Apply the same setting to SH_MEM_CONFIG and VM_CONTEXT1_CNTL. This
makes the noretry param no longer KFD-specific. On GFX10 I'm not
changing SH_MEM_CONFIG in this commit because GFX10 has different
retry behaviour in the SQ and I don't have a way to test it at the
moment.
Suggested-by: Christian König <Christian.Koenig@amd.com>
CC: Philip Yang <Philip.Yang@amd.com>
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by : Shaoyun.liu < Shaoyun.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
clang complains that 'false' is a not a pointer:
drivers/gpu/drm/amd/amdgpu/../display/dc/dcn20/dcn20_resource.c:2428:10: error: expression which evaluates to zero treated as a null pointer constant of type 'struct pipe_ctx *' [-Werror,-Wnon-literal-null-conversion]
return false;
Changing it to 'NULL' looks like the right thing that will shut up
the warning and make it easier to read, while not changing behavior.
Fixes: 7ed4e6352c ("drm/amd/display: Add DCN2 HW Sequencer and Resource")
Reviewed-by: Nathan Chancellor <natechancellor@gmail.com>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
As previously fixed for dml in commit 4769278e5c ("amdgpu/dc/dml:
Support clang option for stack alignment") and calcs in commit
cc32ad8f55 ("amdgpu/dc/calcs: Support clang option for stack
alignment"), dcn20 uses an option that is not available with clang:
clang: error: unknown argument: '-mpreferred-stack-boundary=4'
scripts/Makefile.build:281: recipe for target 'drivers/gpu/drm/amd/amdgpu/../display/dc/dcn20/dcn20_resource.o' failed
Use the same trick that we have in the other two files.
Fixes: 7ed4e6352c ("drm/amd/display: Add DCN2 HW Sequencer and Resource")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
add pstate mclk(uclk) support.
Signed-off-by: Kevin Wang <kevin1.wang@amd.com>
Reviewed-by: Evan Quan <evan.quan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
in the smu module, use the smu_xxxclk type to identify the CLK type
use SMU_SCLK, SMU_MCLK to replace PP_SCLK, PP_MCLK.
Signed-off-by: Kevin Wang <kevin1.wang@amd.com>
Reviewed-by: Feifei Xu <Feifei.Xu@amd.com>
Reviewed-by: Evan Quan <evan.quan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Starting from navi10, driver should send Key Database Load command
to bootloader before loading sys_drv and sos
Signed-off-by: John Clements <John.Clements@amd.com>
Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Xiaojie Yuan <xiaojie.yuan@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
The command will be sent to psp bootloader from driver
to ask psp bootloader to exerise tOS, sys_drv and kdb loading
Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Xiaojie Yuan <xiaojie.yuan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
As the clock is already logged, also log the voltage.
Signed-off-by: Paul Menzel <pmenzel@molgen.mpg.de>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
The drm_audio_component can be used to give pin ELD notifications
directly to the sound driver. This fixes audio endpoints disappearing
due to missing unsolicited notifications.
[How]
Send the notification via the audio component whenever we enable or
disable audio state on a stream. This matches what i915 does with
their drm_audio_component and what Takashi Iwai's proposed hack for
radeon/amdpgu did.
This is a bit delayed in when the notification actually occurs, however.
We wait until after all the programming is complete rather than sending
the notification mid sequence.
Particular care is needed for the get ELD callback since it can happen
outside the locking and fencing DRM does for atomic commits.
Cc: Leo Li <sunpeng.li@amd.com>
Cc: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
In order to give pin notifications to the sound driver from DM we need
to know whether audio is enabled on a stream and what pin it's using
from DC.
[How]
Expose the instance via stream status if it's a mapped resource for
the stream. It will be -1 if there's no audio mapped.
Cc: Leo Li <sunpeng.li@amd.com>
Cc: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
The cp hang occurs in OCL conformance test only on supermicro
platform which has 40 cores and the test generates 40 threads.
The root cause is race condition in non-protected flags.
The fix is to add flags of is_evicted and is_active(init_mqd())
into protected area.
Signed-off-by: Eric Huang <JinhuiEric.Huang@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Use the IP discovery table rather than hardcoding the
settings in the driver.
Reviewed-by: Xiaojie Yuan <xiaojie.yuan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
ID of DCE_HWIP from vbios is DMU_HWID,
mismatch cause null pointer crash in navi10 modprobe.
Reviewed-by: Feifei Xu <Feifei.Xu@amd.com>
Reviewed-by: Xiaojie Yuan <xiaojie.yuan@amd.com>
Signed-off-by: tiancyin <tianci.yin@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
when we set profile_peak to sysfs:power_dpm_force_performance_level,
we gets the wrong socclk level and mclk level.this patch fix this issue.
Signed-off-by: Kenneth Feng <kenneth.feng@amd.com>
Reviewed-by: Feifei Xu <Feifei.Xu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This is expected to fix some mode1 reset failures. And this
affects SMU part only as the timeout setting for other parts
is controlled by a different macro.
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>
With mm_nodes larger than 4GB, byte_count in amdgpu_fill_buffer would
overflow.
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
We need to serialize access to the psp ring if there are multiple
callers at runtime.
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Query the metrics table for the current uclk activity.
Reviewed-by: Evan Quan <evan.quan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
New parameters added for navi lack documentation.
Reviewed-by: Xiaojie Yuan <xiaojie.yuan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This RELEASE_MEM use has the Release semantic, which means we should write
back but not invalidate. Invalidations only make sense with the Acquire
semantic (ACQUIRE_MEM), or when RELEASE_MEM is used to do the combined
Acquire-Release semantic, which is a barrier, not a fence.
The undesirable side effect of doing invalidations for the Release semantic
is that it invalidates caches while shaders are running, because the Release
can execute in the middle of the next IB.
UMDs should use ACQUIRE_MEM at the beginning of IBs. Doing cache
invalidations for a fence (like in this case) doesn't do anything
for correctness.
Signed-off-by: Marek Olšák <marek.olsak@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
On 32-bit architectures, dividing a 64-bit integer in the kernel
leads to a link error:
ERROR: "__udivdi3" [drivers/gpu/drm/amd/amdgpu/amdgpu.ko] undefined!
ERROR: "__divdi3" [drivers/gpu/drm/amd/amdgpu/amdgpu.ko] undefined!
Change the two recently introduced instances to a multiply+shift
operation that is also much cheaper on 32-bit architectures.
We can do that here, since both of them are really 32-bit numbers
that change a few percent.
Fixes: bedbbe6af4 ("drm/amd/display: Move link functions from dc to dc_link")
Fixes: f18bc4e53a ("drm/amd/display: update calculated bounding box logic for NV")
Acked-by: Slava Abramov <slava.abramov@amd.com>
Tested-by: Slava Abramov <slava.abramov@amd.com>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Split it between navi10 and newer and everything before
navi10.
Reviewed-by: Xiaojie Yuan <xiaojie.yuan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Need to add appropriate ifdef.
Acked-by: Leo Li <sunpeng.li@amd.com>
Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
If smu_get_current_rpm() fails, we can't use the output,
as that may be uninitialized:
drivers/gpu/drm/amd/amdgpu/../powerplay/vega20_ppt.c:3023:8: error: variable 'current_rpm' is used uninitialized whenever '?:' condition is false [-Werror,-Wsometimes-uninitialized]
ret = smu_get_current_rpm(smu, ¤t_rpm);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
drivers/gpu/drm/amd/amdgpu/../powerplay/inc/amdgpu_smu.h:735:3: note: expanded from macro 'smu_get_current_rpm'
((smu)->funcs->get_current_rpm ? (smu)->funcs->get_current_rpm((smu), (speed)) : 0)
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~
drivers/gpu/drm/amd/amdgpu/../powerplay/vega20_ppt.c:3024:12: note: uninitialized use occurs here
percent = current_rpm * 100 / pptable->FanMaximumRpm;
^~~~~~~~~~~
drivers/gpu/drm/amd/amdgpu/../powerplay/vega20_ppt.c:3023:8: note: remove the '?:' if its condition is always true
ret = smu_get_current_rpm(smu, ¤t_rpm);
^
drivers/gpu/drm/amd/amdgpu/../powerplay/inc/amdgpu_smu.h:735:3: note: expanded from macro 'smu_get_current_rpm'
((smu)->funcs->get_current_rpm ? (smu)->funcs->get_current_rpm((smu), (speed)) : 0)
^
drivers/gpu/drm/amd/amdgpu/../powerplay/vega20_ppt.c:3020:22: note: initialize the variable 'current_rpm' to silence this warning
uint32_t current_rpm;
Propagate the error code in that case.
Fixes: ee0db82027 ("drm/amd/powerplay: move PPTable_t uses into asic level")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Without this header, we get a compiler error in some configurations:
.../dc/dcn20/dcn20_hwseq.c: In function 'dcn20_hwss_wait_for_blank_complete':
.../dc/dcn20/dcn20_hwseq.c:1493:3: error: implicit declaration of function 'udelay' [-Werror=implicit-function-declaration]
Note: the use of udelay itself may be problematic, as can occupy
the CPU for 200ms in a busy-loop here.
Fixes: 7ed4e6352c ("drm/amd/display: Add DCN2 HW Sequencer and Resource")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
When CONFIG_PERF_EVENTS is disabled, we cannot compile the pmu
portion of the amdgpu driver:
drivers/gpu/drm/amd/amdgpu/amdgpu_pmu.c:48:38: error: no member named 'hw' in 'struct perf_event'
struct hw_perf_event *hwc = &event->hw;
~~~~~ ^
drivers/gpu/drm/amd/amdgpu/amdgpu_pmu.c:51:13: error: no member named 'attr' in 'struct perf_event'
if (event->attr.type != event->pmu->type)
~~~~~ ^
...
Use conditional compilation for this file.
Fixes: 9c7c85f7ea ("drm/amdgpu: add pmu counters")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
clang warns (trimmed for brevity):
drivers/gpu/drm/amd/amdgpu/../powerplay/vega20_ppt.c:3023:8: warning:
variable 'current_rpm' is used uninitialized whenever '?:' condition is
false [-Wsometimes-uninitialized]
ret = smu_get_current_rpm(smu, ¤t_rpm);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
smu_get_current_rpm expands to a ternary operator conditional on
smu->funcs->get_current_rpm being not NULL. When this is false,
current_rpm will be uninitialized. Zero initialize current_rpm to
avoid using random stack values if that ever happens.
Fixes: ee0db82027 ("drm/amd/powerplay: move PPTable_t uses into asic level")
Link: https://github.com/ClangBuiltLinux/linux/issues/588
Signed-off-by: Nathan Chancellor <natechancellor@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
clang warns (trimmed for brevity):
drivers/gpu/drm/amd/amdgpu/../powerplay/smu_v11_0.c:1098:10: warning:
variable 'freq' is used uninitialized whenever '?:' condition is false
[-Wsometimes-uninitialized]
ret = smu_get_current_clk_freq_by_table(smu, clk_id, &freq);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
If get_current_clk_freq_by_table is ever NULL, freq will fail to be
properly initialized. Zero initialize it to avoid using uninitialized
stack values.
smu_get_current_clk_freq_by_table expands to a ternary operator
conditional on smu->funcs->get_current_clk_freq_by_table being not NULL.
When this is false, freq will be uninitialized. Zero initialize freq to
avoid using random stack values if that ever happens.
Fixes: e36182490d ("drm/amd/powerplay: fix dpm freq unit error (10KHz -> Mhz)")
Link: https://github.com/ClangBuiltLinux/linux/issues/585
Signed-off-by: Nathan Chancellor <natechancellor@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
clang warns:
drivers/gpu/drm/amd/amdgpu/../powerplay/navi10_ppt.c:601:33: warning:
suggest braces around initialization of subobject [-Wmissing-braces]
static SmuMetrics_t metrics = {0};
^
{}
drivers/gpu/drm/amd/amdgpu/../powerplay/navi10_ppt.c:905:26: warning:
suggest braces around initialization of subobject [-Wmissing-braces]
SmuMetrics_t metrics = {0};
^
{}
2 warnings generated.
One way to fix these warnings is to add additional braces like clang
suggests; however, there has been a bit of push back from some
maintainers[1][2], who just prefer memset as it is unambiguous, doesn't
depend on a particular compiler version[3], and properly initializes all
subobjects. Do that here so there are no more warnings.
[1]: https://lore.kernel.org/lkml/022e41c0-8465-dc7a-a45c-64187ecd9684@amd.com/
[2]: https://lore.kernel.org/lkml/20181128.215241.702406654469517539.davem@davemloft.net/
[3]: https://lore.kernel.org/lkml/20181116150432.2408a075@redhat.com/
Fixes: 98e1a543c7 ("drm/amd/powerplay: add function get current clock freq interface for navi10")
Fixes: ab43c4bf1c ("drm/amd/powerplay: fix fan speed show error (for hwmon pwm)")
Link: https://github.com/ClangBuiltLinux/linux/issues/583
Signed-off-by: Nathan Chancellor <natechancellor@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
clang warns:
In file included from drivers/gpu/drm/amd/amdgpu/nv.c:53:
drivers/gpu/drm/amd/amdgpu/../amdgpu/mes_v10_1.h:24:9: warning:
'__MES_V10_1_H__' is used as a header guard here, followed by #define of
a different macro [-Wheader-guard]
#ifndef __MES_V10_1_H__
^~~~~~~~~~~~~~~
drivers/gpu/drm/amd/amdgpu/../amdgpu/mes_v10_1.h:25:9: note:
'__MES_v10_1_H__' is defined here; did you mean '__MES_V10_1_H__'?
