summaryrefslogtreecommitdiffstats
path: root/drivers/gpu/nvgpu/gk20a/channel_gk20a.c
Commit message (Collapse)AuthorAge
* gpu: nvgpu: WAR for bad GPFIFO entries from userspaceAlex Waterman2015-06-11
| | | | | | | | | | | | | | | | | | Userspace sometimes sends GPFIFO entries with a zero length. This is a problem when the address of the pushbuffer of zero length is larger than 32 bits. The high bits are interpreted as an opcode and either triggers an operation that should not happen or is trated as invalid. Oddly, this WAR is only necessary on simulation. Change-Id: I8be007c50f46d3e35c6a0e8512be88a8e68ee739 Signed-off-by: Alex Waterman <alexw@nvidia.com> Reviewed-on: http://git-master/r/753379 Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Tested-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: http://git-master/r/755814 Reviewed-by: Automatic_Commit_Validation_User
* gpu: nvgpu: Do not pre-allocate cmdbuf entriesTerje Bergstrom2015-06-09
| | | | | | | | | | | | | | | Do not preallocate cmdbuf tracking entries. Allocate them only when needed. Bug 200104160 Change-Id: I12f8392723c301a368af1e280893ff993480477f Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: http://git-master/r/743953 Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: Konsta Holtta <kholtta@nvidia.com> Reviewed-by: Deepak Nibade <dnibade@nvidia.com> Reviewed-on: http://git-master/r/755148
* gpu: nvgpu: add per-channel refcountingKonsta Holtta2015-06-09
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Add reference counting for channels, and wait for reference count to get to 0 in gk20a_channel_free() before actually freeing the channel. Also, change free channel tracking a bit by employing a list of free channels, which simplifies the procedure of finding available channels with reference counting. Each use of a channel must have a reference taken before use or held by the caller. Taking a reference of a wild channel pointer may fail, if the channel is either not opened or in a process of being closed. Also, add safeguards for protecting accidental use of closed channels, specifically, by setting ch->g = NULL in channel free. This will make it obvious if freed channel is attempted to be used. The last user of a channel might be the deferred interrupt handler, so wait for deferred interrupts to be processed twice in the channel free procedure: once for providing last notifications to the channel and once to make sure there are no stale pointers left after referencing to the channel has been denied. Finally, fix some races in channel and TSG force reset IOCTL path, by pausing the channel scheduler in gk20a_fifo_recover_ch() and gk20a_fifo_recover_tsg(), while the affected engines have been identified, the appropriate MMU faults triggered, and the MMU faults handled. In this case, make sure that the MMU fault does not attempt to query the hardware about the failing channel or TSG ids. This should make channel recovery more safe also in the regular (i.e., not in the interrupt handler) context. Bug 1530226 Bug 1597493 Bug 1625901 Bug 200076344 Bug 200071810 Change-Id: Ib274876908e18219c64ea41e50ca443df81d957b Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Signed-off-by: Konsta Holtta <kholtta@nvidia.com> Signed-off-by: Sami Kiminki <skiminki@nvidia.com> Reviewed-on: http://git-master/r/448463 (cherry picked from commit 3f03aeae64ef2af4829e06f5f63062e8ebd21353) Reviewed-on: http://git-master/r/755147 Reviewed-by: Automatic_Commit_Validation_User
* gpu: nvgpu: Use vmalloc only when size >4KTerje Bergstrom2015-06-06
| | | | | | | | | | | | | | | When allocation size is 4k or below, we should use kmalloc. vmalloc should be used only for larged allocations. Introduce nvgpu_alloc, which checks the size, and decides the API to use. Change-Id: I593110467cd319851b27e57d1bfe8d228d3f2909 Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: http://git-master/r/743974 (cherry picked from commit 7f56aa1f0ecafbfde7286353b60e25e494674d26) Reviewed-on: http://git-master/r/753276 Reviewed-by: Automatic_Commit_Validation_User
* gpu: nvgpu: cyclestats mode E snapshots supportLeonid Moiseichuk2015-06-06
| | | | | | | | | | | | | | | | | | | | | | | That is a kernel supporting code for cyclestats mode E. Cyclestats mode E implemented following Windows-design in user-space and required the following operations to be implemented: - attach a client for shared hardware buffer of device - detach client from shared hardware buffer - flush means copy of available data from hardware buffer to private client buffers according to perfmon IDs assigned for clients - perfmon IDs management for user-space clients - a NVGPU_GPU_FLAGS_SUPPORT_CYCLE_STATS_SNAPSHOT capability added Bug 1573150 Change-Id: I9e09f0fbb2be5a95c47e6d80a2e23fa839b46f9a Signed-off-by: Leonid Moiseichuk <lmoiseichuk@nvidia.com> Reviewed-on: http://git-master/r/740653 (cherry picked from commit 79fe89fd4cea39d8ab9dbef0558cd806ddfda87f) Reviewed-on: http://git-master/r/753274 Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu:nvgpu: update channel_setup_ramfc interfaceSeshendra Gadagottu2015-06-02
| | | | | | | | | | | | | Pass flags parameter to channel_setup_ramfc for indicating nvgpu_alloc_gpfifo_args characteristics. Bug 1645628 Change-Id: Ia40b37c5c7b208d459aa84f1b022036dd5e1b599 Signed-off-by: Seshendra Gadagottu <sgadagottu@nvidia.com> Reviewed-on: http://git-master/r/744526 Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: fix channel leak with immediate closeDeepak Nibade2015-05-19
| | | | | | | | | | | | | | | | | | | If a GPU channel is closed immediately after opening without performing any operation on it, we leak that channel e.g. below command leaks a channel echo > /dev/nvhost-gpu Fix this leak by releasing the channel before returning Change-Id: I2598e3cabec6996cb1cf8066a1e6d7d5864ae02b Signed-off-by: Deepak Nibade <dnibade@nvidia.com> Reviewed-on: http://git-master/r/743235 (cherry picked from commit 428771509b4431ebe88e38061b495cabc5192327) Reviewed-on: http://git-master/r/744279 Reviewed-by: Automatic_Commit_Validation_User GVS: Gerrit_Virtual_Submit Reviewed-by: Sachin Nikam <snikam@nvidia.com>
* gpu: nvgpu: check sync existence in channel updateKonsta Holtta2015-05-18
| | | | | | | | | | | | | | | The channel sync object can get deleted before all channel updates have finished if the channel is freed before them, so work around a null dereference by testing if the sync exists. Channel and/or c->sync refcounting would be necessary for proper fix. Bug 200076344 Change-Id: Ica8ef2df9cd95cfa593cd4f41768dbb6641357b2 Signed-off-by: Konsta Holtta <kholtta@nvidia.com> Reviewed-on: http://git-master/r/734266 Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: Do not send WFI when finishing channelTerje Bergstrom2015-05-05
| | | | | | | | | | | | | | | | | The channel teardown process sends a WFI method to ensure that all work has been completed. But we also preempt the channel a while later, which also ensures that all work is completed. Remove the code for submitting WFI, and rely on preemption to handle idling the pipe. Change-Id: I2af029184440ee73e70d377f15690ddaf9b8599f Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: http://git-master/r/735067 Reviewed-on: http://git-master/r/737527 Reviewed-by: Alexander Van Brunt <avanbrunt@nvidia.com> Tested-by: Alexander Van Brunt <avanbrunt@nvidia.com> Reviewed-by: Automatic_Commit_Validation_User
* gpu: nvgpu: Reconfigure instance block with syncptTerje Bergstrom2015-05-05
