[Prev][Next][Index][Thread]
Re: x86 minimum processor
> Date: Wed, 20 Dec 2000 12:05:33 +1100 (EST)
> From: Matthew Palmer <mjp16@ieee.uow.edu.au>
> To: Mike Hibler <mike@fast.cs.utah.edu>
> Subject: Re: x86 minimum processor
>
> 0x113a9b <oskit_base_irq_handler+11>: rdtsc
>
> Excellent guess on your part. Time to update the documentation to recommend
> not using a 486 (above and beyond the usual "they're slow, don't use them"
> admonishment, of course).
>
Doh! We have left in some performance monitoring code. It has since been
pulled out. I've included a diff vs. the "Labor Day" (09/01/2000) release.
Don't know if this fixes all problems though. I'm trying to scare up a
working 486 box to test on...
We should be making a new release RSN as well.
Making the world safe for 486s,
Mike
-----------
Index: base_irq_inittab.S
===================================================================
RCS file: /n/fast/usr/lsrc/flux/CVS/oskit/kern/x86/pc/base_irq_inittab.S,v
retrieving revision 1.24
retrieving revision 1.25
diff -u -r1.24 -r1.25
--- base_irq_inittab.S 2000/08/22 16:46:36 1.24
+++ base_irq_inittab.S 2000/09/11 15:16:19 1.25
@@ -25,18 +25,6 @@
#define TS_EIP 56
#define TS_CS 60
- .data
- .globl EXT(interrupt_start_high)
-LEXT(interrupt_start_high)
- .long 0
- .globl EXT(interrupt_start_low)
-LEXT(interrupt_start_low)
- .long 0
-tmp1:
- .long 0
-tmp2:
- .long 0
-
.text
/*
@@ -50,13 +38,6 @@
GATE_ENTRY(BASE_IRQ_MASTER_BASE + (num), 0f, ACC_PL_K|ACC_INTR_GATE) ;\
P2ALIGN(TEXT_ALIGN) ;\
0: ;\
- movl %eax,tmp1 ;\
- movl %edx,tmp2 ;\
- rdtsc ;\
- movl %eax,EXT(interrupt_start_low) ;\
- movl %edx,EXT(interrupt_start_high) ;\
- movl tmp1,%eax ;\
- movl tmp2,%edx ;\
pushl $(irq) /* error code = irq vector */ ;\
pushl $BASE_IRQ_MASTER_BASE + (num) /* trap number */ ;\
pusha /* save general registers */ ;\
Follow-Ups: