Re: Sun Leo (ZX)
Robert Reif wrote:
One thing you might want to try is disabling leo_switch_from_graph.
That is a recent code change.
I've not found info on the change yet, but commenting out the couple of
places where it is called doesn't improve things. With extra diagnostics
in all public routines, I see:
Remapping the kernel... done.
Booting Linux...
[MarkMLl] Debugging output to prom_printf()
[MarkMLl] Entered leo_init()
[MarkMLl] In leo_init(), matched option
[MarkMLl] In leo_probe()
[MarkMLl] Entered leo_init_one()
[MarkMLl] In leo_fixup_var_rgb()
[MarkMLl] of_getintprop_default(linebytes, 1152) returned 1152
[MarkMLl] fbsize 1040384
[MarkMLl] In leo_init_wids()
[MarkMLl] In leo_wid_put()
[MarkMLl] In leo_wait()
[MarkMLl] In leo_wid_put()
[MarkMLl] In leo_wait()
[MarkMLl] In leo_wid_put()
[MarkMLl] In leo_wait()
[MarkMLl] In leo_wid_put()
[MarkMLl] In leo_wait()
[MarkMLl] In leo_init_hw()
[MarkMLl] In leo_blank()
[MarkMLl] Enter leo_init_fix()
[MarkMLl] Using strlcpy() to copy 16 bytes from SUNW,leo
[MarkMLl] Done strlcpy()
[MarkMLl] Exit leo_init_fix()
That includes diagnostics before the spinlock in leo_switch_from_graph()
so it's not being called from elsewhere, it was previously being called
from leo_init_hw().
I've not so far managed to compile 2.4 on the system as installed, so I
think I'll revert- for investigative purposes only- to a distro based on
2.4 and see how that runs.
I am walking out the door now on a short holiday and won't be back until
Monday evening so I can't try this myself.
Enjoy your break :-)
--
Mark Morgan Lloyd
markMLl .AT. telemetry.co .DOT. uk
[Opinions above are the author's, not those of his employers or colleagues]
Reply to: