[PATCH 4/5] drm/i915/dram: bypass fsb/mem freq detection on dg2 and no display

Rodrigo Vivi rodrigo.vivi at intel.com
Fri Aug 15 15:43:03 UTC 2025


On Wed, Aug 06, 2025 at 04:52:06PM +0300, Jani Nikula wrote:
> On Tue, 05 Aug 2025, Rodrigo Vivi <rodrigo.vivi at intel.com> wrote:
> > On Thu, Jul 31, 2025 at 05:21:24PM +0300, Jani Nikula wrote:
> >> Non-display now calls the intel_fsb_freq() and intel_mem_freq()
> >> functions, so we don't have to have the frequencies initialized for dg2
> >> or non-display cases.
> 
> Is this enough amendmend for the commit message:
> 
> """
> This is in preparation for unifying the pre-gen9 handling in dram info.
> 
> DG2 remains a special case as described in commit 5eb6bf0b44e7
> ("drm/i915/dg2: Don't read DRAM info").
> """
> 
> Or do you want more?

this is great, with that

Reviewed-by: Rodrigo Vivi <rodrigo.vivi at intel.com>

> 
> BR,
> Jani.
> 
> >> 
> >> Signed-off-by: Jani Nikula <jani.nikula at intel.com>
> >> ---
> >>  drivers/gpu/drm/i915/soc/intel_dram.c | 5 ++++-
> >>  1 file changed, 4 insertions(+), 1 deletion(-)
> >> 
> >> diff --git a/drivers/gpu/drm/i915/soc/intel_dram.c b/drivers/gpu/drm/i915/soc/intel_dram.c
> >> index 193e7f71a356..d896fb67270f 100644
> >> --- a/drivers/gpu/drm/i915/soc/intel_dram.c
> >> +++ b/drivers/gpu/drm/i915/soc/intel_dram.c
> >> @@ -725,10 +725,13 @@ int intel_dram_detect(struct drm_i915_private *i915)
> >>  	struct dram_info *dram_info;
> >>  	int ret;
> >>  
> >> +	if (IS_DG2(i915) || !HAS_DISPLAY(i915))
> >> +		return 0;
> >> +
> >>  	detect_fsb_freq(i915);
> >>  	detect_mem_freq(i915);
> >
> > but they will only be set to zero no? do we really need to avoid it?
> > if so, perhaps make this change earlier?
> >
> > Also I wonder what's special in DG2, but not in BMG...
> >
> >>  
> >> -	if (GRAPHICS_VER(i915) < 9 || IS_DG2(i915) || !HAS_DISPLAY(i915))
> >> +	if (GRAPHICS_VER(i915) < 9)
> >
> > what about the old gen remaining here? at least deserves a comment on why
> > it needs the upper calls, but not the call bellow? or should the upper
> > calls be in another function/flow?
> >
> >>  		return 0;
> >>  
> >>  	dram_info = drmm_kzalloc(&i915->drm, sizeof(*dram_info), GFP_KERNEL);
> >> -- 
> >> 2.39.5
> >> 
> 
> -- 
> Jani Nikula, Intel


More information about the Intel-xe mailing list