Aren/NetBSD
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
187,922 Commits 606 Branches 0 Tags 3.1 GiB
753d396585
Commit Graph

1122 Commits

Author SHA1 Message Date
pooka
0cadacacab * create /dev on the rump rootfs automatically 
* make rump_vfs_makedevnodes() take a full basepath instead of hardcoding
  an assumption that the device node is created in /dev
  + the caller is responsible that the directory the nodes are being
    created in exists
 2009-12-03 15:06:04 +00:00
pooka
89d988455a Do bounds-checking before adding host offset. Otherwise the host 
offset would be counted in with "size" and incorrect operation
would ensue.
 2009-12-03 14:05:46 +00:00
pooka
2ce80d0210 Remove last remnants of the long-ago-properly-fixed RUMP_LMUTEX_MAGIC hack. 2009-12-03 13:12:16 +00:00
pooka
4535fac711 Make sure node lengths match in addition to strncmp(). 
(I thought i fixed this already once?)
 2009-12-03 12:54:30 +00:00
pooka
f703feb95e Call spec_init() to avoid device open lossage. 2009-12-03 12:52:42 +00:00
pooka
1e98588f3b Decide it's not worth the fuss to have rumpfs as a module and just 
hardcode attach into rump_vfs_init().  Saves us from a lot of
pingpong init bouncing from one component to another to get the order
right.
 2009-12-03 12:35:34 +00:00
pooka
53ddf73f75 Soft-fail xcall thread creation to make RUMP_THREADS=0 work again. 2009-12-03 12:16:36 +00:00
pooka
c92b36b85b fix inverted comparison 2009-12-02 17:18:59 +00:00
uebayasi
5eef0d3167 Declare global pointers as extern, otherwise they become common symbols. 
Fix mips build in lib/librumpnet.
 2009-12-02 08:31:56 +00:00
pooka
9fd1bf7eef call cache_cpu_init() for all cpus 2009-12-01 09:56:59 +00:00
pooka
612479bd5e Include cpu crosscall support (instead of stubbing it out). 2009-12-01 09:52:29 +00:00
pooka
62e84772b7 Almost there for virtual CPU MP support: 
* support bound kernel threads
* bind softint threads to specific virtual cpus
  + remove now-unnecessary locks from softint code

Now, if we only had MI CPU_INFO_FOREACH() .... (hi rmind ;)
 2009-12-01 09:50:51 +00:00
pooka
03735afa20 whitespace 2009-12-01 09:04:21 +00:00
pooka
d772942437 test for error in creating root vnode before using it 2009-11-30 12:32:13 +00:00
pooka
0c5879110f Use genfs_statvfs() for now. 2009-11-30 11:18:22 +00:00
pooka
ea0564ea28 fix comment 2009-11-30 11:14:58 +00:00
pooka
307d8134c0 support lookup of pathname component "." 2009-11-30 10:11:09 +00:00
pooka
2a0ed023df Remove no longer necessary vfs weak symbols. 2009-11-27 17:55:04 +00:00
pooka
ee0457e379 include subr_kobj_vfs 2009-11-27 17:54:24 +00:00
pooka
8102fe7341 Move rootfs-related init from init_main() to vfs_mountroot(). 
Reduces code re-written in rump.
 2009-11-27 16:43:51 +00:00
pooka
ebb48144a9 Now that Makefile.rump was changed and everything gets built in 
update builds too, flip the allocator define to prefer the kernel
pool/kmem instead of malloc(3).  Use malloc(3) only if
RUMP_USE_UNREAL_ALLOCATORS is defined.
 2009-11-27 13:45:15 +00:00
pooka
67cd12f07f Include arch/include/Makefile.inc for some compiler flags. 2009-11-27 13:36:30 +00:00
pooka
b318442ec7 Make sure rumpfs is attached on all host platforms. 2009-11-26 21:04:42 +00:00
pooka
aee2ff8e8d For rumpfs, do mountroot instead of the bunch of homegrown hacks 
currently there.  Still needs a little massage to get the kernel
interfaces right and avoid copypaste especially from main().

Also, move it a bit more into the direction of a real file system
(finally!) by giving it a vfsops.  Most ops are still unimplemented,
though.
 2009-11-26 20:58:51 +00:00
pooka
f631bc6ff9 remember to call pipe_init() 2009-11-26 17:36:22 +00:00
pooka
9c1300e7ee Provide some sort of cv_is_valid(). What a silly routine. 2009-11-26 17:29:34 +00:00
pooka
fea060dfd3 include sys_pipe.c 2009-11-26 17:24:52 +00:00
pooka
7bcd019a9d regen for rump_sys_pipe() 2009-11-26 17:24:36 +00:00
pooka
c845b495c5 regen: retval -> retval[2] 2009-11-26 16:38:01 +00:00
pooka
32e96a0e50 * load symbols only up to actual size of symbol table instead of size 
we allocated for storing the table
* trimm unused end off of table before passing to the kernel
 2009-11-26 15:44:26 +00:00
pooka
d2b375e830 Shuffle types around and add some random typecasts to make lint 
and LP64 gcc sign comparison shut up.  I'm not sure what benefit
this brings apart from introducing some more bugs for everyone's
debugging pleasure.
 2009-11-26 10:57:26 +00:00
pooka
2dc78d8839 Remove assym.h. It's no longer required by hppa since spcopy is 
not included anymore.
 2009-11-26 10:42:04 +00:00
pooka
b3d7587523 update a bit 2009-11-26 10:10:50 +00:00
pooka
300f534bd8 Build the kernel symbol table in rumpuser bootstrap and feed it to 
the rump kernel.  After this change it is possible to use the
in-kernel linker and rump_sys_modctl() to load kernel modules at
runtime.