| | | | | | | | | | | | Resetup RAMFC once sync point id is allocated for a channel. Change-Id: Idbac406bea1c94c89ef587dda08fddc740c1fadb Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: http://git-master/r/711302 Reviewed-on: http://git-master/r/737526 Reviewed-by: Alexander Van Brunt <avanbrunt@nvidia.com> Tested-by: Alexander Van Brunt <avanbrunt@nvidia.com> Reviewed-by: Automatic_Commit_Validation_User
* gpu: nvgpu: Sem wakeup to post eventTerje Bergstrom2015-04-04
| | | | | | | | Add posting a channel event whenever we do a wakeup due to semaphore. Change-Id: Id1765123de93bcbc0822af7926d7f4e9919ffe10 Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: http://git-master/r/726420
* gpu: nvgpu: Fix the GPFIFO submit conditionGagan Grover2015-04-04
| | | | | | | | | | | | | | | | GPFIFO HW with fifo size N can actually accept N-1 entries. Also, kernel can insert two extra entries, before and after the user GPFIFOs. So, GPFIFO with N size can accept max N-3 user gpfifos. Bug 1613125 Change-Id: I173526afb70dddf1b2b9ec0a99890335c81f0e02 Signed-off-by: Gagan Grover <ggrover@nvidia.com> Reviewed-on: http://git-master/r/725380 Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: Increase GPFIFO priv cmd queue sizeTerje Bergstrom2015-04-04
| | | | | | | | | We're doing two sync point increments, so we need to allocate space for it. Change-Id: I663abb18a930eb3955379d5f8dd1b08c8fa56897 Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: http://git-master/r/719884
* gpu: nvgpu: improve error print in check_gp_put()Deepak Nibade2015-04-04
| | | | | | | | | | | | | | Improve error print in check_gp_put() and dump put pointer stored in channel's gpfifo structure and the put pointer that we read from memory Bug 200089835 Change-Id: I7e854398811330a43d9d9bf86bf29c5986ab22b7 Signed-off-by: Deepak Nibade <dnibade@nvidia.com> Reviewed-on: http://git-master/r/722536 Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: Use common allocator for GPFIFOTerje Bergstrom2015-04-04
| | | | | | | | | | | | Reduce amount of duplicate code around memory allocation by using common helpers, and common data structure for storing results of allocations. Bug 1605769 Change-Id: I81701427ae29b298039a77f1634af9c14237812e Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: http://git-master/r/719872
* gpu: nvgpu: Use common allocator for cmd queueTerje Bergstrom2015-04-04
| | | | | | | | | | | | Reduce amount of duplicate code around memory allocation by using common helpers, and common data structure for storing results of allocations. Bug 1605769 Change-Id: If93063acbbfaa92aef530208241988427b5df8eb Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: http://git-master/r/719871
* gpu: nvgpu: Use gk20a_mem_phys instead of sg_physTerje Bergstrom2015-04-04
| | | | | | | | | There were still a couple of places using sg_phys directly. Use new gk20a_mem_phys() to make the code shorter. Change-Id: I6eb9b14e0c14a27ec39bacd06ab24e31e99769ca Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: http://git-master/r/717502
* gpu: nvgpu: use vmalloc for temp gpfifo in submitKonsta Holtta2015-04-04
| | | | | | | | | | | | | | | Use vmalloc instead of kzalloc for the temporary gpfifo buffer copied from userspace in the ioctl when submitting gpfifos. The data may be too big for kzalloc, and it doesn't need to be physically contiguous. Bug 1617747 Bug 200081843 Change-Id: I66a43d17eb13a2783bc1f0598a38abbf330b2ba6 Signed-off-by: Konsta Holtta <kholtta@nvidia.com> Reviewed-on: http://git-master/r/715207 Reviewed-by: Sami Kiminki <skiminki@nvidia.com> Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: Implement common allocator and mem_descTerje Bergstrom2015-04-04
| | | | | | | | | | Introduce mem_desc, which holds all information needed for a buffer. Implement helper functions for allocation and freeing that use this data type. Change-Id: I82c88595d058d4fb8c5c5fbf19d13269e48e422f Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: http://git-master/r/712699
* gpu: nvgpu: protect channel ioctls with a mutexKonsta Holtta2015-04-04
| | | | | | | | | | | | | | Add a big mutex for protecting the channel during ioctls, in case the userspace uses the same channel from several threads at once. The lock is taken during all operations except CHANNEL_WAIT, which could deadlock. Bug 1603482 Change-Id: Ibed962eadc9f00645abd54413dde9aaee00377ab Signed-off-by: Konsta Holtta <kholtta@nvidia.com> Reviewed-on: http://git-master/r/678871 Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: Per-chip PBDMA signatureTerje Bergstrom2015-04-04
| | | | | | | | PBDMA HW signature depends on the chip. Change-Id: If57d721d9bb77a090f967930a1aa2037bf4a16fe Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: http://git-master/r/672922
* gpu: nvgpu: Install fd after no errors can happenTerje Bergstrom2015-04-04
| | | | | | | | | | | | | | fd_install() should be called only once all other initialization is done and no errors can happen. Bug 1589104 Change-Id: I822511a64d4c6fa59c8e772a896dbd6818459c97 Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: http://git-master/r/706928 Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: Seshendra Gadagottu <sgadagottu@nvidia.com> Reviewed-by: Deepak Nibade <dnibade@nvidia.com>
* gpu: nvgpu: More robust recoveryTerje Bergstrom2015-04-04
| | | | | | | | | | | | | Make recovery a more straightforward process. When we detect a fault, trigger MMU fault, and wait for it to trigger, and complete recovery. Also reset engines before aborting channel to ensure no stray sync point increments can happen. Change-Id: Iac685db6534cb64fe62d9fb452391f43100f2999 Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: http://git-master/r/709060 (cherry picked from commit 95c62ffd9ac30a0d2eb88d033dcc6e6ff25efd6f) Reviewed-on: http://git-master/r/707443
* gpu: nvgpu: TLB invalidate after map/unmapTerje Bergstrom2015-04-04
| | | | | | | | | | | Always invalidate TLB after mapping or unmapping, and remove the delayed TLB invalidate. Change-Id: I6df3c5c1fcca59f0f9e3f911168cb2f913c42815 Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Signed-off-by: Deepak Nibade <dnibade@nvidia.com> Reviewed-on: http://git-master/r/696413 Reviewed-by: Automatic_Commit_Validation_User
* gpu: nvgpu: add open channel ioctl to ctrl nodeKonsta Holtta2015-04-04
| | | | | | | | | | | | | | | Add the ioctl to open a new gpu channel to also the control node for improved process startup performance, in addition to the current open ioctl in the channel node. The new channel fd creation is refactored to a separate function which is called from both ctrl and channel ioctls. Bug 1604952 Change-Id: I3357ceec694c0e6d7a85807183884324cb725d3a Signed-off-by: Konsta Holtta <kholtta@nvidia.com> Reviewed-on: http://git-master/r/679516 Reviewed-by: Sami Kiminki <skiminki@nvidia.com> Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: unify instance block initializationKonsta Holtta2015-04-04
| | | | | | | | | | | Create gk20a_init_inst_block() to reduce reg write clutter when initializing instance blocks, which is done in several places. Change-Id: Idcb8b604851a849e0bb6abce5743c9f4cbf98033 Signed-off-by: Konsta Holtta <kholtta@nvidia.com> Reviewed-on: http://git-master/r/672434 Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: gk20a: Optimize gpfifo entry copyJanne Hellsten2015-04-04
| | | | | | | | | | | | | | | Use memcpy for copying gpfifo inputs into the gpfifo ring buffer. This speeds up one command buffer heavy benchmark from 82 FPS to 86 FPS. Speed up is due to a) faster memory move and b) zero tracing overhead when PB tracing is disabled. Bug 1550886 Change-Id: If95ebff53745bbf59edeac32ad4f32f10f1ea7ee Signed-off-by: Janne Hellsten <jhellsten@nvidia.com> Reviewed-on: http://git-master/r/676967 Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: gk20a: Add a gpfifo wait trace pointJanne Hellsten2015-04-04
