Commit Graph

206 Commits

Author SHA1 Message Date
Giacomo Tesio 7f1ba356f1 kern: devproc: forbid open(OWRITE) of notes in Broken proceses 2017-12-15 02:58:09 +01:00
Giacomo Tesio c48d0e46ec kern: do not read wdir size on Broken processes 2017-12-15 02:58:09 +01:00
Giacomo Tesio e0b8a369d0 kern: align stack pointer to 16 byte in notify 2017-12-15 02:58:09 +01:00
Giacomo Tesio a3918b108d kern: awake must not interrupt flushes in dev9p
While sending Tflush after a wakeup, we might call sleep.

Since the awake was not yet consumed, the new sleep was interrupted
causing a new Tflush, in an infinite loop that consumes all tags.
2017-12-15 02:58:09 +01:00
Giacomo Tesio 799f99a40f kern: fix pprint on invalid address at syspread 2017-12-15 02:58:09 +01:00
Giacomo Tesio b73c2bce14 kern: awake & dev9p: do not interrupt Tflush to avoid consuming all tags 2017-12-15 02:58:09 +01:00
Giacomo Tesio 3946a06a98 kern: dev9p: Rwrite's count can be greater than Twrite's one 2017-12-15 02:58:09 +01:00
Giacomo Tesio 726217c766 kern: fix proc ref counting in procread 2017-12-15 02:58:09 +01:00
Giacomo Tesio df9c0d8fe4 kern: awake: fix syscall() check for awake_awakened(up) 2017-12-15 02:58:09 +01:00
Giacomo Tesio 2b62ed8010 kern: keep reading /proc/n/syscall after process death, to get all messages 2017-12-15 02:58:09 +01:00
Giacomo Tesio 9aea97a51f kern: fix fmtrwdata on nil buffers 2017-12-15 02:58:09 +01:00
Giacomo Tesio 8ea71b1c66 kern: fix very old bug in syscall tracing 2017-12-15 02:58:09 +01:00
Giacomo Tesio 7ae04d282f kern: devproc: add ppid; enable `rm -e noteid` 2017-12-15 02:56:57 +01:00
Giacomo Tesio 26bd1d8dc4 kern: fix working dir after ns clone
After sending the "clone" request to /proc/$pid/ns, the current process
will be moved to the working directory of $pid.

Otherwise the current process could still access the file and folders
on the service providing its dot, even if such service was not visible to $pid.

Also, keeping the original working directory means that the 'cd' line of the
current process's ns file would be different from that of $pid: the two ns
files must match since the command asked for a "clone".

See also https://marc.info/?l=9fans&m=150893734909969&w=2
2017-10-25 02:06:29 +02:00
Giacomo Tesio 2836ae4d66 kern: improve procfdprint and procfds in devproc 2017-10-25 01:35:34 +02:00
Giacomo Tesio c17d5bdb18 kern: cleanup procread for Qnote 2017-10-25 01:33:51 +02:00
Giacomo Tesio 70d5961022 kern: add missing poperror() to awake_gc_proc 2017-10-25 01:30:07 +02:00
Giacomo Tesio 430df16543 kernel: ignore errors on awake_gc_proc
Calling resrcwait in awake_gc_proc (that is called from pexit and thus
from sys_exits) means that we have to ignore errors that could cause
the syscall to return.

The funny part here was that one single test for newlib showed the
issue by faulting in a note handler AFTER completing with success:

	/arch/amd64/qa/lib/newlib/sigchld/213-sigqueue

The fact was that in the executable, the function notifier() was
located just after the exits() function, whose last line was a call
to _exits().

Thus, when the _exits returned, the notifier() code was executed.
2017-10-25 01:22:01 +02:00
Giacomo Tesio 7c017f5059 kern: fix %r13 management in syscallentry 2017-10-25 01:18:57 +02:00
Giacomo Tesio 90c3415176 kernel: fix _procfdprint to use chan's flags 2017-10-25 01:17:31 +02:00
Giacomo Tesio ba11888753 kern: proc, dup: _procfdprint OCEXEC and ORCLOSE
In _procfdprint print additional informations in the second column:

	E is printed if the file was open with OCEXEC flag
	D is printed if the file was open with ORCLOSE flag

Example output:
  3  rE  9   46 (0000000000000001 0 00)  8192       13 /dev/cons
2017-10-25 01:16:45 +02:00
Giacomo Tesio 9f8050aa4a libc: move exits to its own translation unit (another good idea from 9front) 2017-10-25 01:11:34 +02:00
Giacomo Tesio 2a4b3f26d6 cmds: add ns/clone
usage: ns/clone pid cmd [args...]

It starts cmd with provided arguments in a clone of the namespace
available to pid.
2017-10-25 01:02:32 +02:00
Giacomo Tesio 9d04fba8f2 cmds: rename `ns` to `ns/cat` 2017-10-25 00:22:26 +02:00
Giacomo Tesio 320e6e6f35 kern: clone $pid's namespace writing #p/$pid/ns
Any process X can get a new copy of the namespace of a target
process Y by writing the string "clone" to the ns file of Y.

The same user must own both processes.
The process writing the ns file must be allowed to mount.

The Pgrp of the calling process is then replaced with a new copy of
the Pgrp of the target process.

After the operation, any change done by X to its own namespace does
not affect Y. Also, if mount was forbidden for Y, it will also be
forbidden for X after the clone.
2017-10-24 23:28:14 +02:00
Giacomo Tesio 9202b04504 rc: use some uppercase variables (PATH, USER...) 2017-10-22 13:22:37 +02:00
Giacomo Tesio fab9c1d80f libc: introduce <envvars.h>
The new header envvars.h contains the names of commonly used
environment variables, such as $user, $ifs, $path and so on.

These defines are useful for Jehanne core applications that
use them to comunicate some values.

They are not strictly required, but having such defines we
can easily change the naming convention (from lowercase to uppercase).
2017-10-20 01:22:48 +02:00
Giacomo Tesio 65d37f432f rc: rename $cflag in $rccmd 2017-10-20 00:14:43 +02:00
Giacomo Tesio 9b0b9794c3 rc: write ENV_PATH and ENV_CDPATH as a colon separated string 2017-10-19 00:25:34 +02:00
Giacomo Tesio a5c14e60d7 rc: replace forgotten ENV_RCARGLIST 2017-10-19 00:16:19 +02:00
Giacomo Tesio d33221de34 rc: allow colons as separator in $path and $cdpath 2017-10-18 23:48:42 +02:00
Giacomo Tesio 0ef55939e8 rc: rm plan9.c (fix previous commit) 2017-10-18 22:41:31 +02:00
Giacomo Tesio 4017b73ab2 rc: define constants for rc variable names 2017-10-18 22:38:30 +02:00
Giacomo Tesio b42f8bd4a8 rc: mv plan9.c jehanne.c 2017-10-18 01:20:52 +02:00
Giacomo Tesio 085f5dfa34 rc: import 9front's improvements 2017-10-18 01:10:06 +02:00
Giacomo Tesio 26a2f0a1be libposix: fix typos in get_noteid/set_noteid 2017-09-08 00:24:37 +02:00
Giacomo Tesio e206015f7f libposix: add getpgid 2017-09-07 23:27:33 +02:00
Giacomo Tesio 1f3203e6d3 libposix: add rename 2017-09-07 23:25:34 +02:00
Giacomo Tesio e3ff5b1004 libposix: add getpgrp; draft setsid, getsid and setpgid 2017-09-07 00:10:34 +02:00
Giacomo Tesio e168dcb7fa libposix: add umask 2017-09-06 22:59:05 +02:00
Giacomo Tesio 61f0293e39 libposix: fake get/set functions for uid, gid, euid, egid, pgid, sid, reuid... 2017-09-05 23:53:40 +02:00
Giacomo Tesio 4be2181f89 libposix: add readlink and readlinkat (both failing with EINVAL) 2017-09-03 23:46:18 +02:00
Giacomo Tesio c5aaa72bbb libposix: add dup, dup2, getrusage, fstat and lstat 2017-09-03 17:51:44 +02:00
Giacomo Tesio 4115f92ff6 libc: make times return -1 on io error 2017-09-03 17:43:02 +02:00
Giacomo Tesio 0bc2e60c72 libposix: stub chown family (always returning 0)
I do not expect chown, fchownat and lchown to be much used in
UNIX softwares that we care to port.

We stub the functions in libposix so that we can refer them from
standard C libaries (such as newlib).

We will implement them (parsing /cfg/users to determinate uid and gid)
when it will be actually needed from a software ported to Jehanne.
2017-08-31 00:50:31 +02:00
Giacomo Tesio 3e533d234a cmds: port vt from 9front 2017-08-30 23:07:07 +02:00
Giacomo Tesio 713eb8843f libc: simplify access; libposix: let access lie
There are a few issues with Plan 9's `access`:

- it has side effects: to test the actual access (that the file
  servers can allow or deny according to complex custom rules)
  it opens and then closes the file, allocating (and disposing) the fd
- it does not work on directories, since
  - they cannot be opened for writing, despite the fact that to
    create a file in a directory you must be granted write access on
    that directory
  - they cannot be opened for execution, despite the fact that to
    access a file in a directory you must be granted execution access
    on that directory

Despite the fact that `access` (even on UNIX) is a violation of the
"tell, don't ask" principle (the access could be forbidden just after
its successful return, making subsequent `open` fail anyway), this
fact smells of a little design error in the file interface.

So, right now we choose to let the libposix's `access` lie on directories:
it will always return 0 on AWRITE and AEXEC for them, accepting that
a successive create/mkdir may fail.

However, a cleaner file API and protocol should allow a simpler `access`
to be implemented for directories too.
2017-08-29 00:17:51 +02:00
Giacomo Tesio 56287a4337 libc: better access on directories 2017-08-28 23:04:32 +02:00
Giacomo Tesio 59e881c04d libc: improve access(AWRITE) and access(AEXEC) on directories 2017-08-28 23:04:30 +02:00
Giacomo Tesio b78613615d libposix: add access 2017-08-28 23:04:12 +02:00