aboutsummaryrefslogtreecommitdiff
path: root/drivers/gpu/drm/amd
AgeCommit message (Collapse)AuthorFilesLines
2017-10-06drm/amd/powerplay: fix typo on avfs disableEvan Quan1-1/+1
Signed-off-by: Evan Quan <[email protected]> Acked-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-10-06drm/amd/powerplay: get raven sclk and mclk levels (v2)Evan Quan1-1/+47
v2: squash in rebase fix (Tom) Signed-off-by: Evan Quan <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Reviewed-by: Junwei Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-10-06drm/amd/powerplay: get raven current sclk and mclk (v2)Evan Quan1-2/+26
v2: squash in rebase fix (Tom) Signed-off-by: Evan Quan <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Reviewed-by: Junwei Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-10-06drm/amd/powerplay: get raven max/min gfx clocks (v2)Evan Quan2-0/+22
v2: squash in rebase fix (Tom) Signed-off-by: Evan Quan <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Reviewed-by: Junwei Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-10-06drm/amd/powerplay: added new raven ppsmc messagesEvan Quan1-1/+6
Signed-off-by: Evan Quan <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Reviewed-by: Junwei Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-10-06drm/amd/powerplay: fixed wrong return value on error (v2)Evan Quan1-4/+4
v2: squash in typo fix (Tom) Signed-off-by: Evan Quan <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-10-06drm/amdgpu: Fixed a potential circular lockozeng2-2/+18
The dead circular lock senario captured is as followed. The idea of the fix is moving read_user_wptr outside of acquire_queue...release_queue critical section [ 63.477482] WARNING: possible circular locking dependency detected [ 63.484091] 4.12.0-kfd-ozeng #3 Not tainted [ 63.488531] ------------------------------------------------------ [ 63.495146] HelloWorldLoop/2526 is trying to acquire lock: [ 63.501011] (&mm->mmap_sem){++++++}, at: [<ffffffff911898ce>] __might_fault+0x3e/0x90 [ 63.509472] but task is already holding lock: [ 63.515716] (&adev->srbm_mutex){+.+...}, at: [<ffffffffc0484feb>] lock_srbm+0x2b/0x50 [amdgpu] [ 63.525099] which lock already depends on the new lock. [ 63.533841] the existing dependency chain (in reverse order) is: [ 63.541839] -> #2 (&adev->srbm_mutex){+.+...}: [ 63.548178] lock_acquire+0x6d/0x90 [ 63.552461] __mutex_lock+0x70/0x8c0 [ 63.556826] mutex_lock_nested+0x16/0x20 [ 63.561603] gfx_v8_0_kiq_resume+0x1039/0x14a0 [amdgpu] [ 63.567817] gfx_v8_0_hw_init+0x204d/0x2210 [amdgpu] [ 63.573675] amdgpu_device_init+0xdea/0x1790 [amdgpu] [ 63.579640] amdgpu_driver_load_kms+0x63/0x220 [amdgpu] [ 63.585743] drm_dev_register+0x145/0x1e0 [ 63.590605] amdgpu_pci_probe+0x11e/0x160 [amdgpu] [ 63.596266] local_pci_probe+0x40/0xa0 [ 63.600803] pci_device_probe+0x134/0x150 [ 63.605650] driver_probe_device+0x2a1/0x460 [ 63.610785] __driver_attach+0xdc/0xe0 [ 63.615321] bus_for_each_dev+0x5f/0x90 [ 63.619984] driver_attach+0x19/0x20 [ 63.624337] bus_add_driver+0x40/0x270 [ 63.628908] driver_register+0x5b/0xe0 [ 63.633446] __pci_register_driver+0x5b/0x60 [ 63.638586] rtsx_pci_switch_output_voltage+0x1d/0x20 [rtsx_pci] [ 63.645564] do_one_initcall+0x4c/0x1b0 [ 63.650205] do_init_module+0x56/0x1ea [ 63.654767] load_module+0x208c/0x27d0 [ 63.659335] SYSC_finit_module+0x96/0xd0 [ 63.664058] SyS_finit_module+0x9/0x10 [ 63.668629] entry_SYSCALL_64_fastpath+0x1f/0xbe [ 63.674088] -> #1 (reservation_ww_class_mutex){+.+.+.