]> Devi Nivas Git - cs3210-lab1.git/log
cs3210-lab1.git
9 years agoAdd mikecat's argp test
Frans Kaashoek [Mon, 26 Sep 2016 11:54:02 +0000 (07:54 -0400)]
Add mikecat's argp test

9 years agofix memory region validation in argptr()
MikeCAT [Sat, 19 Dec 2015 13:52:11 +0000 (22:52 +0900)]
fix memory region validation in argptr()

9 years agoTypo (Thanks to Austin Liew)
Frans Kaashoek [Sun, 25 Sep 2016 12:27:26 +0000 (08:27 -0400)]
Typo (Thanks to Austin Liew)

9 years agoMerge branch 'master' of g.csail.mit.edu:xv6-dev
Frans Kaashoek [Mon, 19 Sep 2016 11:02:08 +0000 (07:02 -0400)]
Merge branch 'master' of g.csail.mit.edu:xv6-dev

9 years agoCheck result of write (thans to Alexander Kapshuk <alexander.kapshuk@gmail)
Frans Kaashoek [Mon, 19 Sep 2016 11:01:30 +0000 (07:01 -0400)]
Check result of write (thans to Alexander Kapshuk <alexander.kapshuk@gmail)

9 years agoFix a small README typo
Tej Chajed [Fri, 16 Sep 2016 13:11:23 +0000 (09:11 -0400)]
Fix a small README typo

9 years agoRemove left-over print statements
Frans Kaashoek [Thu, 15 Sep 2016 23:16:30 +0000 (19:16 -0400)]
Remove left-over print statements

9 years agoMerge branch 'master' of g.csail.mit.edu:xv6-dev
Robert Morris [Thu, 15 Sep 2016 16:56:17 +0000 (12:56 -0400)]
Merge branch 'master' of g.csail.mit.edu:xv6-dev

9 years agochange allocproc() back to acquiring the ptable.lock.
Robert Morris [Thu, 15 Sep 2016 16:12:05 +0000 (12:12 -0400)]
change allocproc() back to acquiring the ptable.lock.
fix bugs where caller incorrectly released lock on error return path.

9 years agoxx
Robert Morris [Thu, 15 Sep 2016 16:01:52 +0000 (12:01 -0400)]
xx

9 years agoDelete two left-over print statements
Frans Kaashoek [Thu, 15 Sep 2016 01:13:09 +0000 (21:13 -0400)]
Delete two left-over print statements

9 years agoUpdate comment a bit.
Frans Kaashoek [Wed, 14 Sep 2016 17:01:53 +0000 (13:01 -0400)]
Update comment a bit.

9 years agoAdd sleeplock.[ch] to listing
Frans Kaashoek [Wed, 14 Sep 2016 12:43:57 +0000 (08:43 -0400)]
Add sleeplock.[ch] to listing

9 years agoone more BUSY
Frans Kaashoek [Tue, 13 Sep 2016 18:51:44 +0000 (14:51 -0400)]
one more BUSY

9 years agosleeplock files
Frans Kaashoek [Mon, 12 Sep 2016 01:01:24 +0000 (21:01 -0400)]
sleeplock files

9 years agoReplace I_BUSY with sleep locks
Frans Kaashoek [Mon, 12 Sep 2016 00:59:57 +0000 (20:59 -0400)]
Replace I_BUSY with sleep locks

9 years agoMove retry loop/sleep/wakeup in bio.c into sleeping locks
Frans Kaashoek [Mon, 12 Sep 2016 00:17:22 +0000 (20:17 -0400)]
Move retry loop/sleep/wakeup in bio.c into sleeping locks

9 years agoStraight replacement of B_BUSY with a sleeping lock.
Frans Kaashoek [Sun, 11 Sep 2016 21:24:04 +0000 (17:24 -0400)]
Straight replacement of B_BUSY with a sleeping lock.

9 years agoMerge branch 'master' of g.csail.mit.edu:xv6-dev
Frans Kaashoek [Fri, 9 Sep 2016 01:20:26 +0000 (21:20 -0400)]
Merge branch 'master' of g.csail.mit.edu:xv6-dev

9 years agox
Frans Kaashoek [Fri, 9 Sep 2016 01:20:17 +0000 (21:20 -0400)]
x

9 years agouse asm() for lock release, not a C assignment
Robert Morris [Thu, 8 Sep 2016 18:45:20 +0000 (14:45 -0400)]
use asm() for lock release, not a C assignment

9 years agothis might fix the reported deadlock, though I can't reproduce it.
Robert Morris [Thu, 8 Sep 2016 18:22:38 +0000 (14:22 -0400)]
this might fix the reported deadlock, though I can't reproduce it.

