Commit Graph

1348 Commits

Author SHA1 Message Date
pjd
648f58f532 Try to stabilize ZFS with regard to memory consumption:
- Allow to shrink ARC down to 16MB (instead of 64MB).
- Set arc_max to 1/2 of kmem_map by default.
- Start freeing things earlier when low memory situation is detected.
- Serialize execution of arc_lowmem().

I decided to setup minimum ZFS memory requirements to 512MB of RAM and 256MB of
kmem_map size. If there is less RAM or kmem_map, a warning will be printed.
World is cruel, be no better. In other words: modern file system requires
modern hardware:)

From ZFS administration guide:

"Currently the minimum amount of memory recommended to install a Solaris
 system is 512 Mbytes. However, for good ZFS performance, at least one
 Gbyte or more of memory is recommended."
2007-04-10 02:35:57 +00:00
pjd
a26ee9422b Instead of detecting if lock is already initialized based on standard 1 bit
check, use more accurate 13 bits check. We had too many false-positives with
the standard check.

Reported by:	mlaier
2007-04-09 01:05:31 +00:00
pjd
e47fb9eabd Extend kobj compatibility KPI to support operating on files before and
after the root file system is mounted.
This is one of the changes that will allow to put root file system on ZFS.
2007-04-08 23:57:08 +00:00
pjd
2b260dcd5e MFp4: Synchronize with recent OpenSolaris changes. 2007-04-08 16:29:25 +00:00
scottl
1320f9f144 Add the CAM 'SG' peripheral device. This device implements a subset of the
Linux SCSI SG passthrough device API.  The intention is to allow for both
running of Linux apps that want to talk to /dev/sg* nodes, and to facilitate
porting of apps from Linux to FreeBSD.  As such, both native and linuxolator
entry points and definitions are provided.

Caveats:
 - This does not support the procfs and sysfs nodes that the Linux SG
   driver provides.  Some Linux apps may rely on these for operation,
   others may only use them for informational purposes.
 - More ioctls need to be implemented.
 - Linux uses a naming scheme of "sg[a-z]" for devices, while FreeBSD uses a
   scheme of "sg[0-9]".  Devfs aliasis (symlinks) are automatically created
   to link the two together.  However, tools like camcontrol only see the
   native names.
 - Some operations were originally designed to return byte counts or other
   data directly as the syscall return value.  The linuxolator doesn't appear
   to support this well, so this driver just punts for these cases.

Now that the driver is in place, others are welcome to add missing
functionality.  Thanks to Roman Divacky for pushing this work along.
2007-04-07 19:40:58 +00:00
jkim
bbfc500036 Fix kernel module dependency. linprocfs depends on sysvmsg and sysvsem.
Submitted by:	nork
2007-04-06 18:15:56 +00:00
pjd
6be01b7ba0 We have strcasecmp() in libkern now. 2007-04-06 11:18:57 +00:00
pjd
3b005d3302 Please welcome ZFS - The last word in file systems.
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/)
2007-04-06 01:09:06 +00:00
rwatson
765a83fd79 Replace custom file descriptor array sleep lock constructed using a mutex
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
2007-04-04 09:11:34 +00:00
jkim
b204c9cc13 MFP4: Turn emul_lock into a mutex.
Submitted by:	rdivacky
2007-04-02 18:38:13 +00:00
jkim
5657d675b3 Use underlying structures instead of kernel_sysctlbyname() for msginfo and
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
2007-03-30 17:56:44 +00:00
jkim
19e6b257f6 MFP4: Linux futex support for amd64.
Initial patch was submitted by kib and additional work was done
by Divacky Roman.

Tested by:	emulation
2007-03-30 01:07:28 +00:00
julian
93fc8e768e Implement the openat() linux syscall
Submitted by:	Roman Divacky (rdivacky@)
MFC after:	2 weeks
2007-03-29 02:11:46 +00:00
des
3d3e50beaf Add a pn_destroy field to pfs_node. This field points to a destructor
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>
2007-03-12 12:16:52 +00:00
rwatson
fe063bb84d In translate_path_major_minor(), do not calculate otherwise unused 'fp'
variable, avoiding an extra locking of the file descriptor array.
2007-03-06 07:39:12 +00:00
jkim
bad5e1f23b MFP4: 113090, 113130, 113132
Add Linux kernel version strings to /proc/sys/kernel.
2007-03-02 01:10:26 +00:00
jkim
554fb0a678 MFP4: 115220, 115222
- Fix style(9) and reduce diff between amd64 and i386.
- Prefix Linuxulator macros with LINUX_ to prevent future collision.
2007-03-02 00:08:47 +00:00
netchild
249ecc9078 MFp4 (110541):
Sync with rev 1.7 in NetBSD.

	Obtained from:	NetBSD
2007-02-25 12:43:07 +00:00
netchild
40358f3b01 MFp4 (110523, parts which apply cleanly):
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.
2007-02-25 12:40:35 +00:00
netchild
888f5e57b2 Partial MFp4 of 114977:
Whitespace commit: Fix grammar, spelling and punctuation.

Submitted by:	"Scot Hetzel" <swhetzel@gmail.com>
2007-02-24 16:49:25 +00:00
netchild
902cc4aeba MFp4 (114193 (i386 part), 114194, 114195, 114200):
- 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
2007-02-23 22:39:26 +00:00
netchild
d0c562e57c MFp4 (part of 114132):
- Fix a LOR caused by holding emul_lock and proctree_lock at once.

