Hi, On Sat, Nov 11, 2023 at 5:10 AM Doug Anderson <dianders@xxxxxxxxxx> wrote: > > Hi, > > On Fri, Nov 10, 2023 at 1:46 AM Cong Yang > <yangcong5@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxx> wrote: > > > > The refresh reported by modotest is 60.46Hz, and the actual measurement > > s/modotets/modetest/ > > > is 60.01Hz, which is outside the expected tolerance. Adjust hporch and > > pixel clock to fix it. After repair, modetest and actual measurement were > > all 60.01Hz. > > Can you explain this more? Why was the rate that modetest reported > different from the actual measured rate? This feels like it's a > problem with your MIPI controller not being able to accurately make > the rate. Is that it? modetest refresh = Pixel CLK/ htotal * vtotal measurement HS->LP cycle time = Vblanking According to the vendor's feedback, the actual measured frame rate is not only affected by Htotal/Vtotal/pixel clock, Lane-num/PixelBit/LineTime also affected. It seems that if change to a different SOC platform, may need to readjust these parameters. > > If so then this is a bit of a hack. Someone else using the same panel > might have a MIPI controller that can make slightly different clock > rates. I think you're currently the only user of the panel, so maybe > this isn't too terrible (would love to hear other people's advice). > > Assuming this is actually the problem there are probably at least > several different ways to solve this. One that comes to mind is the > solution we ended up with for eDP where we allowed specifying some of > this stuff in the device tree, though that might cause a whole pile of > debates... > > In any case, as I said above this patch is probably OK if you're the > only user of this panel, but it might be at least good to add > something to the commit message? OK, I'll add some instructions in V2. Thanks. > > -Doug
