AMD-GFX Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: "Timur Kristóf" <timur.kristof@gmail.com>
To: amd-gfx@lists.freedesktop.org, alexander.deucher@amd.com,
	Alex Hung <alex.hung@amd.com>,
	Harry Wentland <Harry.Wentland@amd.com>,
	Roman Li <Roman.Li@amd.com>, Leo Li <sunpeng.li@amd.com>,
	David Airlie <airlied@gmail.com>,
	Mario Limonciello <mario.limonciello@amd.com>,
	Ivan Lipski <ivan.lipski@amd.com>, Melissa Wen <mwen@igalia.com>
Subject: Re: [PATCH 05/14] drm/amd/display: Set max supported display clock without max_clks_by_state
Date: Thu, 30 Apr 2026 14:28:27 +0200	[thread overview]
Message-ID: <3591364.LZWGnKmheA@timur-max> (raw)
In-Reply-To: <a16bce0b-6d7c-49a8-85c0-77458dd4bacb@igalia.com>

On 2026. április 29., szerda 22:24:38 közép-európai nyári idő Melissa Wen 
wrote:
> On 23/04/2026 16:15, Timur Kristóf wrote:
> > The max_clks_by_state was based on hardcoded values, which are
> > not really used anywhere, only to know the maximum clock.
> > Just hardcode the same maximum clock for each DCE version.
> > 
> > Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
> > ---
> > 
> >   .../amd/display/dc/clk_mgr/dce100/dce_clk_mgr.c  | 16 +++++++++++-----
> >   1 file changed, 11 insertions(+), 5 deletions(-)
> > 
> > diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce100/dce_clk_mgr.c
> > b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce100/dce_clk_mgr.c index
> > 2ba341df7fffd..bef9a72f3382f 100644
> > --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce100/dce_clk_mgr.c
> > +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce100/dce_clk_mgr.c
> > @@ -391,9 +391,7 @@ static void dce_update_clocks(struct clk_mgr
> > *clk_mgr_base,> 
> >   			struct dc_state *context,
> >   			bool safe_to_lower)
> >   
> >   {
> > 
> > -	struct clk_mgr_internal *clk_mgr_dce =
> > TO_CLK_MGR_INTERNAL(clk_mgr_base);
> > -	const int max_disp_clk =
> > -		clk_mgr_dce-
>max_clks_by_state[DM_PP_CLOCKS_STATE_PERFORMANCE].display_
> > clk_khz; +	const int max_disp_clk =
> > clk_mgr_base->clks.max_supported_dispclk_khz;> 
> >   	int patched_disp_clk = MIN(max_disp_clk,
> >   	context->bw_ctx.bw.dce.dispclk_khz);
> >   	
> >   	if (should_set_clock(safe_to_lower, patched_disp_clk,
> >   	clk_mgr_base->clks.dispclk_khz)) {> 
> > @@ -445,8 +443,16 @@ void dce_clk_mgr_construct(
> > 
> >   	clk_mgr->dprefclk_ss_divider = 1000;
> >   	clk_mgr->ss_on_dprefclk = false;
> > 
> > -	base->clks.max_supported_dispclk_khz =
> > -		clk_mgr-
>max_clks_by_state[DM_PP_CLOCKS_STATE_PERFORMANCE].display_clk_
> > khz; +	if (ctx->dce_version >= DCE_VERSION_12_0)
> > +		base->clks.max_supported_dispclk_khz = 1133000;
> > +	else if (ctx->dce_version >= DCE_VERSION_11_2)
> > +		base->clks.max_supported_dispclk_khz = 1108000;
> 
> For DCE 11.2, I see ClocksStatePerformance is 1132000 instead of
> 1108000, right?

Hi Melissa,

For DCE11.2, nobody really knows what the maximum supported display clock is.
There are different values hardcoded in different parts of the code base.

dce112_max_clks_by_state says it's 1132 MHz
bw_calcs says it's 1108 MHz
and dce112_update_clocks() adds 15%

In this patch, I chose to go for 1108 MHz to match bw_calcs, but I can edit 
that if you feel that 1132 MHz is better. What do you think?