Submitted by:	rdivacky
2007-02-23 22:29:24 +00:00
jhb
e21fb2b555 Use 'pause' in several places rather than trying to tsleep() on NULL (which
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.
2007-02-23 16:25:08 +00:00
piso
6a2ffa86e5 o break newbus api: add a new argument of type driver_filter_t to
bus_setup_intr()

o add an int return code to all fast handlers

o retire INTR_FAST/IH_FAST

For more info: http://docs.freebsd.org/cgi/getmsg.cgi?fetch=465712+0+current/freebsd-current

Reviewed by: many
Approved by: re@
2007-02-23 12:19:07 +00:00
kib
de1264b042 Remove extern int hz; use proper include file instead. 2007-02-02 08:58:16 +00:00
kib
8f812418c1 Introduce some more SO_ option equivalents from Linux to FreeBSD.
The msg variable in linux_recvmsg() was not initialized.
Copy it from userspace.

Submitted by: rdivacky
2007-02-01 13:36:19 +00:00
kib
3f2b6c010a No need to lock emul_lock in exit_group() because em->shared
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
2007-02-01 13:33:33 +00:00
kib
02650398d1 No need to synchronize linux_schedtail with linux_proc_init.
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
2007-02-01 13:29:27 +00:00
netchild
1542e0642c Use a printf-modifier which doesn't need a cast.
Submitted by:	scottl
2007-01-21 13:18:52 +00:00
netchild
023c3ce346 Fix tinderbox build on amd64. 2007-01-20 19:32:23 +00:00
netchild
42392e7a0b MFp4 (113077, 113083, 113103, 113124, 113097):
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
2007-01-20 14:58:59 +00:00
netchild
d1c3c94c60 Ooops, fix the ratelimit. 2007-01-20 11:31:14 +00:00
netchild
22ce9c2574 Convert a KASSERT into a runtime warning (rate limited) + failsafe fallback.
Because of a stupid bug (also fixed with this commit) the KASSERT was
triggered when runnung the linux top.

Pointy hat to:	netchild
2007-01-20 11:07:41 +00:00
kib
a40cd17e13 Add support for LINUX_O_DIRECT, LINUX_O_DIRECT and LINUX_O_NOFOLLOW flags
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]
2007-01-18 09:32:08 +00:00
netchild
79662cd11c MFp4 (112379):
Implement SETALL/GETALL IPC primitives. This fixes some LTP testcases and
LabView is able to proceed a little bit further.

Submitted by:	rdivacky
2007-01-14 16:34:43 +00:00
netchild
48557a3c88 MFp4 (112705):
Inherit setting of the default emulation version to the jails.

Pointed out by:	jhb
Submitted by:	rdivacky
2007-01-14 16:07:01 +00:00
netchild
bda4ef3aed MFp4 (112646):
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
2007-01-07 19:30:19 +00:00
netchild
d49d13b83f MFp4 (112535):
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
2007-01-07 19:20:17 +00:00
netchild
e7d3b7ff43 MFp4 (112534):
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
2007-01-07 19:14:06 +00:00
netchild
10ec776aee MFp4 (112499):
Protect em->shared with the lock in case of CLONE_THREAD.

Submitted by:	rdivacky
2007-01-07 19:09:20 +00:00
netchild
977ef4a8bc MFp4 (112498):
Rename the locking flags to EMUL_DOLOCK and EMUL_DONTLOCK to prevent confusion.

Submitted by:	rdivacky
2007-01-07 19:00:38 +00:00
delphij
fb54bd8574 Fix amd64 build.
Submitted by:	Divacky Roman <xdivac02 stud fit vutbr cz>
2007-01-01 14:47:45 +00:00
netchild
33166d619b MFp4 (111746, 108671, 108945, 112352):
- add linux utimes syscall [1]
 - add linux rt_sigtimedwait syscall [2]

Submitted by:	"Scot Hetzel" <swhetzel@gmail.com> [1]
Submitted by:	Bruce Becker <hostmaster@whois.gts.net> [2]
PR:		93199 [2]
2006-12-31 13:16:00 +00:00
netchild
0a7aa5486d MFp4:
- semi-automatic style fixes
2006-12-31 12:42:55 +00:00
netchild
938b19deb7 MFp4 (111746+):
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
2006-12-31 12:39:10 +00:00
netchild
e9779a0351 MFp4:
- semi-automatic style fixes
 - spelling fixes in comments
 - add some comments
2006-12-31 11:56:16 +00:00
sam
874c920cbc add entry points required by newer broadcom wireless driver
PR:		kern/106131
Submitted by:	Scot Hetzel
MFC after:	2 weeks
2006-12-25 17:04:41 +00:00
netchild
c8fb50b5e6 MFP4 (110956):
Add definition for LINUX_MSG_INFO.

This fixes the tinderbox errors.

Submitted by:	rdivacky
2006-12-21 13:11:06 +00:00
jkim
3b05cb0c58 MFP4: 109655
- 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.
2006-12-20 20:17:35 +00:00
jkim
32be611658 MFP4: 110179
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().
2006-12-20 20:08:45 +00:00