linux/drivers/gpu/drm/amd/display/amdgpu_dm, branch v5.6

drm/amd/display: Fix pageflip event race condition for DCN.

2020-03-19T20:18:45Z

Commit '16f17eda8bad ("drm/amd/display: Send vblank and user events at vsartup for DCN")' introduces a new way of pageflip completion handling for DCN, and some trouble. The current implementation introduces a race condition, which can cause pageflip completion events to be sent out one vblank too early, thereby confusing userspace and causing flicker: prepare_flip_isr(): 1. Pageflip programming takes the ddev->event_lock. 2. Sets acrtc->pflip_status == AMDGPU_FLIP_SUBMITTED 3. Releases ddev->event_lock. --> Deadline for surface address regs double-buffering passes on target pipe. 4. dc_commit_updates_for_stream() MMIO programs the new pageflip into hw, but too late for current vblank. => pflip_status == AMDGPU_FLIP_SUBMITTED, but flip won't complete in current vblank due to missing the double-buffering deadline by a tiny bit. 5. VSTARTUP trigger point in vblank is reached, VSTARTUP irq fires, dm_dcn_crtc_high_irq() gets called. 6. Detects pflip_status == AMDGPU_FLIP_SUBMITTED and assumes the pageflip has been completed/will complete in this vblank and sends out pageflip completion event to userspace and resets pflip_status = AMDGPU_FLIP_NONE. => Flip completion event sent out one vblank too early. This behaviour has been observed during my testing with measurement hardware a couple of time. The commit message says that the extra flip event code was added to dm_dcn_crtc_high_irq() to prevent missing to send out pageflip events in case the pflip irq doesn't fire, because the "DCH HUBP" component is clock gated and doesn't fire pflip irqs in that state. Also that this clock gating may happen if no planes are active. This suggests that the problem addressed by that commit can't happen if planes are active. The proposed solution is therefore to only execute the extra pflip completion code iff the count of active planes is zero and otherwise leave pflip completion handling to the pflip irq handler, for a more race-free experience. Note that i don't know if this fixes the problem the original commit tried to address, as i don't know what the test scenario was. It does fix the observed too early pageflip events though and points out the problem introduced. Fixes: 16f17eda8bad ("drm/amd/display: Send vblank and user events at vsartup for DCN") Reviewed-by: Nicholas Kazlauskas Signed-off-by: Mario Kleiner Signed-off-by: Alex Deucher

drm/amdgpu/display: navi1x copy dcn watermark clock settings to smu resume from s3 (v2)

2020-03-05T14:42:08Z

This interface is for dGPU Navi1x. Linux dc-pplib interface depends on window driver dc implementation. For Navi1x, clock settings of dcn watermarks are fixed. the settings should be passed to smu during boot up and resume from s3. boot up: dc calculate dcn watermark clock settings within dc_create, dcn20_resource_construct, then call pplib functions below to pass the settings to smu: smu_set_watermarks_for_clock_ranges smu_set_watermarks_table navi10_set_watermarks_table smu_write_watermarks_table For Renoir, clock settings of dcn watermark are also fixed values. dc has implemented different flow for window driver: dc_hardware_init / dc_set_power_state dcn10_init_hw notify_wm_ranges set_wm_ranges For Linux smu_set_watermarks_for_clock_ranges renoir_set_watermarks_table smu_write_watermarks_table dc_hardware_init -> amdgpu_dm_init dc_set_power_state --> dm_resume therefore, linux dc-pplib interface of navi10/12/14 is different from that of Renoir. v2: add missing unlock in error case Signed-off-by: Hersen Wu Reviewed-by: Evan Quan Signed-off-by: Alex Deucher

drm/amd/display: Clear link settings on MST disable connector

2020-03-05T14:42:08Z

[Why] If we have a single MST display and we disconnect it, we dont disable that link. This causes the old link settings to still exist Now on a replug for MST we think its a link loss and will try to reallocate mst payload which will fail, throwing warning below. [ 129.374192] [drm] Failed to updateMST allocation table forpipe idx:0 [ 129.374206] ------------[ cut here ]------------ [ 129.374284] WARNING: CPU: 14 PID: 1710 at drivers/gpu/drm/amd/amdgpu/../dal-dev/dc/core/dc_link.c:3153 dc_link_allocate_mst_payload+0x1f7/0x220 [amdgpu] [ 129.374285] Modules linked in: amdgpu(OE) amd_iommu_v2 gpu_sched ttm drm_kms_helper drm fb_sys_fops syscopyarea sysfillrect sysimgblt binfmt_misc nls_iso8859_1 edac_mce_amd snd_hda_codec_realtek snd_hda_codec_generic ledtrig_audio kvm snd_hda_codec_hdmi snd_hda_intel snd_intel_nhlt snd_hda_codec irqbypass snd_hda_core snd_hwdep snd_pcm snd_seq_midi snd_seq_midi_event snd_rawmidi crct10dif_pclmul snd_seq crc32_pclmul ghash_clmulni_intel snd_seq_device snd_timer snd aesni_intel eeepc_wmi crypto_simd asus_wmi joydev cryptd sparse_keymap input_leds soundcore video glue_helper wmi_bmof mxm_wmi k10temp ccp mac_hid sch_fq_codel parport_pc ppdev lp parport ip_tables x_tables autofs4 hid_generic usbhid hid igb i2c_algo_bit ahci dca i2c_piix4 libahci gpio_amdpt wmi gpio_generic [ 129.374318] CPU: 14 PID: 1710 Comm: kworker/14:2 Tainted: G W OE 5.4.0-rc7bhawan+ #480 [ 129.374318] Hardware name: System manufacturer System Product Name/PRIME X370-PRO, BIOS 0515 03/30/2017 [ 129.374397] Workqueue: events dm_irq_work_func [amdgpu] [ 129.374468] RIP: 0010:dc_link_allocate_mst_payload+0x1f7/0x220 [amdgpu] [ 129.374470] Code: 52 20 e8 1c 63 ad f4 48 8b 5d d0 65 48 33 1c 25 28 00 00 00 b8 01 00 00 00 75 16 48 8d 65 d8 5b 41 5c 41 5d 41 5e 41 5f 5d c3 <0f> 0b e9 fa fe ff ff e8 ed 5b d6 f3 41 0f b6 b6 c4 02 00 00 48 c7 [ 129.374471] RSP: 0018:ffff9f9141e7fcc0 EFLAGS: 00010246 [ 129.374472] RAX: 0000000000000000 RBX: ffff91ef0762f800 RCX: 0000000000000000 [ 129.374473] RDX: 0000000000000005 RSI: ffffffffc0c4a988 RDI: 0000000000000004 [ 129.374474] RBP: ffff9f9141e7fd10 R08: 0000000000000005 R09: 0000000000000000 [ 129.374475] R10: 0000000000000002 R11: 0000000000000001 R12: ffff91eebd510c00 [ 129.374475] R13: ffff91eebd510e58 R14: ffff91ef052c01b8 R15: 0000000000000006 [ 129.374476] FS: 0000000000000000(0000) GS:ffff91ef0ef80000(0000) knlGS:0000000000000000 [ 129.374477] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 129.374478] CR2: 000055623ea01d50 CR3: 0000000408a8c000 CR4: 00000000003406e0 [ 129.374479] Call Trace: [ 129.374550] dc_link_reallocate_mst_payload+0x12e/0x150 [amdgpu] [ 129.374617] dc_link_handle_hpd_rx_irq+0x6d4/0x6e0 [amdgpu] [ 129.374693] handle_hpd_rx_irq+0x77/0x310 [amdgpu] [ 129.374768] dm_irq_work_func+0x53/0x70 [amdgpu] [ 129.374774] process_one_work+0x1fd/0x3f0 [ 129.374776] worker_thread+0x255/0x410 [ 129.374778] kthread+0x121/0x140 [ 129.374780] ? process_one_work+0x3f0/0x3f0 [ 129.374781] ? kthread_park+0x90/0x90 [ 129.374785] ret_from_fork+0x22/0x40 [How] when we disable MST we should clear the cur link settings (lane_count=0 is good enough). This will cause us to not reallocate payloads earlier than expected and not throw the warning Signed-off-by: Bhawanpreet Lakha Reviewed-by: Hersen Wu Acked-by: Rodrigo Siqueira Signed-off-by: Alex Deucher

drm/amdgpu/display: clean up hdcp workqueue handling

2020-02-19T16:03:24Z

Use the existence of the workqueue itself to determine when to enable HDCP features rather than sprinkling asic checks all over the code. Also add a check for the existence of the hdcp workqueue in the irq handling on the off chance we get and HPD RX interrupt with the CP bit set. This avoids a crash if the driver doesn't support HDCP for a particular asic. Fixes: 96a3b32e67236f ("drm/amd/display: only enable HDCP for DCN+") Bug: https://bugzilla.kernel.org/show_bug.cgi?id=206519 Reviewed-by: Bhawanpreet Lakha Signed-off-by: Alex Deucher

drm/amd/display: Fix psr static frames calculation

2020-02-11T17:10:41Z

[Why] Driver crash with psr feature enabled due to divide-by-zero error. This is a regression after rework to calculate static screen frame number entry time. [How] Correct order of operations to avoid divide-by-zero. Signed-off-by: Roman Li Reviewed-by: Zhan Liu Acked-by: Bhawanpreet Lakha Signed-off-by: Alex Deucher

