This patch adds van gogh to amd_asic_type enum and amdgpu_asic_name[].
v2: add missing comma
Signed-off-by: Huang Rui <ray.huang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
support both direct and indirect accessor in unified
helper functions.
v2: Retire indirect mmio access via mm_index/data
Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Kevin Wang <kevin1.wang@amd.com>
Reviewed-by: Guchun Chen <guchun.chen@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Add helper function in order to remove RREG32/WREG32
in current pcie_rreg/wreg function for soc15 and
onwards adapters.
PCIE_INDEX/DATA pairs are used to access regsiters
outside of mmio bar in the helper functions.
The new helper functions help remove the recursion
of amdgpu_mm_rreg/wreg from pcie_rreg/wreg and
provide the oppotunity to centralize direct and
indirect access in a single function.
v2: Fixed typo and refine the comments
v3: Remove unnecessary volatile local variable
Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Kevin Wang <kevin1.wang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Guchun Chen <guchun.chen@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
In FLR routine, init_data_exchange is called at reset_sriov
while fini_data_exchange is not. This will duplicating work
thread.
So call fini_data_exchange before reset for SRIOV
Signed-off-by: Tiecheng Zhou <Tiecheng.Zhou@amd.com>
Signed-off-by: Bokun Zhang <Bokun.Zhang@amd.com>
Reviewed-by: Monk Liu <monk.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
- Refactor the driver code to use amdgpu_virt_read_pf2vf_data
and amdgpu_virt_write_vf2pf_data instead of writing all code in
one function (which is the old amdgpu_virt_init_data_exchange)
- Adding a new transaction method for VF2PF message between host
and guest driver. Guest side will periodically update VF2PF
message in the framebuffer.
In the new header, we include guest ucode information, guest
framebuffer usage, and engine usage
- Clean up the old macros since they will cause compile error if
the new transaction method is used
v2: squash in build fix
Signed-off-by: Bokun Zhang <Bokun.Zhang@amd.com>
Reviewed-by: Monk Liu <monk.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This will allow us to have different defaults per asic
in a future patch.
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Luben Tuikov <luben.tuikov@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Remove gpu_info fw support for sienna_cichlid etc., since the
information can be retrieved from discovery binary.
Signed-off-by: Jiansong Chen <Jiansong.Chen@amd.com>
Reviewed-by: Likun Gao <Likun.Gao@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
We never unmapped the regiser BAR on failure.
Reported-by: kernel test robot <lkp@intel.com>
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Reviewed-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Not being able to create amdgpu sysfs attributes
is not a fatal error warranting not to continue
to try to bring up the display. Thus, if we get
an error trying to create amdgpu sysfs attrs,
report it and continue on to try to bring up
a display.
Signed-off-by: Luben Tuikov <luben.tuikov@amd.com>
Acked-by: Slava Abramov <slava.abramov@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
XGMI support is more complicated than single device support as
questions of synchronization between the device recovering from
PCI error and other members of the hive are required.
Leaving this for next round.
Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cache the PCI state on boot and before each case where we might
loose it.
v2: Add pci_restore_state while caching the PCI state to avoid
breaking PCI core logic for stuff like suspend/resume.
v3: Extract pci_restore_state from amdgpu_device_cache_pci_state
to avoid superflous restores during GPU resets and suspend/resumes.
v4: Style fixes.
Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Wait for HW/PSP initiated ASIC reset to complete before
starting the recovery operations.
v2: Remove typo
Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
DPC recovery involves ASIC reset just as normal GPU recovery so block
SW GPU schedulers and wait on all concurrent GPU resets.
Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
At this point the ASIC is already post reset by the HW/PSP
so the HW not in proper state to be configured for suspension,
some blocks might be even gated and so best is to avoid touching it.
v2: Rename in_dpc to more meaningful name
Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
The DRM device is a static member of
the amdgpu device structure and as such
always exists, so long as the PCI and
thus the amdgpu device exist.
Signed-off-by: Luben Tuikov <luben.tuikov@amd.com>
Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
When GPU is in reset, its status isn't stable and ring buffer also need
be reset when resuming. Therefore driver should protect GPU recovery
thread from ring buffer accessed by other threads. Otherwise GPU will
randomly hang during recovery.
v2: correct indent
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Dennis Li <Dennis.Li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Fixes below compiler warnings:
CC [M] drivers/gpu/drm/amd/amdgpu/amdgpu_device.o
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c:381:1: warning: ‘static’ is not at beginning of declaration [-Wold-style-declaration]
381 | void static inline amdgpu_mm_wreg_mmio(struct amdgpu_device *adev, uint32_t reg, uint32_t v, uint32_t acc_flags)
| ^~~~
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c:381:1: warning: ‘inline’ is not at beginning of declaration [-Wold-style-declaration]
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c: In function ‘amdgpu_device_fini’:
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c:3381:6: warning: variable ‘r’ set but not used [-Wunused-but-set-variable]
3381 | int r;
| ^
Signed-off-by: Nirmoy Das <nirmoy.das@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Optimize code to iterate less loops in
amdgpu_device_ip_reinit_early_sriov()
Signed-off-by: Jiawei <Jiawei.Gu@amd.com>
Reviewed-by: Emily.Deng <Emily.Deng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Virtual display is non-atomic so report false to avoid checking
atomic state and other atomic things at runtime.
v2: squash into the sr-iov check
Acked-by: Nirmoy Das <nirmoy.das@amd.com>
Acked-by: Guchun Chen <guchun.chen@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This allows us to add asic specific workarounds for atom
asic init while keeping the adev specifics out of the
atombios parser code.
Acked-by: Nirmoy Das <nirmoy.das@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
a) Embed struct drm_device into struct amdgpu_device.
b) Modify the inline-f drm_to_adev() accordingly.
c) Modify the inline-f adev_to_drm() accordingly.
d) Eliminate the use of drm_device.dev_private,
in amdgpu.
e) Switch from using drm_dev_alloc() to
drm_dev_init().
f) Add a DRM driver release function, which frees
the container amdgpu_device after all krefs on
the contained drm_device have been released.
v2: Split out adding adev_to_drm() into its own
patch (previous commit), making this patch
more succinct and clear. More detailed commit
description.
v3: squash in fix to call drmm_add_final_kfree()
to avoid a warning.
Signed-off-by: Luben Tuikov <luben.tuikov@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Add a static inline adev_to_drm() to obtain
the DRM device pointer from an amdgpu_device pointer.
Signed-off-by: Luben Tuikov <luben.tuikov@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Get the amdgpu_device from the DRM device by use
of an inline function, drm_to_adev(). The inline
function resolves a pointer to struct drm_device
to a pointer to struct amdgpu_device.
v2: Use a typed visible static inline function
instead of an invisible macro.
Signed-off-by: Luben Tuikov <luben.tuikov@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Change to dynamically create and release hive info object,
which help driver support more hives in the future.
v2:
Change to save hive object pointer in adev, to avoid locking
xgmi_mutex every time when calling amdgpu_get_xgmi_hive.
v3:
1. Change type of hive object pointer in adev from void* to
amdgpu_hive_info*.
2. remove unnecessary variable initialization.
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Dennis Li <Dennis.Li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Using dev_xxx instead of DRM_xxx/pr_xxx to indicate which device
of a hive is the message for.
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Dennis Li <Dennis.Li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
in single gpu system, if driver reenter gpu recovery,
amdgpu_device_lock_adev will return false, but hive is
nullptr now.
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Dennis Li <Dennis.Li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
clients don't need reset-lock for synchronization when no
GPU recovery.
v2:
change to return the return value of down_read_killable.
v3:
if GPU recovery begin, VF ignore FLR notification.
Reviewed-by: Monk Liu <monk.liu@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Dennis Li <Dennis.Li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
if other threads have holden the reset lock, recovery will
fail to try_lock. Therefore we introduce atomic hive->in_reset
and adev->in_gpu_reset, to avoid reentering GPU recovery.
v2:
drop "? true : false" in the definition of amdgpu_in_reset
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Dennis Li <Dennis.Li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Only for no job running test case need to do recover in
flr notification.
For having job in mirror list, then let guest driver to
hit job timeout, and then do recover.
Signed-off-by: jqdeng <Emily.Deng@amd.com>
Acked-by: Nirmoy Das <nirmoy.das@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
The whole approach wasn't thought through till the end.
We already had a reset lock like this in the past and it caused the same problems like this one.
Completely revert the patch for now and add individual trylock protection to the hardware access functions as necessary.
This reverts commit df9c8d1aa2.
Signed-off-by: Christian König <christian.koenig@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cover the implementation details from outside(of power). Also preparing
for expanding this to swSMU.
Signed-off-by: Evan Quan <evan.quan@amd.com>
Acked-by: Nirmoy Das <nirmoy.das@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Backmerging drm-next into drm-misc-next for nouveau and panel updates.
Resolves a conflict between ttm and nouveau, where struct ttm_mem_res got
renamed to struct ttm_resource.
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Assigning false to block->status.hw overwrites PSP's previous
hardware status, which causes the PSP to Resume operation after
hardware init.
Remove this assignment and let the PSP execute Resume operation
when it is told to.
v2: Remove the braces.
v3: Modify the description.
Signed-off-by: Liu ChengZhe <ChengZhe.Liu@amd.com>
Reviewed-by: Luben Tuikov <luben.tuikov@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
There is a spelling mistake in a dev_warn message. Fix it.
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
It's related to the memory manager so move it there.
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
what:
the MQD's save and restore of KCQ (kernel compute queue)
cost lots of clocks during world switch which impacts a lot
to multi-VF performance
how:
introduce a paramter to control the number of KCQ to avoid
performance drop if there is no kernel compute queue needed
notes:
this paramter only affects gfx 8/9/10
v2:
refine namings
v3:
choose queues for each ring to that try best to cross pipes evenly.
v4:
fix indentation
some cleanupsin the gfx_compute_queue_acquire()
v5:
further fix on indentations
more cleanupsin gfx_compute_queue_acquire()
TODO:
in the future we will let hypervisor driver to set this paramter
automatically thus no need for user to configure it through
modprobe in virtual machine
Signed-off-by: Monk Liu <Monk.Liu@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
When GPU executes recovery and retriving bad GPU tag
from external eerpom device, the recovery will be broken
and error message is printed as well for user's awareness.
v2: Refine warning message in threshold reaching case, and
fix spelling typo.
v3: Fix explicit calling of bad gpu.
v4: Rename function names.
Signed-off-by: Guchun Chen <guchun.chen@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
When retrieving bad gpu tag from eeprom, GPU init should
fail as the GPU needs to be retired for further check.
v2: Fix spelling typo, correct the condition to detect
bad gpu tag and refine error message.
v3: Refine function argument name.
v4: Fix missing check of returning value of i2c
initialization error case.
v5: Use dev_err to print PCI information in dmesg instead
of DRM_ERROR.
Signed-off-by: Guchun Chen <guchun.chen@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Assigning false to block->status.hw overwrites PSP's previous
hardware status, which causes the PSP to Resume operation after
hardware init.
Remove this assignment and let the PSP execute Resume operation
when it is told to.
v2: Remove the braces.
v3: Modify the description.
Signed-off-by: Liu ChengZhe <ChengZhe.Liu@amd.com>
Reviewed-by: Luben Tuikov <luben.tuikov@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
amdgpu_device.c requires changes for SI chipsets support
si.c require changes for Display Manager IP block enabling
[How]
amdgpu_device.c: add SI families in amdgpu_device_asic_has_dc_support()
si.c: changes in si_set_ip_blocks() for Display Manager IP blocks enablement
(v1) NOTE: As per Kaveri and older amdgpu.dc=1 kernel cmdline is required
(v2) fix for bc011f9350 ("drm/amdgpu: Change SI/CI gfx/sdma/smu init sequence")
remove CHIP_HAINAN support since it does not have physical DCE6 module
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Mauro Rossi <issor.oruam@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
when GPU hang, driver has multi-paths to enter amdgpu_device_gpu_recover,
the atomic adev->in_gpu_reset and hive->in_reset are used to avoid
re-entering GPU recovery.
During GPU reset and resume, it is unsafe that other threads access GPU,
which maybe cause GPU reset failed. Therefore the new rw_semaphore
adev->reset_sem is introduced, which protect GPU from being accessed by
external threads during recovery.
v2:
1. add rwlock for some ioctls, debugfs and file-close function.
2. change to use dqm->is_resetting and dqm_lock for protection in kfd
driver.
3. remove try_lock and change adev->in_gpu_reset as atomic, to avoid
re-enter GPU recovery for the same GPU hang.
v3:
1. change back to use adev->reset_sem to protect kfd callback
functions, because dqm_lock couldn't protect all codes, for example:
free_mqd must be called outside of dqm_lock;
[ 1230.176199] Hardware name: Supermicro SYS-7049GP-TRT/X11DPG-QT, BIOS 3.1 05/23/2019
[ 1230.177221] Call Trace:
[ 1230.178249] dump_stack+0x98/0xd5
[ 1230.179443] amdgpu_virt_kiq_reg_write_reg_wait+0x181/0x190 [amdgpu]
[ 1230.180673] gmc_v9_0_flush_gpu_tlb+0xcc/0x310 [amdgpu]
[ 1230.181882] amdgpu_gart_unbind+0xa9/0xe0 [amdgpu]
[ 1230.183098] amdgpu_ttm_backend_unbind+0x46/0x180 [amdgpu]
[ 1230.184239] ? ttm_bo_put+0x171/0x5f0 [ttm]
[ 1230.185394] ttm_tt_unbind+0x21/0x40 [ttm]
[ 1230.186558] ttm_tt_destroy.part.12+0x12/0x60 [ttm]
[ 1230.187707] ttm_tt_destroy+0x13/0x20 [ttm]
[ 1230.188832] ttm_bo_cleanup_memtype_use+0x36/0x80 [ttm]
[ 1230.189979] ttm_bo_put+0x1be/0x5f0 [ttm]
[ 1230.191230] amdgpu_bo_unref+0x1e/0x30 [amdgpu]
[ 1230.192522] amdgpu_amdkfd_free_gtt_mem+0xaf/0x140 [amdgpu]
[ 1230.193833] free_mqd+0x25/0x40 [amdgpu]
[ 1230.195143] destroy_queue_cpsch+0x1a7/0x270 [amdgpu]
[ 1230.196475] pqm_destroy_queue+0x105/0x260 [amdgpu]
[ 1230.197819] kfd_ioctl_destroy_queue+0x37/0x70 [amdgpu]
[ 1230.199154] kfd_ioctl+0x277/0x500 [amdgpu]
[ 1230.200458] ? kfd_ioctl_get_clock_counters+0x60/0x60 [amdgpu]
[ 1230.201656] ? tomoyo_file_ioctl+0x19/0x20
[ 1230.202831] ksys_ioctl+0x98/0xb0
[ 1230.204004] __x64_sys_ioctl+0x1a/0x20
[ 1230.205174] do_syscall_64+0x5f/0x250
[ 1230.206339] entry_SYSCALL_64_after_hwframe+0x49/0xbe
2. remove try_lock and introduce atomic hive->in_reset, to avoid
re-enter GPU recovery.
v4:
1. remove an unnecessary whitespace change in kfd_chardev.c
2. remove comment codes in amdgpu_device.c
3. add more detailed comment in commit message
4. define a wrap function amdgpu_in_reset
v5:
1. Fix some style issues.
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Suggested-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Suggested-by: Christian König <christian.koenig@amd.com>
Suggested-by: Felix Kuehling <Felix.Kuehling@amd.com>
Suggested-by: Lijo Lazar <Lijo.Lazar@amd.com>
Suggested-by: Luben Tukov <luben.tuikov@amd.com>
Signed-off-by: Dennis Li <Dennis.Li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Plumb DC support for navy flounder through.
Signed-off-by: Bhawanpreet Lakha <Bhawanpreet.Lakha@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Add the asic family and IP blocks for navy flounder.
Signed-off-by: Jiansong Chen <Jiansong.Chen@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
If we are in RAS triggered situation and
BACO isn't support, emergency restart is needed,
and this code is only needed for some specific
cases(vega20 with given smu fw version).
After we add smu mode1 reset for sienna cichlid, we
need to share AMD_RESET_METHOD_MODE1 with psp mode1 reset,
so in amdgpu_device_gpu_recover, we need differentiate
which mode1 reset we are using, then decide if it's
a full reset and then decide if emergency restart is needed,
the logic will become much more complex.
After discussion with Hawking, move emergency restart logic
to an independent function.
Signed-off-by: Likun Gao <Likun.Gao@amd.com>
Signed-off-by: Wenhui Sheng <Wenhui.Sheng@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cleanup of phase1 suspend code to reduce unnecessary indentation.
Signed-off-by: Nirmoy Das <nirmoy.das@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This reverts commit 2eee0229f6.
Fallback to a stable base until we have a correct new one
Signed-off-by:Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
issue:
originally we kickoff IB test asynchronously with driver's
init, thus
the IB test may still running when the driver loading
done (modprobe amdgpu done).
if we shutdown VM immediately after amdgpu driver
loaded then GPU may
hang because the IB test is still running
fix:
flush the delayed_init routine at the bottom of device_init
to avoid driver loading done prior to the IB test completes
Signed-off-by: Monk Liu <Monk.Liu@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
After we move request full access before set
ip blocks, we can merge atombios init block
of SRIOV and baremetal together.
Signed-off-by: Wenhui Sheng <Wenhui.Sheng@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
From SIENNA_CICHLID, HW introduce a new protection
feature which can control the FB, doorbell and MMIO
write access for VF, so guest driver should request
full access before ip discovery, or we couldn't access
ip discovery data in FB.
Signed-off-by: Wenhui Sheng <Wenhui.Sheng@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
The release_firmware() function is NULL tolerant so we do not need
to check for NULL param before calling it.
Signed-off-by: Nirmoy Das <nirmoy.das@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
the module parameter sched_hw_submission is probably from user mode,
and the kernel need to check whether it is legal.
1. align hw sched jobs to power of 2 and set minimum number is 2.
2. use kernel api is_power_of_2() to simplify driver code.
Signed-off-by: Kevin Wang <kevin1.wang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Silence documentation warnings by correcting kernel-doc comments.
./drivers/gpu/drm/amd/amdgpu/amdgpu_device.c:3388: warning: Excess function parameter 'suspend' description in 'amdgpu_device_suspend'
./drivers/gpu/drm/amd/amdgpu/amdgpu_device.c:3485: warning: Excess function parameter 'resume' description in 'amdgpu_device_resume'
./drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c:418: warning: Excess function parameter 'tbo' description in 'amdgpu_vram_mgr_del'
./drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c:418: warning: Excess function parameter 'place' description in 'amdgpu_vram_mgr_del'
./drivers/gpu/drm/amd/amdgpu/amdgpu_gtt_mgr.c:279: warning: Excess function parameter 'tbo' description in 'amdgpu_gtt_mgr_del'
./drivers/gpu/drm/amd/amdgpu/amdgpu_gtt_mgr.c:279: warning: Excess function parameter 'place' description in 'amdgpu_gtt_mgr_del'
./drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h:332: warning: Function parameter or member 'hdcp_workqueue' not described in 'amdgpu_display_manager'
./drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h:332: warning: Function parameter or member 'cached_dc_state' not described in 'amdgpu_display_manager'
Signed-off-by: Colton Lewis <colton.w.lewis@protonmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
No need to do it again.
Acked-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
The bounding box is still needed by Navi12, temporarily read it from gpu_info
firmware. Should be droped when DAL no longer needs it.
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Xiaojie Yuan <xiaojie.yuan@amd.com>
Reviewed-by: Feifei Xu <Feifei.Xu@amd.com>
Signed-off-by: Tianci.Yin <tianci.yin@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Jerry (Fangzhi) Zuo <Jerry.Zuo@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Rather than checking of the variable is enabled and the
chip is the right family check for the presence of the
discovery table.
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Acked-by: Evan Quan <evan.quan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
The GPU info firmware is only applicable at bring up when the
IP discovery table is not present. If it's available, use that
first and then fallback to parsing the gpu info firmware.
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Acked-by: Evan Quan <evan.quan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
User can check and set the enablement of throttling logging and
the interval between each logging.
V2: simplify the sysfs interface(no string parsing)
V3: add proper lock protection on updating throttling_logging_rs.interval
V4: documentation cosmetic per Luben's suggestion
Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Luben Tuikov <luben.tuikov@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Add some APU flags to simplify handling of different APU
variants. It's easier to understand the special cases
if we use names flags rather than checking device ids and
silicon revisions.
v2: rebase on latest code
Acked-by: Evan Quan <evan.quan@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
We need to get the silicon revision id before we parse
the firmware in order to load the correct gpu info firmware
for raven2 variants.
Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/1103
Acked-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Evan Quan <evan.quan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Move it into the fw_info function since it's logically part
of the same functionality.
Acked-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Evan Quan <evan.quan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
When GPU got timeout, it would notify an interested part
of an opportunity to dump info before actual GPU reset.
A usermode app would open 'autodump' node under debugfs system
and poll() for readable/writable. When a GPU reset is due,
amdgpu would notify usermode app through wait_queue_head and give
it 10 minutes to dump info.
After usermode app has done its work, this 'autodump' node is closed.
On node closure, amdgpu gets to know the dump is done through
the completion that is triggered in release().
There is no write or read callback because necessary info can be
obtained through dmesg and umr. Messages back and forth between
usermode app and amdgpu are unnecessary.
v2: (1) changed 'registered' to 'app_listening'
(2) add a mutex in open() to prevent race condition
v3 (chk): grab the reset lock to avoid race in autodump_open,
rename debugfs file to amdgpu_autodump,
provide autodump_read as well,
style and code cleanups
v4: add 'bool app_listening' to differentiate situations, so that
the node can be reopened; also, there is no need to wait for
completion when no app is waiting for a dump.
v5: change 'bool app_listening' to 'enum amdgpu_autodump_state'
add 'app_state_mutex' for race conditions:
(1)Only 1 user can open this file node
(2)wait_dump() can only take effect after poll() executed.
(3)eliminated the race condition between release() and
wait_dump()
v6: removed 'enum amdgpu_autodump_state' and 'app_state_mutex'
removed state checking in amdgpu_debugfs_wait_dump
Improve on top of version 3 so that the node can be reopened.
v7: move reinit_completion into open() so that only one user
can open it.
v8: remove complete_all() from amdgpu_debugfs_wait_dump().
Signed-off-by: Jiange Zhao <Jiange.Zhao@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Create sysfs file using attributes.
Signed-off-by: Nirmoy Das <nirmoy.das@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This should speed up debugging VRAM access a lot.
v2: add HDP flush/invalidate
Unrevert: RAS issue at root of the issue has been addressed
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Acked-by: Jonathan Kim <Jonathan.Kim@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Kent Russell <kent.russell@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
When building with Clang:
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c:4160:53: warning: overflow in
expression; result is -294967296 with type 'long' [-Winteger-overflow]
expires = ktime_get_mono_fast_ns() + NSEC_PER_SEC * 4L;
^
1 warning generated.
Multiplication happens first due to order of operations and both
NSEC_PER_SEC and 4 are long literals so the expression overflows. To
avoid this, make 4 an unsigned long long literal, which matches the
type of expires (u64).
Fixes: 3f12acc8d6 ("drm/amdgpu: put the audio codec into suspend state before gpu reset V3")
Link: https://github.com/ClangBuiltLinux/linux/issues/1017
Signed-off-by: Nathan Chancellor <natechancellor@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
At default, the autosuspend delay of audio controller is 3S. If the
gpu reset is triggered within 3S(after audio controller idle),
the audio controller may be unable into suspended state. Then
the sudden gpu reset will cause some audio errors. The change
here is targeted to resolve this.
However if the audio controller is in use when the gpu reset
triggered, this change may be still not enough to put the
audio controller into suspend state. Under this case, the
gpu reset will still proceed but there will be a warning
message printed("failed to suspend display audio").
V2: limit this for BACO and mode1 reset only
V3: try 1st to use pm_runtime_autosuspend_expiration() to
query how much time is left. Use default setting on
failure
Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Implement an accessor of adev->tmz.enabled. Let not
code around access it as "if (adev->tmz.enabled)"
as the organization may change. Instead...
Recruit "bool amdgpu_is_tmz(adev)" to return
exactly this Boolean value. That is, this function
is now an accessor of an already initialized and
set adev and adev->tmz.
Add "void amdgpu_gmc_tmz_set(adev)" to check and
set adev->gmc.tmz_enabled at initialization
time. After which one uses "bool
amdgpu_is_tmz(adev)" to query whether adev
supports TMZ.
Also, remove circular header file include.
v2: Remove amdgpu_tmz.[ch] as requested.
v3: Move TMZ into GMC.
Signed-off-by: Luben Tuikov <luben.tuikov@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Add a function to check tmz capability with kernel parameter and ASIC type.
v2: use a per device tmz variable instead of global amdgpu_tmz.
v3: refine the comments for the function. (Luben)
v4: add amdgpu_tmz.c/h for future use.
Signed-off-by: Huang Rui <ray.huang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
The '>' expression itself is bool, no need to convert it to bool again.
This fixes the following coccicheck warning:
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c:3004:68-73: WARNING:
conversion to bool not needed here
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Jason Yan <yanaijie@huawei.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
CG/PG ungate is already performed in ip_suspend_phase1. Otherwise,
the CG/PG ungate will be performed twice. That will cause gfxoff
disablement is performed twice also on runpm enter while gfxoff
enablemnt once on rump exit. That will put gfxoff into disabled
state.
Signed-off-by: Evan Quan <evan.quan@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This sequence change should be safe as what did in ip_suspend_phase1
is to suspend DCE only. And this is a prerequisite for coming
redundant cg/pg ungate dropping.
Signed-off-by: Evan Quan <evan.quan@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
If set error query ready in amdgpu_ras_late_init, which will
cause some IP blocks aren't initialized, but their error query
is ready.
Signed-off-by: Dennis Li <Dennis.Li@amd.com>
Reviewed-by: Guchun Chen <guchun.chen@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Make code more readable.
Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This is basically just some code cosmetic. The current design
for XGMI setup gput reset is to operate on current device(adev)
first and then on other devices from the hive(by another 'for' loop).
But actually we can do some sort to the device list(to put current
device 1st position) and handle all the devices in a single 'for'
loop.
V2: added missing hive->hive_lock protection
Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
As for XGMI setup, it should be performed on other devices
from the hive also.
Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
As for XGMI setup, it needs to be performed on
all the devices from the same hive.
Signed-off-by: Evan Quan <evan.quan@amd.com>
Acked-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
clean up unused variable:
1. ring_lru_list
2. ring_lru_list_lock
related-commit:
drm/amdgpu: remove ring lru handling
Signed-off-by: Kevin Wang <kevin1.wang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This is convenient for multiple teams to obtain the information. Also,
add device info by using dev_info().
Signed-off-by: Yong Zhao <Yong.Zhao@amd.com>
Reviewed-by: Dennis Li <Dennis.Li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Vega20 arbitrates pstate at hive level and not device level. Last peer to
remote buffer unmap could drop P-State while another process is still
remote buffer mapped.
With this fix, P-States still needs to be disabled for now as SMU bug
was discovered on synchronous P2P transfers. This should be fixed in the
next FW update.
Signed-off-by: Jonathan Kim <Jonathan.Kim@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This reverts commit c12b84d6e0.
The original patch causes a RAS event and subsequent kernel hard-hang
when running the KFDMemoryTest.PtraceAccessInvisibleVram on VG20 and
Arcturus
dmesg output at hang time:
[drm] RAS event of type ERREVENT_ATHUB_INTERRUPT detected!
amdgpu 0000:67:00.0: GPU reset begin!
Evicting PASID 0x8000 queues
Started evicting pasid 0x8000
qcm fence wait loop timeout expired
The cp might be in an unrecoverable state due to an unsuccessful queues preemption
Failed to evict process queues
Failed to suspend process 0x8000
Finished evicting pasid 0x8000
Started restoring pasid 0x8000
Finished restoring pasid 0x8000
[drm] UVD VCPU state may lost due to RAS ERREVENT_ATHUB_INTERRUPT
amdgpu: [powerplay] Failed to send message 0x26, response 0x0
amdgpu: [powerplay] Failed to set soft min gfxclk !
amdgpu: [powerplay] Failed to upload DPM Bootup Levels!
amdgpu: [powerplay] Failed to send message 0x7, response 0x0
amdgpu: [powerplay] [DisableAllSMUFeatures] Failed to disable all smu features!
amdgpu: [powerplay] [DisableDpmTasks] Failed to disable all smu features!
amdgpu: [powerplay] [PowerOffAsic] Failed to disable DPM!
[drm:amdgpu_device_ip_suspend_phase2 [amdgpu]] *ERROR* suspend of IP block <powerplay> failed -5
Signed-off-by: Kent Russell <kent.russell@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
The system reboot failed as some IP blocks enter power gate before perform
hw resource destory. Meanwhile use unify interface to set device CGPG to ungate
state can simplify the amdgpu poweroff or reset ungate guard.
Fixes: 487eca11a3 ("drm/amdgpu: fix gfx hang during suspend with video playback (v2)")
Signed-off-by: Prike Liang <Prike.Liang@amd.com>
Tested-by: Mengbing Wang <Mengbing.Wang@amd.com>
Tested-by: Paul Menzel <pmenzel@molgen.mpg.de>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Let format prefixes take care of printing the module name
through pr_fmt and dev_fmt definitions.
Signed-off-by: Aurabindo Pillai <mail@aurabindo.in>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Vram lost counter is wrongly increased by two during baco reset.
V2: assumed vram lost for mode1 reset on all ASICs
Signed-off-by: Evan Quan <evan.quan@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
add indirect access support to registers outside of
mmio bar.
Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
all the register access through kiq is redirected
to amdgpu_kiq_rreg/amdgpu_kiq_wreg
Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
those are not needed anymore
Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
the workaround is not needed for soc15 ASICs except
for vega10. it is even not needed with latest vega10
vbios.
Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
The system will be hang up during S3 suspend because of SMU is pending
for GC not respose the register CP_HQD_ACTIVE access request.This issue
root cause of accessing the GC register under enter GFX CGGPG and can
be fixed by disable GFX CGPG before perform suspend.
v2: Use disable the GFX CGPG instead of RLC safe mode guard.
