document the new bo range manager interface.
Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
We need the unlocked variant for the new codepath introduced to fix the
race condition in master recently.
Signed-off-by: Dave Airlie <airlied@redhat.com>
Nouveau will need this on GeForce 8 and up to account for the GPU
reordering physical VRAM for some memory types.
Reviewed-by: Jerome Glisse <jglisse@redhat.com>
Acked-by: Thomas Hellström <thellstrom@vmware.com>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Existing core code/drivers call drm_mm_put_block on ttm_mem_reg.mm_node
directly. Future patches will modify TTM behaviour in such a way that
ttm_mem_reg.mm_node doesn't necessarily belong to drm_mm.
Reviewed-by: Jerome Glisse <jglisse@redhat.com>
Acked-by: Thomas Hellström <thellstrom@vmware.com>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
I wrote this for the prime sharing work, but I also noticed other external
non-upstream drivers from a large company carrying a similiar patch, so I
may as well ship it in master.
Signed-off-by: Dave Airlie <airlied@redhat.com>
* 'drm-for-2.6.35' of git://git.kernel.org/pub/scm/linux/kernel/git/airlied/drm-2.6: (207 commits)
drm/radeon/kms/pm/r600: select the mid clock mode for single head low profile
drm/radeon: fix power supply kconfig interaction.
drm/radeon/kms: record object that have been list reserved
drm/radeon: AGP memory is only I/O if the aperture can be mapped by the CPU.
drm/radeon/kms: don't default display priority to high on rs4xx
drm/edid: fix typo in 1600x1200@75 mode
drm/nouveau: fix i2c-related init table handlers
drm/nouveau: support init table i2c device identifier 0x81
drm/nouveau: ensure we've parsed i2c table entry for INIT_*I2C* handlers
drm/nouveau: display error message for any failed init table opcode
drm/nouveau: fix init table handlers to return proper error codes
drm/nv50: support fractional feedback divider on newer chips
drm/nv50: fix monitor detection on certain chipsets
drm/nv50: store full dcb i2c entry from vbios
drm/nv50: fix suspend/resume with DP outputs
drm/nv50: output calculated crtc pll when debugging on
drm/nouveau: dump pll limits entries when debugging is on
drm/nouveau: bios parser fixes for eDP boards
drm/nouveau: fix a nouveau_bo dereference after it's been destroyed
drm/nv40: remove some completed ctxprog TODOs
...
It's unused and buggy in its current form, since it can place a bo
in the reserved state without removing it from lru lists.
Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
All TTM driver have been converted to new io_mem_reserve/free
interface which allow driver to choose and return proper io
base, offset to core TTM for ioremapping if necessary. This
patch remove what is now deadcode.
V2 adapt to match with change in first patch of the patchset
V3 update after io_mem_reserve/io_mem_free callback balancing
V4 adjust to minor cleanup
V5 remove the needs ioremap flag
V6 keep the ioremapping facility in TTM
[airlied- squashed driver removals in here also]
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
Reviewed-by: Thomas Hellstrom <thellstrom@vmware.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
On fault the driver is given the opportunity to perform any operation
it sees fit in order to place the buffer into a CPU visible area of
memory. This patch doesn't break TTM users, nouveau, vmwgfx and radeon
should keep working properly. Future patch will take advantage of this
infrastructure and remove the old path from TTM once driver are
converted.
V2 return VM_FAULT_NOPAGE if callback return -EBUSY or -ERESTARTSYS
V3 balance io_mem_reserve and io_mem_free call, fault_reserve_notify
is responsible to perform any necessary task for mapping to succeed
V4 minor cleanup, atomic_t -> bool as member is protected by reserve
mecanism from concurent access
V5 the callback is now responsible for iomapping the bo and providing
a virtual address this simplify TTM and will allow to get rid of
TTM_MEMTYPE_FLAG_NEEDS_IOREMAP
V6 use the bus addr data to decide to ioremap or this isn't needed
but we don't necesarily need to ioremap in the callback but still
allow driver to use static mapping
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
Reviewed-by: Thomas Hellstrom <thellstrom@vmware.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
There is case where we want to be able to wait only for the
GPU while not waiting for other buffer to be unreserved. This
patch split the no_wait argument all the way down in the whole
ttm path so that upper level can decide on what to wait on or
not.
[airlied: squashed these 4 for bisectability reasons.]
drm/radeon/kms: update to TTM no_wait splitted argument
drm/nouveau: update to TTM no_wait splitted argument
drm/vmwgfx: update to TTM no_wait splitted argument
[vmwgfx patch: Reviewed-by: Thomas Hellstrom <thellstrom@vmware.com>]
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
Acked-by: Thomas Hellstrom <thellstrom@vmware.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Now that the drm core can do this, lets just use it, split the code out
so TTM doesn't have to drag all of drmP.h in.
