better parsing of sh commands (copied from jos sh)
cat: read from 1 if no args
sbrk system call, but untested
getpid system call
moved locks in keyboard intr, but why do we get intr w. null characters from keyboard?

renumber system calls (run gmake clean!)
skeleton for dup system call

cd in shell
nits in mkdir, ls, etc.

mkdir
ls shows stat info for each dir entry

stat

don't hold fd table lock across idecref() (latter does block i/o)
idecref calls iput() in case last ref -> freeing inode
dir size is 512 * # blocks, so readi/writei &c work
unlink deletes dirent even if ip->nlink > 0

no more cons_put system calls
usertests tests two processes writing files

primitive ls

remove console init code from userfs

working unlink, but doesn't free dir blocks that become empty
remove out-of-date comment in ioapic

sh accepts 0-argument commands (like userfs)
reads from console

fix acquire() to cli() *before* incrementing nlock
make T_SYSCALL a trap gate, not an interrupt gate
sadly, various crashes if you hold down a keyboard key...

convert userfs to use printf
bfree
ifree
writei
start on unlink

O_RDWR, etc.
create file

set fd to writeable on open for write

checkpoint: write(fd,"hello\n",6) where fd is a console dev almost works

give cpu1 a TSS and gdt for when it enters scheduler()
and a pseudo proc[] entry for each cpu
cpu0 waits for each other cpu to start up
read() for files

ioapic

iget
mkfs makes a file system image
put this in your .bochsrc:
ata0-slave: type=disk, mode=flat, path="fs.img", cylinders=1024, heads=1, spt=1

Attempt to clean up newproc somewhat.

Also remove all calls to memcpy in favor of
memmove, which has defined semantics when
the ranges overlap.  The fact that memcpy was
working in console.c to scroll the screen is not
guaranteed by all implementations.

New scheduler.

Removed cli and sti stack in favor of tracking
number of locks held on each CPU and explicit
conditionals in spinlock.c.

Change fetchint, fetcharg, and putint to return -1 on error, 0 on success.
They had been returning 0 on error, 1 on success, but all the callers
were checking for return value < 0.