Signed-off-by: Prike Liang <Prike.Liang@amd.com>
Tested-by: Mengbing Wang <Mengbing.Wang@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[PATCH 2/2]
kfd_pre_reset will free mem_objs allocated by kfd_gtt_sa_allocate
Without this change, sriov tdr code path will never free those
allocated memories and get memory leak.
Signed-off-by: Jack Zhang <Jack.Zhang1@amd.com>
Reviewed-by: Monk Liu <monk.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This reverts commit 5161bba4311f in order to split it into two
different patches, and this will make it easier to understand.
[PATCH 1/2]
porting to gfx10 from
commit 1b0bfcff46 ("drm/amdgpu: Avoid destroy hqd when GPU is on reset")
Originally, MEC is touched
without GPU initialized first.
Signed-off-by: Jack Zhang <Jack.Zhang1@amd.com>
Reviewed-by: Monk Liu <monk.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Generate HW IP's sched_list in amdgpu_ring_init() instead of
amdgpu_ctx.c. This makes amdgpu_ctx_init_compute_sched(),
ring.has_high_prio and amdgpu_ctx_init_sched() unnecessary.
This patch also stores sched_list for all HW IPs in one big
array in struct amdgpu_device which makes amdgpu_ctx_init_entity()
much more leaner.
v2:
fix a coding style issue
do not use drm hw_ip const to populate amdgpu_ring_type enum
v3:
remove ctx reference and move sched array and num_sched to a struct
use num_scheds to detect uninitialized scheduler list
v4:
use array_index_nospec for user space controlled variables
fix possible checkpatch.pl warnings
Signed-off-by: Nirmoy Das <nirmoy.das@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
kfd_pre_reset will free mem_objs allocated by kfd_gtt_sa_allocate
Without this change, sriov tdr code path will never free those allocated
memories and get memory leak.
v2:add a bugfix for kiq ring test fail
Signed-off-by: Jack Zhang <Jack.Zhang1@amd.com>
Reviewed-by: Monk Liu <monk.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
extend compute lockup timeout to 60000 for SR-IOV.
Reviewed-by: Emily Deng <Emily.Deng@amd.com>
Signed-off-by: Jiawei <Jiawei.Gu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
if host support new handshake we only need to enter
fullaccess_mode in ip_init() part, otherwise we need
to do it before reading vbios (becuase host prepares vbios
for VF only after received REQ_GPU_INIT event under
legacy handshake)
Signed-off-by: Monk Liu <Monk.Liu@amd.com>
Reviewed-by: Emily Deng <Emily.Deng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
what:
1)move timtout setting before ip_early_init to reduce exclusive mode
cost for SRIOV
2)move ip_discovery_init() to inside of amdgpu_discovery_reg_base_init()
it is a prepare for the later upcoming patches.
why:
in later upcoming patches we would use a new mailbox event --
"req_gpu_init_data", which is a callback hooked in adev->virt.ops and
this callback send a new event "REQ_GPU_INIT_DAT" to host to notify
host to do some preparation like "IP discovery/vbios on the VF FB"
and this callback must be:
A) invoked after set_ip_block() because virt.ops is configured during
set_ip_block()
B) invoked before ip_discovery_init() becausen ip_discovery_init()
need host side prepares everything in VF FB first.
current place of ip_discovery_init() is before we can invoke callback
of adev->virt.ops, thus we must move ip_discovery_init() to a place
after the adev->virt.ops all settle done, and the perfect place is in
amdgpu_discovery_reg_base_init()
Signed-off-by: Monk Liu <Monk.Liu@amd.com>
Reviewed-by: Emily Deng <Emily.Deng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
by this new handshake host side can prepare vbios/ip-discovery
and pf&vf exchange data upon recieving this request without
stopping world switch.
this way the world switch is less impacted by VF's exclusive mode
request
Signed-off-by: Monk Liu <Monk.Liu@amd.com>
Reviewed-by: Emily Deng <Emily.Deng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
added flag to ras context to indicate if ras query functionality is ready
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: John Clements <john.clements@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
we need to move virt detection much earlier because:
1) HW team confirms us that RCC_IOV_FUNC_IDENTIFIER will always
be at DE5 (dw) mmio offset from vega10, this way there is no
need to implement detect_hw_virt() routine in each nbio/chip file.
for VI SRIOV chip (tonga & fiji), the BIF_IOV_FUNC_IDENTIFIER is at
0x1503
2) we need to acknowledged we are SRIOV VF before we do IP discovery because
the IP discovery content will be updated by host everytime after it recieved
a new coming "REQ_GPU_INIT_DATA" request from guest (there will be patches
for this new handshake soon).
Signed-off-by: Monk Liu <Monk.Liu@amd.com>
Reviewed-by: Emily Deng <Emily.Deng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Allow for reading of information like manufacturer, product number
and serial number from the FRU chip. Report the serial number as
the new sysfs file serial_number. Note that this only works on
server cards, as consumer cards do not feature the FRU chip, which
contains this information.
v2: Add documentation to amdgpu.rst, add helper functions,
rename functions for consistency, fix bad starting offset
v3: Remove testing definitions
Signed-off-by: Kent Russell <kent.russell@amd.com>
Reviewed-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
MMHub EDC becomes dirty after BACO reset
EDC registers should be cleared early on in reset phase
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: John Clements <john.clements@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
what changed:
1)provide new implementation interface for the rlcg access path
2)put SQ_CMD/SQ_IND_INDEX to GFX9 RLCG path to let debugfs's reg_op
function can access reg that need RLCG path help
now even debugfs's reg_op can used to dump wave.
tested-by: Monk Liu <monk.liu@amd.com>
tested-by: Zhou pengju <pengju.zhou@amd.com>
Signed-off-by: Zhou pengju <pengju.zhou@amd.com>
Signed-off-by: Monk Liu <Monk.Liu@amd.com>
Reviewed-by: Emily Deng <Emily.Deng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This can fix the baco reset failure seen on Navi10.
And this should be a low risk fix as the same sequence
is already used for system suspend/resume.
Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
1)for gfx IB test we shouldn't insert DE meta data
2)we should make sure IB test finished before we
send event 3 to hypervisor otherwise the IDLE from
event 3 will preempt IB test, which is not designed
as a compatible structure for MCBP
Signed-off-by: Monk Liu <Monk.Liu@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
We've moved the debugfs handling into a centralized place
so we can remove the legacy load an unload callbacks.
Tested-by: Thomas Zimmermann <tzimmermann@suse.de>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
In order to remove the load and unload drm callbacks,
we need to reorder the init sequence to move all the drm
debugfs file handling. Do this for firmware.
Tested-by: Thomas Zimmermann <tzimmermann@suse.de>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
In order to remove the load and unload drm callbacks,
we need to reorder the init sequence to move all the drm
debugfs file handling. Do this for register access files.
Tested-by: Thomas Zimmermann <tzimmermann@suse.de>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
In order to remove the load and unload drm callbacks,
we need to reorder the init sequence to move all the drm
debugfs file handling. Do this for gem.
Tested-by: Thomas Zimmermann <tzimmermann@suse.de>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
to amdgpu_debugfs_fini. It will be used for other things in
the future.
Tested-by: Thomas Zimmermann <tzimmermann@suse.de>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Since emulators are slower, sometime some operations like flushing tlb
through FM need more than twice the regular timout of 100ms, so increase
the timeout to 1s on emulators.