Previously loading modules at runtime was possible only through
using the host system ld.so.  Note that it is still preferred to
use shared libs when possible, since they are PIC and n virtual
kernels will only require one copy of r/o segments.  However, when
there is no access to source code, a binary kernel module is the
only thing available ...
 2009-11-26 09:50:38 +00:00
pooka
4f28ef3250 regen: rump_kernelfsym_load() 2009-11-26 09:21:16 +00:00
pooka
ff492c8ebe Add rump_kernelfsym_load(), which loads the kernel symbol table. 2009-11-26 09:20:07 +00:00
pooka
7d11849a4a regen: rump_sys_modctl() 2009-11-26 09:02:38 +00:00
pooka
282ff84bcb Include kern_module_vfs and the modctl() syscall. The modctl() 
syscall is here because module loading is a vfs operation.
Theoretically you shouldn't be able to have modules without vfs
support.
 2009-11-26 07:30:24 +00:00
pooka
4686e2c38a * report partition size in blocks instead of bytes 
* "support" DIOCCACHESYNC.
 2009-11-25 15:01:28 +00:00
pooka
3d17105af3 Include link.h instead of the NetBSD-specific link_elf.h. Use 
dlinfo() only if __ELF__ is defined.
 2009-11-23 14:39:35 +00:00
pooka
09dbb89b44 If cpu_disklabel includes struct dkbad, define __HAVE_DISKLABEL_DKBAD. 
This allows use of subr_disk_mbr on all archs.  Default to it for
the rump disk component.  No functional change for regular kernels.
(The other option would've been to include dkbad in disklabels
everywhere, but arguably this approach has less possible side-effects,
especially given that wedges and related magic will take over the
world any second now).
 2009-11-23 13:40:08 +00:00
pooka
f8d9830a1a Add rf_paritymap.c, which is now required for raidframe. 2009-11-23 12:23:20 +00:00
pooka
bb7f50bcca kern_physio is already provided by rumpvfs 2009-11-23 11:20:25 +00:00
pooka
02a11f095f Generate fictional disklabel and return that from ioctl() instead 
of the halfwitted semi-host semi-virtual ioctl nonsense that was
here previously.
 2009-11-20 17:48:52 +00:00
pooka
071c062533 include components required by smbfs 2009-11-20 15:37:44 +00:00
pooka
59f20ea12d Hmm, in addition to leaving out -lc, -nostdlibs also leaves out 
libgcc, and causes problems on some platforms.  Currently there is
no easy way to say "link against libgcc but not against libc", so
just comment out -nostdlibs until a better way is invented.
 2009-11-20 12:24:17 +00:00
pooka
8e8d412248 Create async i/o "interrupt" thread from within the kernel so that 
it gets a kernel thread context.
 2009-11-19 14:44:58 +00:00
pooka
1de15b17c4 * don't open device when calling getfileinfo(), it might need to open 
the device and will f-a-i-l if it's a block device
* use rumpuser_ioctl() as little as possible (going away completely
  wewy wewy soon)
* improve some variable scoping
 2009-11-19 13:46:55 +00:00
pooka
e7c51ca379 .. umm, and use correct device value for previous. 
And use 64bit arithmetic.
 2009-11-19 13:42:29 +00:00
pooka
69eddfb6c5 Improve host file size detection for devices (i.e. make it work). 
Use ioctl() on NetBSD since it appears to be the only way to get
the info.  On other platforms, attempt lseek() and if that doesn't
work, punt, and hope someone comes along with a better implementation
later.
 2009-11-19 13:25:48 +00:00
pooka
1fb57ddbb5 Don't link rump kernel libs against libc. They have no host system 
dependencies apart from indirect ones via rumpuser.
 2009-11-17 16:17:06 +00:00
pooka
f7dd9c5812 Don't build rumpuser if NORUMPUSER is set. 2009-11-17 15:36:28 +00:00
pooka
8ef9411fc9 init global cv only once 2009-11-17 15:23:42 +00:00
pooka
53e3c13474 set return value properly 2009-11-11 16:50:17 +00:00
pooka
9359f96adb support timeouts in tsleep 2009-11-11 16:47:50 +00:00
pooka
ee4c8e1012 Make rumpuser_cv_timedwait take two int64's instead timespec to 
uncouple it from the timespec layout.  Also, change return value
to zero for "timeout didn't expire" and non-zero for "timeout
expired".  This decouples the interface from errno assignments.
 2009-11-11 16:46:50 +00:00
pooka
303e90e2c8 Give this a makefile for easy rebuild of all the components when 
the kernel ABI changes.
 2009-11-11 13:54:54 +00:00
pooka
886a3cee40 init uvm readahead 2009-11-10 17:02:36 +00:00
pooka
054e6b6e93 Code which uses pages fetched by the pager needs PGO_SYNCIO for 
obvious reasons.
 2009-11-10 16:55:12 +00:00
pooka
644009f9cb Hash out soft interrupts to be a little closer to real softints: 
* split them into levels
  * allow only one per level to be active at a time
  * fire softints only when we are unscheduling from a CPU instead
    of immediately in softint_schedule().  this will later morph
    into return from interrupt, but that part isn't done yet.
 2009-11-09 19:16:18 +00:00
pooka
19cde6b50d adjust comment for previous 2009-11-09 19:02:49 +00:00
pooka
1029997ffd set LP_INTR for KTHREAD_INTR 2009-11-09 19:00:52 +00:00
pooka
f1be958892 Add scheduling points around the iothread call to biodone() 2009-11-09 18:00:26 +00:00
nakayama
62117cce6e - make this compile on sparc. 
- shrink blanks to fit in 80 columns.
 2009-11-09 14:35:38 +00:00
dsl
26a8a91ded Fix stub prototype 2009-11-07 12:08:35 +00:00
pooka
3ad682caa6 Enable kernel kmem/vmem/pool/pool_cache by default again instead 
of malloc(3) allocators.
 2009-11-06 16:16:59 +00:00
pooka
83a2f995b9 In case a temp thread is exiting, borrow lwp0 for the respective 
kmem_free().
 2009-11-06 16:15:16 +00:00
pooka
9d391c932b account creds for processes, not lwp's 2009-11-06 15:22:33 +00:00
pooka
6abd8e3cdf lie that we are never in a softintr (i.e. add stub) 2009-11-06 15:22:16 +00:00
pooka
f5b3a079d0 Loop over dso list while loading succeeds (brute force dependency handling). 2009-11-05 14:13:03 +00:00
pooka
2a244b850d enqueue loaded modules and prevent duplicates 2009-11-05 14:10:53 +00:00
pooka
947f23c334 Pull all relegating memory allocators under a common roof in memalloc.c 
(forgot to commit the new file)
 2009-11-04 20:38:58 +00:00
pooka
d502da58ec misc_stub and emul have been the same thing for a looong time now, 
so just move the few remaining routines in misc_stub to emul.
 2009-11-04 19:21:51 +00:00
pooka
c03306bc0d Give the kthread->pthread interface emulation its own module. 2009-11-04 19:17:53 +00:00
pooka
8ec96b27ff trim trailing whitespace 2009-11-04 18:27:40 +00:00
pooka
8ef52102c2 Pull all relegating memory allocators under a common roof in memalloc.c 2009-11-04 18:25:36 +00:00
pooka
83dbe94e8b Implement yield()/preempt() now that there is a CPU scheduler. 2009-11-04 18:11:11 +00:00
pooka
080522e1ce move copy-related routines to their own module 2009-11-04 17:01:45 +00:00
pooka
a69d555865 Use std. uiomove() & friends. 2009-11-04 16:55:20 +00:00
pooka
510038040b Use kern_mutex_obj.c directly instead of copypasting code. 2009-11-04 13:32:39 +00:00
pooka
332ad24cfd Weak aliazeize usermount_common_policy() to uncouple rumpkern 
linkage from rumpvfs once again after the secmodel changes.
(temp solution, should find a better one)
 2009-11-03 20:25:31 +00:00
pooka
652a56fde0 Reorder enough of lwp/cpu initialization to have enough context 
for pool/kmem init to work again with the stock versions (as opposed
to rump malloc relegations).
 2009-11-03 20:22:33 +00:00
pooka
341072b737 move module to SRCS where it logically belongs. no functional change. 2009-11-03 18:44:15 +00:00
pooka
2a63153751 regen: add AB_* macros 2009-11-03 18:23:15 +00:00
pooka
ef8c3e068e include AB_* macros from reboot.h 2009-11-03 18:22:44 +00:00
pooka
b32585a644 Add rump_boot_set/gethowto(), which can be called before rump_init() 
to control .... *tadaa* boothowto.
 2009-11-03 18:22:16 +00:00
pooka
99e178a1c1 Pass modinit to rumpuser_dl_module_bootstrap() as a function pointer 
to avoid linker rump -> rumpuser -> rump dependency which is annoying
redundancy in static linking.
 2009-10-24 11:36:59 +00:00
pooka
c9f53946a5 unused rumpuser_realpath is really dead 2009-10-24 11:29:55 +00:00
rmind
10758cb0f9 Simplify pmap_remove() a little by avoiding pmap_do_remove() layer, since 
possibility to skip wired mappings is not needed anymore.  Apart from that,
no functional differences are intended.
 2009-10-22 19:50:55 +00:00
rmind
3e88e8e75c Sync rump with kernel changes. 2009-10-21 23:13:53 +00:00
rmind
40cf6f3659 Remove uarea swap-out functionality: 
- Addresses the issue described in PR/38828.
- Some simplification in threading and sleepq subsystems.
- Eliminates pmap_collect() and, as a side note, allows pmap optimisations.
- Eliminates XS_CTL_DATA_ONSTACK in scsipi code.
- Avoids few scans on LWP list and thus potentially long holds of proc_lock.
- Cuts ~1.5k lines of code.  Reduces amd64 kernel size by ~4k.
- Removes __SWAP_BROKEN cases.

