Buster has End of Life from our perspective for over a year, and hits Debian's end of Long Term Support in 5 days (June 30th, 2024).
The 5.10 kernel hasn't been updated over 2 years.
Update if you want any significant level of support.
Are you sure that you have "dtoverlay=vc4-kms-v3d" in config.txt? Memory says that Buster defaulted to fkms.
The CVT algorithm for timings has limitations on what it can represent, generally not allowing resolutions that aren't a multiple of 8. It is also fairly conservative on timings (ie big blanking periods).
"cvt 1500 256 60" gives me the mode line of "1504x256 30.00 1504 1544 1688 1872 256 259 269 272 -hsync +vsync", so
30MHz clock, 1504x256, hfp 40, hsync 144, hbp 184, vfp 3, vsync 10, vbp 3.
So it doesn't match the timing in your EDID anyway.
If using vc4-kms-v3d on a recent kernel, then the mode you quoted from the EDID should be read by default and used. Ensure that "disable_fw_kms_setup=1" is present in config.txt to avoid the firmware trying to guide the kernel as to the mode to use.
The 5.10 kernel hasn't been updated over 2 years.
Update if you want any significant level of support.
Are you sure that you have "dtoverlay=vc4-kms-v3d" in config.txt? Memory says that Buster defaulted to fkms.
The CVT algorithm for timings has limitations on what it can represent, generally not allowing resolutions that aren't a multiple of 8. It is also fairly conservative on timings (ie big blanking periods).
"cvt 1500 256 60" gives me the mode line of "1504x256 30.00 1504 1544 1688 1872 256 259 269 272 -hsync +vsync", so
30MHz clock, 1504x256, hfp 40, hsync 144, hbp 184, vfp 3, vsync 10, vbp 3.
So it doesn't match the timing in your EDID anyway.
If using vc4-kms-v3d on a recent kernel, then the mode you quoted from the EDID should be read by default and used. Ensure that "disable_fw_kms_setup=1" is present in config.txt to avoid the firmware trying to guide the kernel as to the mode to use.
Statistics: Posted by 6by9 — Tue Jun 25, 2024 1:36 pm