This website requires JavaScript.
34295f461a
group locks into structs they protect. few naming nits.
rsc
2009-05-31 05:12:21 +0000
949e55902b
usys.S: put before init.c, STUB -> SYSCALL
rsc
2009-05-31 02:14:04 +0000
90bab90832
exec sanity check
rsc
2009-05-31 02:11:27 +0000
f3685aa391
simplify
rsc
2009-05-31 02:07:51 +0000
7f399ccaa4
use ftable like btable and itable
rsc
2009-05-31 02:07:26 +0000
ce72cadbe0
fix bug fix
rsc
2009-05-31 01:53:08 +0000
13ae8808c4
tidy fs.c; bmap callers always expected allocation
rsc
2009-05-31 01:44:20 +0000
f12551b599
inode reuse bug.
rsc
2009-05-31 01:34:46 +0000
0badeaa29f
bio.c: use struct like icache does
rsc
2009-05-31 01:29:17 +0000
c47bc4fd54
ide.c: make names more regular
rsc
2009-05-31 01:29:01 +0000
7b644318dd
clean up %fs %gs use
rsc
2009-05-31 01:12:08 +0000
e97519a6d2
sync with c; .text is implied
rsc
2009-05-31 01:00:38 +0000
ba6cd8a685
drop NADDRS and INDIRECT; too many names
rsc
2009-05-31 00:59:37 +0000
030a47736f
tab police
rsc
2009-05-31 00:39:17 +0000
215738336a
move fork into proc.c
rsc
2009-05-31 00:38:51 +0000
dae9b0d48a
missed this file in last checkin
rsc
2009-05-31 00:31:20 +0000
19333efb9e
Some proc cleanup, moving some of copyproc into allocproc.
rsc
2009-05-31 00:28:45 +0000
0c7f483838
move T_DIR etc into stat.h; move type up
rsc
2009-05-31 00:26:10 +0000
74afa70d30
Add serial port input/output. Delete parallel port output. Works well with qemu -nographic mode.
rsc
2009-05-31 00:24:11 +0000
0ca9ca0c55
slight rearrangement; sleep on b not buf.
rsc
2009-05-31 00:23:00 +0000
d827c8f3c8
Add MIT license file; update xv6-rev2.tar.gz with LICENSE. Other files unchanged.
rsc
2009-05-26 16:58:33 +0000
2157576107
be consistent: no underscores in function names
rsc
2009-03-08 22:07:13 +0000
b7f653dc49
xv6: boot loader adjustments
rsc
2009-03-08 21:41:30 +0000
8220135362
xv6: use stosb for memset
rsc
2009-03-08 21:38:30 +0000
c396d065d3
xv6/x86.h: add stosb, fix bugs in insl/outsl (rep not repne)
rsc
2009-03-08 21:27:57 +0000
4003e9bed8
xv6/x86.h: inline assembly cleanup
rsc
2009-03-08 20:56:38 +0000
8eb20827c2
Explain why itrunc() is OK w.r.t. crashes.
rtm
2008-10-20 18:05:48 +0000
15a2693637
try harder to get directory refcounts right
kolya
2008-10-17 12:42:13 +0000
f3e87bc838
make mkdir crash-safer, as noticed by many students on midterm
kolya
2008-10-16 15:18:49 +0000
e1626709d6
free the indirect block
kolya
2008-10-16 15:00:23 +0000
e670a48358
as rtm noticed, avoid doubly-freeing pipe page
kolya
2008-10-15 17:42:56 +0000
deca9fef83
indent
kolya
2008-10-15 05:15:32 +0000
c100d9ee2d
cleaner swtch.S
kolya
2008-10-15 05:14:10 +0000
228e500a0c
save cpus.intena in sched(), so we get the right EFLAGS.IF value once a timer-preempted kernel thread resumes execution in trap() after yield(). otherwise the kernel could get an arbitrary number of nested timer intrs.
kolya
2008-10-15 05:01:39 +0000
15ce79de14
check cp->killed before returning to user from a timer interrupt
kolya
2008-10-15 04:57:02 +0000
00ce31c09c
update .cvsignore
kolya
2008-10-15 04:54:24 +0000
289bbef733
minor cleanup
kolya
2008-10-12 20:19:16 +0000
f73b81205e
minor cleanup with STUB macro
kolya
2008-10-12 19:54:11 +0000
c780dbf967
include explicitly initialized globals (int x = 0;) in cross-refs, also thanks to greg price.
kolya
2008-10-12 18:45:30 +0000
6186836d28
fix: when two names only differed in case, crossrefs only included one of them. thanks to greg price.
kolya
2008-10-12 18:33:14 +0000
2ce40d7088
be clear what the root i-number is
rtm
2008-10-08 18:57:13 +0000
be38c8413e
document lock->locked=0 vs xchg(&lock->locked, 0)
rtm
2008-09-28 10:53:54 +0000
0159c8bb94
trap 9 does not push ec: fix from Greg Price <price@mit.edu>
kolya
2008-09-24 18:20:56 +0000
c7317d4dc7
always save and restore %fs, %gs to ensure old segment entries are never accessible to user from the hidden CPU segment registers.
