summaryrefslogtreecommitdiff
path: root/BUGS
diff options
context:
space:
mode:
authorrsc <rsc>2006-09-17 19:19:01 +0000
committerrsc <rsc>2006-09-17 19:19:01 +0000
commitc25e0a5ff86739517925f9aa71234d33d87ca7a8 (patch)
tree555d802172a978e095a287412df3bd9f0349edd2 /BUGS
parent05eecb6313b55f0e1ff26faf673066450482aa5c (diff)
downloadxv6-labs-c25e0a5ff86739517925f9aa71234d33d87ca7a8.tar.gz
xv6-labs-c25e0a5ff86739517925f9aa71234d33d87ca7a8.tar.bz2
xv6-labs-c25e0a5ff86739517925f9aa71234d33d87ca7a8.zip
accumulate bugs
Diffstat (limited to 'BUGS')
-rw-r--r--BUGS36
1 files changed, 36 insertions, 0 deletions
diff --git a/BUGS b/BUGS
new file mode 100644
index 0000000..a051750
--- /dev/null
+++ b/BUGS
@@ -0,0 +1,36 @@
+missing bootmain.c from printout
+funny \ alignment in mmu.h
+
+main.c:
+ should use static int bcpu to avoid having live
+ stack variable across the asm volatile that
+ changes %esp and %ebp.
+
+init.c:
+ should while(wait() >= 0);
+ not just wait();
+
+proc.c:
+ comment at top of scheduler() should say
+ via longjmp back to the scheduler
+ not
+ via longjmp back to the top of scheduler
+
+ sched should panic if state == RUNNING.
+
+ forkret comment should say "Return" not "return"
+
+ as a consequence of the implementation of proc_kill,
+ any loop calling sleep should check for p->killed
+ and be able to break out with an error return.
+
+ can swap procdump up after proc_kill
+ and then have proc_exit and proc_wait on same sheet
+
+ proc_exit should wake up proc[1] if any
+ reparenting was done.
+
+trap.c
+ if spin locks are ever held without cli,
+ trap should call yield() only if nlock == 0
+