Signed-off-by: Yong Zhao <Yong.Zhao@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
UAPI Changes:
- lima: Add support for heap buffers
Cross-subsystem Changes:
Core Changes:
- Implement mode_config mode_valid for memory constrained drivers
- Bus format negociation between bridges
- Consolidate fake vblank events for drivers without vblank interrupts
- drm/bufs: dma_alloc related cleanups
- drm/dp_mst: Various fixes
- drm/print: New drm_device based print helpers
- Thomas is a drm-misc maintainer now!
Driver Changes:
- DPMS cleanups for atomic drivers
- Removal of owner field in SPI tinydrm drivers
- Removal of explicit dependency on DT for tinydrm drivers
- Conversion to YAML schemas for DT bindings
- tidss: New driver
- virtio: various reworks and fixes
- Our usual dozen or so new panels or bridges
-----BEGIN PGP SIGNATURE-----
iHUEABYIAB0WIQRcEzekXsqa64kGDp7j7w1vZxhRxQUCXkEjOgAKCRDj7w1vZxhR
xeaDAQD+1MludG4RmfQhATe4jTsPC1r2x63OF2CA0ChMGHXJyQEA8qqQ+8y1Cd/u
PZ3PpcTl4qYYHgzJ6FwW7kDPTvlaZQE=
=IJAt
-----END PGP SIGNATURE-----
Merge tag 'drm-misc-next-2020-02-10' of git://anongit.freedesktop.org/drm/drm-misc into drm-next
drm-misc-next for 5.7:
UAPI Changes:
- lima: Add support for heap buffers
Cross-subsystem Changes:
Core Changes:
- Implement mode_config mode_valid for memory constrained drivers
- Bus format negociation between bridges
- Consolidate fake vblank events for drivers without vblank interrupts
- drm/bufs: dma_alloc related cleanups
- drm/dp_mst: Various fixes
- drm/print: New drm_device based print helpers
- Thomas is a drm-misc maintainer now!
Driver Changes:
- DPMS cleanups for atomic drivers
- Removal of owner field in SPI tinydrm drivers
- Removal of explicit dependency on DT for tinydrm drivers
- Conversion to YAML schemas for DT bindings
- tidss: New driver
- virtio: various reworks and fixes
- Our usual dozen or so new panels or bridges
Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Maxime Ripard <maxime@cerno.tech>
Link: https://patchwork.freedesktop.org/patch/msgid/20200210093421.xu4sofldm6wm6xq6@gilmour.lan
So far the kfd driver implemented same routines for runtime and system
wide suspend and resume (s2idle or mem). During system wide suspend the
kfd aquires an atomic lock that prevents any more user processes to
create queues and interact with kfd driver and amd gpu. This mechanism
created problem when amdgpu device is runtime suspended with BACO
enabled. Any application that relies on kfd driver fails to load because
the driver reports a locked kfd device since gpu is runtime suspended.
However, in an ideal case, when gpu is runtime suspended the kfd driver
should be able to:
- auto resume amdgpu driver whenever a client requests compute service
- prevent runtime suspend for amdgpu while kfd is in use
This change refactors the amdgpu and amdkfd drivers to support BACO and
runtime power management.
Reviewed-by: Oak Zeng <oak.zeng@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This should speed up debugging VRAM access a lot.
v2: add HDP flush/invalidate
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Acked-by: Jonathan Kim <Jonathan.Kim@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Only write the _HI register when necessary.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Acked-by: Jonathan Kim <Jonathan.Kim@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
For sriov and pp_onevf_mode, do not send message to set smu
status, because smu doesn't support these messages under VF.
Besides, it should skip smu_suspend when pp_onevf_mode is disabled.
Signed-off-by: Jack Zhang <Jack.Zhang1@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Acked-by: Evan Quan <evan.quan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Since drm_global_mutex is a true global mutex across devices, we don't
want to acquire it unless absolutely necessary. For maintaining the
device local open_count, we can use atomic operations on the counter
itself, except when making the transition to/from 0. Here, we tackle the
easy portion of delaying acquiring the drm_global_mutex for the final
release by using atomic_dec_and_mutex_lock(), leaving the global
serialisation across the device opens.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Thomas Hellström (VMware) <thomas_os@shipmail.org>
Reviewed-by: Thomas Hellström <thellstrom@vmware.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200124130107.125404-1-chris@chris-wilson.co.uk
Better clean that up before some automation starts to complain about it
Signed-off-by: Nirmoy Das <nirmoy.das@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Reading some registers by mmio will result in hang when GPU is in
"gfxoff" state.This problem can be solved by GPU in "ring command
packages" way.
Signed-off-by: chen gong <curry.gong@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Move amdgpu_virt_kiq_rreg/amdgpu_virt_kiq_wreg function to amdgpu_gfx.c,
and rename them to amdgpu_kiq_rreg/amdgpu_kiq_wreg.Make it generic and
flexible.
Signed-off-by: chen gong <curry.gong@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Initialize notifier_lock.
Bug: https://gitlab.freedesktop.org/drm/amd/issues/1016
Reviewed-by: Feifei Xu <Feifei.Xu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: xinhui pan <xinhui.pan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
support check if dirver should try gpu recovery for
arcturus
Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Guchun Chen <guchun.chen@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This can save users much troubles. As they do not
actually need to care whether swSMU or traditional
powerplay routine should be used.
V2: apply the fixes to vi.c and cik.c also
V3: squash in oops fix
Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Under sriov and pp_onevf mode,
1.take resume instead of hw_init for smc recover to avoid
potential memory leak.
2.add return condition inside smc resume function for
sriov_pp_onevf_mode and pm_enabled param.
Signed-off-by: Jack Zhang <Jack.Zhang1@amd.com>
Acked-by: Evan Quan <evan.quan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Fixes coccicheck warning:
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c:3961:1-19: WARNING: Assignment of 0/1 to bool variable
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c:3981:1-19: WARNING: Assignment of 0/1 to bool variable
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: zhengbin <zhengbin13@huawei.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Fixes coccicheck warning:
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c:1036:5-8: Unneeded variable: "ret". Return "0" on line 1079
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Ma Feng <mafeng.ma@huawei.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Previously there is no VCN1 type ID in psp gfx interface. Also add VCN ip
block type unless the reinit after FLR for sriov would fail.
Signed-off-by: Jane Jian <Jane.Jian@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This is to avoid queueing jobs to same CPU during XGMI hive reset
because there is a strict timeline for when the reset commands
must reach all the GPUs in the hive.
Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Reviewed-by: Le Ma <Le.Ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Use task barrier in XGMI hive to synchronize ASIC resets
across devices in XGMI hive.
v2: Return right away with a warning if no xgmi hive, update doc.
Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Reviewed-by: Le Ma <Le.Ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
In preparation for doing XGMI reset synchronization using task barrier.
Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Reviewed-by: Le Ma <Le.Ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
issues:
MEC is ruined by the amdkfd_pre_reset after VF FLR done
fix:
amdkfd_pre_reset() would ruin MEC after hypervisor finished the VF FLR,
the correct sequence is do amdkfd_pre_reset before VF FLR but there is
a limitation to block this sequence:
if we do pre_reset() before VF FLR, it would go KIQ way to do register
access and stuck there, because KIQ probably won't work by that time
(e.g. you already made GFX hang)
so the best way right now is to simply remove it.
Signed-off-by: Monk Liu <Monk.Liu@amd.com>
Reviewed-by: Emily Deng <Emily.Deng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This sched array can be passed on to entity creation routine
instead of manually creating such sched array on every context creation.
v2: squash in missing break fix
Signed-off-by: Nirmoy Das <nirmoy.das@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm_sched_entity_init() takes drm gpu scheduler list instead of
drm_sched_rq list. This makes conversion of drm_sched_rq list
to drm gpu scheduler list unnecessary
Signed-off-by: Nirmoy Das <nirmoy.das@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
The log will be useful for easily getting the CU info on various
emulation models or ASICs.
Signed-off-by: Yong Zhao <Yong.Zhao@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This makes it easier to figure out whether the kernel parameter has been
taken into account.
Signed-off-by: Simon Ser <contact@emersion.fr>
Cc: Harry Wentland <hwentlan@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Originally, due to the restriction from PSP and SMU, VF has
to send message to hypervisor driver to handle powerplay
change which is complicated and redundant. Currently, SMU
and PSP can support VF to directly handle powerplay
change by itself. Therefore, the old code about the handshake
between VF and PF to handle powerplay will be removed and VF
will use new the registers below to handshake with SMU.
mmMP1_SMN_C2PMSG_101: register to handle SMU message
mmMP1_SMN_C2PMSG_102: register to handle SMU parameter
mmMP1_SMN_C2PMSG_103: register to handle SMU response
v2: remove module parameter pp_one_vf
v3: fix the parens
v4: forbid vf to change smu feature
v5: use hwmon_attributes_visible to skip sepicified hwmon atrribute
v6: change skip condition at vega10_copy_table_to_smc
Signed-off-by: Yintian Tao <yttao@amd.com>
Acked-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Otherwise next err_event_athub error cannot call gpu reset. And following
resume sequence will not be affected by this flag.
v2: create function to clear amdgpu_ras_in_intr for modularity of ras driver
Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This athub fatal error can be recovered by baco without system-level reboot,
so add a mode to use baco for the recovery. Not affect the default psp reset
situations for now.
Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Currently each XGMI node reset wq does not run in parrallel if bound to same
cpu. Make change to bound the xgmi_reset_work item to different cpus.
XGMI requires all nodes enter into baco within very close proximity before
any node exit baco. So schedule the xgmi_reset_work wq twice for enter/exit
baco respectively.
To use baco for XGMI, PMFW supported for baco on XGMI needs to be involved.
The case that PSP reset and baco reset coexist within an XGMI hive never exist
and is not in the consideration.
v2: define use_baco flag to simplify the code for xgmi baco sequence
Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This operation is needed when baco entry/exit for ras recovery
Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
As will call unload kms when initialize fail, and the unload kms will
send event 3 and 4, so don't need event 3 and 4 in device init.
Signed-off-by: Emily Deng <Emily.Deng@amd.com>
Reviewed-by: Zhan Liu <zhan.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
The documentation says the that PCI core handles this
for you unless you choose to implement it. Just rely
on the PCI core to handle the pci specific bits.
Reviewed-by: Zhan Liu <zhan.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Originally we only supported runtime pm on PX/HG laptops
so vga_switcheroo and runtime pm are sort of entangled.
Attempt to logically separate them.
Reviewed-by: Evan Quan <evan.quan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
BACO - Bus Active, Chip Off
Will be used for runtime pm. Entry will enter the BACO
state (chip off). Exit will exit the BACO state (chip on).
Reviewed-by: Evan Quan <evan.quan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
BACO - Bus Active, Chip Off
BOCO - Bus Off, Chip Off
To better match what we are checking for and to align with
amdgpu_device_supports_baco.
BOCO is used on PowerXpress/Hybrid Graphics systems and BACO
is used on desktop dGPU boards.
v2: fix typo in documentation
Reviewed-by: Evan Quan <evan.quan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
BACO - Bus Active, Chip Off
To check if a device supports BACO or not. This will be
used in determining when to enable runtime pm.
Reviewed-by: Evan Quan <evan.quan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
There is one regression from 042f3d7b745cd76aa
To put flush_delayed_work after adev->shutdown = true
which will make amdgpu_ih_process not response the irq
At last, all ib ring tests will be failed just like below
[drm] amdgpu: finishing device.
[drm] Fence fallback timer expired on ring gfx
[drm] Fence fallback timer expired on ring comp_1.0.0
[drm] Fence fallback timer expired on ring comp_1.1.0
[drm] Fence fallback timer expired on ring comp_1.2.0
[drm] Fence fallback timer expired on ring comp_1.3.0
[drm] Fence fallback timer expired on ring comp_1.0.1
amdgpu 0000:00:07.0: [drm:amdgpu_ib_ring_tests [amdgpu]] *ERROR* IB test failed on comp_1.1.1 (-110).
amdgpu 0000:00:07.0: [drm:amdgpu_ib_ring_tests [amdgpu]] *ERROR* IB test failed on comp_1.2.1 (-110).
amdgpu 0000:00:07.0: [drm:amdgpu_ib_ring_tests [amdgpu]] *ERROR* IB test failed on comp_1.3.1 (-110).
amdgpu 0000:00:07.0: [drm:amdgpu_ib_ring_tests [amdgpu]] *ERROR* IB test failed on sdma0 (-110).
amdgpu 0000:00:07.0: [drm:amdgpu_ib_ring_tests [amdgpu]] *ERROR* IB test failed on sdma1 (-110).
amdgpu 0000:00:07.0: [drm:amdgpu_ib_ring_tests [amdgpu]] *ERROR* IB test failed on uvd_enc_0.0 (-110).
amdgpu 0000:00:07.0: [drm:amdgpu_ib_ring_tests [amdgpu]] *ERROR* IB test failed on vce0 (-110).
[drm:amdgpu_device_delayed_init_work_handler [amdgpu]] *ERROR* ib ring test failed (-110).
v2: replace cancel_delayed_work_sync() with flush_delayed_work()
Signed-off-by: Yintian Tao <yttao@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
By using JPEG IP block type
Signed-off-by: Leo Liu <leo.liu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Fixes gcc warning:
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c:1954: warning: Function
parameter or member 'state' not described in 'amdgpu_device_set_cg_state'
Fixes: e3ecdffac9 ("drm/amdgpu: add documentation for amdgpu_device.c")
Signed-off-by: yu kuai <yukuai3@huawei.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Since dcn20 and dcn21 are under dcn1 it doesnt make sense to
have it named dcn1.
Change it to "dcn" to make it generic
Signed-off-by: Bhawanpreet Lakha <Bhawanpreet.Lakha@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
DCN21 is stable enough to be build by default. So drop the flags.
[How]
Remove them using the unifdef tool. The following commands were executed
in sequence:
$ find -name '*.c' -exec unifdef -m -DCONFIG_DRM_AMD_DC_DCN2_1 -UCONFIG_TRIM_DRM_AMD_DC_DCN2_1 '{}' ';'
$ find -name '*.h' -exec unifdef -m -DCONFIG_DRM_AMD_DC_DCN2_1 -UCONFIG_TRIM_DRM_AMD_DC_DCN2_1 '{}' ';'
In addition:
* Remove from kconfig, and replace any dependencies with DCN1_0.
* Remove from any makefiles.
* Fix and cleanup Renoir definitions in dal_asic_id.h
* Expand DCN1 ifdef to include DCN21 code in the following files:
* clk_mgr/clk_mgr.c: dc_clk_mgr_create()
* core/dc_resources.c: dc_create_resource_pool()
* gpio/hw_factory.c: dal_hw_factory_init()
* gpio/hw_translate.c: dal_hw_translate_init()
Signed-off-by: Bhawanpreet Lakha <Bhawanpreet.Lakha@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
DCN2 and DSC are stable enough to be build by default. So drop the flags.
[How]
Remove them using the unifdef tool. The following commands were executed
in sequence:
$ find -name '*.c' -exec unifdef -m -DCONFIG_DRM_AMD_DC_DSC_SUPPORT -DCONFIG_DRM_AMD_DC_DCN2_0 -UCONFIG_TRIM_DRM_AMD_DC_DCN2_0 '{}' ';'
$ find -name '*.h' -exec unifdef -m -DCONFIG_DRM_AMD_DC_DSC_SUPPORT -DCONFIG_DRM_AMD_DC_DCN2_0 -UCONFIG_TRIM_DRM_AMD_DC_DCN2_0 '{}' ';'
In addition:
* Remove from kconfig, and replace any dependencies with DCN1_0.
* Remove from any makefiles.
* Fix and cleanup NV defninitions in dal_asic_id.h
* Expand DCN1 ifdef to include DCN2 code in the following files:
* clk_mgr/clk_mgr.c: dc_clk_mgr_create()
* core/dc_resources.c: dc_create_resource_pool()
* dce/dce_dmcu.c: dcn20_*lock_phy()
* dce/dce_dmcu.c: dcn20_funcs
* dce/dce_dmcu.c: dcn20_dmcu_create()
* gpio/hw_factory.c: dal_hw_factory_init()
* gpio/hw_translate.c: dal_hw_translate_init()
Signed-off-by: Leo Li <sunpeng.li@amd.com>
Signed-off-by: Bhawanpreet Lakha <Bhawanpreet.Lakha@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
flush/cancel delayed works before doing finalization
to avoid concurrently requests.
Signed-off-by: Jesse Zhang <zhexi.zhang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
P-state switch should be performed after all devices from the hive
get initialized.
Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Feifei Xu <Feifei.Xu@amd.com>
Reviewed-by: Jonathan Kim <Jonathan.Kim@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Otherwise, the feature enablement will be skipped due to wrong count.
Fixes: beff74bc6e ("drm/amdgpu: fix a race in GPU reset with IB test (v2)")
Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Pstate settings should be performed after all device of the
XGMI setup get initialized.
Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
PSP lost connection when err_event_athub occurs. These cleanup work can be
skipped in BACO reset.
v2: squash in missing include (Alex)
Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Hawking Zhang <hawking.zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Correct the "_LENTH" mispelling in the AMDGPU_MAX_TIMEOUT_PARAM_LENGTH
constant.
Signed-off-by: Wambui Karuga <wambui.karugax@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
For Arcturus the I2C traffic is done through SMU tables and so
we must postpone RAS recovery init to after they are ready
which is in amdgpu_device_ip_hw_init_phase2.
Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Guchun Chen <guchun.chen@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
add a generic helper function for accessing framebuffer via MMIO
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Luben Tuikov <luben.tuikov@amd.com>
Signed-off-by: Tianci.Yin <tianci.yin@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Move it into the caller. There are cases were we don't
want it. We need it for hibernation, but we don't need
it for runtime pm, so drop it for runtime pm.
Reviewed-by: Evan Quan <evan.quan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
for amdgpu_device_suspend. This follows the logic
in the resume path.
Reviewed-by: Evan Quan <evan.quan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
When index is 1, need to set compute ring timeout for sriov and passthrough.
Signed-off-by: Emily Deng <Emily.Deng@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
It's only used in amdgpu_device.c and the naming also
reflects that. Move it there.
Reviewed-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Currently, every single piece of code in amdgpu that loops through
connectors does it incorrectly and doesn't use the proper list iteration
helpers, drm_connector_list_iter_begin() and
drm_connector_list_iter_end(). Yeesh.
So, do that.
Cc: Juston Li <juston.li@intel.com>
Cc: Imre Deak <imre.deak@intel.com>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Harry Wentland <hwentlan@amd.com>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Lyude Paul <lyude@redhat.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
When compute fence did not signal, compute ring cannot detect hardware hang
because its timeout value is set to be infinite by default.
In SR-IOV and passthrough mode, if user does not declare custome timeout
value for compute ring, then use gfx ring timeout value as default. So
that when there is a ture hardware hang, compute ring can detect it.
Signed-off-by: Jesse Zhang <zhexi.zhang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
except soc_bounding_box which is not integrated in discovery table yet
Signed-off-by: Xiaojie Yuan <xiaojie.yuan@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Set mp1 state properly for SW SMU suspend/reset routine.
Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
define sched_policy in case CONFIG_HSA_AMD is not
enabled, with this there is no need to check for CONFIG_HSA_AMD
else where in driver code.
Suggested-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Shirish S <shirish.s@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
If CONFIG_HSA_AMD is not set, build fails:
drivers/gpu/drm/amd/amdgpu/amdgpu_device.o: In function `amdgpu_device_ip_early_init':
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c:1626: undefined reference to `sched_policy'
Use CONFIG_HSA_AMD to guard this.
Fixes: 1abb680ad371 ("drm/amdgpu: disable gfxoff while use no H/W scheduling policy")
Signed-off-by: Shirish S <shirish.s@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
While gfxoff is enabled, the mmVM_XXX registers will be 0xfffffff while the GFX
is in "off" state. KFD queue creattion doesn't use ring based method, so it will
trigger a VM fault.
Signed-off-by: Huang Rui <ray.huang@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
As more and more new asics start to reuse the old device IDs before
launch, there is a need to quickly override the existing asic type
corresponding to the reused device ID through a kernel parameter. With
this, engineers no longer need to rely on local hack patches,
facilitating cooperation across teams.
Signed-off-by: Yong Zhao <Yong.Zhao@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
ras recovery_init should be called after ttm init,
bad page reserve should be put in front of gpu reset since i2c
may be unstable during gpu reset.
add cleanup for recovery_init and recovery_fini
v2: add more comment and print.
remove cancel_work_sync in recovery_init.
Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Guchun Chen <guchun.chen@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This optimizes out the pci device id usage in KFD and makes the code
more maintainable.
Signed-off-by: Yong Zhao <Yong.Zhao@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
In case of RAS error allow user configure auto system
reboot through ras_ctrl.
This is also part of the temproray work around for the RAS
hang problem.
v4: Use latest kernel API for disk sync.
Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Problem:
Under certain conditions, when some IP bocks take a RAS error,
we can get into a situation where a GPU reset is not possible
due to issues in RAS in SMU/PSP.
Temporary fix until proper solution in PSP/SMU is ready:
When uncorrectable error happens the DF will unconditionally
broadcast error event packets to all its clients/slave upon
receiving fatal error event and freeze all its outbound queues,
err_event_athub interrupt will be triggered.
In such case and we use this interrupt
to issue GPU reset. THe GPU reset code is modified for such case to avoid HW
reset, only stops schedulers, deatches all in progress and not yet scheduled
job's fences, set error code on them and signals.
Also reject any new incoming job submissions from user space.
All this is done to notify the applications of the problem.
v2:
Extract amdgpu_amdkfd_pre/post_reset from amdgpu_device_lock/unlock_adev
Move amdgpu_job_stop_all_jobs_on_sched to amdgpu_job.c
Remove print param from amdgpu_ras_query_error_count
v3:
Update based on prevoius bug fixing patch to properly call amdgpu_amdkfd_pre_reset
for other XGMI hive memebers.
Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Acked-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Issue 1:
In XGMI case amdgpu_device_lock_adev for other devices in hive
was called to late, after access to their repsective schedulers.
So relocate the lock to the begining of accessing the other devs.
Issue 2:
Using amdgpu_device_ip_need_full_reset to switch the device list from
all devices in hive to the single 'master' device who owns this reset
call is wrong because when stopping schedulers we iterate all the devices
in hive but when restarting we will only reactivate the 'master' device.
Also, in case amdgpu_device_pre_asic_reset conlcudes that full reset IS
needed we then have to stop schedulers for all devices in hive and not
only the 'master' but with amdgpu_device_ip_need_full_reset we
already missed the opprotunity do to so. So just remove this logic and
always stop and start all schedulers for all devices in hive.
Also minor cleanup and print fix.
v4: Minor coding style fix.
Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Acked-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This should be checked at all places job is accessed.
Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Enable DC support for renoir.
Acked-by: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: Roman Li <Roman.Li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
