aboutsummaryrefslogtreecommitdiffstats
path: root/drivers/gpu/drm/radeon/radeon_vm.c
AgeCommit message (Collapse)AuthorFilesLines
2015-02-02drm/radeon: properly set vm fragment size for TN/RLAlex Deucher1-2/+4
Should be the same as cayman. We don't use VM by default on NI parts so this isn't critical. Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Cc: stable@vger.kernel.org
2015-01-22drm/radeon: Remove rdev->gart.pages_addr arrayMichel Dänzer1-4/+2
radeon_vm_map_gart can use rdev->gart.pages_entry instead. Also move the masking of the page address to radeon_vm_map_gart from its callers. Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-12-03drm/radeon: add spinlock for BO_VA status protection (v2)Christian König1-6/+30
Stop using the VM mutex for this v2: fix typo in comment Signed-off-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-12-03drm/radeon: fence PT updates as sharedChristian König1-9/+19
The BO_VA contains everything necessary. Signed-off-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-12-03drm/radeon: rename radeon_cs_reloc to radeon_bo_listChristian König1-3/+3
Better match what it is actually doing. Signed-off-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-12-03drm/radeon: drop the handle from radeon_cs_relocChristian König1-2/+0
It's only used for duplicate check and that can be done on the original as well. Signed-off-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-12-03drm/radeon drop gobj from radeon_cs_relocChristian König1-2/+0
It's only used once after initializing and that ptr can be calculated from the BO as well. Signed-off-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-11-20drm/radeon: sync PT updates as shared v2Christian König1-1/+8
Only invalidating PTEs needs to be executed synchronized to using the PT. v2: fix sync to uses Signed-off-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-11-20drm/radeon: sync PD updates as sharedChristian König1-1/+1
We never invalidate PD entries and making them valid can run with other users in parallel. Signed-off-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-11-20drm/radeon: fence BO_VAs manuallyChristian König1-11/+6
This allows us to finally remove the VM fence and so allow concurrent use of it from different engines. Signed-off-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-11-20drm/radeon: use one VMID for each ringChristian König1-23/+36
Use multiple VMIDs for each VM, one for each ring. That allows us to execute flushes separately on each ring, still not ideal cause in a lot of cases rings can share IDs. Signed-off-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-11-20drm/radeon: track VM update fences separatelyChristian König1-12/+13
Note for each fence if it's a VM page table update or not. This allows us to determine the last VM update in a sync object and so to figure out if we need to flush the TLB or not. Signed-off-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-11-20drm/radeon: fence PT updates manually v2Christian König1-24/+41
This allows us to add the real execution fence as shared. v2: fix typo Signed-off-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-11-20drm/radeon: split semaphore and sync object handling v2Christian König1-2/+2
Previously we just allocated space for four hardware semaphores in each software semaphore object. Make software semaphore objects represent only one hardware semaphore address again by splitting the sync code into it's own object. v2: fix typo in comment Signed-off-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-11-20drm/radeon: remove unnecessary VM syncsChristian König1-2/+0
The PD/PTs reservation object now contains everything needed. Signed-off-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-11-20drm/radeon: stop re-reserving the BO in radeon_vm_bo_set_addrChristian König1-2/+2
That's useless when all callers drop the reservation immediately after calling the function. Signed-off-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-11-20drm/radeon: rework vm_flush parametersChristian König1-1/+2
Use ring structure instead of index and provide vm_id and pd_addr separately. Signed-off-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-10-28drm/radeon: Use drm_malloc_ab instead of kmalloc_arrayMichel Dänzer1-2/+2
Should avoid kmalloc failures due to large number of array entries. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=81991 Reviewed-by: Dave Airlie <airlied@redhat.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Cc: stable@vger.kernel.org
2014-10-03drm/radeon: export reservation_object from dmabuf to ttmMaarten Lankhorst1-2/+3
Adds an extra argument to radeon_bo_create, which is only used in radeon_prime.c. Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-10-03drm/radeon: cope with foreign fences inside the reservation objectMaarten Lankhorst1-2/+2
