aboutsummaryrefslogtreecommitdiff
path: root/drivers/gpu
AgeCommit message (Collapse)AuthorFilesLines
2017-01-09drm/i915: don't open code the pdpe/pml4e clearingMatthew Auld1-14/+2
Now that it's obvious what the helpers do, we can simplify the code somewhat by using them when clearing the pdpe/pml4e with the relevant scratch entry. Cc: Joonas Lahtinen <[email protected]> Cc: Chris Wilson <[email protected]> Signed-off-by: Matthew Auld <[email protected]> Reviewed-by: Michał Winiarski <[email protected]> Signed-off-by: Joonas Lahtinen <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2017-01-09drm/i915: s/gen8_setup_page_directory_pointer/gen8_setup_pml4e/Matthew Auld1-5/+5
The function name gen8_setup_page_directory_pointer is misleading, and only serves to confuse the reader, it's not setting up a pdp, but rather encoding a specific pml4e with a given pdp. Cc: Joonas Lahtinen <[email protected]> Cc: Chris Wilson <[email protected]> Signed-off-by: Matthew Auld <[email protected]> Reviewed-by: Chris Wilson <[email protected]> Signed-off-by: Joonas Lahtinen <[email protected]>
2017-01-09drm/i915: s/gen8_setup_page_directory/gen8_setup_pdpe/Matthew Auld1-5/+5
The function name gen8_setup_page_directory is misleading, and only serves to confuse the reader, it's not setting up a pd, but rather encoding a specific pdpe with a given pd. Cc: Joonas Lahtinen <[email protected]> Cc: Chris Wilson <[email protected]> Signed-off-by: Matthew Auld <[email protected]> Reviewed-by: Chris Wilson <[email protected]> Signed-off-by: Joonas Lahtinen <[email protected]>
2017-01-09drm: fix compilations issues introduced by "drm: allow to use mmuless SoC"Benjamin Gaignard11-12/+12
Removing MMU configuration flag from DRM make few automatic build failed when they answer yes to all flags. Add asm/vga.h file on Blackfin architecture to not broke compilation. Signed-off-by: Benjamin Gaignard <[email protected]> Fixes: 62a0d98a188c ("drm: allow to use mmuless SoC") Signed-off-by: Daniel Vetter <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2017-01-09drm: Change the return type of the unload hook to voidGabriel Krisman Bertazi32-69/+38
The integer returned by the unload hook is ignored by the drm core, so let's make it void. This patch was created using the following Coccinelle semantic script (except for the declaration and comment in drm_drv.h): Compile-tested only. // <smpl> @ get_name @ struct drm_driver drv; identifier fn; @@ drv.unload = fn; @ replace_type @ identifier get_name.fn; @@ - int + void fn (...) { ... } @ remove_return_param @ identifier get_name.fn; @@ void fn (...) { <... if (...) return - ... ; ...> } @ drop_final_return @ identifier get_name.fn; @@ void fn (...) { ... - return 0; } // </smpl> Suggested-by: Daniel Vetter <[email protected]> Signed-off-by: Gabriel Krisman Bertazi <[email protected]> Acked-by: Christian König <[email protected]>. Signed-off-by: Daniel Vetter <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2017-01-09drm: add more document for drm_crtc_from_index()Shawn Guo2-2/+5
Add a bit more document for function drm_crtc_from_index() to cross link it with drm_crtc_from_index(), and explain that the function is useful in vblank code. While at it, add cross link comment for drm_plane_from_index() as well. Signed-off-by: Shawn Guo <[email protected]> Signed-off-by: Daniel Vetter <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2017-01-09drm: remove useless parameters from drm_pick_cmdline_mode functionVincent Abriou2-5/+4
drm_pick_cmdline_mode width and height parameters are useless. Just remove them. Cc: Daniel Vetter <[email protected]> Cc: Jani Nikula <[email protected]> Signed-off-by: Vincent Abriou <[email protected]> Signed-off-by: Daniel Vetter <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2017-01-09drm/i915/gvt: cleanup GFP flagsJike Song1-24/+15
