summaryrefslogtreecommitdiffstats
path: root/drivers/gpu/drm/i915/gvt/cmd_parser.c
Commit message (Collapse)AuthorAgeFilesLines
* drm/i915/gvt: Fix MI_FLUSH_DW parsing with correct index checkZhenyu Wang2019-04-031-1/+1
| | | | | | | | | | | | | | | | | | | | | | commit 13bcb80b7ee79431fce361e060611134cb19e209 upstream. When MI_FLUSH_DW post write hw status page in index mode, the index value is in dword step and turned into address offset in cmd dword1. As status page size is 4K, so can't exceed that. This fixed upper bound check in cmd parser code which incorrectly stopped VM for reason of invalid MI_FLUSH_DW write index. v2: - Fix upper bound as 4K page size because index value is address offset. Fixes: be1da7070aea ("drm/i915/gvt: vGPU command scanner") Cc: stable@vger.kernel.org # v4.10+ Cc: "Zhao, Yan Y" <yan.y.zhao@intel.com> Reviewed-by: Yan Zhao <yan.y.zhao@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* drm/i915/gvt: return error on cmd accessZhao Yan2018-08-141-1/+1
| | | | | | | | | | If a register is not cmd accessible, should not just print error message. Return error here so as not to deliver this cmd. v2: return -EBADRQC to align with return value elsewhere. (kevin tian) Signed-off-by: Zhao Yan <yan.y.zhao@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
* BackMerge v4.18-rc7 into drm-nextDave Airlie2018-07-301-0/+23
|\ | | | | | | | | | | | | rmk requested this for armada and I think we've had a few conflicts build up. Signed-off-by: Dave Airlie <airlied@redhat.com>
| * drm/i915/gvt: update vreg on inhibit context lri commandHang Yuan2018-07-051-0/+23
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Commit cd7e 61b9"init mmio by lri command in vgpu inhibit context" initializes registers saved/restored in context with its vreg value through lri command in ring buffer. It relies on vreg got updated on every guest access. There is a case found that Linux guest uses lri command in inhibit-ctx to update the register. This patch adds vreg update on this case. v2: move mmio_attribute functions to gvt.h (Zhenyu) v3: use mask_mmio_write in vreg update v4: refine codes and add more comments (Zhenyu) Fixes: cd7e61b9("drm/i915/gvt: init mmio by lri command in vgpu inhibit context") Signed-off-by: Hang Yuan <hang.yuan@linux.intel.com> Signed-off-by: Weinan Li <weinan.z.li@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
* | Merge tag 'drm-intel-next-2018-06-20' of ↵Dave Airlie2018-06-281-25/+18
|\ \ | |/ |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://anongit.freedesktop.org/drm/drm-intel into drm-next Chris is doing many reworks that allow us to get full-ppgtt supported on all platforms back to HSW. As well many other fix and improvements, Including: - Use GEM suspend when aborting initialization (Chris) - Change i915_gem_fault to return vm_fault_t (Chris) - Expand VMA to Non gem object entities (Chris) - Improve logs for load failure, but quite logging on fault injection to avoid noise on CI (Chris) - Other page directory handling fixes and improvements for gen6 (Chris) - Other gtt clean-up removing redundancies and unused checks (Chris) - Reorder aliasing ppgtt fini (Chris) - Refactor of unsetting obg->mm.pages (Chris) - Apply batch location restrictions before pinning (Chris) - Ringbuffer fixes for context restore (Chris) - Execlist fixes on freeing error pointer on allocation error (Chris) - Make closing request flush mandatory (Chris) - Move GEM sanitize from resume_early to resume (Chris) - Improve debug dumps (Chris) - Silent compiler for selftest (Chris) - Other execlists changes to improve hangcheck and reset. - Many gtt page directory fixes and improvements (Chris) - Reorg context workarounds (Chris) - Avoid ERR_PTR dereference on selftest (Chris) Other GEM related work: - Stop trying to reset GPU if reset failed (Mika) - Add HW workaround for KBL to fix GPU reset (Mika) - Fix context ban and hang accounting for client (Mika) - Fixes on OA perf (Michel, Jani) - Refactor on GuC log mechanisms (Piotr) - Enable provoking vertex fix on Gen9 system (Kenneth) More ICL patches for Display enabling: - ICL - 10-bit support for HDMI (RK) - ICL - Start adding TBT PLL (Paulo) - ICL - DDI HDMK level selection (Manasi) - ICL - GMBUS GPIO pin mapping fix (Mahesh) - ICL - Adding DP_AUX_E support (James) - ICL - Display interrupts handling (DK) Other display fixes and improvements: - Fix sprite destination color keying on SKL+ (Ville) - Fixes and improvements on PCH detection, specially for non PCH systems (Jani) - Document PCH_NOP (Lucas) - Allow DBLSCAN user modes with eDP/LVDS/DSI (Ville) - Opregion and ACPI cleanup and organization (Jani) - Kill delays when activation psr (Rodrigo) - ...and a consequent fix of the psr activation flow (DK) - Fix HDMI infoframe setting (Imre) - Fix Display interrupts and modes on old gens (Ville) - Start switching to kernel unsigned int types (Jani) - Introduction to Amber Lake and Whiskey Lake platforms (Jose) - Audio clock fixes for HBR3 (RK) - Standardize i915_reg.h definitions according to our doc and checkpatch (Paulo) - Remove unused timespec_to_jiffies_timeout function (Arnd) - Increase the scope of PSR wake fix for other VBTs out there (Vathsala) - Improve debug msgs with prop name/id (Ville) - Other clean up on unecessary cursor size defines (Ville) - Enforce max hdisplay/hblank_start limits on HSW/BDW (Ville) - Make ELD pointers constant (Jani) - Fix for PSR VBT parse (Colin) - Add warn about unsupported CDCLK rates (Imre) Signed-off-by: Dave Airlie <airlied@redhat.com> # gpg: Signature made Thu 21 Jun 2018 07:12:10 AM AEST # gpg: using RSA key FA625F640EEB13CA # gpg: Good signature from "Rodrigo Vivi <rodrigo.vivi@intel.com>" # gpg: aka "Rodrigo Vivi <rodrigo.vivi@gmail.com>" # gpg: WARNING: This key is not certified with a trusted signature! # gpg: There is no indication that the signature belongs to the owner. # Primary key fingerprint: 6D20 7068 EEDD 6509 1C2C E2A3 FA62 5F64 0EEB 13CA Link: https://patchwork.freedesktop.org/patch/msgid/20180625165622.GA21761@intel.com
| * drm/i915/gvt: Enable cmd_parser support for BXT.Colin Xu2018-06-131-25/+14
| | | | | | | | | | | | | | | | | | | | | | Handle BXT cmd_parser as SKL/KBL. v2: All supported platforms share the same routines. Remove the platform check by now and let is_supported_device() be the gate keeper. Signed-off-by: Colin Xu <colin.xu@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
| * drm/i915/gvt: Add MEDIA_POOL_STATE for BXT.Colin Xu2018-06-131-0/+4
| | | | | | | | | | | | | | As referred in PRM for Broxton Graphics on 01.org Signed-off-by: Colin Xu <colin.xu@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
* | Merge tag 'gvt-fixes-2018-04-19' of https://github.com/intel/gvt-linux into ↵Jani Nikula2018-06-071-0/+1
|\ \ | |/ |/| | | | | | | | | | | | | | | | | | | | | | | drm-intel-next-fixes gvt-fixes-2018-04-19 - cmd parser error path mem leak fix (Colin) - fix dp aux header validation (Changbin) - sanity check on pfn after vfio pin page (Changbin) - fix msi eventfd put (Xiong) Signed-off-by: Jani Nikula <jani.nikula@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180419073948.4mojv7xaxxvfuyud@zhen-hp.sh.intel.com
| * drm/i915/gvt: fix memory leak of a cmd_entry struct on error exit pathColin Ian King2018-04-111-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | The error exit path when a duplicate is found does not kfree and cmd_entry struct and hence there is a small memory leak. Fix this by kfree'ing it. Detected by CoverityScan, CID#1370198 ("Resource Leak") Fixes: be1da7070aea ("drm/i915/gvt: vGPU command scanner") Signed-off-by: Colin Ian King <colin.king@canonical.com> Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
* | Merge branch 'drm-intel-next-queued' into gvt-nextZhi Wang2018-05-141-0/+1
|\ \ | | | | | | | | | Signed-off-by: Zhi Wang <zhi.a.wang@intel.com>
| * \ Merge drm/drm-next into drm-intel-next-queuedJani Nikula2018-05-021-0/+1
| |\ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | Need d224985a5e31 ("sched/wait, drivers/drm: Convert wait_on_atomic_t() usage to the new wait_var_event() API") in dinq to be able to fix https://bugs.freedesktop.org/show_bug.cgi?id=106085. Signed-off-by: Jani Nikula <jani.nikula@intel.com>
| | * \ Merge tag 'gvt-fixes-2018-04-03' of https://github.com/intel/gvt-linux into ↵Joonas Lahtinen2018-04-041-0/+1
| | |\ \ | | | |/ | | |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | drm-intel-next-fixes gvt-fixes-2018-04-03 - fix unhandled vfio ioctl return value (Gerd) - no-op user interrupt for vGPU (Zhipeng) - fix ggtt dma unmap (Changbin) - fix warning in fb decoder (Xiong) - dmabuf drm_format_mod fix (Tina) - misc cleanup Signed-off-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180403072835.kltk47gcwy7kuenv@zhen-hp.sh.intel.com