9 years agoBe a bit more explicit what kind of suggestions are are interested.
Frans Kaashoek [Thu, 8 Sep 2016 17:09:09 +0000 (13:09 -0400)]
Be a bit more explicit what kind of suggestions are are interested.

9 years agoMake runoff happy
Frans Kaashoek [Fri, 2 Sep 2016 19:27:28 +0000 (15:27 -0400)]
Make runoff happy

9 years agoFix a few lines that runoff is complaining about that they are too long
Frans Kaashoek [Fri, 2 Sep 2016 18:59:00 +0000 (14:59 -0400)]
Fix a few lines that runoff is complaining about that they are too long

9 years agoack xi
Frans Kaashoek [Fri, 2 Sep 2016 12:43:33 +0000 (08:43 -0400)]
ack xi

9 years agoMerge branch 'master' of g.csail.mit.edu:xv6-dev
Frans Kaashoek [Fri, 2 Sep 2016 12:36:14 +0000 (08:36 -0400)]
Merge branch 'master' of g.csail.mit.edu:xv6-dev

9 years agoAPIC IDs may not be consecutive and start from zero, so we cannot really use it
Frans Kaashoek [Fri, 2 Sep 2016 12:31:13 +0000 (08:31 -0400)]
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.

9 years agocomment about sched() saving/restoring cpu->intena
Robert Morris [Fri, 2 Sep 2016 09:40:54 +0000 (05:40 -0400)]
comment about sched() saving/restoring cpu->intena

9 years agocurproc -> proc in a comment
Robert Morris [Thu, 1 Sep 2016 21:21:37 +0000 (17:21 -0400)]
curproc -> proc in a comment

9 years agobetter comment
Frans Kaashoek [Fri, 26 Aug 2016 12:46:13 +0000 (08:46 -0400)]
better comment

9 years agoack
Frans Kaashoek [Fri, 26 Aug 2016 12:22:24 +0000 (08:22 -0400)]
ack

9 years agoset iomb to forbid i/o instructions from user space
Frans Kaashoek [Fri, 26 Aug 2016 12:20:11 +0000 (08:20 -0400)]
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

9 years agox
Frans Kaashoek [Thu, 25 Aug 2016 13:14:52 +0000 (09:14 -0400)]
x

9 years agoRemove trailing white space with:
Frans Kaashoek [Thu, 25 Aug 2016 13:13:00 +0000 (09:13 -0400)]
Remove trailing white space with:
 for f in *.{h,c}; do sed -i .sed 's/[[:blank:]]*$//' $f; done
(Thanks to Nicolás Wolovick)

9 years agoack mikecat
Frans Kaashoek [Thu, 25 Aug 2016 11:12:33 +0000 (07:12 -0400)]
ack mikecat

9 years agoadd check for wrapping of address + size in exec()
MikeCAT [Sat, 19 Dec 2015 15:04:52 +0000 (00:04 +0900)]
add check for wrapping of address + size in exec()

9 years agoadd alignment check of virtual address to exec()
MikeCAT [Sat, 19 Dec 2015 14:43:56 +0000 (23:43 +0900)]
add alignment check of virtual address to exec()

9 years agop2v -> P2V
Robert Morris [Wed, 24 Aug 2016 17:40:06 +0000 (13:40 -0400)]
p2v -> P2V

9 years agox
Frans Kaashoek [Fri, 19 Aug 2016 11:20:08 +0000 (07:20 -0400)]
x

9 years agoCoding style (thanks to phf).
Frans Kaashoek [Fri, 19 Aug 2016 01:02:05 +0000 (21:02 -0400)]
Coding style (thanks to phf).

9 years agoFixed QEMU command line avoids raw image warning.
Peter H. Froehlich [Sat, 26 Sep 2015 21:38:04 +0000 (17:38 -0400)]
Fixed QEMU command line avoids raw image warning.

9 years agoIt will be for 2016
Frans Kaashoek [Thu, 18 Aug 2016 18:00:48 +0000 (14:00 -0400)]
It will be for 2016

9 years agoUpdate based on a scan of my 6828 folder
Frans Kaashoek [Thu, 18 Aug 2016 17:37:32 +0000 (13:37 -0400)]
Update based on a scan of my 6828 folder

9 years agoRemoved useless variable.
Peter H. Froehlich [Sat, 3 Oct 2015 08:43:05 +0000 (04:43 -0400)]
Removed useless variable.

9 years agoAgain, QEMU command line to avoid image warning.
Peter H. Froehlich [Sat, 3 Oct 2015 05:55:43 +0000 (01:55 -0400)]
Again, QEMU command line to avoid image warning.

9 years agoRemove unused variable, nits.
Peter H. Froehlich [Sun, 27 Sep 2015 03:10:25 +0000 (23:10 -0400)]
Remove unused variable, nits.

9 years agoRemoved useless function and prototypes.
Peter H. Froehlich [Sat, 26 Sep 2015 22:47:39 +0000 (18:47 -0400)]
Removed useless function and prototypes.

9 years agoThanks to Warren Toomey for RDMUL and WRMUL.
Frans Kaashoek [Thu, 18 Aug 2016 11:42:13 +0000 (07:42 -0400)]
Thanks to Warren Toomey for RDMUL and WRMUL.

9 years agoSmall change to support RWMUL and WRMUL. Now xv6 truly works with a block size
Frans Kaashoek [Thu, 18 Aug 2016 11:39:03 +0000 (07:39 -0400)]
Small change to support RWMUL and WRMUL.  Now xv6 truly works with a block size
that is a multiple of the sector size.

9 years agohold ptable.lock for the entire process creation
Robert Morris [Sat, 13 Aug 2016 07:44:13 +0000 (03:44 -0400)]
hold ptable.lock for the entire process creation
however, processes still modify their own p->* without locking

9 years agoseparate atomic test-and-set from memory barrier.
Robert Morris [Fri, 12 Aug 2016 11:03:35 +0000 (07:03 -0400)]
separate atomic test-and-set from memory barrier.
* use xchg only for its atomicness.
* use __sync_synchronize() for both CPU and compiler barrier.

9 years agonothing much
Robert Morris [Thu, 11 Aug 2016 19:41:53 +0000 (15:41 -0400)]
nothing much

9 years agomain comments
Robert Morris [Thu, 11 Aug 2016 17:55:13 +0000 (13:55 -0400)]
main comments

9 years agonits
Robert Morris [Wed, 10 Aug 2016 16:06:02 +0000 (12:06 -0400)]
nits

9 years agomore comments in entryother.S
Robert Morris [Wed, 10 Aug 2016 15:35:28 +0000 (11:35 -0400)]
more comments in entryother.S

9 years agoeven more explicit that entry.S is linked into kernel, unlike bootasm.S.
Robert Morris [Wed, 10 Aug 2016 15:22:36 +0000 (11:22 -0400)]
even more explicit that entry.S is linked into kernel, unlike bootasm.S.

9 years agotiny clarifications to some gdt code.
Robert Morris [Wed, 10 Aug 2016 14:51:14 +0000 (10:51 -0400)]
tiny clarifications to some gdt code.

9 years agoshell nits
Robert Morris [Mon, 8 Aug 2016 17:06:38 +0000 (13:06 -0400)]
shell nits
fix Regehr complaint

10 years agonits
Frans Kaashoek [Sun, 15 Nov 2015 18:40:42 +0000 (13:40 -0500)]
nits

10 years agoRemove console input.lock
Nathaniel Wesley Filardo [Sat, 7 Nov 2015 05:37:23 +0000 (00:37 -0500)]
Remove console input.lock

Use cons.lock for everything.  This eliminates the possibility that two CPUS
independently, simultaneously manipulate the CRTC in cgaputc.

10 years agoAdd sanity check panic for CRTC interactions
Nathaniel Wesley Filardo [Sat, 7 Nov 2015 05:38:40 +0000 (00:38 -0500)]
Add sanity check panic for CRTC interactions

10 years agoMerge pull request #3 from beordle/master
kaashoek [Sun, 15 Nov 2015 17:51:54 +0000 (12:51 -0500)]
Merge pull request #3 from beordle/master

Fix comments

10 years agoMerge pull request #5 from kehao95/master
kaashoek [Sun, 15 Nov 2015 17:45:53 +0000 (12:45 -0500)]
Merge pull request #5 from kehao95/master

Update URL for 2014

10 years agoUpdate URL for 2014
kehao95 [Fri, 13 Nov 2015 16:14:31 +0000 (00:14 +0800)]
Update URL for 2014

10 years agoFix comments
Jindong Zhang [Fri, 18 Sep 2015 15:36:45 +0000 (23:36 +0800)]
Fix comments

10 years agoFixed broken build on Arch Linux
Ayan Shafqat [Thu, 13 Aug 2015 02:11:39 +0000 (22:11 -0400)]
Fixed broken build on Arch Linux

10 years agocoding style (per Jeremías Feltan <jeremias.feltan@gmail.com>)
Frans Kaashoek [Tue, 30 Jun 2015 20:23:56 +0000 (16:23 -0400)]
coding style (per Jeremías Feltan <jeremias.feltan@gmail.com>)

10 years agoUpdate memide to reflect recent changes to fs and buf
Frans Kaashoek [Sat, 27 Jun 2015 17:38:03 +0000 (13:38 -0400)]
Update memide to reflect recent changes to fs and buf

10 years agoPick up where i left off in april:
Frans Kaashoek [Sat, 27 Jun 2015 16:39:13 +0000 (12:39 -0400)]
Pick up where i left off in april:
- move log into metadata part of disk, so that marking
that the log's blocks are in use falls out for free
- superblock describes the whole disk (sizes and offets)
- sizes and offsets are computed in one place (mkfs) and
the rest of the code refers to the superblock for these values,
instead of recomputing them.

10 years agoOne ide command for nsector_per_block generates nsector_per_block interrupts,
Frans Kaashoek [Sat, 11 Apr 2015 00:39:40 +0000 (20:39 -0400)]
One ide command for nsector_per_block generates nsector_per_block interrupts,
instead of 1.   ideintr() should wait for nsector_per_block interrupts before
waking up the reader; now it wakes up the reader after 1 interrupt, which
apparently means that the first sector is available of the command instead of all.
For now keep BSIZE to 512.

10 years agocheck blockno passed to idestart
Frans Kaashoek [Fri, 10 Apr 2015 11:15:06 +0000 (07:15 -0400)]
check blockno passed to idestart

10 years agoDisentangle block size from the disk's sector size. Set block size to 1024 to show
Frans Kaashoek [Fri, 3 Apr 2015 12:22:02 +0000 (08:22 -0400)]
Disentangle block size from the disk's sector size.  Set block size to 1024 to show
that they can be different.  Clean up mkfs, simplifying specifying fs parameters,
remove some redundancy between fs and mkfs, and fix disk layout bugs. Call blocks
in the file system blocks instead of sectors.  Passes usertests for different
block sizes.

10 years agoFix missing NUL-terminator in grep
Austin Clements [Wed, 25 Mar 2015 00:54:39 +0000 (20:54 -0400)]
Fix missing NUL-terminator in grep

Currently, grep read()s into a buffer and then uses the buffer as a
string. Since there's no NUL-terminator, this can cause it to falsely
identify line breaks and matches from leftover data on earlier lines
and, if a line fills up the entire buffer, to read past the end of the
buffer.

Fix this by NUL-terminating any data returned by read().

Thanks to Keiichi Watanabe for the report.

10 years agofix buf where concurrent fs syscalls race on log entries
Cody Cutler [Thu, 19 Feb 2015 16:51:47 +0000 (11:51 -0500)]
fix buf where concurrent fs syscalls race on log entries

11 years agoIncrease VM memory when running qemu-memfs
Austin Clements [Fri, 3 Oct 2014 13:44:26 +0000 (09:44 -0400)]
Increase VM memory when running qemu-memfs

Presumably because of the pre-loaded file system, the default RAM is
insufficient when running kernelmemfs

11 years agoMake kernelmemfs depend on kernel.ld
Austin Clements [Fri, 3 Oct 2014 13:43:44 +0000 (09:43 -0400)]
Make kernelmemfs depend on kernel.ld

11 years agoFix kernelmemfs linking
Austin Clements [Fri, 3 Oct 2014 13:39:09 +0000 (09:39 -0400)]
Fix kernelmemfs linking

Commit 9aa0337d introduced a linker script for linking the kernel image,
but only changed the kernel target to use it.  kernelmemfs was still
using the old linker command line arguments, which weren't sufficient.

11 years agoggdb -> gdwarf-2 for gcc4.8
Frans Kaashoek [Tue, 23 Sep 2014 11:39:32 +0000 (07:39 -0400)]
ggdb -> gdwarf-2 for gcc4.8

11 years agowhoops. forgot date.h
Cody Cutler [Fri, 12 Sep 2014 22:41:38 +0000 (18:41 -0400)]
whoops. forgot date.h

11 years agocmosgetdate() for system-call homework
Cody Cutler [Fri, 12 Sep 2014 21:18:57 +0000 (17:18 -0400)]
cmosgetdate() for system-call homework

the day of reckoning has come for the debug port "Shutdown" hack.

instead of mucking with ACPI or using a new hack, the student will now write
sys_date() using the cmosgetdate() helper.

