Commit Graph

244 Commits

Author SHA1 Message Date
Giacomo Tesio 04962f0637 libsec: fix #3: out-of-bound access in aesXCBCmac (CID 155914) 2017-01-17 21:05:59 +01:00
Giacomo Tesio 90f3d2ce31 lib9p2000: cleanup redundant checks in convM2S
CID 49221 (#1 of 1): Identical code for different branches (IDENTICAL_BRANCHES)
identical_branches: The same code is executed when the condition p == NULL is true or false, because the code in the if-then branch and after the if statement is identical. Should the if statement be removed?
2017-01-17 21:05:15 +01:00
Giacomo Tesio 663b002529 libsec: mark DigestState as stack var in ccpoly_*
CID 160099 and CID 160100 (#1 of 1): Free of address-of expression (BAD_FREE)
address_free: ccpolylen frees address of ds
2017-01-17 21:04:15 +01:00
Giacomo Tesio 3cfe366cc0 libsec: fix out of bound write (CID 155904)
In aesXCBCmac fix (potential) out of bound write in padding.

CID 155904 (#1 of 1): Out-of-bounds write (OVERRUN)
7. overrun-local: Overrunning array of 16 bytes at byte offset 16 by dereferencing pointer p2++.
2017-01-17 21:03:32 +01:00
Giacomo Tesio 06376d5859 libc.h: mark _assert as noreturn
Indeed the `assert` macro calls the _assert function only when the assertion failed.
2017-01-17 21:02:30 +01:00
Giacomo Tesio 6daa49324a kernel: umem: fix frees in segments_fill
CID 155453 (#1 of 1): Free of address-of expression (BAD_FREE)
address_free: free frees address of (segments + 0).pages.
2017-01-17 00:26:23 +01:00
Giacomo Tesio 20f15cbe5a kernel: fix not null terminated string in options()
CID 155471 (#1 of 1): Buffer not null terminated (BUFFER_SIZE_WARNING)67.
buffer_size_warning: Calling strncpy with a maximum size argument of 256 bytes on destination array envcopy of size 256 bytes might leave the destination string unterminated.
2017-01-17 00:18:18 +01:00
Giacomo Tesio 41a9489dd4 libsec: fix des56to64 access to key array
CID 155910 (#1 of 1): Out-of-bounds access (OVERRUN)1.

