On Wed, 17 Jan 2001, Ian Chilton wrote: > Determined physical RAM map: > memory: 00001000 @ 00000000 (reserved) > memory: 00001000 @ 00001000 (reserved) > memory: 001d6000 @ 08002000 (reserved) > memory: 00568000 @ 081d8000 (usable) > memory: 000c0000 @ 08740000 (ROM data) > memory: 05800000 @ 08800000 (usable) [...] > Freeing prom memory: 768kb freed > Freeing prom memory: 768kb freed Good -- I was worrying if the code is able to free areas mared as "ROM data". > root:~# cat /proc/iomem > 00000000-00000fff : reserved > 00001000-00001fff : reserved > 08002000-081d7fff : reserved > 08002000-08188297 : Kernel code > 0819c300-081b23bf : Kernel data > 081d8000-0873ffff : System RAM > 08740000-087fffff : System RAM > 08800000-0dffffff : System RAM It looks consistent with what was reported previously -- good. Thanks for the report. From your other report I see the system works with debugging code enabled as well. It's consistent with last Florian's, too. Therefore I'm considering the previous Florian's failure report an accident. I have no idea why swap doesn't work -- I'm actually running NFS-rooted and unless I twiddle with the NBD thingy, I will not be able to test swap anytime soon. The double output looks like a problem with printk. Ralf, I recall you made a few changes related to printk on SGI recently -- could you please look into it? Maciej -- + Maciej W. Rozycki, Technical University of Gdansk, Poland + +--------------------------------------------------------------+ + e-mail: macro@ds2.pg.gda.pl, PGP key available +