| | | | | | | | | | | | | | | Add a couple of trace points for tracking when we need to wait for space in the gpfifo ring buffer. This wait can introduce significant latencies to rendering with large gpfifo entry inputs so it's good to be able to measure how often this path is taken. Bug 1592391 Bug 1550886 Change-Id: I7f362e9c307eeffeeecaaba268ef2e3613e54597 Signed-off-by: Janne Hellsten <jhellsten@nvidia.com> Reviewed-on: http://git-master/r/674021 Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: correct channel open sequenceSeshendra Gadagottu2015-04-04
| | | | | | | | | | | | | | Corrected sequence to bind and enable channel only afer channel gpfifo alloction done. Bug 1591647 Change-Id: I539458d1b666c0403cca1abcf8271b9c8c09f52c Signed-off-by: Seshendra Gadagottu <sgadagottu@nvidia.com> Reviewed-on: http://git-master/r/671208 Reviewed-by: Automatic_Commit_Validation_User GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: unify instance block creationKonsta Holtta2015-04-04
| | | | | | | | | | | Reduce copypaste code in instance block allocation and deletion with functions purposed for that. Change-Id: I2c8ae6a317ac89e2c857dde4296cb4316b8aaafe Signed-off-by: Konsta Holtta <kholtta@nvidia.com> Reviewed-on: http://git-master/r/668698 Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: check gpfifo size against requestSam Payne2015-04-04
| | | | | | | | | | | | | | | | | if a request is submitted larger than the allocated fifo, an error is returned immediately rather than waiting for timeout while enough space becomes available in the fifo (timeout will not trigger in this case) bug 1563401 Change-Id: I264dee2673dc8722034881f9e7db7bb137a8c0c8 Signed-off-by: Sam Payne <spayne@nvidia.com> Reviewed-on: http://git-master/r/665113 Reviewed-by: Automatic_Commit_Validation_User GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: protect channel abort with submit lockKonsta Holtta2015-04-04
| | | | | | | | | | | | Bug 200065789 Change-Id: I59eb93c7929a77cd4de4be40fd7902cd05e536c7 Signed-off-by: Konsta Holtta <kholtta@nvidia.com> Reviewed-on: http://git-master/r/665655 (cherry-picked from commit 4ee1893926557b01d7058a0a4c1c23e4476d7668) Reviewed-on: http://git-master/r/668850 Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: Arto Merilainen <amerilainen@nvidia.com>
* Revert "gpu: nvgpu: Enable syncpt reclaim only on gm20b"Timo Alho2015-03-18
| | | | | | | | | | This reverts commit 8eefb93c21934b101d7f423c38d9ea384a45fad6. Bug 1585422 Change-Id: I217e0ffe6c230ee3c63d9aec1c48ce9c41770468 Signed-off-by: Timo Alho <talho@nvidia.com> Reviewed-on: http://git-master/r/659426
* gpu: nvgpu: Enable syncpt reclaim only on gm20bTerje Bergstrom2015-03-18
| | | | | | | | | | | | | gm20b has more channels than sync points. We use aggressive reclaim of sync points to offset that. Disable aggressive reclaim for gk20a because it is not needed there. Bug 1583849 Change-Id: I2a74b0504150a54cb8a97016effe20c5d905ac95 Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: http://git-master/r/657095 Reviewed-by: Deepak Nibade <dnibade@nvidia.com>
* gpu: nvgpu: Physical page bits to be per chipTerje Bergstrom2015-03-18
| | | | | | | | | | Retrieve number of physical page bits based on chip. Bug 1567274 Change-Id: I5a0f6a66be37f2cf720d66b5bdb2b704cd992234 Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: http://git-master/r/601700
* gpu: nvgpu: add trace event for channel updateKonsta Holtta2015-03-18
| | | | | | | | | | Bug 200052943 Change-Id: Ied6454bbfb5df9ab29497ecbf2aac495f6d89362 Signed-off-by: Konsta Holtta <kholtta@nvidia.com> Reviewed-on: http://git-master/r/602887 Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: flush update notifiers on suspendKonsta Holtta2015-03-18
| | | | | | | | | | | | | | If any gk20a update callback work is pending, wait for them to finish before going to suspend. Bug 200052943 Change-Id: Ib469db6e29d13ae26aaca5fceb1ccd20f18bfc3c Signed-off-by: Konsta Holtta <kholtta@nvidia.com> Reviewed-on: http://git-master/r/601034 Reviewed-by: Automatic_Commit_Validation_User GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: fix sparse warningsDeepak Nibade2015-03-18
| | | | | | | | | | | | | | | | | | | Fix below sparse warnings : warning: Using plain integer as NULL pointer warning: symbol <variable/funcion> was not declared. Should it be static? warning: Initializer entry defined twice Also, remove dead functions Bug 1573254 Change-Id: I29d71ecc01c841233cf6b26c9088ca8874773469 Signed-off-by: Deepak Nibade <dnibade@nvidia.com> Reviewed-on: http://git-master/r/593363 Reviewed-by: Amit Sharma (SW-TEGRA) <amisharma@nvidia.com> Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: Sachin Nikam <snikam@nvidia.com>
* gpu: nvgpu: protect channel update callback accessKonsta Holtta2015-03-18
| | | | | | | | | | | | | | | | | Protect callback races from spurious gk20a channel updates by testing if the channel update callback still exists when in the scheduled work (instead of only when scheduling the work to the queue), and by canceling the work when the channel is freed. Protect access to the callback and its data by accessing them together inside spinlock-protected regions. Bug 200051384 Change-Id: Ib4e1571c35f662195e1dec1e362df32ddc099eb3 Signed-off-by: Konsta Holtta <kholtta@nvidia.com> Reviewed-on: http://git-master/r/592026 Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: kernel support for suspending/resuming SMssujeet baranwal2015-03-18
| | | | | | | | | | | | | | | Kernel support for allowing a GPU debugger to suspend and resume SMs. Invocation of "suspend" on a given channel will suspend all SMs if the channel is resident, else remove the channel form the runlist. Similarly, "resume" will either resume all SMs if the channel was resident, or re-enable the channel in the runlist. Change-Id: I3b4ae21dc1b91c1059c828ec6db8125f8a0ce194 Signed-off-by: sujeet baranwal <sbaranwal@nvidia.com> Signed-off-by: Mayank Kaushik <mkaushik@nvidia.com> Reviewed-on: http://git-master/r/552115 GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
* Revert "gpu: nvgpu: GR and LTC HAL to use const structs"Sam Payne2015-03-18
| | | | | | | | | | | This reverts commit 41b82e97164138f45fbdaef6ab6939d82ca9419e. Change-Id: Iabd01fcb124e0d22cd9be62151a6552cbb27fc94 Signed-off-by: Sam Payne <spayne@nvidia.com> Reviewed-on: http://git-master/r/592221 Tested-by: Hoang Pham <hopham@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Mitch Luban <mluban@nvidia.com>
* gpu: nvgpu: GR and LTC HAL to use const structsTerje Bergstrom2015-03-18
| | | | | | | | | | | Convert GR and LTC HALs to use const structs, and initialize them with macros. Bug 1567274 Change-Id: Ia3f24a5eccb27578d9cba69755f636818d11275c Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: http://git-master/r/590371
* gpu: nvgpu: Sync gk20a and gm20b headersTerje Bergstrom2015-03-18
| | | | | | | | | | | | | Synchronize gk20a and gm20b headers. All registers which were added to gk20a are now added to gm20b, and some registers that are unused are removed. Bug 1567274 Change-Id: Ia3b7958c148e495cbff420ee56bb448db0f58680 Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: http://git-master/r/590313 GVS: Gerrit_Virtual_Submit
* gpu: nvgpu: add update callback to gk20a channelKonsta Holtta2015-03-18
| | | | | | | | | | | | Add support for a callback function with user data pointer to be scheduled from the end of gk20a_channel_update. The function and its private data are supplied when opening a new channel. Change-Id: Ib6b408855ea60d46a6a114a69c01904703019572 Signed-off-by: Konsta Holtta <kholtta@nvidia.com> Reviewed-on: http://git-master/r/552014 Reviewed-by: Arto Merilainen <amerilainen@nvidia.com> Tested-by: Arto Merilainen <amerilainen@nvidia.com>
