SMP in 2.2: linux/Documentation/SMP.txt

Tigran Aivazian (
Sat, 23 May 1998 16:45:13 -0400 (EDT)

Perhaps a patch is in order, I'm not sure if this is all correct. Someone
who knows more will want to revise this further, I just gutted out all the
lines I think are fixed. Perhaps the whole document needs to be nuked.

Gregory Maxwell

--- ./old/linux/Documentation/SMP.txt Sat May 23 17:24:55 1998
+++ ./linux/Documentation/SMP.txt Sat May 23 17:34:15 1998
@@ -1,25 +1,22 @@
SMP support for Linux with up to 16 processors using the Intel MP

- This is experimental. Back up your disks first. Experience is that
-it is basically stable in its current (inefficient form).
+SMP on x86/Linux is now an official feature and is not exprtimental.
+Experimental SMP support for other architectures is underway.
+Please view linux/Documentation/smp for more information about enabling

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 Clean up message pass.
-o Test for B stepping processors.
-o Clean up processor specific/independent split.
o Document it all. [PARTLY DONE]
o Halt other CPU's on reset/panic doesn't always work.
o Don't waste page at 4K - don't need it now.(watch the GDT code).
-o Dump bootup pages once booted somehow.
o Clean up warnings/volatiles.
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 Test IO-APIC on various hardware.
+o Seperate IO-APIC from SMP.

To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to