}: [ 63.681257] lock_acquire+0x6d/0x90 [ 63.685551] __ww_mutex_lock.constprop.11+0x8c/0xed0 [ 63.691426] ww_mutex_lock+0x67/0x70 [ 63.695802] amdgpu_verify_access+0x6d/0x100 [amdgpu] [ 63.701743] ttm_bo_mmap+0x8e/0x100 [ttm] [ 63.706615] amdgpu_bo_mmap+0xd/0x60 [amdgpu] [ 63.711814] amdgpu_mmap+0x35/0x40 [amdgpu] [ 63.716904] mmap_region+0x3b5/0x5a0 [ 63.721255] do_mmap+0x400/0x4d0 [ 63.725260] vm_mmap_pgoff+0xb0/0xf0 [ 63.729625] SyS_mmap_pgoff+0x19e/0x260 [ 63.734292] SyS_mmap+0x1d/0x20 [ 63.738199] entry_SYSCALL_64_fastpath+0x1f/0xbe [ 63.743681] -> #0 (&mm->mmap_sem){++++++}: [ 63.749641] __lock_acquire+0x1401/0x1420 [ 63.754491] lock_acquire+0x6d/0x90 [ 63.758750] __might_fault+0x6b/0x90 [ 63.763176] kgd_hqd_load+0x24f/0x270 [amdgpu] [ 63.768432] load_mqd+0x4b/0x50 [amdkfd] [ 63.773192] create_queue_nocpsch+0x535/0x620 [amdkfd] [ 63.779237] pqm_create_queue+0x34d/0x4f0 [amdkfd] [ 63.784835] kfd_ioctl_create_queue+0x282/0x670 [amdkfd] [ 63.790973] kfd_ioctl+0x310/0x4d0 [amdkfd] [ 63.795944] do_vfs_ioctl+0x90/0x6e0 [ 63.800268] SyS_ioctl+0x74/0x80 [ 63.804207] entry_SYSCALL_64_fastpath+0x1f/0xbe [ 63.809607] other info that might help us debug this: [ 63.818026] Chain exists of: &mm->mmap_sem --> reservation_ww_class_mutex --> &adev->srbm_mutex [ 63.830382] Possible unsafe locking scenario: [ 63.836605] CPU0 CPU1 [ 63.841364] ---- ---- [ 63.846123] lock(&adev->srbm_mutex); [ 63.850061] lock(reservation_ww_class_mutex); [ 63.857475] lock(&adev->srbm_mutex); [ 63.864084] lock(&mm->mmap_sem); [ 63.867657] *** DEADLOCK *** [ 63.873884] 3 locks held by HelloWorldLoop/2526: [ 63.878739] #0: (&process->mutex){+.+.+.}, at: [<ffffffffc06e1a9a>] kfd_ioctl_create_queue+0x24a/0x670 [amdkfd] [ 63.889543] #1: (&dqm->lock){+.+...}, at: [<ffffffffc06eedeb>] create_queue_nocpsch+0x3b/0x620 [amdkfd] [ 63.899684] #2: (&adev->srbm_mutex){+.+...}, at: [<ffffffffc0484feb>] lock_srbm+0x2b/0x50 [amdgpu] [ 63.909500] stack backtrace: [ 63.914187] CPU: 3 PID: 2526 Comm: HelloWorldLoop Not tainted 4.12.0-kfd-ozeng #3 [ 63.922184] Hardware name: AMD Carrizo/Gardenia, BIOS WGA5819N_Weekly_15_08_1 08/19/2015 [ 63.930865] Call Trace: [ 63.933464] dump_stack+0x85/0xc9 [ 63.936999] print_circular_bug+0x1f9/0x207 [ 63.941442] __lock_acquire+0x1401/0x1420 [ 63.945745] ? lock_srbm+0x2b/0x50 [amdgpu] [ 63.950185] lock_acquire+0x6d/0x90 [ 63.953885] ? __might_fault+0x3e/0x90 [ 63.957899] __might_fault+0x6b/0x90 [ 63.961699] ? __might_fault+0x3e/0x90 [ 63.965755] kgd_hqd_load+0x24f/0x270 [amdgpu] [ 63.970577] load_mqd+0x4b/0x50 [amdkfd] [ 63.974745] create_queue_nocpsch+0x535/0x620 [amdkfd] [ 63.980242] pqm_create_queue+0x34d/0x4f0 [amdkfd] [ 63.985320] kfd_ioctl_create_queue+0x282/0x670 [amdkfd] [ 63.991021] kfd_ioctl+0x310/0x4d0 [amdkfd] [ 63.995499] ? kfd_ioctl_destroy_queue+0x70/0x70 [amdkfd] [ 64.001234] do_vfs_ioctl+0x90/0x6e0 [ 64.005065] ? up_read+0x1a/0x40 [ 64.008496] SyS_ioctl+0x74/0x80 [ 64.011955] entry_SYSCALL_64_fastpath+0x1f/0xbe [ 64.016863] RIP: 0033:0x7f4b3bd35f07 [ 64.020696] RSP: 002b:00007ffe7689ec38 EFLAGS: 00000246 ORIG_RAX: 0000000000000010 [ 64.028786] RAX: ffffffffffffffda RBX: 00000000002a2000 RCX: 00007f4b3bd35f07 [ 64.036414] RDX: 00007ffe7689ecb0 RSI: 00000000c0584b02 RDI: 0000000000000005 [ 64.044045] RBP: 00007f4a3212d000 R08: 00007f4b3c919000 R09: 0000000000080000 [ 64.051674] R10: 00007f4b376b64b8 R11: 0000000000000246 R12: 00007f4a3212d000 [ 64.059324] R13: 0000000000000015 R14: 0000000000000064 R15: 00007ffe7689ef50 Signed-off-by: Oak Zeng <[email protected]> Reviewed-by: Felix Kuehling <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-10-06drm/amdgpu: add FENCE_TO_HANDLE ioctl that returns syncobj or sync_fileMarek Olšák4-1/+66
for being able to convert an amdgpu fence into one of the handles. Mesa will use this. Reviewed-by: Dave Airlie <[email protected]> Signed-off-by: Marek Olšák <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-10-06drm/amd/powerplay: delete flag PP_VALIDRex Zhu2-5/+1
don't need to check pp_valid, all pp export functions are moved to ip_funcs and pp_funcs. so just need to check the function point. Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Rex Zhu <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-10-06drm/amd/powerplay: move set_clockgating_by_smu to pp func tableRex Zhu5-40/+54
Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Rex Zhu <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-10-06drm/amd/powerplay: tidy up ret checks in amd_powerplay.c (v3)Rex Zhu1-38/+38
v2: squash in regression fix (Rex) v3: Squash in regression fix (Rex) Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Rex Zhu <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-10-06amdgpu/dc: use kref for dc_state.Dave Airlie3-16/+16
I'm not a huge fan of those copying around refcounts bits, might want to consider alternates, but this should work for now. Signed-off-by: Dave Airlie <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-10-06amdgpu/dc: convert dc_sink to kref.Dave Airlie2-11/+11
Refcounts use krefs. Signed-off-by: Dave Airlie <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-10-06amdgpu/dc: convert dc_stream_state to kref.Dave Airlie2-12/+12
Signed-off-by: Dave Airlie <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-10-06amdgpu/dc: use kref for dc_plane_state.Dave Airlie2-11/+11
Signed-off-by: Dave Airlie <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-10-06amdgpu/dc: convert dc_gamma to kref reference counting.Dave Airlie2-13/+10
Rolling your own reference counting is frowned upon. Signed-off-by: Dave Airlie <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-10-06amdgpu/dc: convert dc_transfer to use a kref.Dave Airlie3-9/+12
Rolling your own atomic ref counts is frowned upon. Signed-off-by: Dave Airlie <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-10-06amdgpu/dc: kill a bunch of dead code.Dave Airlie3-150/+0
None of this code is used currently. Signed-off-by: Dave Airlie <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-10-06amdgpu/dc: set a bunch of functions to static.Dave Airlie8-11/+10
All of these are unused outside the file they are in. Signed-off-by: Dave Airlie <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-10-06amdgpu/dc: kill some deadcode in dc core.Dave Airlie2-96/+4
Moves one function to an inline, but kills a bunch of deadcode. Signed-off-by: Dave Airlie <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-10-06amdgpu/dc: fix indentation on a couple of returns.Dave Airlie1-2/+2
These were misaligned. found by the cocci ifcol script. Signed-off-by: Dave Airlie <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-10-06amdgpu/dm: don't use after free.Dave Airlie1-1/+0
This dereference acrtc after freeing it. Found by the kfree cocci script. Signed-off-by: Dave Airlie <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-10-06amdgpu/dc: kfree already checks for NULL.Dave Airlie5-22/+9
Don't bother checking for it. Found with the cocci ifnullfree.cocci script. Signed-off-by: Dave Airlie <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-10-06amdgpu/dc: fix a bunch of misc whitespace.Dave Airlie6-55/+55
This just aligns a few things with kernel style. Signed-off-by: Dave Airlie <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-10-06amdgpu/dc: drop hw_sequencer_types.hDave Airlie5-37/+0
This isn't used or required. Signed-off-by: Dave Airlie <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-10-06amdgpu/dc: drop dce110_types.hDave Airlie1-30/+0
Doesn't appear to be used. Signed-off-by: Dave Airlie <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-10-06amdgpu/dc: use kernel ilog2 for log_2.Dave Airlie2-11/+4
This should produce the same result. Signed-off-by: Dave Airlie <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-10-06amdgpu/dc: don't memset after kzalloc.Dave Airlie1-2/+0
We allocate this struct zeroed, so don't need to memset in the constructor. Signed-off-by: Dave Airlie <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-10-06amdgpu/dc: inline dal grph object id functions.Dave Airlie2-77/+56
This is worth 400 bytes. Signed-off-by: Dave Airlie <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-10-03BackMerge tag 'v4.14-rc3' into drm-nextDave Airlie5-6/+211
Linux 4.14-rc3 Requested by Daniel for the tracing build fix in fixes.