In gvt, almost all memory allocations are in sleepable contexts. It's fault-prone to use GFP_ATOMIC everywhere. Replace it with GFP_KERNEL wherever possible. Signed-off-by: Jike Song <[email protected]> Signed-off-by: Zhenyu Wang <[email protected]>
2017-01-09drm/i915: Update DRIVER_DATE to 20170109Daniel Vetter1-2/+2
Signed-off-by: Daniel Vetter <[email protected]>
2017-01-09drm/bridge: analogix dp: Fix runtime PM state on driver bindMarek Szyprowski1-0/+7
Analogix_dp_bind() can be called from component framework, which doesn't guarantee proper runtime PM state of the device during bind operation, so ensure that device is runtime active before doing any register access. This ensures that the power domain, to which DP module belongs, is turned on. While at it, also fix the unbalanced call to phy_power_on() in analogix_dp_bind() function. This patch solves the following kernel oops on Samsung Exynos5250 Snow board: Unhandled fault: imprecise external abort (0x406) at 0x00000000 pgd = c0004000 [00000000] *pgd=00000000 Internal error: : 406 [#1] PREEMPT SMP ARM Modules linked in: CPU: 0 PID: 75 Comm: kworker/0:2 Not tainted 4.9.0 #1046 Hardware name: SAMSUNG EXYNOS (Flattened Device Tree) Workqueue: events deferred_probe_work_func task: ee272300 task.stack: ee312000 PC is at analogix_dp_enable_sw_function+0x18/0x2c LR is at analogix_dp_init_dp+0x2c/0x50 ... [<c03fcb38>] (analogix_dp_enable_sw_function) from [<c03fa9c4>] (analogix_dp_init_dp+0x2c/0x50) [<c03fa9c4>] (analogix_dp_init_dp) from [<c03fab6c>] (analogix_dp_bind+0x184/0x42c) [<c03fab6c>] (analogix_dp_bind) from [<c03fdb84>] (component_bind_all+0xf0/0x218) [<c03fdb84>] (component_bind_all) from [<c03ed64c>] (exynos_drm_load+0x134/0x200) [<c03ed64c>] (exynos_drm_load) from [<c03d5058>] (drm_dev_register+0xa0/0xd0) [<c03d5058>] (drm_dev_register) from [<c03d66b8>] (drm_platform_init+0x58/0xb0) [<c03d66b8>] (drm_platform_init) from [<c03fe0c4>] (try_to_bring_up_master+0x14c/0x188) [<c03fe0c4>] (try_to_bring_up_master) from [<c03fe188>] (component_add+0x88/0x138) [<c03fe188>] (component_add) from [<c0403a38>] (platform_drv_probe+0x50/0xb0) [<c0403a38>] (platform_drv_probe) from [<c0402470>] (driver_probe_device+0x1f0/0x2a8) [<c0402470>] (driver_probe_device) from [<c0400a54>] (bus_for_each_drv+0x44/0x8c) [<c0400a54>] (bus_for_each_drv) from [<c04021f8>] (__device_attach+0x9c/0x100) [<c04021f8>] (__device_attach) from [<c04018e8>] (bus_probe_device+0x84/0x8c) [<c04018e8>] (bus_probe_device) from [<c0401d1c>] (deferred_probe_work_func+0x60/0x8c) [<c0401d1c>] (deferred_probe_work_func) from [<c012fc14>] (process_one_work+0x120/0x318) [<c012fc14>] (process_one_work) from [<c012fe34>] (process_scheduled_works+0x28/0x38) [<c012fe34>] (process_scheduled_works) from [<c0130048>] (worker_thread+0x204/0x4ac) [<c0130048>] (worker_thread) from [<c01352c4>] (kthread+0xd8/0xf4) [<c01352c4>] (kthread) from [<c0107978>] (ret_from_fork+0x14/0x3c) Code: e59035f0 e5935018 f57ff04f e3c55001 (f57ff04e) ---[ end trace 3d1d0d87796de344 ]--- Reviewed-by: Sean Paul <[email protected]> Signed-off-by: Marek Szyprowski <[email protected]> Signed-off-by: Archit Taneja <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2017-01-09drm/i915/gvt/kvmgt: return meaningful error for vgpu creating failureJike Song1-4/+10
The vgpu_create() routine we called returns meaningful errors to indicate failures, so we'd better to pass it to our caller, the mdev framework, whereby the sysfs is able to tell userspace what happened. Signed-off-by: Jike Song <[email protected]> Signed-off-by: Zhenyu Wang <[email protected]>
2017-01-09drm/i915/gvt: cleanup opregion memory allocation codeJike Song2-6/+5