Signed-off-by: Dave Airlie <airlied@redhat.com>
Fix function prototype to match its actual usage and implementation.
drivers/gpu/drm/ttm/ttm_bo_util.c:341:10: error: symbol 'ttm_io_prot' redeclared with different type (originally declared at include/drm/ttm/ttm_bo_driver.h:911) - incompatible argument 1 (different signedness)
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Cc: David Airlie <airlied@linux.ie>
Signed-off-by: Dave Airlie <airlied@redhat.com>
This is needed for a bugfix in the vmwgfx driver.
Drivers may have GPU bindings on buffers that core TTM is not aware of,
and TTM may view those buffers as ordinary system memory buffers.
Add a notifier to such drivers when TTM is about to move the buffer
contents out to swappable memory. The driver must then release any
private GPU bindings on those buffers.
Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Return -ERESTARTSYS instead of -ERESTART when interrupted by a signal.
The -ERESTARTSYS is converted to an -EINTR by the kernel signal layer
before returned to user-space.
Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
This change allow driver to pass sorted memory placement,
from most prefered placement to least prefered placement.
In order to avoid long function prototype a structure is
used to gather memory placement informations such as range
restriction (if you need a buffer to be in given range).
Range restriction is determined by fpfn & lpfn which are
the first page and last page number btw which allocation
can happen. If those fields are set to 0 ttm will assume
buffer can be put anywhere in the address space (thus it
avoids putting a burden on the driver to always properly
set those fields).
This patch also factor few functions like evicting first
entry of lru list or getting a memory space. This avoid
code duplication.
V2: Change API to use placement flags and array instead
of packing placement order into a quadword.
V3: Make sure we set the appropriate mem.placement flag
when validating or allocation memory space.
[Pending Thomas Hellstrom further review but okay
from preliminary review so far].
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Common resources, like memory accounting and swap lists should be
global and not per device. Introduce a struct ttm_bo_global to
accomodate this, and register it with sysfs. Add a small sysfs interface
to return the number of active buffer objects.
Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
Signed-off-by: Dave Airlie <airlied@linux.ie>
This adds new set/get tiling interfaces where the pitch
and macro/micro tiling enables can be set. Along with
a flag to decide if this object should have a surface when mapped.
The only thing we need to allocate with a mapped surface should be
the frontbuffer. Note rotate scanout shouldn't require one, and
back/depth shouldn't either, though mesa needs some fixes.
It fixes the TTM interfaces along Thomas's suggestions, and I've tested
the surface stealing code with two X servers and not seen any lockdep issues.
I've stopped tiling the fbcon frontbuffer, as I don't see there being
any advantage other than testing, I've left the testing commands in there,
just flip the fb_tiled to true in radeon_fb.c
Open: Can we integrate endian swapping in with this?
Future features:
texture tiling - need to relocate texture registers TXOFFSET* with tiling info.
This also merges Michel's cleanup surfaces regs at init time patch
even though it makes sense on its own, this patch really relies on it.
Some PowerMac firmwares set up a tiling surface at the beginning of VRAM
which messes us up otherwise.
that patch is:
Signed-off-by: Michel Dänzer <daenzer@vmware.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
This add support for using dma32 memory on gpus that really need it.
Currently IGPs are left without DMA32 but we might need to change
that unless we can fix rs690.
Signed-off-by: Dave Airlie <airlied@redhat.com>
TTM is a GPU memory manager subsystem designed for use with GPU
devices with various memory types (On-card VRAM, AGP,
PCI apertures etc.). It's essentially a helper library that assists
the DRM driver in creating and managing persistent buffer objects.
TTM manages placement of data and CPU map setup and teardown on
data movement. It can also optionally manage synchronization of
data on a per-buffer-object level.
TTM takes care to provide an always valid virtual user-space address
to a buffer object which makes user-space sub-allocation of
big buffer objects feasible.
TTM uses a fine-grained per buffer-object locking scheme, taking
care to release all relevant locks when waiting for the GPU.
Although this implies some locking overhead, it's probably a big
win for devices with multiple command submission mechanisms, since
the lock contention will be minimal.
TTM can be used with whatever user-space interface the driver
chooses, including GEM. It's used by the upcoming Radeon KMS DRM driver
and is also the GPU memory management core of various new experimental
DRM drivers.
Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Jerome Glisse