| | | * drm/i915/gvt: Make MI_USER_INTERRUPT nop in cmd parserZhipeng Gong2018-03-281-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | GVT-g dispatches request to host i915 and depends on i915 notify ring interrupt mechanism to check completion of request. For now MI_USER_INTERRUPT in guest requests is passed through in GVT-g cmd parser and i915 does not use it, which causes unnecessary interrupt handling in i915. On the other hand, if several requests from guest are combined into one request in and contain MI_USER_INTERRUPT in the middle of combined request. GVT-g still has to wait on the whole request to complete to inject user interrupts to guest. This patch makes all the MI_USER_INTERRUPT nop to save some interrupt handling. Here is test result to run glmark2 on guest for 10 seconds: host master interrupts number is reduced from 16021 to 11162 host user interrupts number is reduced from 7936 to 3536 v2: - revise commit message. (Kevin) Reviewed-by: Kevin Tian <kevin.tian@intel.com> Signed-off-by: Zhipeng Gong <zhipeng.gong@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
* | | | drm/i915/gvt: let force_to_nonpriv cmd handler only valid for LRI cmdZhao Yan2018-05-141-3/+11
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | the cmd_reg_handler() is called by cmds LRM, PIPE_CTRL, SRM... for LRM, SRM, we cannot get write data in a simple way. On other side, the force_to_nonpriv reigsters will only be written in LRI in current drivers. so we don't want to bother the handler to handle those memory access cmds, just leave a print message here. Signed-off-by: Zhao Yan <yan.y.zhao@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
* | | | drm/i915/gvt: do not return error on handling force_to_nonpriv registersZhao Yan2018-05-141-1/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Return error will cause vm hang and enter failsafe mode. However, we don't want that happen on detecting an wrong force_to_nonpriv register write. Therefore, we just omit the wrong write or patch it to default value. v2: only return 0 on detecting lri write of registers outside whitelist, but still return error on other error conditions. (zhenyu wang) Signed-off-by: Zhao Yan <yan.y.zhao@intel.com> Reviewed-by: Zhang Yulei <yulei.zhang@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
* | | | drm/i915/gvt: let NOPID be the default value of force_to_nonpriv registersZhao Yan2018-05-141-1/+8
|/ / / | | | | | | | | | | | | | | | | | | | | | Each ring has a NOPID register and currently they are regarded as default value of force_to_nonpriv registers in guest drivers Signed-off-by: Zhao Yan <yan.y.zhao@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
* / / drm/i915/gvt: scan non-privileged batch buffer for debug purposeZhao Yan2018-04-231-13/+42
|/ / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | For perfomance purpose, scanning of non-privileged batch buffer is turned off by default. But for debugging purpose, it can be turned on via debugfs. After scanning, we submit the original non-privileged batch buffer into hardware, so that the scanning is only a peeking window of guest submitted commands and will not affect the execution results. v4: - refine debugfs print format&content (zhenyu wang) - print engine id instread of engine name to prevent potential memory leak in debugfs warning message. (zhenyu wang) v3: - change vgpu->scan_nonprivbb from type bool to u32, so it is able to selectively turn on/off scanning of non-privileged batch buffer on engine level. e.g. if vgpu->scan_nonprivbb=3, then it will scan non-privileged batch buffer on engine 0 and 1. - in debugfs interface to set vgpu->scan_nonprivbb, print warning message to warn user and explicitly tell state change in kernel log (zhenyu wang) v2: - rebase - update comments for start_gma_offset (henry) Signed-off-by: Zhao Yan <yan.y.zhao@intel.com> Reviewed-by: Zhenyu Wang <zhenyuw@linux.intel.com> Signed-off-by: Zhi Wang <zhi.a.wang@intel.com>
* / drm/i915/gvt: Correct the privilege shadow batch buffer addressfred gao2018-03-151-0/+8
|/ | | | | | | | | | | | | | | | | | | | | | | | | Once the ring buffer is copied to ring_scan_buffer and scanned, the shadow batch buffer start address is only updated into ring_scan_buffer, not the real ring address allocated through intel_ring_begin in later copy_workload_to_ring_buffer. This patch is only to set the right shadow batch buffer address from Ring buffer, not include the shadow_wa_ctx. v2: - refine some comments. (Zhenyu) v3: - fix typo in title. (Zhenyu) v4: - remove the unnecessary comments. (Zhenyu) - add comments in bb_start_cmd_va update. (Zhenyu) Fixes: 0a53bc07f044 ("drm/i915/gvt: Separate cmd scan from request allocation") Cc: stable@vger.kernel.org # v4.15 Cc: Zhenyu Wang <zhenyuw@linux.intel.com> Cc: Yulei Zhang <yulei.zhang@intel.com> Signed-off-by: fred gao <fred.gao@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
* BackMerge tag 'v4.15-rc8' into drm-nextDave Airlie2018-01-181-2/+2
|\ | | | | | | | | | | | | Linux 4.15-rc8 Daniel requested this for so the intel CI won't fall over on drm-next so often.