overrun-buffer-val: Overrunning buffer pointed to by key of 7 bytes by passing it to a function which accesses it at byte offset 63.
2017-01-17 00:01:08 +01:00
Giacomo Tesio e2b5953d99 libc: handle positive fd on "always-failing" creates (brk and dup) 2017-01-16 23:53:05 +01:00
Giacomo Tesio 042e206d13 kernel: devrtc: fix use of uninitialized array in rtcextract
CID 156229: Uninitialized scalar variable (UNINIT)9.
uninit_use: Using uninitialized value bcdclock[0].
2017-01-16 02:00:10 +01:00
Giacomo Tesio 237b7709ae kernel: umem: fix use of uninitialized fptr in freelist_pop
CID 156230 (#1 of 1): Uninitialized pointer read (UNINIT)9.
uninit_use: Using uninitialized value fptr.
2017-01-15 01:59:41 +01:00
Giacomo Tesio 1cb4610ed2 kernel: udp: fix read of uninitialized array
CID 156231 (#1 of 1): Uninitialized scalar variable (UNINIT)6.
uninit_use_in_call: Using uninitialized element of array laddr when calling memcmp.
2017-01-15 01:35:28 +01:00
Giacomo Tesio f2f95a935f kernel: fix use of uninitialized value in freelist_pop
CID 156233 (#1 of 1): Uninitialized pointer read (UNINIT)5. uninit_use: Using uninitialized value fimg.
2017-01-15 01:03:35 +01:00
Giacomo Tesio c4fb99ec38 qa: fix a few issues detected by coverity 2017-01-14 17:58:33 +01:00
Giacomo Tesio d4157c7e24 improve README.md and LICENSE.md 2017-01-14 12:39:05 +01:00
Giacomo Tesio 87a1c929ea doc: draft coding style 2017-01-12 23:19:02 +01:00
Giacomo Tesio 0de2dfc95b kernel: fix /dev/swap and `stats -m` 2017-01-12 00:41:14 +01:00
Giacomo Tesio 4e8d766fe8 doc: drafted hacking/overview.md 2017-01-12 00:41:01 +01:00
Giacomo Tesio d1cc947cd9 kernel: fix fd mode print in devproc 2017-01-12 00:39:50 +01:00
Giacomo Tesio e93aafc028 qa: build everything with -Werror 2017-01-12 00:38:56 +01:00
Giacomo Tesio 51d12f1f34 kernel: fix #0/brk/ qid (it's a special directory, not a special file) 2017-01-09 00:44:01 +01:00
Giacomo Tesio 01562d4ddb qa: rfork(RFNAMEG) on tests that are going to rfork(RFNOMNT) later 2017-01-06 23:04:04 +01:00
Giacomo Tesio 6872a8483b devtools: continuous-build.sh: restore toolchain cache in travis-ci 2017-01-06 18:49:12 +01:00
Giacomo Tesio d7d987c746 devtools: update 2017-01-06 03:48:47 +01:00
Giacomo Tesio a8e2247e20 kernel: drop SysNsec references from syscall() 2017-01-06 01:40:04 +01:00
Giacomo Tesio 4676c65a3d kernel: move nsec to libc 2017-01-06 00:56:22 +01:00
Giacomo Tesio 2508de40ea kernel: move sysdup to libc 2017-01-06 00:56:22 +01:00
Giacomo Tesio 882e33b091 qa: fix warnings 2017-01-06 00:54:42 +01:00
Giacomo Tesio c342234d74 kernel: print faults to process standard error 2017-01-06 00:54:04 +01:00
Giacomo Tesio 410e6d1fa1 cflags: set --sysroot=$JEHANNE 2017-01-06 00:53:25 +01:00
Giacomo Tesio d24c2b3606 rio: fix wloc call to sys/call 2017-01-04 23:34:44 +01:00
Giacomo Tesio e8562ff537 cmd/pict: import 9front's graphical tools 2017-01-03 22:31:09 +01:00
Giacomo Tesio 890f126abc kernel: fix usb after changing OREAD/OWRITE values
OREAD and OWRITE are used as array indexes assuming that OREAD was zero
and OWRITE was one. Thus each related allocation reserved just 2 slot and
even Ep struct in usb.h reserved just 2 int for toggles.

Since OREAD is now 1 and OWRITE is 2 we have to allocate/reserve 3 slot
as long as we use them as array indexes (which we could change in the future).

Unfortunately this means we waste the index zero in those arrays that will
always be unused. This also means that, to loop in such arrays we must begin
with OREAD as index zero is always empty.

PRO-MEMORIA: if/when we introduce the walk() syscall, OSTAT might turn useless.
In that case we might remove it and thus consider to move back OREAD/OWRITE
to 0/1 respectively (which might or might not be a good idea, to be analyzed).
2017-01-03 01:27:08 +01:00
Giacomo Tesio c1eb65b35e kernel: move null and zero from #c to #0 2017-01-03 01:27:08 +01:00
Giacomo Tesio 8e7f47c743 kernel/boot: bind #0 wherever #c is bound 2017-01-03 01:27:08 +01:00
Giacomo Tesio ac15346de7 kernel: make #0/pid, #0/ppid and #0/pgrpid readable 2017-01-03 01:27:07 +01:00
Giacomo Tesio da77086023 mount: update dev9p name (#9 instead of #M) 2016-12-31 17:19:40 +01:00
Giacomo Tesio ba80ac9649 lib/namespace: bind #0 to /dev 2016-12-31 00:46:34 +01:00
Giacomo Tesio 37541724d0 kernel: add #0/brk and move brk_() to libc 2016-12-31 00:43:02 +01:00
Giacomo Tesio 0bbd79e0a5 kernel: waitdebugger on lapictimerset if apic->div == 0 2016-12-26 02:18:19 +01:00
Giacomo Tesio 71ea62eb5b rm -e prints the return value of remove syscalls
Some devices return useful info on specific file remove (eg #0/pid, #0/ppid...)
so we need a tool to get such info.