According to the spec, ACPI OpRegion must be placed at a physical address below 4G. That is, for a vGPU it must be associated with a GPA below 4G, but on host side, it doesn't matter where the backing pages actually are. So when allocating pages from host, the GFP_DMA32 flag is unnecessary. Also the allocation is from a sleepable context, so GFP_ATOMIC is also unnecessary. This patch also removes INTEL_GVT_OPREGION_PORDER and use get_order() instead. Signed-off-by: Jike Song <[email protected]> Signed-off-by: Zhenyu Wang <[email protected]>
2017-01-09drm/i915/gvt: destroy the allocated idr on vgpu creating failuresJike Song1-1/+3
Once idr_alloc gets called data is allocated within the idr list, if any error occurs afterwards, we should undo that by idr_remove on the error path. Signed-off-by: Jike Song <[email protected]> Signed-off-by: Zhenyu Wang <[email protected]>
2017-01-09drm/i915/gvt: init/destroy vgpu_idr properlyJike Song1-1/+7
An idr should be initialized before use and destroyed afterwards. Signed-off-by: Jike Song <[email protected]> Signed-off-by: Zhenyu Wang <[email protected]>
2017-01-09drm/i915/gvt: dec vgpu->running_workload_num after the workload is really doneChangbin Du1-5/+5
The vgpu->running_workload_num is used to determine whether a vgpu has any workload running or not. So we should make sure the workload is really done before we dec running_workload_num. Function complete_current_workload is not the right place to do it, since this function is still processing the workload. This patch move the dec op afterward. v2: move dec op before wake_up(&scheduler->workload_complete_wq) (Min He) Signed-off-by: Changbin Du <[email protected]> Reviewed-by: Min He <[email protected]> Signed-off-by: Zhenyu Wang <[email protected]>
2017-01-09drm/i915/gvt: fix use after free for workloadChangbin Du1-2/+2
In the function workload_thread(), we invoke complete_current_workload() to cleanup the just processed workload (workload will be freed there). So we cannot access workload->req after that. This patch move complete_current_workload() afterward. Signed-off-by: Changbin Du <[email protected]> Signed-off-by: Zhenyu Wang <[email protected]>
2017-01-09drm/i915/gvt: remove duplicated definitionZhenyu Wang1-7/+0
Remove duplicated definition for resource size in aperture_gm.c which are already defined in gvt.h. Need only one to take effect. Signed-off-by: Zhenyu Wang <[email protected]>
2017-01-09drm/i915/gvt: adjust high memory size for default vGPU typeZhenyu Wang1-2/+2
Previous high mem size initialized for vGPU type was too small which caused failure for some VMs. This trys to take minimal value of 384MB for each VM and enlarge default high mem size to make guest driver happy. Signed-off-by: Zhenyu Wang <[email protected]>
2017-01-09drm/i915/gvt: print correct value for untracked mmioPei Zhang1-15/+16
In function intel_vgpu_emulate_mmio_read, the untracked mmio register is dumped through kernel log, but the register value is not correct. This patch fixes this issue. V2: fix the fromat warning from checkpatch.pl. Signed-off-by: Pei Zhang <[email protected]> Signed-off-by: Zhenyu Wang <[email protected]>
2017-01-09drm/i915/gvt: always use readq and writeqChangbin Du1-13/+2
The readq and writeq are already offered by drm_os_linux.h. So we can use them directly whithout dectecting their presence. This patch removed the duplicated code. Signed-off-by: Changbin Du <[email protected]> Signed-off-by: Zhenyu Wang <[email protected]>
2017-01-09drm/i915/gvt: fix return value in mul_force_wake_writeChangbin Du1-1/+1
All mmio handlers should return a negetive value for failure, not 1. Signed-off-by: Changbin Du <[email protected]> Signed-off-by: Zhenyu Wang <[email protected]>
2017-01-09drm/i915/gvt: fix error handing of tlb_control emulationChangbin Du1-4/+2
Return ealier for a invalid access, else it would false set tlb flag for RCS. Signed-off-by: Changbin Du <[email protected]> Signed-off-by: Zhenyu Wang <[email protected]>
2017-01-09drm/i915/gvt: verify functions types in new_mmio_info()Nicolas Iooss1-2/+3
The current prototype of new_mmio_info() uses void* for parameters read and write, which are functions with precise calling conventions (argument types and return type). Write down these conventions in new_mmio_info() definition. This has been reported by the following warnings when clang is used to build the kernel: drivers/gpu/drm/i915/gvt/handlers.c:124:21: error: pointer type mismatch ('void *' and 'int (*)(struct intel_vgpu *, unsigned int, void *, unsigned int)') [-Werror,-Wpointer-type-mismatch] info->read = read ? read : intel_vgpu_default_mmio_read; ^ ~~~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~ drivers/gpu/drm/i915/gvt/handlers.c:125:23: error: pointer type mismatch ('void *' and 'int (*)(struct intel_vgpu *, unsigned int, void *, unsigned int)') [-Werror,-Wpointer-type-mismatch] info->write = write ? write : intel_vgpu_default_mmio_write; ^ ~~~~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ This allows the compiler to detect that sbi_ctl_mmio_write() returns a "bool" value instead of an expected "int" one. Fix this. Signed-off-by: Nicolas Iooss <[email protected]> Signed-off-by: Zhenyu Wang <[email protected]>
2017-01-09Merge tag 'drm-misc-next-2016-12-30' of ↵Dave Airlie194-1351/+3953
git://anongit.freedesktop.org/git/drm-misc into drm-next First -misc pull for 4.11: - drm_mm rework + lots of selftests (Chris Wilson) - new connector_list locking+iterators - plenty of kerneldoc updates - format handling rework from Ville - atomic helper changes from Maarten for better plane corner-case handling in drivers, plus the i915 legacy cursor patch that needs this - bridge cleanup from Laurent - plus plenty of small stuff all over - also contains a merge of the 4.10 docs tree so that we could apply the dma-buf kerneldoc patches It's a lot more than usual, but due to the merge window blackout it also covers about 4 weeks, so all in line again on a per-week basis. The more annoying part with no pull request for 4 weeks is managing cross-tree work. The -intel pull request I'll follow up with does conflict quite a bit with -misc here. Longer-term (if drm-misc keeps growing) a drm-next-queued to accept pull request for the next merge window during this time might be useful. I'd also like to backmerge -rc2+this into drm-intel next week, we have quite a pile of patches waiting for the stuff in here. * tag 'drm-misc-next-2016-12-30' of git://anongit.freedesktop.org/git/drm-misc: (126 commits) drm: Add kerneldoc markup for new @scan parameters in drm_mm drm/mm: Document locking rules drm: Use drm_mm_insert_node_in_range_generic() for everyone drm: Apply range restriction after color adjustment when allocation drm: Wrap drm_mm_node.hole_follows drm: Apply tight eviction scanning to color_adjust drm: Simplify drm_mm scan-list manipulation drm: Optimise power-of-two alignments in drm_mm_scan_add_block() drm: Compute tight evictions for drm_mm_scan drm: Fix application of color vs range restriction when scanning drm_mm drm: Unconditionally do the range check in drm_mm_scan_add_block() drm: Rename prev_node to hole in drm_mm_scan_add_block() drm: Fix O= out-of-tree builds for selftests drm: Extract struct drm_mm_scan from struct drm_mm drm: Add asserts to catch overflow in drm_mm_init() and drm_mm_init_scan() drm: Simplify drm_mm_clean() drm: Detect overflow in drm_mm_reserve_node() drm: Fix kerneldoc for drm_mm_scan_remove_block() drm: Promote drm_mm alignment to u64 drm: kselftest for drm_mm and restricted color eviction ...