| * Merge tag 'gvt-fixes-2018-01-08' of https://github.com/intel/gvt-linux into ↵Jani Nikula2018-01-081-2/+2
| |\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | drm-intel-fixes gvt-fixes-2018-01-08 - clear shadow entry for post-sync (Zhi) - fix stack out-of-bound warning in cmd parser (Changbin) Signed-off-by: Jani Nikula <jani.nikula@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180108061130.ucwtumhpbfbu4psu@zhen-hp.sh.intel.com
| | * drm/i915/gvt: Fix stack-out-of-bounds bug in cmd parserChangbin Du2018-01-081-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | for_each_set_bit() only accepts variable of type unsigned long, and we can not cast it from smaller types. [ 16.499365] ================================================================== [ 16.506655] BUG: KASAN: stack-out-of-bounds in find_first_bit+0x1d/0x70 [ 16.513313] Read of size 8 at addr ffff8803616cf510 by task systemd-udevd/180 [ 16.521998] CPU: 0 PID: 180 Comm: systemd-udevd Tainted: G U O 4.15.0-rc3+ #14 [ 16.530317] Hardware name: Dell Inc. OptiPlex 7040/0Y7WYT, BIOS 1.2.8 01/26/2016 [ 16.537760] Call Trace: [ 16.540230] dump_stack+0x7c/0xbb [ 16.543569] print_address_description+0x6b/0x290 [ 16.548306] kasan_report+0x28a/0x370 [ 16.551993] ? find_first_bit+0x1d/0x70 [ 16.555858] find_first_bit+0x1d/0x70 [ 16.559625] intel_gvt_init_cmd_parser+0x127/0x3c0 [i915] [ 16.565060] ? __lock_is_held+0x8f/0xf0 [ 16.568990] ? intel_gvt_clean_cmd_parser+0x10/0x10 [i915] [ 16.574514] ? __hrtimer_init+0x5d/0xb0 [ 16.578445] intel_gvt_init_device+0x2c3/0x690 [i915] [ 16.583537] ? unregister_module_notifier+0x20/0x20 [ 16.588515] intel_gvt_init+0x89/0x100 [i915] [ 16.592962] i915_driver_load+0x1992/0x1c70 [i915] [ 16.597846] ? __i915_printk+0x210/0x210 [i915] [ 16.602410] ? wait_for_completion+0x280/0x280 [ 16.606883] ? lock_downgrade+0x2c0/0x2c0 [ 16.610923] ? __pm_runtime_resume+0x46/0x90 [ 16.615238] ? acpi_dev_found+0x76/0x80 [ 16.619162] ? i915_pci_remove+0x30/0x30 [i915] [ 16.623733] local_pci_probe+0x74/0xe0 [ 16.627518] pci_device_probe+0x208/0x310 [ 16.631561] ? pci_device_remove+0x100/0x100 [ 16.635871] ? __list_add_valid+0x29/0xa0 [ 16.639919] driver_probe_device+0x40b/0x6b0 [ 16.644223] ? driver_probe_device+0x6b0/0x6b0 [ 16.648696] __driver_attach+0x11d/0x130 [ 16.652649] bus_for_each_dev+0xe7/0x160 [ 16.656600] ? subsys_dev_iter_exit+0x10/0x10 [ 16.660987] ? __list_add_valid+0x29/0xa0 [ 16.665028] bus_add_driver+0x31d/0x3a0 [ 16.668893] driver_register+0xc6/0x170 [ 16.672758] ? 0xffffffffc0ad8000 [ 16.676108] do_one_initcall+0x9c/0x206 [ 16.679984] ? initcall_blacklisted+0x150/0x150 [ 16.684545] ? do_init_module+0x35/0x33b [ 16.688494] ? kasan_unpoison_shadow+0x31/0x40 [ 16.692968] ? kasan_kmalloc+0xa6/0xd0 [ 16.696743] ? do_init_module+0x35/0x33b [ 16.700694] ? kasan_unpoison_shadow+0x31/0x40 [ 16.705168] ? __asan_register_globals+0x82/0xa0 [ 16.709819] do_init_module+0xe7/0x33b [ 16.713597] load_module+0x4481/0x4ce0 [ 16.717397] ? module_frob_arch_sections+0x20/0x20 [ 16.722228] ? vfs_read+0x13b/0x190 [ 16.725742] ? kernel_read+0x74/0xa0 [ 16.729351] ? get_user_arg_ptr.isra.17+0x70/0x70 [ 16.734099] ? SYSC_finit_module+0x175/0x1b0 [ 16.738399] SYSC_finit_module+0x175/0x1b0 [ 16.742524] ? SYSC_init_module+0x1e0/0x1e0 [ 16.746741] ? __fget+0x157/0x240 [ 16.750090] ? trace_hardirqs_on_thunk+0x1a/0x1c [ 16.754747] entry_SYSCALL_64_fastpath+0x23/0x9a [ 16.759397] RIP: 0033:0x7f8fbc837499 [ 16.762996] RSP: 002b:00007ffead76c138 EFLAGS: 00000246 ORIG_RAX: 0000000000000139 [ 16.770618] RAX: ffffffffffffffda RBX: 0000000000000012 RCX: 00007f8fbc837499 [ 16.777800] RDX: 0000000000000000 RSI: 000056484e67b080 RDI: 0000000000000012 [ 16.784979] RBP: 00007ffead76b140 R08: 0000000000000000 R09: 0000000000000021 [ 16.792164] R10: 0000000000000012 R11: 0000000000000246 R12: 000056484e67b460 [ 16.799345] R13: 00007ffead76b120 R14: 0000000000000005 R15: 0000000000000000 [ 16.808052] The buggy address belongs to the page: [ 16.812876] page:00000000dc4b8c1e count:0 mapcount:0 mapping: (null) index:0x0 [ 16.820934] flags: 0x17ffffc0000000() [ 16.824621] raw: 0017ffffc0000000 0000000000000000 0000000000000000 00000000ffffffff [ 16.832416] raw: ffffea000d85b3e0 ffffea000d85b3e0 0000000000000000 0000000000000000 [ 16.840208] page dumped because: kasan: bad access detected [ 16.847318] Memory state around the buggy address: [ 16.852143] ffff8803616cf400: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 [ 16.859427] ffff8803616cf480: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 f1 f1 [ 16.866708] >ffff8803616cf500: f1 f1 04 f4 f4 f4 f3 f3 f3 f3 00 00 00 00 00 00 [ 16.873988] ^ [ 16.877770] ffff8803616cf580: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 [ 16.885042] ffff8803616cf600: 00 00 00 00 00 00 00 00 00 00 00 00 f1 f1 f1 f1 [ 16.892312] ================================================================== Signed-off-by: Changbin Du <changbin.du@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
