ZFS file system was ported from OpenSolaris operating system. The code in under
CDDL license.
I'd like to thank all SUN developers that created this great piece of software.
Supported by: Wheel LTD (http://www.wheel.pl/)
Supported by: The FreeBSD Foundation (http://www.freebsdfoundation.org/)
Supported by: Sentex (http://www.sentex.net/)
and flags with an sxlock. This leads to a significant and measurable
performance improvement as a result of access to shared locking for
frequent lookup operations, reduced general overhead, and reduced overhead
in the event of contention. All of these are imported for threaded
applications where simultaneous access to a shared file descriptor array
occurs frequently. Kris has reported 2x-4x transaction rate improvements
on 8-core MySQL benchmarks; smaller improvements can be expected for many
workloads as a result of reduced overhead.
- Generally eliminate the distinction between "fast" and regular
acquisisition of the filedesc lock; the plan is that they will now all
be fast. Change all locking instances to either shared or exclusive
locks.
- Correct a bug (pointed out by kib) in fdfree() where previously msleep()
was called without the mutex held; sx_sleep() is now always called with
the sxlock held exclusively.
- Universally hold the struct file lock over changes to struct file,
rather than the filedesc lock or no lock. Always update the f_ops
field last. A further memory barrier is required here in the future
(discussed with jhb).
- Improve locking and reference management in linux_at(), which fails to
properly acquire vnode references before using vnode pointers. Annotate
improper use of vn_fullpath(), which will be replaced at a future date.
In fcntl(), we conservatively acquire an exclusive lock, even though in
some cases a shared lock may be sufficient, which should be revisited.
The dropping of the filedesc lock in fdgrowtable() is no longer required
as the sxlock can be held over the sleep operation; we should consider
removing that (pointed out by attilio).
Tested by: kris
Discussed with: jhb, kris, attilio, jeff
seminfo because kernel_sysctlbyname() is slow. There is no dependency
problem since linux module depends on both sysvmsg and sysvsem and linprocfs
depends on it in turn.
Pointed out by: des
Reviewed by: des
function which is called from pfs_destroy() before the node is reclaimed.
Modify pfs_create_{dir,file,link}() to accept a pointer to a destructor
function in addition to the usual attr / fill / vis pointers.
This breaks both the programming and binary interfaces between pseudofs
and its consumers. It is believed that there are no pseudofs consumers
outside the source tree, so that the impact of this change is minimal.
Submitted by: Aniruddha Bohra <bohra@cs.rutgers.edu>
semi-automatic style(9)
The futex stuff already differs a lot (only a small part does not differ)
from NetBSD, so we are already way off and can't apply changes from NetBSD
automatically. As we need to merge everything by hand already, we can even
make the files comply to our world order.
- Dont "return" in linux_clone() after we forked the new process in a case
of problems.
- Move the copyout of p2->p_pid outside the emul_lock coverage in
linux_clone().
- Cache the em->pdeath_signal in a local variable and move the copyout
out of the emul_lock coverage.
- Move the free() out of the emul_shared_lock coverage in a preparation
to switch emul_lock to non-sleepable lock (mutex).
Submitted by: rdivacky
triggers a KASSERT) or local variables. In the case of kern_ndis, the
tsleep() actually used a common sleep address (curproc) making it
susceptible to a premature wakeup.
cannot change (because its referenced by curthread). This fixes
a LOR caused by acquiring emul_shared_lock while holding emul_lock.
Fix typo in comment.
Submitted by: rdivacky
p->p_emuldata is properly initialized in the time when the child can run.
Do not set p->p_emuldata to NULL when the process is exiting.
It does not make any sense and only costs 2 mutex operations.
Do not lock emul_data to unlock it on the very next line.
Comment on possible race while there.
Reparent all procs that are part of a threading group but not its leaders
to init and SIGCHLD init to finish the zombies off. This fixes zombies
left after opera's exit. [1]
There is no need to lock p_em in the linux_proc_init CLONE_THREAD
case because the process cannot change the address of the p_em->shared
because its currently running this code path.
Move assigning of em->shared outside emul_shared_lock.