2017-01-09Merge branch 'drm-fixes-4.10' of git://people.freedesktop.org/~agd5f/linux ↵Dave Airlie20-105/+145
into drm-fixes Fixes for 4.10: - Polaris 12 support - Add new amd-gfx mailing list to MAINTAINERS file - UVD clockgating fix - SI dpm fixes * 'drm-fixes-4.10' of git://people.freedesktop.org/~agd5f/linux: drm/amdgpu: drop verde dpm quirks drm/radeon: drop verde dpm quirks drm/radeon: update smc firmware selection for SI drm/amdgpu: update si kicker smc firmware drm/amd/powerplay: extend smu's response timeout time. drm/amdgpu: remove static integer for uvd pp state drm/amd/amdgpu: add Polaris12 PCI ID drm/amdgpu/powerplay: add Polaris12 support drm/amd/amdgpu: add Polaris12 support (v3) MAINTAINERS: Update mailing list for radeon and amdgpu
2017-01-09Merge branch 'msm-fixes-4.10' of ↵Dave Airlie3-9/+21
git://people.freedesktop.org/~robclark/linux into drm-fixes A few fixes for 4.10.. the first fixes a long-standing logic bug, that by luck (ie. size of packets written into RB for a submit) wasn't hit on a3xx/a4xx but was causing intermittent GPU lockups on a5xx. And a couple other robustness issues that Jordan noticed. * 'msm-fixes-4.10' of git://people.freedesktop.org/~robclark/linux: drm/msm: Verify that MSM_SUBMIT_BO_FLAGS are set drm/msm: Put back the vaddr in submit_reloc() drm/msm: Ensure that the hardware write pointer is valid
2017-01-09Merge tag 'meson-drm-fixes-for-4.10' of ↵Dave Airlie3-1/+23
git://people.freedesktop.org/~narmstrong/linux into drm-fixes - plan atomic check oops fix - fix CVBS init when HDMI is configured by bootloader - fix CVBS VDAC disable * tag 'meson-drm-fixes-for-4.10' of git://people.freedesktop.org/~narmstrong/linux: drm/meson: Fix CVBS VDAC disable drm/meson: Fix CVBS initialization when HDMI is configured by bootloader drm/meson: Fix plane atomic check when no crtc for the plane
2017-01-09Merge tag 'tilcdc-4.10-fixes' of https://github.com/jsarha/linux into drm-fixesDave Airlie1-9/+18
tilcdc fixes for v4.10. * tag 'tilcdc-4.10-fixes' of https://github.com/jsarha/linux: drm: tilcdc: simplify the recovery from sync lost error on rev1
2017-01-09Merge tag 'drm-misc-fixes-2017-01-04' of ↵Dave Airlie1-1/+3
git://anongit.freedesktop.org/git/drm-misc into drm-fixes single drm fix. * tag 'drm-misc-fixes-2017-01-04' of git://anongit.freedesktop.org/git/drm-misc: drm: Clean up planes in atomic commit helper failure path
2017-01-06drm/amdgpu: drop verde dpm quirksAlex Deucher1-13/+0
Port of radeon change to amdgpu. Acked-by: Edward O'Callaghan <[email protected]> Signed-off-by: Alex Deucher <[email protected]> Cc: [email protected]
2017-01-06drm/radeon: drop verde dpm quirksAlex Deucher1-13/+0
fixes: https://bugs.freedesktop.org/show_bug.cgi?id=98897 https://bugs.launchpad.net/bugs/1651981 Acked-by: Edward O'Callaghan <[email protected]> Signed-off-by: Alex Deucher <[email protected]> Cc: [email protected] Cc: Adrian Fiergolski <[email protected]>
2017-01-06drm/radeon: update smc firmware selection for SIAlex Deucher1-32/+28
Use the appropriate smc firmware for each chip revision. Using the wrong one can cause stability issues. Acked-by: Edward O'Callaghan <[email protected]> Signed-off-by: Alex Deucher <[email protected]> Cc: [email protected]
2017-01-06drm/amdgpu: update si kicker smc firmwareFlora Cui1-29/+28
Use the appropriate smc firmware for each chip revision. Using the wrong one can cause stability issues. Acked-by: Edward O'Callaghan <[email protected]> Signed-off-by: Flora Cui <[email protected]> Reviewed-by: Junwei Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]> Cc: [email protected]
2017-01-06drm/amd/powerplay: extend smu's response timeout time.Rex Zhu1-1/+1
Signed-off-by: Rex Zhu <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-01-06drm/amdgpu: remove static integer for uvd pp stateYintian Tao1-5/+0
At two gpu core condition, static integer will cause that second gpu core uvd state setting will be directly skipped due to the first one setting Signed-off-by: Yintian Tao <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-01-06drm/amd/amdgpu: add Polaris12 PCI IDJunwei Zhang1-0/+7