Not the whole world is a radeon! :-) Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-09-11drm/radeon: add the infrastructure for concurrent buffer accessChristian König1-12/+4
This allows us to specify if we want to sync to the shared fences of a reservation object or not. Signed-off-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-09-11drm/ttm: allow fence to be added as sharedChristian König1-0/+3
This patch adds a new flag to the ttm_validate_buffer list to add the fence as shared to the reservation object. Signed-off-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-09-02drm/ttm: flip the switch, and convert to dma_fenceMaarten Lankhorst1-3/+12
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com>
2014-09-01drm/ttm: add interruptible parameter to ttm_eu_reserve_buffersMaarten Lankhorst1-1/+1
It seems some drivers really want this as a parameter, like vmwgfx. Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com>
2014-08-26Merge branch 'drm-next-3.18' of git://people.freedesktop.org/~agd5f/linux into drm-nextDave Airlie1-0/+3
radeon userptr support. * 'drm-next-3.18' of git://people.freedesktop.org/~agd5f/linux: drm/radeon: allow userptr write access under certain conditions drm/radeon: add userptr flag to register MMU notifier v3 drm/radeon: add userptr flag to directly validate the BO to GTT drm/radeon: add userptr flag to limit it to anonymous memory v2 drm/radeon: add userptr support v8 Conflicts: drivers/gpu/drm/radeon/radeon_prime.c
2014-08-19drm/radeon: re-enable selective GPUVM flushingAlex Deucher1-3/+1
Now that the PFP and ME synchronization is fixed, we can enable this again reliably. Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Tested-by: Michel Dänzer <michel.daenzer@amd.com>
2014-08-18drm/radeon: Only flush HDP cache for indirect buffers from userspaceMichel Dänzer1-3/+3
It isn't necessary for command streams generated by the kernel (at least not while we aren't storing ring or indirect buffers in VRAM). Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-15drm/radeon: Always flush VM again on < CIKMichel Dänzer1-1/+3
Not doing this causes piglit hangs[0] on my Cape Verde card. No issues on Bonaire and Kaveri though. [0] Same symptoms as those fixed on CIK by 'drm/radeon: set VM base addr using the PFP v2'. Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-15drm/radeon: add a check for allocation failure (v2)Dan Carpenter1-0/+4
We can easily return -ENOMEM here if kzalloc() fails. v2: agd5f: drop the vm mutex Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-11drm/radeon: add userptr support v8Christian König1-0/+3
This patch adds an IOCTL for turning a pointer supplied by userspace into a buffer object. It imposes several restrictions upon the memory being mapped: 1. It must be page aligned (both start/end addresses, i.e ptr and size). 2. It must be normal system memory, not a pointer into another map of IO space (e.g. it must not be a GTT mmapping of another object). 3. The BO is mapped into GTT, so the maximum amount of memory mapped at all times is still the GTT limit. 4. The BO is only mapped readonly for now, so no write support. 5. List of backing pages is only acquired once, so they represent a snapshot of the first use. Exporting and sharing as well as mapping of buffer objects created by this function is forbidden and results in an -EPERM. v2: squash all previous changes into first public version v3: fix tabs, map readonly, don't use MM callback any more v4: set TTM_PAGE_FLAG_SG so that TTM never messes with the pages, pin/unpin pages on bind/unbind instead of populate/unpopulate v5: rebased on 3.17-wip, IOCTL renamed to userptr, reject any unknown flags, better handle READONLY flag, improve permission check v6: fix ptr cast warning, use set_page_dirty/mark_page_accessed on unpin v7: add warning about it's availability in the API definition v8: drop access_ok check, fix VM mapping bits Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> (v4) Reviewed-by: Jérôme Glisse <jglisse@redhat.com> (v4) Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-05drm/radeon: update IB size estimation for VMChristian König1-13/+30
That should allow us to allocate bigger BOs. Signed-off-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-05drm/radeon: split PT setup in more functionsChristian König1-14/+53
Move the decision what to use into the common VM code. Signed-off-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-05drm/radeon: take a BO reference on VM cleanupChristian König1-2/+6
This closes a small window where the GPU might have accessed freed up memory. Signed-off-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-05drm/radeon: try to enable VM flushing once moreChristian König1-2/+1
