On Wed, 06 May 2015, "Jindal, Sonika" <sonika.jindal@xxxxxxxxx> wrote: > On 5/5/2015 7:28 PM, Jani Nikula wrote: >> On Tue, 05 May 2015, Damien Lespiau <damien.lespiau@xxxxxxxxx> wrote: >>> On Tue, May 05, 2015 at 02:39:48PM +0300, Jani Nikula wrote: >>>> There's still the question whether we can default to using the vbt value >>>> if that can be broken... >>> >>> This is a workaround for SDPs loaded with a "default" VBT which doesn't >>> correspond to the full fleet of SPDs. For actual product, it's really >>> expected to have a VBT with a correct low vswing bit set, otherwise the >>> panel won't lit up (even on Windows!). >>> >>> It's probably a good idea to add a comment saying that's it's a >>> workaround for early platform with generic-ish VBT. Something that we >>> should remove in a bit. >> >> Thanks for the clarification, Damien. That settles the default then. I'd >> still like the module parameter to be effective even if changed runtime. >> > This is used while preparing ddi buffers to program ddi_buf_trans during > driver load. It will not really have any effect at runtime until power > well is disabled/re-enabled. Right, missed that part, sorry. > PLease let me know if you still see the need of moving the check > directly in prepare_ddi_buffers. No, it would be confusing. And I guess we'll remove this parameter afterwards anyway? Please just change the module parameter permissions to read only, so it's not possible to change it dynamically via sysfs. BR, Jani. > > Regards, > Sonika >> BR, >> Jani. >> >> >> >>> >>> -- >>> Damien >> -- Jani Nikula, Intel Open Source Technology Center _______________________________________________ Intel-gfx mailing list Intel-gfx@xxxxxxxxxxxxxxxxxxxxx http://lists.freedesktop.org/mailman/listinfo/intel-gfx