Signed-off-by: Junwei Zhang <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Reviewed-by: Christian König <[email protected]> Reviewed-by: Ken Wang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-01-06drm/amdgpu/powerplay: add Polaris12 supportJunwei Zhang4-2/+5
Signed-off-by: Junwei Zhang <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Reviewed-by: Christian König <[email protected]> Reviewed-by: Ken Wang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-01-06drm/amd/amdgpu: add Polaris12 support (v3)Junwei Zhang11-10/+76
v2: agd: squash in various fixes v3: agd: squash in: drm/amdgpu: remove unnecessary smc sk firmware for polaris12 Signed-off-by: Junwei Zhang <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Reviewed-by: Christian König <[email protected]> Reviewed-by: Ken Wang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-01-06Merge tag 'vfio-v4.10-rc3' of git://github.com/awilliam/linux-vfioLinus Torvalds1-12/+12
Pull VFIO fixes from Alex Williamson: - Add mtty sample driver properly into build system (Alex Williamson) - Restore type1 mapping performance after mdev (Alex Williamson) - Fix mdev device race (Alex Williamson) - Cleanups to the mdev ABI used by vendor drivers (Alex Williamson) - Build fix for old compilers (Arnd Bergmann) - Fix sample driver error path (Dan Carpenter) - Handle pci_iomap() error (Arvind Yadav) - Fix mdev ioctl return type (Paul Gortmaker) * tag 'vfio-v4.10-rc3' of git://github.com/awilliam/linux-vfio: vfio-mdev: fix non-standard ioctl return val causing i386 build fail vfio-pci: Handle error from pci_iomap vfio-mdev: fix some error codes in the sample code vfio-pci: use 32-bit comparisons for register address for gcc-4.5 vfio-mdev: Make mdev_device private and abstract interfaces vfio-mdev: Make mdev_parent private vfio-mdev: de-polute the namespace, rename parent_device & parent_ops vfio-mdev: Fix remove race vfio/type1: Restore mapping performance with mdev support vfio-mdev: Fix mtty sample driver building
2017-01-06Merge branch 'stable/for-linus-4.10' of ↵Linus Torvalds1-10/+1
git://git.kernel.org/pub/scm/linux/kernel/git/konrad/swiotlb Pull swiotlb fixes from Konrad Rzeszutek Wilk: "This has one fix to make i915 work when using Xen SWIOTLB, and a feature from Geert to aid in debugging of devices that can't do DMA outside the 32-bit address space. The feature from Geert is on top of v4.10 merge window commit (specifically you pulling my previous branch), as his changes were dependent on the Documentation/ movement patches. I figured it would just easier than me trying than to cherry-pick the Documentation patches to satisfy git. The patches have been soaking since 12/20, albeit I updated the last patch due to linux-next catching an compiler error and adding an Tested-and-Reported-by tag" * 'stable/for-linus-4.10' of git://git.kernel.org/pub/scm/linux/kernel/git/konrad/swiotlb: swiotlb: Export swiotlb_max_segment to users swiotlb: Add swiotlb=noforce debug option swiotlb: Convert swiotlb_force from int to enum x86, swiotlb: Simplify pci_swiotlb_detect_override()
2017-01-06swiotlb: Export swiotlb_max_segment to usersKonrad Rzeszutek Wilk1-10/+1
So they can figure out what is the optimal number of pages that can be contingously stitched together without fear of bounce buffer. We also expose an mechanism for sub-users of SWIOTLB API, such as Xen-SWIOTLB to set the max segment value. And lastly if swiotlb=force is set (which mandates we bounce buffer everything) we set max_segment so at least we can bounce buffer one 4K page instead of a giant 512KB one for which we may not have space. Signed-off-by: Konrad Rzeszutek Wilk <[email protected]> Reported-and-Tested-by: Juergen Gross <[email protected]>
2017-01-06drm/i915: Drain freed objects for mmap space exhaustionChris Wilson1-13/+11
As we now use a deferred free queue for objects, simply retiring the active objects is not enough to immediately free them and recover their mmap space - we must now also drain the freed object list. Fixes: fbbd37b36fa5 ("drm/i915: Move object release to a freelist + worker" Signed-off-by: Chris Wilson <[email protected]> Cc: Joonas Lahtinen <[email protected]> Cc: <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected] Reviewed-by: Tvrtko Ursulin <[email protected]>