Let's try to fix bugs related to this instead of just disabling it. Signed-off-by: Christian König <christian.koenig@amd.com> Tested-by: Michel Dänzer <michel.daenzer@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-05drm/radeon: use an intervall tree to manage the VMA v2Alex Deucher1-53/+44
Scales much better than scanning the address range linearly. v2: store pfn instead of address Signed-off-by: Christian König <christian.koenig@amd.com> Tested-by: Michel Dänzer <michel.daenzer@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-05drm/radeon: invalidate moved BOs in the VM (v2)Christian König1-10/+41
Don't wait for the BO to be used again, just update the PT on the next VM use. v2: remove stray semicolon. Signed-off-by: Christian König <christian.koenig@amd.com> Tested-by: Michel Dänzer <michel.daenzer@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-05drm/radeon: add trace_radeon_vm_flushChristian König1-0/+1
Signed-off-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-05drm/radeon: Allow write-combined CPU mappings of BOs in GTT (v2)Michel Dänzer1-2/+6
v2: fix rebase onto drm-fixes Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-07-22drm/radeon: fix error handling in radeon_vm_bo_set_addrChristian König1-0/+4
Signed-off-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-07-21drm/radeon: fix VM IB handlingChristian König1-0/+1
Calling radeon_vm_bo_find on the IB BO during CS is illegal and can lead to an crash. Signed-off-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-07-21drm/radeon: fix handling of radeon_vm_bo_rmv v3Christian König1-24/+58
v3: completely rewritten. We now just remember which areas of the PT to clear and do so on the next command submission. Bug: https://bugs.freedesktop.org/show_bug.cgi?id=79980 Signed-off-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-07-02drm/radeon: page table BOs are kernel allocationsChristian König1-2/+2
Userspace shouldn't be able to access them. Signed-off-by: Christian König <christian.koenig@amd.com> Cc: stable@vger.kernel.org Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-06-09drm/radeon: make vm_block_size a module parameterChristian König1-10/+10
Signed-off-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-06-09drm/radeon: rename alt_domain to allowed_domainsChristian König1-4/+4
And also domain to prefered_domains. That matches better what those values represent. Signed-off-by: Christian König <christian.koenig@amd.com> Cc: Marek Olšák <maraeo@gmail.com> Reviewed-by: Marek Olšák <marek.olsak@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-06-05Merge commit '9e9a928eed8796a0a1aaed7e0b676db86ba84594' into drm-nextDave Airlie1-6/+11
Merge drm-fixes into drm-next. Both i915 and radeon need this done for later patches. Conflicts: drivers/gpu/drm/drm_crtc_helper.c drivers/gpu/drm/i915/i915_drv.h drivers/gpu/drm/i915/i915_gem.c drivers/gpu/drm/i915/i915_gem_execbuffer.c drivers/gpu/drm/i915/i915_gem_gtt.c
2014-06-02drm/radeon: add proper support for RADEON_VM_BLOCK_SIZE v2Christian König1-1/+3
This patch makes it possible to decide how many address bits are spend on the page directory vs the page tables. v2: remove unintended change Signed-off-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-06-02drm/radeon: add large PTE support for NI, SI and CIK v5Christian König1-7/+84
This patch implements support for VRAM page table entry compression. PTE construction is enhanced to identify physically contiguous page ranges and mark them in the PTE fragment field. L1/L2 TLB support is enabled for 64KB (SI/CIK) and 256KB (NI) PTE fragments, significantly improving TLB utilization for VRAM allocations. Linear store bandwidth is improved from 60GB/s to 125GB/s on Pitcairn. Unigine Heaven 3.0 sees an average improvement from 24.7 to 27.7 FPS on default settings at 1920x1200 resolution with vsync disabled. See main comment in radeon_vm.c for a technical description. v2 (chk): rebased and simplified. v3 (chk): add missing hw setup v4 (chk): rebased on current drm-fixes-3.15 v5 (chk): fix comments and commit text Signed-off-by: Jay Cornwall <jay@jcornwall.me> Signed-off-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-06-02drm/radeon: sync page table updatesChristian König1-2/+7
Only necessary if we don't use the same engine for buffer moves and table updates. Signed-off-by: Christian König <christian.koenig@amd.com>
2014-06-02drm/radeon: fix vm buffer size estimationChristian König1-1/+1
Only relevant if we got VM_BLOCK_SIZE>9, but better save than sorry. Signed-off-by: Christian König <christian.koenig@amd.com>
2014-05-30drm/radeon: only allocate necessary size for vm bo listChristian König1-3/+3
No need to always allocate the theoretical maximum here. Signed-off-by: Christian König <christian.koenig@amd.com>
Copyright © 1996 – 2026 Jason A. Donenfeld. All Rights Reverse Engineered.