	rm -e '#0/pid' '#0/ppid'
	#0/pid 65
	#0/ppid 59
2016-12-26 02:18:19 +01:00
Giacomo Tesio 31aa85b01a kernel: move pgrpid to devself 2016-12-26 02:18:19 +01:00
Giacomo Tesio 66373243bd libc: rewrite access() 2016-12-26 02:18:06 +01:00
Giacomo Tesio b862596737 kernel&all: create() syscall only sends Tcreate
In Plan9 the create syscall fallback on a open(OTRUNC) if the
path provided already exists. This is actually a common requirement
as most programs (editors, cat...) simply requires that a file is
there and is empty, and doesn't care overwriting existing contents
(note that this is particularily sensible with something like fossil).

In Jehanne the application is responsible of actually handle this
"file exists" error but libc provides ocreate() to mimic the Plan9
behaviour. Note that ocreate introduce a subtle race too: the path
is walked several times if the file exists, thus it could misbehave
on concurrent namespace changes. However I guess this is not going to
happen often enough to care now.

NOTE we will probably address this rare race too, with a more drammatic change
to syscalls: a new walk() syscall that will provide an unopen fd.
2016-12-24 21:25:20 +01:00
Giacomo Tesio 0c094289e6 9P2000: give a name to NinepMsgType to ease debug 2016-12-24 21:25:05 +01:00
Giacomo Tesio c30fb72b82 kernel: allow negative lengths in pwrite 2016-12-24 21:24:49 +01:00
Giacomo Tesio ca05868e4d kernel: use #9 instead of #M in /proc/n/ns 2016-12-24 21:24:28 +01:00
Giacomo Tesio b2ffd8ceaa continuous-build: run QA tests on travis-ci 2016-12-15 22:47:24 +01:00
Giacomo Tesio 99855d60d6 kernel: move chdir to libc using devself/devproc
Added wdir to devself and devproc:

- read '#0/wdir' to get the working directory of the calling process
  NOTE that a read(fd, nil, -1) will return the negated length
  of the working directory, just in case you want to
  allocate the memory required

- read '/proc/n/wdir' to know the working directory of process n
  (read(fd, nil, -1) still returns the negated length)

- write '#0/wdir' to change the working directory of the calling process
  NOTE: no offset is allowed and the provided string must
  be null terminated

- write '/proc/n/wdir' to change the working directory of process n
  NOTE: no offset is allowed and the provided string must
  be null terminated; moreover if another process change the working
  directory change during the write, the current process will
  receive an error.

In libc updated getwd() and chdir().
Also modified pwd to get advantage of the new file.

To test, run /arch/amd64/qa/kern/wdir.rc or simply try

	% pwd
	/usr/glenda
	% echo -n /tmp > /proc/$pid/wdir
        % pwd
        /tmp
        % cat '#0/wdir' && echo
        /tmp

The expected use cases for wdir in devproc are rio and acme.

Also, note that we could theoretically remove the cd builtin
from rc and simply implement it as a rc function.
We don't do that to preserve rc portability to other OS.
2016-12-15 22:42:01 +01:00
Giacomo Tesio c6de6b66e9 kernel: introduce devself
Devself provides to each process access to its own structures.

So far it contains four files:

- pid
- ppid
- pipes	used to implement pipe(2)
- segments used to implement segattach, segdetach and segfree
2016-12-11 01:26:08 +01:00