From: Stefan N. <neu...@tr...> - 2005-07-10 11:17:17
|
Daniel Sobe <har...@se...> writes: > Am Samstag, den 02.07.2005, 18:11 +0200 schrieb Stefan Neuwirth: >> Just a guess screenbuf is vc->vc_screenbuf, but then you get: >> >> CC drivers/char/vt.o >> drivers/char/vt.c: In function `take_over_console': >> drivers/char/vt.c:1912: error: structure has no member named `vc_size_rows' >> make[2]: *** [drivers/char/vt.o] Error 1 >> make[1]: *** [drivers/char] Error 2 >> make: *** [drivers] Error 2 >> >> Changing vc_size_rows to vc_size_row makes things compilable, don't >> know if it working with the changes... > > I had to fix more things (checked out CVS today) to have the kernel > compile. I found that I had to change "pci_name_bus(name, bus)" to > "pci_proc_domain(bus)" in drivers/pci/proc.c, add a forward declaration > of "struct input_handle *input_find_handle(char *phys_descr);" to > drivers/char/vt_proc.c and to change every occurence of "vt_kmalloced" > to "kmalloced". > > Still, I get an error very early when booting. I cannot capture the full > output; the trace looks like this: > > ret_from_fork > init > spawn_ksoftirqd > > The diff between 2.6.12.2 and the CVS is attached including the compile > fixes. Does anybody have an idea of what is wrong? I recently switched to kernel 2.6.13-rc1-mm1-ruby (compiles an runs without any problem). Due to my very special needs this kernel is the best kernel I've seen in 2.6.x. But still a lot of problems, for example matroxfb and framebuffer console. Stefan |