- 28 May, 2020 40 commits
-
-
Gavin Wan authored
For SRIOV, since the CP_INT_CNTL_RING0 is programed on host side. The Guest should not program CP_INT_CNTL_RING0 again. Signed-off-by:
Gavin Wan <Gavin.Wan@amd.com> Reviewed-by:
Monk Liu <Monk.Liu@amd.com> Acked-by:
Alex Deucher <alexander.deucher@amd.com> Signed-off-by:
-
Aric Cyr authored
Signed-off-by:
Aric Cyr <aric.cyr@amd.com> Acked-by:
Qingqing Zhuo <qingqing.zhuo@amd.com> Signed-off-by:
-
Alvin Lee authored
[Why] When determining synchronzied vblank we don't need to compare the stream with itself [How] If comparing same stream, continue to next iteration Signed-off-by:
Alvin Lee <alvin.lee2@amd.com> Reviewed-by:
Jun Lei <Jun.Lei@amd.com> Acked-by:
-
Anthony Koo authored
[Header Changes] - Combine all interface dependencies between driver and fw into a single header file - Add FW Versioning to the dmub_cmd.h file Signed-off-by:Anthony Koo <Anthony.Koo@amd.com> Acked-by:
-
Hugo Hu authored
[why] We hit an issue which driver reallocate a pipe from desktop bottom pipe to video bottom pipe. In this case, driver need to re-enable plane. [how] Enable plane if container of plane status changed. Signed-off-by:
Hugo Hu <hugo.hu@amd.com> Reviewed-by:
Tony Cheng <Tony.Cheng@amd.com> Acked-by:
-
Anthony Koo authored
[Why] We want to better encapsulate all driver-fw dependencies into a single file. [How] Combine all the headers under inc folder into a single header Signed-off-by:
-
Eric Bernstein authored
[Why] Diagnostics DIO test with eDP not connected is required to run [How] Allow Diagnostics test with eDP not connected to skip link detection but still execute DIO test Signed-off-by:
Eric Bernstein <eric.bernstein@amd.com> Reviewed-by:
Dmytro Laktyushkin <Dmytro.Laktyushkin@amd.com> Acked-by:
-
Paul Hsieh authored
[Why] Unit enter to S4, garbage show on screen when do OPTC blank. [How] Wait for vblank then do OPTC blank Signed-off-by:
Paul Hsieh <paul.hsieh@amd.com> Reviewed-by:
-
Aric Cyr authored
[Why] There are scenarios where no OPP is assigned to an OTG so its value is 0xF which is outside the size of the OPP array causing a potential driver crash. [How] Change the assert to an early return to guard against access. If there's no OPP assigned already, then OTG will be blank anyways so no functionality should be lost. Signed-off-by:
Zhan Liu <Zhan.Liu@amd.com> Acked-by:
-
Alvin Lee authored
[Why] HW team request to disable PG on NV12 (fixing missed cases) [How] Disable dpp and hubp PG Signed-off-by:
-
David Galiffi authored
[WHY] To facilitate DM removing the dependency between dc and the firmware binary. [HOW] Setting the default values to match VBIOS: 64 KB. These values are only used if meta is absent. Signed-off-by:
David Galiffi <David.Galiffi@amd.com> Reviewed-by:
Nicholas Kazlauskas <Nicholas.Kazlauskas@amd.com> Acked-by:
-
Aric Cyr authored
[Why] Link loss currently only retrains and re-enables the stream. This can cause issues for some sinks. [How] When link loss occurs, the link and stream(s) should be completely disabled and then reenabled. Signed-off-by:
Wenjing Liu <Wenjing.Liu@amd.com> Acked-by:
-
Dmytro Laktyushkin authored
Current implementation is slightly inaccurate and will often result in truncation/floor operation decrementing an exact integer output by 1. Only rounded down output is ever expected, just extract the fp exponent for this to increase performance and avoid any truncation issues. Signed-off-by:
-
Bhawanpreet Lakha authored
[Why] Previously we used link signal type to get the caps. We should use the sink signal type [How] Use sink signal type instead of link signal type Signed-off-by:
Bhawanpreet Lakha <Bhawanpreet.Lakha@amd.com> Reviewed-by:
-
Paul Hsieh authored
[Why] The link_status is incorrect cause driver power off eDP when backlight on. Some eDP panels may show garbage on screen. [How] Correct link_status when power off encoder Signed-off-by:
-
Aric Cyr authored
Signed-off-by:
-
Yong Zhao authored
Use words insteads of acronyms for better understanding. Signed-off-by:
Yong Zhao <Yong.Zhao@amd.com> Reviewed-by:
-
Mukul Joshi authored
Track SDMA usage on a per process basis and report it through sysfs. The value in the sysfs file indicates the amount of time SDMA has been in-use by this process since the creation of the process. This value is in microsecond granularity. Signed-off-by:
Mukul Joshi <mukul.joshi@amd.com> Reviewed-by:
Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by:
-
Wenhui Sheng authored
smu_i2c_eeprom_init may be invoked twice or more under sroiv mode, while we don't want to add check if (!amdgpu_sriov_vf) before we invoke smu_i2c_eeprom_init/fini each time, so we check if i2c adapter is already added before we invoke i2c_add_adapter Signed-off-by:
Wenhui Sheng <Wenhui.Sheng@amd.com> Reviewed-by:
Hawking Zhang <Hawking.Zhang@amd.com> Reviewed-by:
Kevin Wang <kevin1.wang@amd.com> Signed-off-by:
-
Wenhui Sheng authored
We don't need SRIOV check after we enable SMC msg filter in SMU11 v2: squash in unused variable fix, unused ids Signed-off-by:
-
Wenhui Sheng authored
1. enable SMC message filter in SRIOV situation 2. return -EACCESS if msg is blocked from smu_msg_get_index 3. if msg is block, always return 0 from smu_v11_0_send_msg_with_param Signed-off-by:
-
Wenhui Sheng authored
1. add smu_11_0_msg_mapping definition 2. add valid info for each SMC message in SRIOV Signed-off-by:
-
Evan Quan authored
Since on early phase of bringup, the SMU IP may be not enabled or supported. Without this, we may hit null pointer dereference on accessing smu->adev. Signed-off-by:
Evan Quan <evan.quan@amd.com> Reviewed-by:
-
Nicholas Kazlauskas authored
[Why] Warnings in the kernel are generally treated as errors. The BREAK_TO_DEBUGGER macro is not a critical error or warning, but rather intended for developer use to help investigate behavior and sequences for other issues. We do still make use of DC_ERROR/ASSERT(0) in various places in the code for things that are genuine issues. Since most developers don't actually KGDB while debugging the kernel these essentially would have no value on their own since the KGDB breakpoint wouldn't trigger - ASSERT(0) was used as a shortcut to get a stacktrace. [How] Turn it into a DRM_DEBUG_DRIVER print instead. We unfortunately lose the stacktrace, but we still do retain some of the useful debug information this offers by having at least the function and line number loggable. If KGDB is supported in the kernel this will still trigger a real breakpoint as well. Cc: Harry Wentland <harry.wentland@amd.com> Cc: Leo Li <sunpeng.li@amd.com> Cc: Bhawanpreet Lakha <Bhawanpreet.Lakha@amd.com> Cc: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com> Signed-off-by:
Harry Wentland <harry.wentland@amd.com> Signed-off-by:
-
Gustavo A. R. Silva authored
The current codebase makes use of one-element arrays in the following form: struct something { int length; u8 data[1]; }; struct something *instance; instance = kmalloc(sizeof(*instance) + size, GFP_KERNEL); instance->length = size; memcpy(instance->data, source, size); but the preferred mechanism to declare variable-length types such as these ones is a flexible array member[1][2], introduced in C99: struct foo { int stuff; struct boo array[]; }; By making use of the mechanism above, we will get a compiler warning in case the flexible array does not occur last in the structure, which will help us prevent some kind of undefined behavior bugs from being inadvertently introduced[3] to the codebase from now on. So, replace the one-element array with a flexible-array member. Also, make use of the new struct_size() helper to properly calculate the size of struct SISLANDS_SMC_SWSTATE. This issue was found with the help of Coccinelle and, audited and fixed _manually_. [1] https://gcc.gnu.org/onlinedocs/gcc/Zero-Length.html [2] https://github.com/KSPP/linux/issues/21 [3] commit 76497732 ("cxgb3/l2t: Fix undefined behaviour") Signed-off-by:Gustavo A. R. Silva <gustavoars@kernel.org> Signed-off-by:
-
Gustavo A. R. Silva authored
The current codebase makes use of one-element arrays in the following form: struct something { int length; u8 data[1]; }; struct something *instance; instance = kmalloc(sizeof(*instance) + size, GFP_KERNEL); instance->length = size; memcpy(instance->data, source, size); but the preferred mechanism to declare variable-length types such as these ones is a flexible array member[1][2], introduced in C99: struct foo { int stuff; struct boo array[]; }; By making use of the mechanism above, we will get a compiler warning in case the flexible array does not occur last in the structure, which will help us prevent some kind of undefined behavior bugs from being inadvertently introduced[3] to the codebase from now on. So, replace the one-element array with a flexible-array member. Also, make use of the new struct_size() helper to properly calculate the size of struct NISLANDS_SMC_SWSTATE. This issue was found with the help of Coccinelle and, audited and fixed _manually_. [1] https://gcc.gnu.org/onlinedocs/gcc/Zero-Length.html [2] https://github.com/KSPP/linux/issues/21 [3] commit 76497732 ("cxgb3/l2t: Fix undefined behaviour") Reviewed-by:Christian König <christian.koenig@amd.com> Signed-off-by:
-
Alex Deucher authored
SI and CIK came before VI and newer asics. Reviewed-by:
-
Alex Deucher authored
Just check if it's an APU. The checks for the ppfuncs are pointless because if we don't have them we can't power up sdma anyway so we shouldn't even be in this code in the first place. I'm not sure about the in_gpu_reset check. This probably needs to be double checked. The fini logic doesn't match the init logic however with that in_gpu_reset check in place which seems odd. Reviewed-by:
-
Alex Deucher authored
Check if mec2 fw exists rather than checking asic types. Reviewed-by:
-
Alex Deucher authored
Just check for APU. Reviewed-by:
-
Alex Deucher authored
Just check for APU. Reviewed-by:
-
Alex Deucher authored
Looks like it should be handled here as well. Reviewed-by:
-
Alex Deucher authored
Looks like renoir should be handled here as well. Reviewed-by:
-
Alex Deucher authored
Just register the a pointer to the backlight device and use that. Unifies the DC and non-DC handling. Acked-by:
-
Christian König authored
Always use the PCI GART instead. We just have to many cases where AGP still causes problems. This means a performance regression for some GPUs, but also a bug fix for some others. Signed-off-by:
-
Guchun Chen authored
This will assist debug in error injection case. Signed-off-by:
Guchun Chen <guchun.chen@amd.com> Reviewed-by:
Tao Zhou <tao.zhou1@amd.com> Signed-off-by:
-
Hua Zhang authored
When smu_i2c_eeprom_init is called on the smu resuming process under sroiv mode, there will be a call trace: [ 436.377690] dump_stack+0x63/0x85 [ 436.377695] kobject_init+0x77/0x90 [ 436.377704] device_initialize+0x28/0x110 [ 436.377708] device_register+0x12/0x20 [ 436.377756] i2c_register_adapter+0xeb/0x400 [ 436.377763] i2c_add_adapter+0x5a/0x80 [ 436.377951] arcturus_i2c_eeprom_control_init+0x60/0x80 [amdgpu] [ 436.378123] smu_resume+0xcc/0x110 [amdgpu] [ 436.378247] amdgpu_device_gpu_recover+0xfb1/0xfc0 [amdgpu] [ 436.378401] amdgpu_job_timedout+0xf2/0x150 [amdgpu] [ 436.378414] drm_sched_job_timedout+0x70/0xc0 [amd_sched] [ 436.378420] ? drm_sched_job_timedout+0x70/0xc0 [amd_sched] [ 436.378430] process_one_work+0x1fd/0x3f0 [ 436.378438] worker_thread+0x34/0x410 [ 436.378444] kthread+0x121/0x140 [ 436.378451] ? process_one_work+0x3f0/0x3f0 [ 436.378456] ? kthread_create_worker_on_cpu+0x70/0x70 [ 436.378464] ret_from_fork+0x35/0x40 This is because smu_i2c_eeprom is not released on gpu recovering. Actually, smu_i2c_eeprom_init/fini are only needed under bare mental mode. Signed-off-by:
Hua Zhang <hua.zhang@amd.com> Reviewed-by:
-
Kevin Wang authored
by default, vega20 will use legacy powerplay driver. in order to maintain the code conveniently in the future, remove the support of vega20 from swsmu. Signed-off-by:
-
Kevin Wang authored
the vega20 asic uses legacy powerplay driver by default. 1. cleanup is_support_sw_smu_xgmi() function. (only use for vega20 xgmi pstate check) 2. by default, the vega20 set xgmi pstate by legacy powerplay routine. Signed-off-by:
-
Likun Gao authored
Change memory training init and finit a common function, as it only have software behavior do not relay on the IP version of PSP. Signed-off-by:
Likun Gao <Likun.Gao@amd.com> Reviewed-by:
-
