I remember some strangeness about the blnclegdisbl. I'll see if I can dig up some more.
-----Original Message----- From: Ville Syrjälä <ville.syrj...@linux.intel.com> Sent: Monday, August 24, 2020 11:05 AM To: Kai-Heng Feng <kai.heng.f...@canonical.com> Cc: Runyan, Arthur J <arthur.j.run...@intel.com>; Vivi, Rodrigo <rodrigo.v...@intel.com>; intel-gfx <intel-gfx@lists.freedesktop.org> Subject: Re: [Regression] "drm/i915: Implement display w/a #1143" breaks HDMI on ASUS GL552VW On Mon, Aug 17, 2020 at 02:17:49PM +0800, Kai-Heng Feng wrote: > > > > On Aug 17, 2020, at 00:22, Runyan, Arthur J <arthur.j.run...@intel.com> > > wrote: > > > > You'll need to read out the DDI_BUF_TRANS_* and DISPIO_CR_TX_BMU_CR0 > > registers at boot before i915 programs them and compare with what driver > > programs. > > Rodrigo can probably show you how. > > Right, I'll wait for a patch then :) To grab the BIOS reg values we just have to make sure the driver doesn't load. Eg. pass something like "modprobe.blacklist=i915,snd_hda_intel 3" to the kernel cmdline (+ whatever other magic ubuntu might require). Confirm with something like "lsmod | grep i915" to make sure the driver didn't sneak in despite our best efforts. Then we can dump the registers with intel_reg from igt-gpu-tools: intel_reg read --count 20 0x64E00 0x64E60 0x64EC0 0x64F20 0x64F80 intel_reg read 0x64000 0x64100 0x64200 0x64300 0x64400 0x6C00C The only somewhat suspicious thing I noticed is that we treat DISPIO_CR_TX_BMU_CR0:tx_blnclegdisbl as a bitmask (bit 23 -> DDI A, bit 24 -> DDI B, etc.) whereas the spec seems to be saying that we should just zero out all the bits of tx_blnclegdisbl when any DDI needs iboost. Art, is our interpretation of the bits correct or just a fairy tale? > > Kai-Heng > > > > > -----Original Message----- > > From: Kai-Heng Feng <kai.heng.f...@canonical.com> > > Sent: Thursday, August 13, 2020 10:14 PM > > To: Runyan, Arthur J <arthur.j.run...@intel.com> > > Cc: Vivi, Rodrigo <rodrigo.v...@intel.com>; Ville Syrjälä > > <ville.syrj...@linux.intel.com>; intel-gfx > > <intel-gfx@lists.freedesktop.org> > > Subject: Re: [Regression] "drm/i915: Implement display w/a #1143" > > breaks HDMI on ASUS GL552VW > > > > Hi, > > > >> On Aug 14, 2020, at 01:56, Runyan, Arthur J <arthur.j.run...@intel.com> > >> wrote: > >> > >> The workaround is freeing up stuck vswing values to let new vswing > >> programming kick in. Maybe the new vswing values are wrong. > >> Try checking the vswing that driver programs against what BIOS/GOP > >> programs. > > > > Do you mean to print out value of I915_READ()? > > val = I915_READ(CHICKEN_TRANS(transcoder)); > > > > Kai-Heng > > > >> > >> -----Original Message----- > >> From: Vivi, Rodrigo <rodrigo.v...@intel.com> > >> Sent: Thursday, August 13, 2020 9:50 AM > >> To: Kai-Heng Feng <kai.heng.f...@canonical.com>; Runyan, Arthur J > >> <arthur.j.run...@intel.com> > >> Cc: Ville Syrjälä <ville.syrj...@linux.intel.com>; intel-gfx > >> <intel-gfx@lists.freedesktop.org> > >> Subject: Re: [Regression] "drm/i915: Implement display w/a #1143" > >> breaks HDMI on ASUS GL552VW > >> > >> Art, any comment here? > >> > >> I just checked and the W/a 1143 is implemented as described, but it is > >> failing HDMI on this hybrid system. > >> > >>> On Aug 12, 2020, at 9:07 PM, Kai-Heng Feng <kai.heng.f...@canonical.com> > >>> wrote: > >>> > >>> Hi, > >>> > >>> There's a regression reported that HDMI output stops working after os > >>> upgrade: > >>> https://bugs.launchpad.net/bugs/1871721 > >>> > >>> Here's the bisect result: > >>> 0519c102f5285476d7868a387bdb6c58385e4074 is the first bad commit > >>> commit 0519c102f5285476d7868a387bdb6c58385e4074 > >>> Author: Ville Syrjälä <ville.syrj...@linux.intel.com> > >>> Date: Mon Jan 22 19:41:31 2018 +0200 > >>> > >>> drm/i915: Implement display w/a #1143 > >>> > >>> Apparently SKL/KBL/CFL need some manual help to get the > >>> programmed HDMI vswing to stick. Implement the relevant > >>> workaround (display w/a #1143). > >>> > >>> Note that the relevant chicken bits live in a transcoder register > >>> even though the bits affect a specific DDI port rather than a > >>> specific transcoder. Hence we must pick the correct transcoder > >>> register instance based on the port rather than based on the > >>> cpu_transcoder. > >>> > >>> Also note that for completeness I included support for DDI A/E > >>> in the code even though we never have HDMI on those ports. > >>> > >>> v2: CFL needs the w/a as well (Rodrigo and Art) > >>> > >>> Cc: Rodrigo Vivi <rodrigo.v...@intel.com> > >>> Cc: Art Runyan <arthur.j.run...@intel.com> > >>> Signed-off-by: Ville Syrjälä <ville.syrj...@linux.intel.com> > >>> Link: > >>> https://patchwork.freedesktop.org/patch/msgid/20180122174131.28046 > >>> -1-ville.syrj...@linux.intel.com > >>> Reviewed-by: Rodrigo Vivi <rodrigo.v...@intel.com> > >>> > >>> > >>> dmesg from drm-tip with drm.debug=0xe can be found here: > >>> https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1871721/comme > >>> nts > >>> / > >>> 64 > >>> > >>> Kai-Heng > >> > >> > > -- Ville Syrjälä Intel _______________________________________________ Intel-gfx mailing list Intel-gfx@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/intel-gfx