Thanks,
Timur


> 
> With the value fixed, this is:
> 
> Reviewed-by: Melissa Wen <mwen@igalia.com>
> 
> > +	else if (ctx->dce_version >= DCE_VERSION_11_0)
> > +		base->clks.max_supported_dispclk_khz = 643000;
> > +	else if (ctx->dce_version >= DCE_VERSION_8_0)
> > +		base->clks.max_supported_dispclk_khz = 625000;
> > +	else
> > +		base->clks.max_supported_dispclk_khz = 600000;
> > 
> >   	dce_clock_read_integrated_info(clk_mgr);
> >   	dce_clock_read_ss_info(clk_mgr);





  reply	other threads:[~2026-04-30 13:19 UTC|newest]

Thread overview: 37+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-04-23 19:15 [PATCH 00/14] drm/amd: Delete defunct DAL power level code Timur Kristóf
2026-04-23 19:15 ` [PATCH 01/14] drm/amd/display: Delete unimplemented dm_pp_apply_power_level_change_request() Timur Kristóf
2026-04-24 14:19   ` Melissa Wen
2026-04-24 16:27     ` Timur Kristóf
2026-04-29 20:02   ` Melissa Wen
2026-04-23 19:15 ` [PATCH 02/14] drm/amd/display: Delete dce_get_required_clocks_state() Timur Kristóf
2026-04-29 20:03   ` Melissa Wen
2026-04-23 19:15 ` [PATCH 03/14] drm/amd/display: Remove min/max clock levels from clk_mgr Timur Kristóf
2026-04-24 14:21   ` Melissa Wen
2026-04-24 16:28     ` Timur Kristóf
2026-04-29 20:05       ` Melissa Wen
2026-04-23 19:15 ` [PATCH 04/14] drm/amd/display: Delete max_clocks_state from dm_pp_static_clock_info Timur Kristóf
2026-04-29 20:06   ` Melissa Wen
2026-04-23 19:15 ` [PATCH 05/14] drm/amd/display: Set max supported display clock without max_clks_by_state Timur Kristóf
2026-04-29 20:24   ` Melissa Wen
2026-04-30 12:28     ` Timur Kristóf [this message]
2026-04-30 18:06       ` Melissa Wen
2026-04-23 19:15 ` [PATCH 06/14] drm/amd/display: Delete max_clks_by_state from DCE clock manager Timur Kristóf
2026-04-29 20:29   ` Melissa Wen
2026-04-23 19:15 ` [PATCH 07/14] drm/amd/display: Delete disp_clk_voltage from integrated info Timur Kristóf
2026-04-29 20:35   ` Melissa Wen
2026-04-30 12:30     ` Timur Kristóf
2026-04-23 19:15 ` [PATCH 08/14] drm/amd/display: Delete dm_pp_clocks_state Timur Kristóf
2026-04-29 20:37   ` Melissa Wen
2026-04-30 12:31     ` Timur Kristóf
2026-04-23 19:15 ` [PATCH 09/14] drm/amd/pm: Delete unused get_display_power_level() function Timur Kristóf
2026-04-30 17:56   ` Melissa Wen
2026-04-23 19:15 ` [PATCH 10/14] drm/amd/pm: Delete dummy get_dal_power_level implementations Timur Kristóf
2026-04-30 18:42   ` Melissa Wen
2026-04-23 19:15 ` [PATCH 11/14] drm/amd/pm: Delete non-functional SMU8 get_dal_power_level implementation Timur Kristóf
2026-04-30 18:53   ` Melissa Wen
2026-04-23 19:15 ` [PATCH 12/14] drm/amd/pm: Delete vddc_dep_on_dal_pwrl Timur Kristóf
2026-04-30 19:03   ` Melissa Wen
2026-04-23 19:15 ` [PATCH 13/14] drm/amd/pm: Delete get_dal_power_level Timur Kristóf
2026-04-30 19:03   ` Melissa Wen
2026-04-23 19:15 ` [PATCH 14/14] drm/amd/pm: Delete PP_DAL_POWERLEVEL Timur Kristóf
2026-04-30 19:04   ` Melissa Wen

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=3591364.LZWGnKmheA@timur-max \
    --to=timur.kristof@gmail.com \
    --cc=Harry.Wentland@amd.com \
    --cc=Roman.Li@amd.com \
    --cc=airlied@gmail.com \
    --cc=alex.hung@amd.com \
    --cc=alexander.deucher@amd.com \
    --cc=amd-gfx@lists.freedesktop.org \
    --cc=ivan.lipski@amd.com \
    --cc=mario.limonciello@amd.com \
    --cc=mwen@igalia.com \
    --cc=sunpeng.li@amd.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox