aeaa308943
change allocproc() back to acquiring the ptable.lock.
...
fix bugs where caller incorrectly released lock on error return path.
2016-09-15 12:12:05 -04:00
469aa8b9b3
xx
2016-09-15 12:01:52 -04:00
33188666da
Delete two left-over print statements
2016-09-14 21:13:09 -04:00
6cec0211d8
Update comment a bit.
2016-09-14 13:01:53 -04:00
da91a3a408
Add sleeplock.[ch] to listing
2016-09-14 08:43:57 -04:00
e85cb90cbf
one more BUSY
2016-09-13 14:51:44 -04:00
564a1cf0cb
sleeplock files
2016-09-11 21:01:24 -04:00
dec637bc59
Replace I_BUSY with sleep locks
2016-09-11 20:59:57 -04:00
2adb7c21dc
Move retry loop/sleep/wakeup in bio.c into sleeping locks
2016-09-11 20:17:22 -04:00
6670d3b5e0
Straight replacement of B_BUSY with a sleeping lock.
2016-09-11 17:24:04 -04:00
551c2f3576
Merge branch 'master' of g.csail.mit.edu:xv6-dev
2016-09-08 21:20:26 -04:00
1b8ccf9482
x
2016-09-08 21:20:17 -04:00
34c2efc1d0
use asm() for lock release, not a C assignment
2016-09-08 14:45:20 -04:00
d63ac118e8
this might fix the reported deadlock, though I can't reproduce it.
2016-09-08 14:22:38 -04:00
761ea5471f
Be a bit more explicit what kind of suggestions are are interested.
2016-09-08 13:09:09 -04:00
5bf3fbee00
Make runoff happy
2016-09-02 15:27:28 -04:00
b7fed77b7b
Fix a few lines that runoff is complaining about that they are too long
2016-09-02 14:59:00 -04:00
5f03799f84
ack xi
2016-09-02 08:43:33 -04:00
b8fd4cba13
Merge branch 'master' of g.csail.mit.edu:xv6-dev
2016-09-02 08:36:14 -04:00
ae15515d80
APIC IDs may not be consecutive and start from zero, so we cannot really use it
...
as a direct index into cpus. Record apicid in struct cpu and have cpunum() look
for it. Replace cpu->id with cpunum() everywhere, and replace cpu->id with cpu->apicid.
Thanks to Xi Wang.
2016-09-02 08:31:13 -04:00
bc8221a59c
comment about sched() saving/restoring cpu->intena
2016-09-02 05:40:54 -04:00
4b2152cc2d
curproc -> proc in a comment
2016-09-01 17:21:37 -04:00
37939f24c2
better comment
2016-08-26 08:46:13 -04:00
27ff70dc08
ack
2016-08-26 08:22:24 -04:00
aff0c8d5c7
set iomb to forbid i/o instructions from user space
...
add to test that they indeed trap in user space
thanks to pablogventura@gmail.com and nwolovick@gmail.com
2016-08-26 08:20:11 -04:00
745a4d31a6
x
2016-08-25 09:14:52 -04:00
7894fcd217
Remove trailing white space with:
...
for f in *.{h,c}; do sed -i .sed 's/[[:blank:]]*$//' $f; done
(Thanks to Nicolás Wolovick)
2016-08-25 09:13:00 -04:00
6de6a3c952
ack mikecat
2016-08-25 07:12:33 -04:00
5625ae4973
add check for wrapping of address + size in exec()
2016-08-25 07:09:46 -04:00
67a7f9597e
add alignment check of virtual address to exec()
2016-08-25 07:09:10 -04:00
a7c03bd914
p2v -> P2V
2016-08-24 13:40:06 -04:00
7a77375d64
x
2016-08-19 07:20:08 -04:00
1fc87f367c
Coding style (thanks to phf).
2016-08-18 21:02:05 -04:00
5faed8ba64
Fixed QEMU command line avoids raw image warning.
2016-08-18 14:30:30 -04:00
0f3d9d9d68
It will be for 2016
2016-08-18 14:00:48 -04:00
987064cb12
Update based on a scan of my 6828 folder
2016-08-18 13:37:32 -04:00
2a117310b5
Removed useless variable.
2016-08-18 11:25:44 -04:00
216fdf846f
Again, QEMU command line to avoid image warning.
2016-08-18 11:20:43 -04:00
1ccb5a6fca
Remove unused variable, nits.
2016-08-18 11:15:56 -04:00
2c60b7f31f
Removed useless function and prototypes.
2016-08-18 11:09:04 -04:00
09396a801a
Thanks to Warren Toomey for RDMUL and WRMUL.
2016-08-18 07:42:13 -04:00
4a3576b810
Small change to support RWMUL and WRMUL. Now xv6 truly works with a block size
...
that is a multiple of the sector size.
2016-08-18 07:39:03 -04:00
19f65413bd
hold ptable.lock for the entire process creation
...
however, processes still modify their own p->* without locking
2016-08-13 03:44:13 -04:00
20d05d4411
separate atomic test-and-set from memory barrier.
...
* use xchg only for its atomicness.
* use __sync_synchronize() for both CPU and compiler barrier.
2016-08-12 07:03:35 -04:00
9c65b32d9e
nothing much
2016-08-11 15:41:53 -04:00
c779cc2be1
main comments
2016-08-11 13:55:13 -04:00
241188276c
nits
2016-08-10 12:06:02 -04:00
3431cd4927
more comments in entryother.S
2016-08-10 11:35:28 -04:00
0a69dc9b17
even more explicit that entry.S is linked into kernel, unlike bootasm.S.
2016-08-10 11:22:36 -04:00
858475e4df
tiny clarifications to some gdt code.
2016-08-10 10:51:14 -04:00
Robert Morris