Tested on x86, mips, acorn32 (thanks <mpumford>) and partly tested on
acorn26 (thanks to <bjh21>).

Discussed on <tech-kern>, reviewed by <ad>.
 2009-10-21 21:11:57 +00:00
pooka
5a1b660dab use autogenerated prototype 2009-10-20 23:29:57 +00:00
pooka
b762b305fb regen 2009-10-20 23:29:39 +00:00
pooka
693ea25ed2 add shmif_create 2009-10-20 23:28:50 +00:00
pooka
17e6aa2d7b printf -> rumpuser_dprintf 
(can't use printf while holding spinlock)
 2009-10-20 23:21:53 +00:00
tron
3ad0ebd5a8 Only allocate a "struct iovec" array from the healp if a reasonably sized 
stack array isn't large enough.
 2009-10-20 12:17:44 +00:00
pooka
0d29473b2c Actually, put uvm_readahead into rumpkern, since while it's 
technically vfs stuff, sys_descrip depends on it and readahead
itself uses only the pager interface.
 2009-10-20 10:42:41 +00:00
pooka
02a39de4a7 fix sign-compare warnings 2009-10-20 00:25:26 +00:00
pooka
fe208f2f7a __weak_alias for uvm_readahead() since it's used also in sys_descrip.c 2009-10-19 22:35:11 +00:00
pooka
639268f462 remove stale uvm_readahead() stub 2009-10-19 22:31:47 +00:00
christos
32ec2634aa treat sun2 like the other losing platforms. 2009-10-19 22:07:29 +00:00
pooka
c8d929f4fb add Arnaud's EuroBSDCon paper 2009-10-19 14:43:46 +00:00
tron
8cd44ca832 Avoid panic if a file system tries to write a chain of more than 32 mbuf-s 
to a socket. This happens e.g. when copying large files to SMBFS.

Code reviewed by Antti Kantee.
 2009-10-18 22:55:56 +00:00
pooka
9646a920b0 Clear PG_FAKE for pages we wrote to. This avoids paging in data 
we already have (and the backend might not yet have) in cases where
fs_bshift > PAGE_SHIFT.
 2009-10-18 00:41:09 +00:00
pooka
a6085c2236 fix off-by-one in sanity check and bump lazy bum magic value 2009-10-17 20:35:52 +00:00
pooka
291f141ce9 for udp sockets crank snd/rcvbufsize to 64k 2009-10-16 23:17:46 +00:00
pooka
afa1f603b7 In case of timeout, remember to remove ourselves from the sleep staleq. 2009-10-16 02:13:54 +00:00
pooka
d0c17feb81 We have real pollsuck() now, so remove this, ahem, less-than-perfect one. 
(XXX: only thing in the kernel using pollsuck is netsmb)
 2009-10-16 00:16:32 +00:00
pooka
254ffabc38 Include sys_select.c for proper select()/poll() support. 2009-10-16 00:14:53 +00:00
pooka
cc056ae92c include rumpcpu_generic 2009-10-15 23:42:40 +00:00
pooka
2f6e67dfd6 When allocating the temporary lwp we must have an lwp context. So 
take turns using lwp0 for this purpose, nothing else uses it.
 2009-10-15 23:15:55 +00:00
pooka
cfc0373f49 rump_get_curlwp() is dead 2009-10-15 16:46:37 +00:00
pooka
c663ed2f3f regen: lwp interface changes 2009-10-15 16:40:15 +00:00
pooka
8559951571 Give lwp usage some much-needed love: stop treating lwp0 as the 
all-sink and make sure each separate thread in rump has its own
lwp.  Happy-go-lucky callers will get scheduled a temporary lwp
on entry, while true lwp connoisseurs may request a stable lwp
for their purposes.  Some more love may be required later down the
road, but for now different threads will stepping on each others
toes.
 2009-10-15 16:39:22 +00:00
pooka
5a586117cf fix previous to make sense 2009-10-15 01:50:08 +00:00
pooka
466e9bcfec adjust for new rump_cpu 2009-10-15 01:35:23 +00:00
pooka
8c6bd3376a forgot to commit with previous batch 2009-10-15 01:29:06 +00:00
pooka
9bb1dbd8d7 rump_cpu is now a pointer 2009-10-15 00:34:05 +00:00
pooka
fcebdb26e0 Add prototype for wait operation variants which do not drop any 
resources (because they should be holding any in the first place).
 2009-10-15 00:33:37 +00:00
pooka
f1f3f36233 regen: scheduling points 2009-10-15 00:32:11 +00:00
pooka
000a197343 regen: scheduling points in rump_pub wrappers 2009-10-15 00:31:25 +00:00
pooka
cd231d0604 Generate scheduling points in rump_pub calls. 2009-10-15 00:29:19 +00:00
pooka
9454f185c4 Add initial work on a rump virtual cpu scheduler. This is necessary 
for kernel code which has been written to avoid MP contention by
using cpu-local storage (most prominently, select and pool_cache).

