Re: kern/59605: boot radeondrmkmsfb0: no data for est. mode 640x480x67

["RVP via gnats" <gnats-admin%NetBSD.org@localhost>]

The following reply was made to PR kern/59605; it has been noted by GNATS.

From: RVP <rvp%SDF.ORG@localhost>
To: gnats-bugs%netbsd.org@localhost
Cc: 
Subject: Re: kern/59605: boot radeondrmkmsfb0: no data for est. mode
 640x480x67
Date: Fri, 29 Aug 2025 06:27:33 +0000 (UTC)

 On Mon, 25 Aug 2025, Michael L. Riechers via gnats wrote:
 
 > No overwrite, the /etc/ttys file came directly from my Netbsd 9.2 /etc.  It
 > had all ttyE.. turned off.
 >
 
 :)
 
 > I turned ttyE[1-3] on, and now it works. ttyE0 still slow and
 > sluggish, perhaps because of a VT100 emulation.
 >
 
 This is a userspace issue, not a slowdown in the kernel. Real terminals, like
 the VT100, need delays for certain operations--like reverse-scroll used by
 page-up in less(1):
 
 ```
 $ infocmp vt100 | fgrep ri=
          rev=\E[7m$<2>, ri=\EM$<5>, rmacs=^O, rmam=\E[?7l, rmkx=\E[?1l\E>,
 $
 ```
 
 These delays can be implemented either by sending NULs or by sleeping. In
 NetBSD 11-BETA + the less there, it's by sleeping for 5 ms for practically
 every line on the screen. And, since NetBSD is not currently tickless, the
 5 ms is rounded up to 1/HZ (10ms).
 
 Set `constty' to `off' and `ttyE0' to `on'. This is what I routinely do on
 any fresh install. TERM=wsvt25 doesn't need any delays.
 
 > Yes, I got that far trying to track this thing down, but can't make
 > heads or tails of who's calling for 640x480x67. 640x480x67 seems to be
 > wrong, but where, why?
 >
 
 You should be able to see this in the EDID data returned by your display.
 
 ```
 # wsconsctl -d edid
 ```
 
 HTH,
 
 -RVP