* gpu: nvgpu: Remove get and put client routinesKonsta Holtta2015-03-18
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | gk20a_get_client() and gk20a_put_client() routines are effectively dead code. The GPU has been using pm_runtime for reference counting whether the device should be turned on or off, and gk20a_get_client() and gk20a_put_client() have had no positive effect on the behaviour. In worst case these functions trigger some issues as they may trigger code paths that should not be run. There is also a race between get/put and busy/idle. This patch removes the functions and reworks as_gk20a.c to correctly use gk20a_busy()/gk20a_idle() where put/get was required. Additionally, finalize_poweron() is moved to gk20a_busy(), similarly as it was with gk20a_get_client(). If pm_runtime is not in use, the device is only powered on and never off. Currently this affects vgpu power management since it does not use pm_runtime yet. Bug 1562096 Change-Id: I3162655f83457e9caccd9264eed36b5d51e60c52 Signed-off-by: Arto Merilainen <amerilainen@nvidia.com> Signed-off-by: Konsta Holtta <kholtta@nvidia.com> Reviewed-on: http://git-master/r/414998 Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: Seshendra Gadagottu <sgadagottu@nvidia.com> Tested-by: Seshendra Gadagottu <sgadagottu@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: Implement 64k large page supportTerje Bergstrom2015-03-18
| | | | | | | | | | | | | | | Implement support for 64kB large page size. Add an API to create an address space via IOCTL so that we can accept flags, and assign one flag for enabling 64kB large page size. Also adds APIs to set per-context large page size. This is possible only on Maxwell, so return error if caller tries to set large page size on Kepler. Default large page size is still 128kB. Change-Id: I20b51c8f6d4a984acae8411ace3de9000c78e82f Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: rename gpu ioctls and structs to nvgpuKonsta Holtta2015-03-18
| | | | | | | | | | | | | | To help remove the nvhost dependency from nvgpu, rename ioctl defines and structures used by nvgpu such that nvhost is replaced by nvgpu. Duplicate some structures as needed. Update header guards and such accordingly. Change-Id: Ifc3a867713072bae70256502735583ab38381877 Signed-off-by: Konsta Holtta <kholtta@nvidia.com> Reviewed-on: http://git-master/r/542620 Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: fix -EINVAL retval in ioctlsKonsta Holtta2015-03-18
| | | | | | | | | | | | Proper error number for invalid request number is EINVAL instead of EFAULT, so change it in ioctl calls. Change-Id: I8fddd34e012700550e9e30fe17ba7152b3a0417b Signed-off-by: Konsta Holtta <kholtta@nvidia.com> Reviewed-on: http://git-master/r/542563 Reviewed-by: Automatic_Commit_Validation_User GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: require bound as before gpfifo submitKonsta Holtta2015-03-18
| | | | | | | | | | | | Channel gpfifo cannot be submitted if the channel has no vm, so add a check for it and bail out if no as is bound. Clean up other similar checks too. Change-Id: Ibb0fe08e44e34bbaaa00ebd02dce6cc4d93ca5d9 Signed-off-by: Konsta Holtta <kholtta@nvidia.com> Reviewed-on: http://git-master/r/538887 Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: add debug events in open and ioctlKonsta Holtta2015-03-18
| | | | | | | | | | | | | Add debug events for measuring the duration of gk20a channel open and channel ioctl calls. Bug 200035111 Change-Id: Ib4837c745e73ee1690814d60bdca021821567153 Signed-off-by: Konsta Holtta <kholtta@nvidia.com> Reviewed-on: http://git-master/r/538768 Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
generated by cgit v1.2.2 (git 2.25.0) at 2024-08-27 06:45:18 -0400