11 years agoOops (thanks David Lawrence)
Frans Kaashoek [Mon, 8 Sep 2014 22:00:53 +0000 (18:00 -0400)]
Oops (thanks David Lawrence)

11 years agono need to generate postscript version
Frans Kaashoek [Fri, 29 Aug 2014 21:10:32 +0000 (17:10 -0400)]
no need to generate postscript version

11 years agoNice layout for a printout of the code
Frans Kaashoek [Fri, 29 Aug 2014 21:06:49 +0000 (17:06 -0400)]
Nice layout for a printout of the code

11 years agoAck all bug reports and patches. Is this all of them?
Frans Kaashoek [Fri, 29 Aug 2014 13:03:29 +0000 (09:03 -0400)]
Ack all bug reports and patches.  Is this all of them?

11 years agoupdate for 2014
Frans Kaashoek [Thu, 28 Aug 2014 21:28:40 +0000 (17:28 -0400)]
update for 2014

11 years agoxx
Robert Morris [Thu, 28 Aug 2014 21:04:47 +0000 (17:04 -0400)]
xx

11 years agoMerge branch 'master' of git+ssh://amsterdam.csail.mit.edu/home/am0/6.828/xv6
Robert Morris [Thu, 28 Aug 2014 20:13:58 +0000 (16:13 -0400)]
Merge branch 'master' of git+ssh://amsterdam.csail.mit.edu/home/am0/6.828/xv6

11 years agowrite log blocks from cache only at end of transaction
Robert Morris [Thu, 28 Aug 2014 20:06:12 +0000 (16:06 -0400)]
write log blocks from cache only at end of transaction

11 years agonits
Robert Morris [Thu, 28 Aug 2014 10:27:01 +0000 (06:27 -0400)]
nits

11 years agoi think this is a working concurrent logging scheme
Robert Morris [Thu, 28 Aug 2014 09:57:47 +0000 (05:57 -0400)]
i think this is a working concurrent logging scheme

11 years agoa start at concurrent FS system calls
Robert Morris [Wed, 27 Aug 2014 21:15:30 +0000 (17:15 -0400)]
a start at concurrent FS system calls

11 years agocomments
Robert Morris [Wed, 27 Aug 2014 18:14:52 +0000 (14:14 -0400)]
comments

11 years agoevery iput() and namei() must be inside a transaction
Robert Morris [Mon, 4 Aug 2014 17:06:48 +0000 (13:06 -0400)]
every iput() and namei() must be inside a transaction

11 years agouse acquire/release to force order for pid=np->pid;np->state=RUNNING
Robert Morris [Mon, 4 Aug 2014 10:13:49 +0000 (06:13 -0400)]
use acquire/release to force order for pid=np->pid;np->state=RUNNING
for bug reported by symingz@gmail.com and cs1100254@cse.iitd.ernet.in

11 years agonewer versions of qemu have different binary names (per Joel Nider <joel@nider.org>)
Frans Kaashoek [Thu, 31 Jul 2014 15:08:28 +0000 (11:08 -0400)]
newer versions of qemu have different binary names (per Joel Nider <joel@nider.org>)

12 years agoCorrect a security bug in copyuvm()
Stephen Tu [Mon, 4 Mar 2013 21:16:54 +0000 (16:16 -0500)]
Correct a security bug in copyuvm()

copyuvm() should not allow new copied pages to inherit more
permissions than the original pages.

13 years agoPrevent extra rebuild of fs.img by keeping intermediate object files.
Cam Tenny [Wed, 24 Oct 2012 22:52:40 +0000 (18:52 -0400)]
Prevent extra rebuild of fs.img by keeping intermediate object files.

13 years agoUse static assert instead of _LP64 (thanks Eddie!)
Frans Kaashoek [Tue, 11 Sep 2012 01:58:18 +0000 (21:58 -0400)]
Use static assert instead of _LP64  (thanks Eddie!)

13 years agoRemove -m32 flag from native gcc compiler
Frans Kaashoek [Fri, 7 Sep 2012 21:39:04 +0000 (17:39 -0400)]
Remove -m32 flag from native gcc compiler
But check that mkfs.c is building with LP64