Noticed by: Scott Robbins <scottro@nyc.rr.com> [1]
Submitted by: rdivacky
Dont expose em->shared to the outside world before its properly
initialized. Might not affect anything but its at least a better
coding style.
Dont expose em via p->p_emuldata until its properly initialized.
This also enables us to get rid of some locking and simplify the
code because we are workin on a local copy.
In linux_fork and linux_vfork create the process in stopped state
to be sure that the new process runs with fully initialized emuldata
structure [1]. Also fix the vfork (both in linux_clone and linux_vfork)
race that could result in never woken up process [2].
Reported by: Scot Hetzel [1]
Suggested by: jhb [2]
Reviewed by: jhb (at least some important parts)
Submitted by: rdivacky
Tested by: Scot Hetzel (on amd64)
Change 2 comments (in the new code) to comply to style(9).
Suggested by: jhb
to open() [1].
Improve locking for accessing session control structures [2].
Try to document (most likely harmless) races in the code [3].
Based on submission by: Intron (intron at intron ac) [1]
Reviewed by: jhb [2]
Discussed with: netchild, rwatson, jhb [3]
Now (ok it's been a while...) that FreeBSD has RLIMIT_AS too, we can use
it in the linuxolator instead of ignoring it.
This fixes a LTP test.
Submitted by: rdivacky
No need to lock prison in a case of linux_use26 because the int
setting is atomic and process cannot leave jail.
Submitted by: kib
Reviewed by: jhb
Requested by: rdivacky
Dont lock em in a case of just using em->shared->group_pid because
the group_pid never changes.
Submitted by: rdivacky
Reviewed by: kib
Glanced at by: jhb
Redo the checking for 2.6 emulation. We now cache the value of
use26 and replace calls to linux_get_osrelease() + parsing with
a call to linux_use26(). Typical path is lockless now.
Pointed out by: kib
This allows to ship RELENG_7_0 with a default osrelease of 2.4.2 and the
possibility to enable 2.6.x emulation without the possible performance
impact of the previous version of the check.
Submitted by: rdivacky
- Move linux_nanosleep() from src/sys/amd64/linux32/linux32_machdep.c to
src/sys/compat/linux/linux_time.c.
- Validate timespec ranges before use as Linux kernel does.
- Fix l_timespec structure.
- Clean up style(9) nits.
Add rudimentary IPC_INFO/MSG_INFO command support for linux_msgctl()
to pacify Linux ipcs(1). While I am here, add more bound checks
for linux_msgsnd() and linux_msgrcv().
Fixes for 'blocking in fifoor state' problem of LTP tests.
linux_*stat*() functions were opening files with O_RDONLY to get
major/minor pair for char/block special files. Unfortunately,
when these functions are used against fifo, it is blocked forever
because there is no writer. Instead, we only open char/block special
files for major/minor conversion. We have to get rid of kern_open()
entirely from translate_path_major_minor() but today is not the day.
While I am here, add checks for errors before calling
translate_path_major_minor().
Use TAILQ_FOREACH_SAFE instead of the unsafe one where an item is removed
from the queue.
This prevents a panic on kldunload.
Submitted by: rdivacky
Tested by: bsam
- Currently LINUX_MAX_COMM_LEN is smaller than MAXCOMLEN, but in case
this will change we have a buffer overflow. Apply some defensive
programming to DTRT when this should happen.
- Use copyinstr() instead of copyin where appropriate.
* Fallback to copyin() in case of ENAMETOOLONG. [1]
* Use the right source and destination (it was wrong before).
- Use strlcpy instead of strcpy.
- Properly lock the read case (PR_GET_NAME) like the write case.
Reviewed by: rwatson (except [1])
Suggested by: rwatson [1]
Add two linprocfs entries for Linux IPC:
/proc/sys/kernel/msgmni -> kern.ipc.msgmni
/proc/sys/kernel/sem -> kern.ipc.semmsl
kern.ipc.semmns
kern.ipc.semopm
kern.ipc.semmni
This fixes msgget03 and semget05 from Linux Test Project (LTP) test suite.
msgctl08 and msgctl09 also use /proc/sys/kernel/msgmni but another fix is
required from p4 (Change 110179).
Requested by: netchild
This fix lets clone02 LTP test pass with 2.6 emulation. In reality 99%
of the cases are that CLONE_VM and CLONE_THREAD are both set so it
seemed to work.
Submitted by: rdivacky
__WCLONE. This fixes it thus fixing skype/teamspeak to not keep zombies
after exit.
Submitted by: rdivacky
Reported by: Bakul Shah (bakul at bitblocks com)
specific privilege names to a broad range of privileges. These may
require some future tweaking.
Sponsored by: nCircle Network Security, Inc.
Obtained from: TrustedBSD Project
Discussed on: arch@
Reviewed (at least in part) by: mlaier, jmg, pjd, bde, ceri,
Alex Lyashkov <umka at sevcity dot net>,
Skip Ford <skip dot ford at verizon dot net>,
Antoine Brodin <antoine dot brodin at laposte dot net>
the correct syscalls.master's $FreeBSD$ tag record and
a make sysent in sys/compat/freebsd32. Thanks Ruslan
for pointing out the steps I missed :-0
Approved by: gnn
work is not just mine, but it is also the works of Peter Lei
and Michael Tuexen. They both are my two key other developers
working on the project.. and they need ata-boy's too:
****
peterlei@cisco.comtuexen@fh-muenster.de
****
I did do a make sysent which updated the
syscall's and sysproto.. I hope that is correct... without
it you don't build since we have new syscalls for SCTP :-0
So go out and look at the NOTES, add
option SCTP (make sure inet and inet6 are present too)
and play with SCTP.
I will see about comitting some test tools I have after I
figure out where I should place them. I also have a
lib (libsctp.a) that adds some of the missing socketapi
functions that I need to put into lib's.. I will talk
to George about this :-)
There may still be some 64 bit issues in here, none of
us have a 64 bit processor to test with yet.. Michael
may have a MAC but thats another beast too..
If you have a mac and want to use SCTP contact Michael
he maintains a web site with a loadable module with
this code :-)
Reviewed by: gnn
Approved by: gnn
dynamic nature (if no native aio code is available, the linux part
returns ENOSYS because of missing requisites) should be solved differently
than it is.
All this will be done in P4.
Not included in this commit is a backout of the changes to the native aio
code (removing static in some places). Those changes (and some more) will
also be needed when the reworked linux aio stuff will reenter the tree.
Requested by: rwatson
Discussed with: rwatson
begun with a repo-copy of mac.h to mac_framework.h. sys/mac.h now
contains the userspace and user<->kernel API and definitions, with all
in-kernel interfaces moved to mac_framework.h, which is now included
across most of the kernel instead.
This change is the first step in a larger cleanup and sweep of MAC
Framework interfaces in the kernel, and will not be MFC'd.
Obtained from: TrustedBSD Project
Sponsored by: SPARTA
Implement the linux_io_* syscalls (AIO). They are only enabled if the native
AIO code is available (either compiled in to the kernel or as a module) at
the time the functions are used. If the AIO stuff is not available there
will be a ENOSYS.
From the submitter:
---snip---
DESIGN NOTES:
1. Linux permits a process to own multiple AIO queues (distinguished by
"context"), but FreeBSD creates only one single AIO queue per process.
My code maintains a request queue (STAILQ of queue(3)) per "context",
and throws all AIO requests of all contexts owned by a process into
the single FreeBSD per-process AIO queue.
When the process calls io_destroy(2), io_getevents(2), io_submit(2) and
io_cancel(2), my code can pick out requests owned by the specified context
from the single FreeBSD per-process AIO queue according to the per-context
request queues maintained by my code.
2. The request queue maintained by my code stores contrast information between
Linux IO control blocks (struct linux_iocb) and FreeBSD IO control blocks
(struct aiocb). FreeBSD IO control block actually exists in userland memory
space, required by FreeBSD native aio_XXXXXX(2).