2017-10-01drm/amd/powerplay: refine code in amd_powerplay.c (v2)Rex Zhu3-48/+25
1. use flag PP_DPM_DISABLED within powerplay notify amdgpu dpm state by cgs interface. 2. delete redundant virtualization check in powerplay v2: squash in fix for hwmgr_init (Rex) Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Rex Zhu <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-10-01drm/amdgpu: use designated initialiser for thermal_irq_src.Dave Airlie1-3/+3
This fixes the 0-day build warning. Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
2017-09-29amdgpu/dc: inline dml_round_to_multipleDave Airlie3-21/+19
turns out to be a win to inline this. Signed-off-by: Dave Airlie <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-09-29amdgpu/dc: rename bios get_image symbol to something more searchable.Dave Airlie4-7/+7
This just makes it easier to find. Signed-off-by: Dave Airlie <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-09-29amdgpu/dc: set some of the link dp code to static.Dave Airlie2-13/+13
These aren't currently used outside this file. Signed-off-by: Dave Airlie <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-09-29amdgpu/dc: inline all the signal_types code.Dave Airlie3-89/+52
This is worth 300 bytes, and one less source file. Signed-off-by: Dave Airlie <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-09-29amdgpu/dc: inline fixed31_32 div_intDave Airlie2-12/+6
Signed-off-by: Dave Airlie <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-09-29amdgpu/dc: use the builtin constant p trick on the 31/32 fixed point.Dave Airlie2-14/+23
This only gets us 100 bytes, but may as well be consistent. Signed-off-by: Dave Airlie <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-09-29amdgpu/dc: allow inlining constant int to fixed a lot better.Dave Airlie2-28/+35
This uses two things that might be Linux specific, __builtin_constant_p (might be gcc) and BUILD_BUG_ON. (maybe other dm's can have something similiar). This saves 4k in the bw calcs code. Signed-off-by: Dave Airlie <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-09-29amdgpu/dc: inline some of the bw_fixed code.Dave Airlie2-108/+96
This results in a ~4.5k code size reduction. Signed-off-by: Dave Airlie <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-09-29amdgpu/dc: move some one line dp functions to inlines.Dave Airlie1-2/+2
Signed-off-by: Dave Airlie <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-09-29amdgpu/dc: hide some unused aux/i2c payload apis.Dave Airlie2-18/+8
I've no idea if these are used on other platforms, but they aren't used outside this file here, so make them static. Drops 300 bytes. Signed-off-by: Dave Airlie <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-09-29amdgpu/dc: inline a bunch of the fixed 31_32 helpers.Dave Airlie2-161/+93
This decreases code size by a few hundred bytes. Signed-off-by: Dave Airlie <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-09-29amdgpu/dc: inline some of the fixed 32_32 fnsDave Airlie2-75/+61
This drops ~400 bytes here. Signed-off-by: Dave Airlie <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-09-29amdgpu/dc: fix construct return values on irq service.Dave Airlie6-45/+22
This just removes more unused return/errors paths. Signed-off-by: Dave Airlie <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-09-29amdgpu: fixup construct to void paths on some more dc objects.Dave Airlie3-32/+11
Signed-off-by: Dave Airlie <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-09-29amdgpu/dc: remove pointless return from build_pipe_hw_paramDave Airlie6-33/+8
This never returned anything else. Signed-off-by: Dave Airlie <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-09-29amdgpu/dc: another round of dce/dcn construct cleanups.Dave Airlie32-108/+47
This removes any remaining pointless return codepaths from the DCE code. Signed-off-by: Dave Airlie <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-09-29amdgpu/dc: cleanup construct returns in gpio.Dave Airlie4-56/+30
This is similiar to previous patches, don't return when we don't need to, also do error checking before allocating memory, makes it simpler to cleanup after. Signed-off-by: Dave Airlie <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-09-29amdgpu/dc: remove pointless returns in the i2caux constructor paths. (v2)Dave Airlie28-246/+100
There was lots of return true, and error checking that was never used in these paths. Just remove it all. v2: I missed one return true. Signed-off-by: Dave Airlie <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
Powered by cgit, Gruvboxed by Blaster4385.