| * | drm/i915/gvt: ensure -ve return value is handled correctlyColin Ian King2017-11-211-1/+1
| |/ | | | | | | | | | | | | | | | | | | | | | | | | | | An earlier fix changed the return type from find_bb_size however the integer return is being assigned to a unsigned int so the -ve error check will never be detected. Make bb_size an int to fix this. Detected by CoverityScan CID#1456886 ("Unsigned compared against 0") Fixes: 1e3197d6ad73 ("drm/i915/gvt: Refine error handling for perform_bb_shadow") Signed-off-by: Colin Ian King <colin.king@canonical.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com> (cherry picked from commit 24f8a29af4afe7c53e08f4afa0c3fa9eb3791b89) Signed-off-by: Jani Nikula <jani.nikula@intel.com>
* | drm/i915/gvt: cleanup usage for typed mmio reg vs. offsetZhenyu Wang2017-12-221-10/+10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | We had previous hack that tried to accept either i915_reg_t or offset value to access vGPU virtual/shadow regs which broke that purpose to be type safe in context. This one trys to explicitly separate the usage of typed mmio reg with real offset. Old vgpu_vreg(offset) helper is used only for offset now with new vgpu_vreg_t(reg) is used for i915_reg_t only. Convert left usage of that to new helper. Also fixed left KASAN warning issues caused by previous hack. v2: rebase, fixup against recent mmio switch change Reviewed-by: Zhi Wang <zhi.a.wang@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
* | drm/i915/gvt: refine mocs save restore policyWeinan Li2017-12-181-0/+19
| | | | | | | | | | | | | | | | | | | | | | Save and restore the mocs regs of one VM in GVT-g burning too much CPU utilization. Add LRI command scan to monitor the change of mocs registers, save the state in vreg, and use delta update policy to restore them. It can obviously reduce the MMIO r/w count, and improve the performance of context switch. Signed-off-by: Weinan Li <weinan.z.li@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
* | drm/i915/gvt: Don't dump partial state in cmd parserChangbin Du2017-11-161-8/+3
| | | | | | | | | | | | | | | | I have seen the cmd parser dump partial odd info. Stop that and only dump the full verbose info when debug enabled. Signed-off-by: Changbin Du <changbin.du@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
* | drm/i915/gvt: Use I915_GTT_PAGE_SIZEZhi Wang2017-11-161-6/+7
| | | | | | | | | | | | | | As there is already an I915_GTT_PAGE_SIZE marco in i915, let GVT-g use it as well. Also this patch re-names some GTT marcos with additional prefix. Signed-off-by: Zhi Wang <zhi.a.wang@intel.com>
* | drm/i915/gvt: Refine shadow batch bufferZhi Wang2017-11-161-39/+37
| | | | | | | | | | | | | | | | | | | | | | | | | | | | 1) Use standard i915 GEM object sequence to access the shadow batch buffer. 2) Manage i915 vma life cycle to solve one FIXME. v2: - Refine code structure. - Refine the usage of GEM APIs. - Add the missing lock/unlock in release_shadow_batch_buffer. Test on my SKL NuC. Signed-off-by: Zhi Wang <zhi.a.wang@intel.com>
* | drm/i915/gvt: Refine find_bb_size()Zhi Wang2017-11-161-16/+14
| | | | | | | | | | | | | | Returns the error code if something is wrong and the size of batch buffer is passed through the pointer. Signed-off-by: Zhi Wang <zhi.a.wang@intel.com>
* | drm/i915/gvt: ensure -ve return value is handled correctlyColin Ian King2017-11-161-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | An earlier fix changed the return type from find_bb_size however the integer return is being assigned to a unsigned int so the -ve error check will never be detected. Make bb_size an int to fix this. Detected by CoverityScan CID#1456886 ("Unsigned compared against 0") Fixes: 1e3197d6ad73 ("drm/i915/gvt: Refine error handling for perform_bb_shadow") Signed-off-by: Colin Ian King <colin.king@canonical.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
* | drm/i915/gvt: Change the return type during command scanfred gao2017-11-161-25/+53