3. It is quite troubling that the function io_getevents() of libaio-0.3.105
needs to use Linux-specific "struct aio_ring", which is a partial mirror
of context in user space. I would rather take the address of context in
kernel as the context ID, but the io_getevents() of libaio forces me to
take the address of the "ring" in user space as the context ID.
To my surprise, one comment line in the file "io_getevents.c" of
libaio-0.3.105 reads:
Ben will hate me for this
REFERENCE:
1. Linux kernel source code: http://www.kernel.org/pub/linux/kernel/v2.6/
(include/linux/aio_abi.h, fs/aio.c)
2. Linux manual pages: http://www.kernel.org/pub/linux/docs/manpages/
(io_setup(2), io_destroy(2), io_getevents(2), io_submit(2), io_cancel(2))
3. Linux Scalability Effort: http://lse.sourceforge.net/io/aio.html
The design notes: http://lse.sourceforge.net/io/aionotes.txt
4. The package libaio, both source and binary:
http://rpmfind.net/linux/rpm2html/search.php?query=libaio
Simple transparent interface to Linux AIO system calls.
5. Libaio-oracle: http://oss.oracle.com/projects/libaio-oracle/
POSIX AIO implementation based on Linux AIO system calls (depending on
libaio).
---snip---
Submitted by: Li, Xiao <intron@intron.ac>
calls are not used by libthr in RELENG_6 and HEAD, it is only used by
the libthr in RELENG-5, the _umtx_op system call can do more incremental
dirty works than these two system calls without having to introduce new
system calls or throw away old system calls when things are going on.
- Linux returns ENOPROTOOPT in a case of not supported opt to setsockopt.
- Return EISDIR in pread() when arg is a directory.
- Return EINVAL instead of EFAULT when namelen is not correct in accept().
- Return EINVAL instead of EACCESS if invalid access mode is entered in
access().
- Return EINVAL instead of EADDRNOTAVAIL in a case of bad salen param
to bind().
Submitted by: rdivacky
Tested with: LTP (vfork01 fails now, but it seems to be a race and
not caused by those changes)
MFC after: 1 week
- add support to power off the system [2]
- check the linux magic values [3]
Submitted by: Marcin Cieslak <saper@SYSTEM.PL> [1,2]
Modelled after: linux man page of the reboot() syscall [3]
Found by: LTP testcase "reboot02" [1]
Tested with: LTP testcase "reboot02" [1,3]
MFC after: 1 week
svr4 code: this code would call centralized sysctl code that does
these checks also.
MFC after: 1 week
Obtained from: TrustedBSD Project
Sponsored by: nCircle Network Security, Inc.
but further on -current (still not successful, but a step into the right
direction).
Sponsored by: Google SoC 2006
Submitted by: rdivacky
Tested by: Paul Mather <paul@gromit.dlib.vt.edu>
handling for amd64 in the common code. The MD parts for amd64 are still
outstanding, but at least this fixes some panics on amd64.
Sponsored by: Google SoC 2006
Submitted by: rdivacky
Tested by: bsam
- protect td->td_proc->p_pid with the proc lock in linux_getpid
in the amd64 (= non i386) case [1]
Sponsored by: Google SoC 2006
Submitted by: rdivacky
Noticed by: netchild [1]
has in its procfs (do a readlink of /proc/self/fd/<nn> to find the pathname
that corresponds to a given file descriptor). Valgrind-3.x needs this
functionality. This is a placeholder only at this time.
- Send the systrace_args files for all the compat ABIs to /dev/null for
now. Right now makesyscalls.sh generates a file with a hardcoded
function name, so it wouldn't work for any of the ABIs anyway. Probably
the function name should be configurable via a 'systracename' variable
and the functions should be stored in a function pointer in the sysvec
structure.
- TLS - complete
- pid/tid mangling - complete
- thread area - complete
- futexes - complete with issues
- clone() extension - complete with some possible minor issues
- mq*/timer*/clock* stuff - complete but untested and the mq* stuff is
disabled when not build as part of the kernel with native FreeBSD mq*
support (module support for this will come later)
Tested with:
- linux-firefox - works, tested
- linux-opera - works, tested
- linux-realplay - doesnt work, issue with futexes
- linux-skype - doesnt work, issue with futexes
- linux-rt2-demo - works, tested
- linux-acroread - doesnt work, unknown reason (coredump) and sometimes
issue with futexes
- various unix utilities in linux-base-gentoo3 and linux-base-fc4:
everything tried worked
On amd64 not everything is supported like on i386, the catchup is planned for
later when the remaining bugs in the new functions are fixed.
