- 21 Oct, 2017 40 commits
-
-
Wenjing Liu authored
[Description] There are many occasions we need to retrieve sink capability and notify connectivity change to os even if edid is not changed on a HPD toggle. (HDMI2.0 display needs re-enable link on every hpd, display changes other capability outside from edid need to be queried again and possibly reset the mode, etc.) In these cases we cannot keep the same sink without letting DM know. Signed-off-by:
Wenjing Liu <Wenjing.Liu@amd.com> Reviewed-by:
Tony Cheng <Tony.Cheng@amd.com> Acked-by:
Harry Wentland <Harry.Wentland@amd.com> Signed-off-by:
Alex Deucher <alexander.deucher@amd.com>
-
Tony Cheng authored
in past program SMU will use all voltage headroom. RV does not if DAL need higher voltage for DCFCLK or DISPCLK, also increase FCLK to improve stutter as voltage is already Signed-off-by:
Charlene Liu <Charlene.Liu@amd.com> Acked-by:
-
Andrew Jiang authored
A recent commit moved the backlight control code along with the register defines, but did not move the power control code. This along with remnant fields in the dce110_link_enc_registers struct made it so that the code still compiled, but any attempts to access the LVTMA_PWRSEQ_STATE register led to reading from an address of 0. This patch corrects that. Also, rename blacklight_control to edp_backlight_control (Typo fix). Signed-off-by:
Andrew Jiang <Andrew.Jiang@amd.com> Reviewed-by:
-
Yue Hin Lau authored
Signed-off-by:
Yue Hin Lau <Yuehin.Lau@amd.com> Reviewed-by:
-
Jerry Zuo authored
In the full update type, need to add ref_count to the newly created dc_state->stream. It made mistake to add ref_count to dc->current_state->stream which keeps adding up without release. Signed-off-by:
Jerry Zuo <Jerry.Zuo@amd.com> Reviewed-by:
Roman Li <Roman.Li@amd.com> Acked-by:
-
Tony Cheng authored
Signed-off-by:
-
Tony Cheng authored
should only lower dpp clock. Signed-off-by:
Yongqiang Sun <yongqiang.sun@amd.com> Acked-by:
-
Tony Cheng authored
experimental change for testing if max line buffer result in better stutter efficiency for 1080p, LB can hold up to 9 line at 10bpcc, potentially add 10 line time of latency hiding. Signed-off-by:
-
Tony Cheng authored
1. reverts commit e67f51012740 ("dc: temp disable DCC on high res.") - default still DCC enabled 2. add debug options to decide how DCC is disabled - disable DCC - disable DCC if DCC requires 128b (aka. half) request -- observed compressed data corruption result in screen corruption in full (256b) request while half (128b) would cause DCN to hang, result in DF hang Signed-off-by: -
Tony Cheng authored
also refactor debug option. now pipe_split_policy are dynamic = no hack around dcn_calcs. will split based on HW recommendation avoid = avoid split if we can support the config with higher voltage avoid_multi_display = allow split with single display output. force_single_disp_pipe_split force single display to pipe split to improve stutter efficiency by using DET buffers using 2 HUBP. Signed-off-by:
-
Charlene Liu authored
Signed-off-by:
-
Jerry Zuo authored
In SST daisy chain scenario, edid is getting read in mst hotplug routine. It is getting conflict with drm send_enum_path_resources kernel thread in terms of i2c bus which is getting locked up in such case. Have edid being read in get_mode hook, instead of in hotplug routine. Signed-off-by:
-
Yongqiang Sun authored
Signed-off-by:
-
Yongqiang Sun authored
In case of two monitor connected and turn off one of the monitors, OTG0 is locked after graphic plane off due to redundant programming front end regs. Signed-off-by:
-
SivapiriyanKumarasamy authored
Signed-off-by:
SivapiriyanKumarasamy <sivapiriyan.kumarasamy@amd.com> Reviewed-by:
-
Wenjing Liu authored
[Description] link type is not updated before mst topology discovery. This causes issue when branch device response to link address after before the start topology discovery event finishes. [Solution] update link type to mst before topology discovery Signed-off-by:
-
Eric Yang authored
When doing SLS, KMD gives us clipped v_addressable with border. This results in bw validation failure. Signed-off-by:
Eric Yang <Eric.Yang2@amd.com> Reviewed-by:
-
Yongqiang Sun authored
Signed-off-by:
-
Hersen Wu authored
Signed-off-by:
Hersen Wu <hersenxs.wu@amd.com> Signed-off-by:
-
Tony Cheng authored
Signed-off-by:
-
Yongqiang Sun authored
Signed-off-by:
-
Charlene Liu authored
Signed-off-by:
Anthony Koo <Anthony.Koo@amd.com> Acked-by:
-
Charlene Liu authored
Signed-off-by:
-
Ken Chalmers authored
Signed-off-by:
Ken Chalmers <ken.chalmers@amd.com> Reviewed-by:
-
Yue Hin Lau authored
Signed-off-by:
-
Yue Hin Lau authored
Signed-off-by:
Dmytro Laktyushkin <Dmytro.Laktyushkin@amd.com> Acked-by:
-
Eric Bernstein authored
Signed-off-by:
Eric Bernstein <eric.bernstein@amd.com> Reviewed-by:
-
Eric Yang authored
A previous changed removed the hack to match mpcc_idd with mi instance. This causes pstate hang on resume from hibernate for yet unknown reason. Add the hack back for now to work around the issue. More debugging required in init_hw to root cause the hang. Signed-off-by:
-
Wenjing Liu authored
[Description] hbr2 compliance eye output is unstable (toggling on and off) with debugger break. This caueses intermittent PHY automation failure. Need to look into the root cause later Signed-off-by:
-
Yue Hin Lau authored
Signed-off-by:
-
Andrew Jiang authored
Signed-off-by:
-
ShihChen Chen authored
Signed-off-by:
ShihChen Chen <ShihChen.Chen@amd.com> Reviewed-by:
-
Bhawanpreet Lakha authored
Split update_planes_and_stream_state (split Software and Hardware programming) as the state is already build, so we only need to program the hardware Signed-off-by:
Andrey Grodzovsky <Andrey.Grodzovsky@amd.com> Signed-off-by:
Bhawanpreet Lakha <Bhawanpreet.Lakha@amd.com> Reviewed-by:
-
Jerry Zuo authored
Prevent NULL pointer on new_stream being added to ctx when added MST connectors cannot be found in existing crtc_state in the chained mode Signed-off-by:
-
Jerry Zuo authored
Needs effort to take care of the fake sink scenario in downstream daisy chain device. Exclude MST from fake sink feature for now. Signed-off-by:
-
Yue Hin Lau authored
Signed-off-by:
-
Leo (Sunpeng) Li authored
During system suspend, we: 1. Cache a duplicate of the current DRM atomic state, which calls hooks to duplicate our internal states. 2. Call hooks to disable all functionalities. 3. System suspended. During resume, we attempt to restore the cached state. However, our interal states are now stale, since step 1 was done before step 2. i.e. our cached state does not reflect the disabled nature of things. This change resolves that by destroying all relevant states to reflect the actual state during resume. Signed-off-by:
Leo (Sunpeng) Li <sunpeng.li@amd.com> Reviewed-by:
-
Eric Bernstein authored
Signed-off-by:
-
Roman Li authored
- fixed wrong index in dce110_validate_surface_sets() - formatted for better readability Signed-off-by:
-
Yue Hin Lau authored
Signed-off-by:
-