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Generally, there are 3 types of errors during command scan: a) some commands might be unknown with EBADRQC; b) some cmd access invalid address with EFAULT; c) some unexpected force nonpriv cmd with EPERM. later the healthy state can be judged through the return error. v2: - remove some internal i915 errors rating. (Zhenyu) v3: - the healthy state is judged through the internal defined return error. (Zhenyu) - force non priv cmd error can be ignored. (Kevin) v4: - reuse standard defined errno instead of recreate, e.g EBADRQC for unknown cmd, EFAULT for invalid address, EPERM for nonpriv. (Zhenyu) v5: - remove some irrelevant code for the patch. - fix typo of vgpu_is_vm_unhealthy. (Zhenyu) v6: - move the healthy check and failsafe code into another patch. (Zhenyu) v7: - polish title and commit message. (Zhenyu) Signed-off-by: fred gao <fred.gao@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
* | drm/i915/gvt: Move ring scan buffers into intel_vgpu_submissionZhi Wang2017-11-161-5/+6
| | | | | | | | | | | | | | Move ring scan buffers into intel_vgpu_submission since they belongs to a part of vGPU submission stuffs. Signed-off-by: Zhi Wang <zhi.a.wang@intel.com>
* | drm/i915/gvt: Rename reserved ring bufferZhi Wang2017-11-161-8/+8
| | | | | | | | | | | | | | | | | | | | | | | | "reserved" means reserve something from somewhere. Actually they are buffers used by command scanner. Rename it to ring_scan_buffer. v2: - Remove the usage of an extra variable. (Zhenyu) Fixes: 0a53bc07f044 ("drm/i915/gvt: Separate cmd scan from request allocation") Signed-off-by: Zhi Wang <zhi.a.wang@intel.com>
* | drm/i915/gvt: Fix a memory leak in cmd_parser.cZhi Wang2017-11-161-4/+6
|/ | | | | | | The pointer points to the original memory can never take the return value of krealloc(). Signed-off-by: Zhi Wang <zhi.a.wang@intel.com>
* Backmerge tag 'v4.14-rc7' into drm-nextDave Airlie2017-11-021-0/+3
|\ | | | | | | | | | | | | | | Linux 4.14-rc7 Requested by Ben Skeggs for nouveau to avoid major conflicts, and things were getting a bit conflicty already, esp around amdgpu reverts.
| * drm/i915/gvt: properly check per_ctx bb valid stateZhenyu Wang2017-10-271-0/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | Need to check valid state for per_ctx bb and bypass batch buffer combine for scan if necessary. Otherwise adding invalid MI batch buffer start cmd for per_ctx bb will cause scan failure, which is taken as -EFAULT now so vGPU would be put in failsafe. This trys to fix that by checking per_ctx bb valid state. Also remove old invalid WARNING that indirect ctx bb shouldn't depend on valid per_ctx bb. Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com> Signed-off-by: Zhi Wang <zhi.a.wang@intel.com>
* | Merge drm-upstream/drm-next into drm-intel-next-queuedJani Nikula2017-09-281-1/+1
|\| | | | | | | | | | | Need MST sideband message transaction to power up/down nodes. Signed-off-by: Jani Nikula <jani.nikula@intel.com>
| * Merge tag 'drm-for-v4.14' of git://people.freedesktop.org/~airlied/linuxLinus Torvalds2017-09-031-4/+4
| |\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Pull drm updates from Dave Airlie: "This is the main drm pull request for 4.14 merge window. I'm sending this early, as my continuing journey into fatherhood is occurring really soon now, I'm going to be mostly useless for the next couple of weeks, though I may be able to read email, I doubt I'll be doing much patch applications or git sending. If anything urgent pops up I've asked Daniel/Jani/Alex/Sean to try and direct stuff towards you. Outside drm changes: Some rcar-du updates that touch the V4L tree, all acks should be in place. It adds one export to the radix tree code for new i915 use case. There are some minor AGP cleanups (don't see that too often). Changes to the vbox driver in staging to avoid breaking compilation. Summary: core: - Atomic helper fixes - Atomic UAPI fixes - Add YCBCR 4:2:0 support - Drop set_busid hook - Refactor fb_helper locking - Remove a bunch of internal APIs - Add a bunch of better default handlers - Format modifier/blob plane property added - More internal header refactoring - Make more internal API names consistent - Enhanced syncobj APIs (wait/signal/reset/create signalled) bridge: - Add Synopsys Designware MIPI DSI host bridge driver tiny: - Add Pervasive Displays RePaper displays - Add support for LEGO MINDSTORMS EV3 LCD i915: - Lots of GEN10/CNL support patches - drm syncobj support - Skylake+ watermark refactoring - GVT vGPU 48-bit ppgtt support - GVT performance improvements - NOA change ioctl - CCS (color compression) scanout support - GPU reset improvements amdgpu: - Initial hugepage support - BO migration logic rework - Vega10 improvements - Powerplay fixes - Stop reprogramming the MC - Fixes for ACP audio on stoney - SR-IOV fixes/improvements - Command submission overhead improvements amdkfd: - Non-dGPU upstreaming patches - Scratch VA ioctl - Image tiling modes - Update PM4 headers for new firmware - Drop all BUG_ONs. nouveau: - GP108 modesetting support. - Disable MSI on big endian. vmwgfx: - Add fence fd support. msm: - Runtime PM