[src/sommerfeld_i386mp_1]: src/sys/arch/i386 update

[sommerfeld <sommerfeld%NetBSD.org@localhost>]

details:   https://anonhg.NetBSD.org/src/rev/770c776f0cfc
branches:  sommerfeld_i386mp_1
changeset: 482485:770c776f0cfc
user:      sommerfeld <sommerfeld%NetBSD.org@localhost>
date:      Wed Jan 30 14:55:05 2002 +0000

description:
update

diffstat:

 sys/arch/i386/MP-TODO |  21 ++++++++++++++++-----
 1 files changed, 16 insertions(+), 5 deletions(-)

diffs (49 lines):

diff -r b0bc9bbb2e83 -r 770c776f0cfc sys/arch/i386/MP-TODO
--- a/sys/arch/i386/MP-TODO     Mon Jan 28 04:21:35 2002 +0000
+++ b/sys/arch/i386/MP-TODO     Wed Jan 30 14:55:05 2002 +0000
@@ -5,11 +5,6 @@
  - change splraise/spllower to not touch local apic, since that's very
 slow (requires cheap per-cpu storage).
 
- - per-cpu GDT and MP-safing of gdt.c.  this seems like severe
-overkill so just one descriptor is different.  While we're in there,
-we *might* want to use per-cpu, more ephemeral, descriptors for TSS
-and LDT.
-
  - quad cpu ppro sporadically jumps into the stack; need to diagnose
 this.
 
@@ -23,6 +18,10 @@
 
  - careful code review of all diffs from -current
 
+ - try to render APM *harmless* (and functional in uniprocessor
+configurations).  it can't possibly work correctly when multiple cpu's
+are running, but..
+
 "would be nice":
 
  - use atomic ops to modify ci_flags and cpu_feature to avoid losing bits.
@@ -38,6 +37,12 @@
  - turn off rdtsc-based microtime on uniprocessor systems which might
 have variable-speed TSC.
 
+ - investigate using per-cpu, more ephemeral, descriptors for TSS and
+LDT.
+
+ - make halt -p work in MULTIPROCESSOR config (the one APM feature
+everyone wants)
+
 -----
 
 DONE:
@@ -71,3 +76,9 @@
 [2001/05/26  20:03:41  sommerfeld]
 answer: microtime was just doing the wrong thing.  fixed.
 
+[2002/01/30  09:52:17  sommerfeld]
+done a while ago:
+
+ - per-cpu GDT and MP-safing of gdt.c.  this seems like severe
+overkill so just one descriptor is different.  
+