#define __MES_v10_1_H__
^~~~~~~~~~~~~~~
__MES_V10_1_H__
1 warning generated.
Capitalize the V.
Fixes: 886f82aa7a ("drm/amdgpu/mes10.1: add ip block mes10.1 (v2)")
Link: https://github.com/ClangBuiltLinux/linux/issues/582
Signed-off-by: Nathan Chancellor <natechancellor@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
the hwmon interface need temperature sensor type support.
1. SENSOR_HOTSPOT_TEMP
2. SENSOR_EDGE_TEMP(SENSOR_GPU_TEMP)
3. SENSOR_MEM_TEMP
Signed-off-by: Kevin Wang <kevin1.wang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
An upcoming change in the hmm_range_register API requires passing in
a pointer to an hmm_mirror instead of mm_struct. To access the
hmm_mirror we need pass bo instead of ttm to amdgpu_ttm_tt_get_user_pages
because mirror is part of amdgpu_mn structure, which is accessible from bo.
v2: fix building without CONFIG_HMM_MIRROR (Arnd)
Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
add mode1 (by psp) reset for navi asic.
Signed-off-by: Kevin Wang <kevin1.wang@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Perform a ras_suspend to disable ras on all IPs to workaround
some ROCm stability issue.
Signed-off-by: xinhui pan <xinhui.pan@amd.com>
Acked-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
add baco reset support for smu11.
it can help gpu do asic reset when gpu recovery.
Signed-off-by: Kevin Wang <kevin1.wang@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
We observed some SMU commands take more time for execution,
so increase waiting time for response.
Signed-off-by: Jack Xiao <Jack.Xiao@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This driver currently contains a repeated 500ms blocking delay call
which causes frequent major buffer underruns in PulseAudio. This patch
fixes this issue by replacing the blocking delay with a non-blocking
sleep call.
Signed-off-by: Yrjan Skrimstad <yrjan@skrimstad.net>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
kmemdup is introduced to duplicate a region of memory in a neat way.
Rather than kmalloc/kzalloc + memcpy, which the programmer needs to
write the size twice (sometimes lead to mistakes), kmemdup improves
readability, leads to smaller code and also reduce the chances of mistakes.
Suggestion to use kmemdup rather than using kmalloc/kzalloc + memcpy.
Reviewed-by: Christian König <Christian.Koenig@amd.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Signed-off-by: Fuqian Huang <huangfq.daxian@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This works around difficult-to-reproduce soft hangs on oversubscribed
runlists.
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Oak Zeng <Oak.Zeng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
New flag to disable an idle runlist optimization that is causing soft
hangs with some diffult-to-reproduce customer workloads. This will
serve as a workaround until the problem can be reproduced and the
root-cause determined.
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Oak Zeng <Oak.Zeng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Restore the code that resets mem->invalid. Othewise so mapping
userptrs after they got an MMU notifiers would always be skipped.
This also avoids unnecessarily calling get_user_pages on BOs that
have not been invalidated since the last try.
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Philip Yang <Philip.Yang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
We don't want eviction fences to trigger when waiting for page table
updates to complete during restore. In theory there shouldn't be any
unsignaled eviction fences in the PD reservation object, but I'm
seeing them in instrumented code for reasons not fully understood.
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Philip Yang <Philip.Yang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Oversubscription of queues or processes results in poor performance
mostly because HWS blinbly schedules busy and idle queues, resulting
in poor occupancy if many queues are idle.
Let users know with a warning message when transitioning from a
non-oversubscribed to an oversubscribed runlist.
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Oak Zeng <Oak.Zeng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Just for cleanup.
Reviewed-by: Prike Liang <Prike.Liang@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Jack Xiao <Jack.Xiao@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
All asics newer than vega10 use client ids, so simplify the
check.
Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Since amdgpu has always requested PCIE atomics, kfd don't
need duplicated PCIE atomics enablement. Referring to amdgpu
request result is enough.
Signed-off-by: Jack Xiao <Jack.Xiao@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
GPU atomics operation depends on PCIE atomics support.
Always enable PCIE atomics ops support in case that
it hasn't been enabled.
Signed-off-by: Jack Xiao <Jack.Xiao@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
The new field in amdgpu device is used to record whether the
system has PCIE atomics support. The field can be exposed to
UMD or kfd whether PCIE atomics have supported.
Signed-off-by: Jack Xiao <Jack.Xiao@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