improvements exynos: - NV12MT support - Refactor KMS drivers imx-drm: - Lock scanout channel to improve memory bw - Cleanups etnaviv: - GEM object population fixes tegra: - Prep work for Tegra186 support - PRIME mmap support sunxi: - HDMI support improvements - HDMI CEC support omapdrm: - HDMI hotplug IRQ support - Big driver cleanup - OMAP5 DSI support rcar-du: - vblank fixes - VSP1 updates arcgpu: - Minor fixes stm: - Add STM32 DSI controller driver dw_hdmi: - Add support for Rockchip RK3399 - HDMI CEC support atmel-hlcdc: - Add 8-bit color support vc4: - Atomic fixes - New ioctl to attach a label to a buffer object - HDMI CEC support - Allow userspace to dictate rendering order on submit ioctl" * tag 'drm-for-v4.14' of git://people.freedesktop.org/~airlied/linux: (1074 commits) drm/syncobj: Add a signal ioctl (v3) drm/syncobj: Add a reset ioctl (v3) drm/syncobj: Add a syncobj_array_find helper drm/syncobj: Allow wait for submit and signal behavior (v5) drm/syncobj: Add a CREATE_SIGNALED flag drm/syncobj: Add a callback mechanism for replace_fence (v3) drm/syncobj: add sync obj wait interface. (v8) i915: Use drm_syncobj_fence_get drm/syncobj: Add a race-free drm_syncobj_fence_get helper (v2) drm/syncobj: Rename fence_get to find_fence drm: kirin: Add mode_valid logic to avoid mode clocks we can't generate drm/vmwgfx: Bump the version for fence FD support drm/vmwgfx: Add export fence to file descriptor support drm/vmwgfx: Add support for imported Fence File Descriptor drm/vmwgfx: Prepare to support fence fd drm/vmwgfx: Fix incorrect command header offset at restart drm/vmwgfx: Support the NOP_ERROR command drm/vmwgfx: Restart command buffers after errors drm/vmwgfx: Move irq bottom half processing to threads drm/vmwgfx: Don't use drm_irq_[un]install ...
| * | drm/i915/gvt: Fix the kernel null pointer errorfred gao2017-08-231-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | once error happens in shadow_indirect_ctx function, the variable wa_ctx->indirect_ctx.obj is not initialized but accessed, so the kernel null point panic occurs. Fixes: 894cf7d15634 ("drm/i915/gvt: i915_gem_object_create() returns an error pointer") Cc: stable@vger.kernel.org # v4.8+ Signed-off-by: fred gao <fred.gao@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
* | | drm/i915/gvt: Refine error handling for perform_bb_shadowfred gao2017-09-081-2/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | fix the wrong return type and return error once the unknown command is scanned. v2: - separate this error handle from healthy rating code. (Zhenyu) Signed-off-by: fred gao <fred.gao@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
* | | drm/i915/gvt: Separate cmd scan from request allocationfred gao2017-09-081-11/+20
| |/ |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Currently i915 request structure and shadow ring buffer are allocated before command scan, so it will have to restore to previous states once any error happens afterwards in the long dispatch_workload path. This patch is to introduce a reserved ring buffer created at the beginning of vGPU initialization. Workload will be coped to this reserved buffer and be scanned first, the i915 request and shadow ring buffer are only allocated after the result of scan is successful. To balance the memory usage and buffer alloc time, the coming bigger ring buffer will be reallocated and kept until more bigger buffer is coming. v2: - use kmalloc for the smaller ring buffer, realloc if required. (Zhenyu) v3: - remove the dynamically allocated ring buffer. (Zhenyu) v4: - code style polish. - kfree previous allocated buffer once kmalloc failed. (Zhenyu) Signed-off-by: fred gao <fred.gao@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
* | drm/i915/gvt: Replace duplicated code with exist functionPing Gao2017-08-101-3/+3
| | | | | | | | | | | | | | | | Use the exist function intel_gvt_ggtt_validate_range to replace these duplicated code that do the same thing. Signed-off-by: Ping Gao <ping.a.gao@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
* | drm/i915/gvt: Factor out scan and shadow from workload dispatchPing Gao2017-08-101-1/+1
|/ | | | | | | | | | | | | | | | | To perform the workload scan and shadow in ELSP writing stage for performance consideration, the workload scan and shadow stuffs should be factored out from dispatch_workload(). v2:Put context pin before i915_add_request; Refine the comments; Rename some APIs; v3:workload->status should set only when error happens. v4:i915_add_request is must to have after i915_gem_request_alloc. Signed-off-by: Ping Gao <ping.a.gao@intel.com> Reviewed-by: Zhi Wang <zhi.a.wang@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
* drm/i915/gvt: Audit the command buffer addressPing Gao2017-07-111-0/+10
| | | | | | | | | The command buffer address in context like ring buffer base address and wa_ctx address need to be audit to make sure they are in the valid GGTT range. Signed-off-by: Ping Gao <ping.a.gao@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
* drm/i915/gvt: Delete gvt_dbg_cmd() in cmd_parser_exec()Xiong Zhang2017-06-081-6/+0
| | | | | | | | | Since cmd message have been recorded in trace, gvt_dbg_cmd isn't necessary. This will reduce much of dmesg as gvt_dbg_cmd is repeated on each workload. Signed-off-by: Xiong Zhang <xiong.y.zhang@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
* drm/i915/gvt: rewrite the trace gvt:gvt_command using trace style approachChangbin Du2017-06-081-47/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The gvt:gvt_command trace involve unnecessary overhead even this trace is not enabled. We need improve it. The kernel trace infrastructure provide a full api to define a trace event. We should leverage them if possible. And one important thing is that a trace point should store raw data but not format string. This patch include two part work: 1) Refactor the gvt_command trace definition, including: o only store raw trace data. o use __dynamic_array() to declare a variable size buffer. o use __print_array() to format raw cmd data. o rename vm_id as vgpu_id. 2) Improve the trace invoking, including: o remove the cycles calculation for handler. We can get this data by any perf tool. o do not make a backup for raw cmd data which just doesn't make sense. With this patch, this trace has no overhead if it is not enabled. And we are trace style now. The final output example: gvt workload 0-211 [000] ...1 120.555964: gvt_command: vgpu1 ring 0: buf_type 0, ip_gma e161e880, raw cmd {0x4000000} gvt workload 0-211 [000] ...1 120.556014: gvt_command: vgpu1 ring 0: buf_type 0, ip_gma e161e884, raw cmd {0x7a000004,0x1004000,0xe1511018,0x0,0x7d,0x0} gvt workload 0-211 [000] ...1 120.556062: gvt_command: vgpu1 ring 0: buf_type 0, ip_gma e161e89c, raw cmd {0x7a000004,0x140000,0x0,0x0,0x0,0x0} gvt workload 0-211 [000] ...1 120.556110: gvt_command: vgpu1 ring 0: buf_type 0, ip_gma e161e8b4, raw cmd {0x10400002,0xe1511018,0x0,0x7d} Signed-off-by: Changbin Du <changbin.du@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
* drm/i915/gvt: use directly assignment for structure copyingChangbin Du2017-04-121-1/+1
| | | | | | | | Let c compiler handle the structure copying. The compiler will use builtin function to handle that. Signed-off-by: Changbin Du <changbin.du@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
* drm/i915/gvt: remove redundant ring id check which cause significant CPU ↵Changbin Du2017-04-121-6/+0
| | | | | | | | | | | | | | misprediction From perf data, found a significant overhead at ring id check in the function get_opcode. This inline function is frequently used. Since Intel static predictor will predict the branch to fall through so the prediction most fail. This is wasting CPU pipeline resource. We do not need check the engine id everywhere, it should be reliable. Signed-off-by: Changbin Du <changbin.du@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
* drm/i915/gvt: remove workload from intel_shadow_wa_ctx structureTina Zhang2017-03-291-8/+17
| | | | | | | | | | | intel_shadow_wa_ctx is a field of intel_vgpu_workload. container_of() can be used to refine the relation-ship between intel_shadow_wa_ctx and intel_vgpu_workload. This patch removes the useless dereference. v2. add "drm/i915/gvt" prefix. (Zhenyu) Signed-off-by: Tina Zhang <tina.zhang@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
* drm/i915/gvt: Add KBL dispatch logic in each function.Xu Han2017-03-291-6/+11
| | | | | | | Extend function dispatch logic to support KBL platform. Signed-off-by: Xu Han <xu.han@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
* drm/i915/gvt: fix error return check for copy_gma_to_hva()Zhenyu Wang2017-03-291-4/+4
| | | | | | | | | | | | | | | | | From commit 73dec95e6ba3 ("drm/i915: Emit to ringbuffer directly"), copy_gma_to_hva() now returns copied data length instead of 0, so need to change error return check for that. Note: Looks this is caused by backmerge conflict resolving, so 4.11-rc4 is not impacted as commit 73dec95e6ba3 ("drm/i915: Emit to ringbuffer directly") is not in 4.11. But need to fix this before I can apply 4.12 stuff against drm-intel-next correctly. Fixes: e5c1ff14757a ("Backmerge tag 'v4.11-rc4' into drm-next") Cc: Dave Airlie <airlied@redhat.com> Cc: Tina Zhang <tina.zhang@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
generated by cgit v1.2.3 (git 2.25.1) at 2024-11-07 00:41:12 +0000