2017-01-06drm/i915: Purge loose pages if we run out of DMA remap spaceChris Wilson1-4/+18
If the DMA remap fails, one cause can be that we have too many objects pinned in a small remapping table, such as swiotlb. (DMA remapping does not trigger the shrinker by itself on its normal failure paths.) So try purging all other objects (using i915_gem_shrink_all(), sparing our own pages as we have yet to assign them to the obj->pages) and try again. If there are no pages to reclaim (and consequently no pages to unmap), the shrinker will report 0 and we fail with -ENOSPC as before. Signed-off-by: Chris Wilson <[email protected]> Reviewed-by: Tvrtko Ursulin <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2017-01-06drm/i915: Fix phys pwrite for struct_mutex-less operationChris Wilson1-30/+4
Since commit fe115628d567 ("drm/i915: Implement pwrite without struct-mutex") the lowlevel pwrite calls are now called without the protection of struct_mutex, but pwrite_phys was still asserting that it held the struct_mutex and later tried to drop and relock it. Fixes: fe115628d567 ("drm/i915: Implement pwrite without struct-mutex") Signed-off-by: Chris Wilson <[email protected]> Cc: Joonas Lahtinen <[email protected]> Cc: <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected] Reviewed-by: Tvrtko Ursulin <[email protected]>
2017-01-06drm/i915: Simplify testing for am-I-the-kernel-context?Chris Wilson5-5/+21
The kernel context (dev_priv->kernel_context) is unique in that it is not associated with any user filp - it is the only one with ctx->file_priv == NULL. This is a simpler test than comparing it against dev_priv->kernel_context which involves some pointer dancing. In checking that this is true, we notice that the gvt context is allocating itself a i915_hw_ppgtt it doesn't use and not flagging that its file_priv should be invalid. Signed-off-by: Chris Wilson <[email protected]> Reviewed-by: Tvrtko Ursulin <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2017-01-06drm/i915: Use range_overflows()Chris Wilson3-4/+8
Replace a few more open-coded overflow checks with the macro. Signed-off-by: Chris Wilson <[email protected]> Reviewed-by: Tvrtko Ursulin <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2017-01-06drm/i915: Use fixed-sized types for stolenChris Wilson3-19/+22
Stolen memory is a hardware resource of known size, so use an accurate fixed integer type rather than the ambiguous variable size_t. This was motivated by the next patch spotting inconsistencies in our types. Signed-off-by: Chris Wilson <[email protected]> Reviewed-by: Tvrtko Ursulin <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2017-01-06drm/i915: Use phys_addr_t for the address of stolen memoryChris Wilson2-24/+27
Though we know the hw is limited to keeping stolen memory inside the first 4GiB, it is clearer to the reader that we are handling physical address if we use phys_addr_t to refer to the base of stolen memory. Signed-off-by: Chris Wilson <[email protected]> Reviewed-by: Tvrtko Ursulin <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2017-01-06drm/i915: Consolidate checks for memcpy-from-wc supportChris Wilson4-3/+19
In order to silence sparse: ../drivers/gpu/drm/i915/i915_gpu_error.c:200:39: warning: Using plain integer as NULL pointer add a helper to check whether we have sse4.1 and that the desired alignment is valid for acceleration. v2: Explain the macros and split the two use cases between i915_has_memcpy_from_wc() and i915_can_memcpy_from_wc(). Reported-by: Ville Syrjälä <[email protected]> Signed-off-by: Chris Wilson <[email protected]> Cc: Ville Syrjälä <[email protected]> Reviewed-by: Tvrtko Ursulin <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2017-01-06drm: crc: Call wake_up_interruptible() each time there is a new CRC entryBenjamin Gaignard1-0/+2
Each time new data has being added in CRC list inform reader by calling wake_up_interruptible(). This should avoid to do it in all drivers. Signed-off-by: Benjamin Gaignard <[email protected]> Cc: Tomeu Vizoso <[email protected]> Cc: Daniel Vetter <[email protected]> Reviewed-by: Tomeu Vizoso <[email protected]> Signed-off-by: Daniel Vetter <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
Powered by cgit, Gruvboxed by Blaster4385.