To test this new stuff, you have to run
sysctl compat.linux.osrelease=2.6.16
to switch back use
sysctl compat.linux.osrelease=2.4.2
Don't switch while running a linux program, strange things may or may not
happen.
Sponsored by: Google SoC 2006
Submitted by: rdivacky
Some suggestions/help by: jhb, kib, manu@NetBSD.org, netchild
Please don't style(9) the NetBSD code, we want to stay in sync. Not imported
on a vendor branch since we need local changes.
Sponsored by: Google SoC 2006
Submitted by: rdivacky
With help from: manu@NetBSD.org
Obtained from: NetBSD (linux_{futex,time}.*)
and vn_fullpath (that call malloc(..., M_WAITOK)) from under the
vm object lock, since sleep is not allowed while holding the mutex.
Being there, wrap VOP_GETATTR call with conditional Giant aquire.
Currently this is (almost) noop because pseudofs is Giant-locked.
Tested by: kris
Approved by: pjd (mentor)
MFC after: 2 weeks
mark system calls as being MPSAFE:
- Stop conditionally acquiring Giant around system call invocations.
- Remove all of the 'M' prefixes from the master system call files.
- Remove support for the 'M' prefix from the script that generates the
syscall-related files from the master system call files.
- Don't explicitly set SYF_MPSAFE when registering nfssvc.
except for s_family (which is read-only once after it is set when the
structure is created).
- Mark svr4_sys_ioctl(), svr4_sys_getmsg(), and svr4_sys_putmsg() MPSAFE.
kern_accept() and accept1(). If another thread closed the new file
descriptor and the first thread later got an error trying to copyout the
socket address, then it would attempt to close the wrong file object. To
fix, add a struct file ** argument to kern_accept(). If it is non-NULL,
then on success kern_accept() will store a pointer to the new file object
there and not release any of the references. It is up to the calling code
to drop the references appropriately (including a call to fdclose() in case
of error to safely handle the aforementioned race). While I'm at it, go
ahead and fix the svr4 streams code to not leak the accept fd if it gets an
error trying to copyout the streams structures.
locked.
- Move all the svr4 socket cache code into svr4_socket.c, specifically
move svr4_delete_socket() over from streams.c. Make the socket cache
entry structure and svr4_head private to svr4_socket.c as a result.
- Add a mutex to protect the svr4 socket cache.
- Change svr4_find_socket() to copy the sockaddr_un struct into a
caller-supplied sockaddr_un rather than giving the caller a pointer to
our internal one. This removes the one case where code outside of
svr4_socket.c could access data in the cache.
- Add an eventhandler for process_exit and process_exec to purge the cache
of any entries for the exiting or execing process.
- Add methods to init and destroy the socket cache and call them from the
svr4 ABI module's event handler.
- Conditionally grab Giant around socreate() in streamsopen().
- Use fdclose() instead of inlining it in streamsopen() when handling
socreate() failure.
- Only allocate a stream structure and attach it to a socket in
streamsopen(). Previously, if a svr4 program performed a stream
operation on an arbitrary socket not opened via the streams device,
we would attach streams state data to it and change f_ops of the
associated struct file while it was in use. The latter was especially
not safe, and if a program wants a stream object it should open it via
the streams device anyway.
- Don't bother locking so_emuldata in the streams code now that we only
touch it right after creating a socket (in streamsopen()) or when
tearing it down when the file is closed.
- Remove D_NEEDGIANT from the streams device as it is no longer needed.
Also, call change_dir() instead of doing part of it inline (this now adds
a mac_check_vnode_chdir() call) to match fchdir() and call
mac_check_vnode_chroot() to match chroot(). Also, use the change_root()
function to do the actual change root to match chroot().
Reviewed by: rwatson