kolya
2008-09-24 01:48:31 +0000
adcd16c3f7
restore std toolprefix
rtm
2008-09-11 10:22:01 +0000
4651d04ad1
omit *.d from tar file
rtm
2008-09-11 10:20:40 +0000
e9ae6f5cbe
add copyright notice
kaashoek
2008-09-09 11:42:44 +0000
f68317533e
make bochsrc work for bochs 2.2.6
kolya
2008-09-03 14:10:47 +0000
35a6cf84ec
make pdf, ps, tarball
kolya
2008-09-03 14:05:52 +0000
e87dca5cc5
nits in index.txt add slides for shell, x86 intro, x86 virtual memory (deleted JOS from slides)
kaashoek
2008-09-03 12:29:44 +0000
f53494c28e
DO NOT MAIL: xv6 web pages
rsc
2008-09-03 04:50:04 +0000
ee3f75f229
simplify growproc
rtm
2008-08-28 17:57:47 +0000
98754d687e
avoid a bug w/ exit()
rtm
2008-08-28 00:53:24 +0000
56082468ea
the old explanation of AP startup might have been correct, but I understand this one.
rtm
2008-08-28 00:52:05 +0000
02cc595f28
clean up circular buffers, so pipe can queue 512 bytes rather than 511
kolya
2008-08-22 00:26:22 +0000
5c5470a2fa
fix obvious printf nits after reading through code
kolya
2008-08-21 23:24:02 +0000
1808b2f1b3
now rev2
rsc
2008-08-20 18:00:35 +0000
150785a1fd
add nice font
rsc
2008-08-20 18:00:24 +0000
9b62657ecd
formatting updates
rsc
2008-08-20 18:00:13 +0000
5b7f8cbe7c
bochs faster
rsc
2008-08-20 17:59:29 +0000
174729fc55
xv6: latest (as of January 2008)
rsc
2008-08-20 17:46:32 +0000
eadbd55af2
oops - wrong bit (vic zandy)
rsc
2007-12-20 18:27:07 +0000
c2258bf4d2
fork minibug
rsc
2007-11-28 20:47:22 +0000
4f06ae0d42
More complete lapic startup (thanks Silas)
rsc
2007-11-28 20:47:10 +0000
a6c4711a38
bda[0xE] is a 16-bit segment number, not a real address. So shift 4.
rsc
2007-11-28 20:17:04 +0000
fd6b029401
proc_wait -> wait
rtm
2007-10-20 18:25:38 +0000
949352af66
Model verifying that wakeup really can be called after release without causing deadlock.
rsc
2007-10-12 04:21:04 +0000
943fd378a1
Incorporate new understanding of/with Intel SMP spec.
rsc
2007-10-01 20:43:15 +0000
9fd9f80431
Re: why cpuid() in locking code?
rsc
2007-09-30 14:30:04 +0000
c840f3ecdc
tricks
rsc
2007-09-30 14:20:47 +0000
af7366c945
interrupts during system calls
rsc
2007-09-27 21:37:45 +0000
ab08960f64
Final word on the locking fiasco?
rsc
2007-09-27 21:25:37 +0000
f97f0d2b3d
cleaner
rsc
2007-09-27 21:02:03 +0000
c95bde8163
yank out stack overflow checking ugliness
rsc
2007-09-27 20:38:53 +0000
4f74de0edc
okay, that was long enough - revert
rsc
2007-09-27 20:32:45 +0000
ce2e751555
test: store curproc at top of stack
rsc
2007-09-27 20:29:50 +0000
aefc13f8ba
nit
rsc
2007-09-27 20:25:32 +0000
3807c1f20b
rename splhi/spllo to pushcli/popcli
rsc
2007-09-27 20:09:40 +0000
39c3fb1b15
overkill: use segments to catch stack overflow (delete before next year)
rsc
2007-09-27 19:39:10 +0000
8c8b748a2f
now spllo is okay
rsc
2007-09-27 19:35:25 +0000
b5dcebdbeb
better lapic writes, suggested by cliff
rsc
2007-09-27 19:33:46 +0000
4721271961
use larger, allocated cpu stacks
rsc
2007-09-27 19:32:43 +0000
0fe118f3f6
don't call it ss - that's the stack segment
rsc
2007-09-27 16:47:50 +0000
c8919e6537
kernel SMP interruptibility fixes.
rsc
2007-09-27 12:58:42 +0000
75506c6655
use console lock
rsc
2007-09-27 12:29:25 +0000
d522571068
make slow bigdir last test
rsc
2007-09-27 12:29:06 +0000
ad12b487b5
changes since two days ago:
rsc
2007-09-27 11:27:04 +0000
b30ab3f5af
use standard bios location
rsc
2007-09-27 05:14:25 +0000
666f58c711
believe it or not, this was working
rsc
2007-09-27 05:13:10 +0000
90d975e9c8
comment bochs nonsense
rsc
2007-09-26 23:32:47 +0000
fbaa7b428e
various comment and print tweaks
rsc
2007-09-26 23:32:00 +0000
56c1a151d2
debugging prints
rsc
2007-09-26 23:24:23 +0000
d5596cd61d
Apparently the initial interrupt count lapic[TICR] must be set *after* initializing the lapic[TIMER] vector.
rsc
2007-09-26 20:34:12 +0000
355073ea9e
oops, interrupts on in syscall traps doesn't work after all
rtm
2007-09-25 16:15:05 +0000
3eda2714e6
tell SETGATE to leave interrupts on for T_SYSCALL panic if unknown fault with CPL=0 (i.e. in kernel)
rtm
2007-09-25 15:23:44 +0000
d7a90d3b5c
This should fix building on FreeBSD
nelhage
2007-09-19 23:49:52 +0000
6197c04208
Fix compilation on 64-bit machines (thanks to andersk for patch)
nelhage
2007-09-18 00:41:34 +0000
411ee74127
fix comments
rtm
2007-09-15 20:05:47 +0000
e9f8419fbe
sh
rsc
2007-09-05 15:55:43 +0000
ab4cedb593
continuous quality management
rtm
2007-08-31 19:55:27 +0000