drm/amd/dm/mst: Ignore payload update failures

2020-02-05T04:30:39Z

Disabling a display on MST can potentially happen after the entire MST topology has been removed, which means that we can't communicate with the topology at all in this scenario. Likewise, this also means that we can't properly update payloads on the topology and as such, it's a good idea to ignore payload update failures when disabling displays. Currently, amdgpu makes the mistake of halting the payload update process when any payload update failures occur, resulting in leaving DC's local copies of the payload tables out of date. This ends up causing problems with hotplugging MST topologies, and causes modesets on the second hotplug to fail like so: [drm] Failed to updateMST allocation table forpipe idx:1 ------------[ cut here ]------------ WARNING: CPU: 5 PID: 1511 at drivers/gpu/drm/amd/amdgpu/../display/dc/core/dc_link.c:2677 update_mst_stream_alloc_table+0x11e/0x130 [amdgpu] Modules linked in: cdc_ether usbnet fuse xt_conntrack nf_conntrack nf_defrag_ipv6 libcrc32c nf_defrag_ipv4 ipt_REJECT nf_reject_ipv4 nft_counter nft_compat nf_tables nfnetlink tun bridge stp llc sunrpc vfat fat wmi_bmof uvcvideo snd_hda_codec_realtek snd_hda_codec_generic snd_hda_codec_hdmi videobuf2_vmalloc snd_hda_intel videobuf2_memops videobuf2_v4l2 snd_intel_dspcfg videobuf2_common crct10dif_pclmul snd_hda_codec videodev crc32_pclmul snd_hwdep snd_hda_core ghash_clmulni_intel snd_seq mc joydev pcspkr snd_seq_device snd_pcm sp5100_tco k10temp i2c_piix4 snd_timer thinkpad_acpi ledtrig_audio snd wmi soundcore video i2c_scmi acpi_cpufreq ip_tables amdgpu(O) rtsx_pci_sdmmc amd_iommu_v2 gpu_sched mmc_core i2c_algo_bit ttm drm_kms_helper syscopyarea sysfillrect sysimgblt fb_sys_fops cec drm crc32c_intel serio_raw hid_multitouch r8152 mii nvme r8169 nvme_core rtsx_pci pinctrl_amd CPU: 5 PID: 1511 Comm: gnome-shell Tainted: G O 5.5.0-rc7Lyude-Test+ #4 Hardware name: LENOVO FA495SIT26/FA495SIT26, BIOS R12ET22W(0.22 ) 01/31/2019 RIP: 0010:update_mst_stream_alloc_table+0x11e/0x130 [amdgpu] Code: 28 00 00 00 75 2b 48 8d 65 e0 5b 41 5c 41 5d 41 5e 5d c3 0f b6 06 49 89 1c 24 41 88 44 24 08 0f b6 46 01 41 88 44 24 09 eb 93 <0f> 0b e9 2f ff ff ff e8 a6 82 a3 c2 66 0f 1f 44 00 00 0f 1f 44 00 RSP: 0018:ffffac428127f5b0 EFLAGS: 00010202 RAX: 0000000000000002 RBX: ffff8d1e166eee80 RCX: 0000000000000000 RDX: ffffac428127f668 RSI: ffff8d1e166eee80 RDI: ffffac428127f610 RBP: ffffac428127f640 R08: ffffffffc03d94a8 R09: 0000000000000000 R10: ffff8d1e24b02000 R11: ffffac428127f5b0 R12: ffff8d1e1b83d000 R13: ffff8d1e1bea0b08 R14: 0000000000000002 R15: 0000000000000002 FS: 00007fab23ffcd80(0000) GS:ffff8d1e28b40000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00007f151f1711e8 CR3: 00000005997c0000 CR4: 00000000003406e0 Call Trace: ? mutex_lock+0xe/0x30 dc_link_allocate_mst_payload+0x9a/0x210 [amdgpu] ? dm_read_reg_func+0x39/0xb0 [amdgpu] ? core_link_enable_stream+0x656/0x730 [amdgpu] core_link_enable_stream+0x656/0x730 [amdgpu] dce110_apply_ctx_to_hw+0x58e/0x5d0 [amdgpu] ? dcn10_verify_allow_pstate_change_high+0x1d/0x280 [amdgpu] ? dcn10_wait_for_mpcc_disconnect+0x3c/0x130 [amdgpu] dc_commit_state+0x292/0x770 [amdgpu] ? add_timer+0x101/0x1f0 ? ttm_bo_put+0x1a1/0x2f0 [ttm] amdgpu_dm_atomic_commit_tail+0xb59/0x1ff0 [amdgpu] ? amdgpu_move_blit.constprop.0+0xb8/0x1f0 [amdgpu] ? amdgpu_bo_move+0x16d/0x2b0 [amdgpu] ? ttm_bo_handle_move_mem+0x118/0x570 [ttm] ? ttm_bo_validate+0x134/0x150 [ttm] ? dm_plane_helper_prepare_fb+0x1b9/0x2a0 [amdgpu] ? _cond_resched+0x15/0x30 ? wait_for_completion_timeout+0x38/0x160 ? _cond_resched+0x15/0x30 ? wait_for_completion_interruptible+0x33/0x190 commit_tail+0x94/0x130 [drm_kms_helper] drm_atomic_helper_commit+0x113/0x140 [drm_kms_helper] drm_atomic_helper_set_config+0x70/0xb0 [drm_kms_helper] drm_mode_setcrtc+0x194/0x6a0 [drm] ? _cond_resched+0x15/0x30 ? mutex_lock+0xe/0x30 ? drm_mode_getcrtc+0x180/0x180 [drm] drm_ioctl_kernel+0xaa/0xf0 [drm] drm_ioctl+0x208/0x390 [drm] ? drm_mode_getcrtc+0x180/0x180 [drm] amdgpu_drm_ioctl+0x49/0x80 [amdgpu] do_vfs_ioctl+0x458/0x6d0 ksys_ioctl+0x5e/0x90 __x64_sys_ioctl+0x16/0x20 do_syscall_64+0x55/0x1b0 entry_SYSCALL_64_after_hwframe+0x44/0xa9 RIP: 0033:0x7fab2121f87b Code: 0f 1e fa 48 8b 05 0d 96 2c 00 64 c7 00 26 00 00 00 48 c7 c0 ff ff ff ff c3 66 0f 1f 44 00 00 f3 0f 1e fa b8 10 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d dd 95 2c 00 f7 d8 64 89 01 48 RSP: 002b:00007ffd045f9068 EFLAGS: 00000246 ORIG_RAX: 0000000000000010 RAX: ffffffffffffffda RBX: 00007ffd045f90a0 RCX: 00007fab2121f87b RDX: 00007ffd045f90a0 RSI: 00000000c06864a2 RDI: 000000000000000b RBP: 00007ffd045f90a0 R08: 0000000000000000 R09: 000055dbd2985d10 R10: 000055dbd2196280 R11: 0000000000000246 R12: 00000000c06864a2 R13: 000000000000000b R14: 0000000000000000 R15: 000055dbd2196280 ---[ end trace 6ea888c24d2059cd ]--- Note as well, I have only been able to reproduce this on setups with 2 MST displays. Changes since v1: * Don't return false when part 1 or part 2 of updating the payloads fails, we don't want to abort at any step of the process even if things fail Reviewed-by: Mikita Lipski Signed-off-by: Lyude Paul Acked-by: Harry Wentland Cc: stable@vger.kernel.org Signed-off-by: Alex Deucher

drm/amd/display: Fix HW/SW state mismatch

2020-02-04T15:37:51Z

[Why] When we disable a connector we don't explicitly remove it from the module so the display is still cached(SW) in the hdcp_module. SST: no issues because we can only have 1 display per link MST: We have x displays per link, now if we disable 1 we don't remove it from the module so the module has x display cached(SW). If we try to enable HDCP, psp verification will fail because we are reporting x displays while the HW only has x-1 display enabled [How] Check the callback for when we disable stream and call remove display. Signed-off-by: Bhawanpreet Lakha Reviewed-by: Rodrigo Siqueira Signed-off-by: Alex Deucher

drm/amd/display: Fix a typo when computing dsc configuration

2020-02-04T15:37:51Z

[why] Remove a backslash symbol accidentally left in increase bpp function when computing mst dsc configuration. Signed-off-by: Mikita Lipski Reviewed-by: Zhan Liu Signed-off-by: Alex Deucher

drm/amd/display: Move drm_dp_mst_atomic_check() to the front of dc_validate_global_state()

2020-01-30T22:15:43Z

[Why] Need to do atomic check first, then validate global state. If not, when connecting both MST and HDMI displays and set a bad mode via xrandr, system will hang. [How] Move drm_dp_mst_atomic_check() to the front of dc_validate_global_state(). Signed-off-by: Zhan Liu Reviewed-by: Mikita Lipski Signed-off-by: Alex Deucher

drm/amd/display: Fix update type for multiple planes

2020-01-22T21:55:28Z

[Why] determine_update_type_for_commit() uses pointers to single instance of local variable to fill scaling/color info for all planes updates. This is a bug, that leads to incorrect update type for commit in case of multiple planes per crtc. Each plane should refer to separate scaling/color data. [How] Use arrays for plane properties. Bundle all properties into a single structure to simplify memory allocation. Signed-off-by: Roman Li Reviewed-by: Nicholas Kazlauskas Acked-by: Bhawanpreet Lakha Signed-off-by: Alex Deucher