Re: moj komputer

Autor: Lech Szychowski (lech7_at_pse.pl)
Data: Thu 05 Oct 1995 - 11:16:21 MET


Tomasz Finke (tom_at_sprocket.ict.pwr.wroc.pl) wrote:

: : Oracle spod $CO :)

: Wlasnie. Ale jak chodzi? Czy ktos to naprawde sprawdzil, czy wszystkim
: sie tylko wydaje, bo chodza pod iBCS inne rzeczy? Jesli wydajnosc

Trzeba wziac Slack 3.0 i sprawdzic. Sam bym sprawdzil, ale mam pod reka
Oracle jedynie na AIX'a :)

: wolalbym uzywac Linuxa, niz SCO. Ale pod Linux'em nie ma np. mozliwosci
: skorzystania z maszyn wieloprocesorowych, nie slyszalem tez o obsludze

Cytuje Documentation/SMP.txt:

===================================================================
SMP support for Linux with up to 32 processors using the Intel MP
specification.

WARNING:
        This is experimental. Back up your disks first. Build only
with gcc2.5.8.

To fix:

o Fix sys_idle to exit/enter kernel state and do hlt's.
o Fix scheduler decisions to reschedule. Per cpu reschedule ?
o Scheduler ignores stick to CPU advantage. Critical for P6!
o FPU state save/restore - fault any process on FPU and do a
        restore on the fault each context switch, do a save if we
        faulted that run of the task when we switch away. [DONE]
o Clean up message pass.
o Test for B stepping processors.
o Clean up processor specific/independant split.
o Document it all. [PARTLY DONE]
o Find the exception/crash bug.
o Halt other CPU's on reset/panic doesn't always work.
o Dont waste page at 4K - dont need it now.(watch the GDT code).
o Dump bootup pages once booted somehow.
o Verify message pass safe for >2 CPU's - should be now it
        is atomic lock based.
o Dont schedule switches between idle tasks (if current->pid==0 &&
        next->pid==0 && next!=current) [IN]
o Clean up warnings/volatiles.
o Send CTRL-ALT-DEL to pid 1 not task[1]! [IN]
o Fix load_TR() for non contiguous processor ids
o Iterate over the slave timer requests if one is lost (keep a count
        per cpu)
o Distribute irq's (locking present just needs the 82489 to be asked
        nicely).
o 486 startup code.
o How to handle mixed FPU/non FPU processors.
o Support 4Mb page mode again
===========================================================================

A wiec wypada jescze troszke poczekac - ale jakies 2 tygodnie temu Linus sam
napisal, ze sie za to bierze.

        Leszek.

+============================================================================+
| lech7_at_pse.pl 2:480/33.7 - REAL programmers use INTEGERS - |
| My personal opinions are just mine. |
| If there is anyone I speak for, then most likely it's me. |
+============================================================================+



To archiwum zostało wygenerowane przez hypermail 2.1.7 : Tue 18 May 2004 - 12:25:32 MET DST