- 27 Oct, 2020 33 commits
-
-
Evan Quan authored
Correct Polaris10 pcie spc cap setting. Signed-off-by:
Evan Quan <evan.quan@amd.com> Acked-by:
Alex Deucher <alexander.deucher@amd.com> Signed-off-by:
-
Evan Quan authored
Correct Polaris10 clk/voltage dependence setup. Signed-off-by:
-
Evan Quan authored
Populate the correct highest vddc setting on Polaris. Signed-off-by:
-
Evan Quan authored
Correct Polaris10 sclk/mclk dpm enablement. Signed-off-by:
-
Evan Quan authored
Correct Polaris10 smc voltage controller setup. Signed-off-by:
-
Evan Quan authored
Correct Polaris10 platformcaps setup. Signed-off-by:
-
Evan Quan authored
Correct Polaris VRconfig setting. Signed-off-by:
-
Evan Quan authored
Correct Polaris10 vddc phase control. Signed-off-by:
-
Evan Quan authored
Correct Polaris10 avfs fuse setting. Signed-off-by:
-
Evan Quan authored
Correct Polaris DIDT enablement. Signed-off-by:
-
Evan Quan authored
Correct powertune table setup for Polaris. Signed-off-by:
-
Evan Quan authored
Correct sclk/mclk SS support checks. Signed-off-by:
-
Evan Quan authored
Add VR shared rail info. Signed-off-by:
-
Evan Quan authored
Add mc register table initialization. Signed-off-by:
-
Evan Quan authored
Enable edc controller table setting. Signed-off-by:
-
Evan Quan authored
Only if the ZeroRPM feature is supported. Signed-off-by:
-
Evan Quan authored
By using the saved copy of ro_range_maximum and ro_range_minimum. Correct the setting for "LdoRefSel". Signed-off-by:
-
Evan Quan authored
Make the settings more precise. Signed-off-by:
-
Evan Quan authored
By moving that in atomfw_read_efuse(). Signed-off-by:
-
Evan Quan authored
Programming AC Timing Parameters is only dependent on MCLK. No need to nest loop for each SCLK DPM level. Signed-off-by:
-
Evan Quan authored
Instead of sharing the fan table setup with Tonga, Polaris has its own fan table setup. Signed-off-by:
-
Evan Quan authored
This is for some special Polaris10 ASICs. Signed-off-by:
-
Evan Quan authored
Correct polaris CG_ACLK_CNTL setting. Signed-off-by:
-
Evan Quan authored
As this is not needed for polaris. Signed-off-by:
-
Evan Quan authored
Add missing VRHotLevel setting. Signed-off-by:
-
Evan Quan authored
Set the BootLinkLevel as the max level. Signed-off-by:
-
Evan Quan authored
Correct the setting for "ActivityLevel". V2: rich the comment Signed-off-by:
-
Evan Quan authored
Correct the settings for "StutterEnable" and "EnabledForActivity". Signed-off-by:
-
Evan Quan authored
Correct Polaris10 sclk table setup. Signed-off-by:
-
Evan Quan authored
Make sure the settings are applied only when voltage controlled by gpio. Signed-off-by:
-
Evan Quan authored
Add missing smc samu table setup. Signed-off-by:
-
Evan Quan authored
Add missing vddc table setup. Signed-off-by:
-
Evan Quan authored
By defining new Macros. Signed-off-by:
-
- 26 Oct, 2020 7 commits
-
-
Takashi Iwai authored
This patch simplifies the ASSERT*() and BREAK_TO_DEBUGGER() macros: - Move the dependency check of CONFIG_KGDB into Kconfig - Unify the kgdb_breakpoint() call - Drop the non-existing CONFIG_HAVE_KGDB Also align the behavior of ASSERT() macro in both cases with and without CONFIG_DEBUG_KERNEL_DC. Acked-by:
Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> Signed-off-by:
Takashi Iwai <tiwai@suse.de> Signed-off-by:
-
Takashi Iwai authored
ASSERT_CRITICAL() invokes kgdb_breakpoint() whenever either CONFIG_KGDB or CONFIG_HAVE_KGDB is set. This, however, may lead to a kernel panic when no kdb stuff is attached, since the kgdb_breakpoint() call issues INT3. It's nothing but a surprise for normal end-users. For avoiding the pitfall, make the kgdb_breakpoint() call only when CONFIG_DEBUG_KERNEL_DC is set. https://bugzilla.opensuse.org/show_bug.cgi?id=1177973 Cc: <stable@vger.kernel.org> Acked-by:
-
Takashi Iwai authored
Currently both error code paths handled in dal_gpio_open_ex() issues ASSERT_CRITICAL(), and this leads to a kernel panic unnecessarily if CONFIG_KGDB is enabled. Since basically both are non-critical errors and can be recovered, drop those assert calls and use a safer one, BREAK_TO_DEBUGGER(), for allowing the debugging, instead. BugLink: https://bugzilla.opensuse.org/show_bug.cgi?id=1177973 Cc: <stable@vger.kernel.org> Acked-by:
-
Sumera Priyadarsini authored
Bool initialisation should use 'true' and 'false' values instead of 0 and 1. Modify amdgpu_amdkfd_gpuvm.c to initialise variable is_imported to false instead of 0. Issue found with Coccinelle. Signed-off-by:
Sumera Priyadarsini <sylphrenadin@gmail.com> Signed-off-by:
-
Zhang Qilong authored
The 'break' is unnecessary because of previous 'return', discard it. Signed-off-by:
Zhang Qilong <zhangqilong3@huawei.com> Signed-off-by:
-
Alex Deucher authored
It looks this was accidently lost in a follow up patch. dc context is large and we don't need contiguous pages. Fixes: e4863f11 ("drm/amd/display: Multi display cause system lag on mode change") Reviewed-by:
Alex Xu (Hello71) <alex_y_xu@yahoo.ca> Tested-by:
-
Derek Lai authored
[Why] For user regamma we're missing this function call to combine user regamma + OS for GAMMA_CS_TFM_1D type. [How] Applied 1D LUT in the mod_color_build_user_regamma. And Set the regamma dirty as updateGamma. Signed-off-by:
Derek Lai <Derek.Lai@amd.com> Acked-by:
Aurabindo Pillai <aurabindo.pillai@amd.com> Signed-off-by:
-