Instead of always assuming rump_cpu, the scheduler must now be run
(and unrun) on all entry points into rump.  Likewise, rumpuser
unruns and re-runs the scheduler around each potentially blocking
operation.  As an optimization, I modified some locking primitives
to try to get the lock without blocking before releasing the cpu.

Also, ltsleep was modified to assume that it is never called without
the biglock held and made to use the biglock as the sleep interlock.
Otherwise there is just too much drama with deadlocks.  If some
kernel code wants to call ltsleep without the biglock, then, *snif*,
it's no longer supported and rump and should be modified to support
newstyle locks anyway.
 2009-10-15 00:28:46 +00:00
pooka
36f718311e whoops, restore probing of ugen{0..3} instead of just ugen2 2009-10-14 23:29:42 +00:00
pooka
a9cd93a788 use rump_pub 2009-10-14 19:14:39 +00:00
pooka
d09e2773d0 "rumppriv" goes back to "rump" per internal interface naming change. 2009-10-14 18:18:53 +00:00
pooka
a6f92a2b8f regen: put all public interfaces created by ifspec into a rump_pub namespace 2009-10-14 18:16:41 +00:00
pooka
abf164854c Actually, adjust previous a bit: instead of preserving the names 
of the external interfaces and namespacing the internal ones to
"rumppriv", put the external ones in a "rump_pub" namespace.  While
this requires adjusting all of the external callers of these
interfaces, it is the right thing to do in the long run, since it
clarifies the structure.
 2009-10-14 18:14:48 +00:00
pooka
aa022ebd2b This wasn't supposed to go in -- the tree is under temporary 
component boundary breakage.
 2009-10-14 17:34:08 +00:00
pooka
8ca7f09f6e g/c some stuff which wasn't supposed to see the light of day 2009-10-14 17:30:52 +00:00
pooka
42972db848 Adjust rump sources for external/internal interfaces. 
No functional change.
 2009-10-14 17:29:19 +00:00
pooka
796b2444f2 generate rump local interfaces 2009-10-14 17:28:13 +00:00
pooka
c1bfb8966b adjust whitespace a bit 2009-10-14 17:26:09 +00:00
pooka
f2c144d595 Create rump public interfaces from description tables. This allows 
us to control and wrap all entry points from "userspace" into rump.
This in turn is necessary for the upcoming rump cpu scheduler.

For each interface "foo" a public wrapper called "rump_foo" is
created.  It calls the internal implementation "rumppriv_foo".  In
case foo is to be called from inside of rump kernel space, the
private interface "rumppriv_foo" is used -- the userspace wrapper
prototypes are not even exported into the rump kernel namespace.
Needless to say, the rump kernel internal interfaces are not exported
for users.

Now, three classes of interfaces fight for control of rump:
  + the noble local control interfaces (which this commit addresses)
  + the insidious rump system calls (which are generated from syscalls.master)
  + and the evil vnode interfaces (which are generated from vnode_if.src)
 2009-10-14 17:17:00 +00:00
pooka
ddc943db02 regen: fix rump varargs syscalls prototypes 2009-10-13 21:57:52 +00:00
pooka
5ac5a300e2 don't define curlwp, it's just (unused) sugar 2009-10-13 20:08:08 +00:00
pooka
b7f550b683 add one more passthrough 2009-10-13 18:36:48 +00:00
pooka
c8c66517fe Bump iovec lazy bum magic value to 32: nfsd likes to write mbufs 
with 17 per chain (previous i'm-too-lazy-for-my-forloop value was
of course 16).
 2009-10-12 02:25:44 +00:00
pooka
8ac0fe74d9 add recvmsg and sendmsg to networking rump syscalls 2009-10-11 23:23:09 +00:00
pooka
7e9bd843aa Include host offset in regular etfs read/write operations (I can't 
imagine it being particularly useful, but let's call file this
under POLA).
 2009-10-11 18:12:51 +00:00
pooka
860f5d7238 Support creating file system sockets (non-sockets not supported in 
VOP_CREATE since I don't want to have to write read/write support
for non-etfs files).
 2009-10-11 17:54:22 +00:00
pooka
fed1ab4ee0 raidframe works well enough now to initialize parity, so install it. 2009-10-11 11:26:40 +00:00
pooka
8d38112c32 add some stubs required by raidframe linkage 2009-10-10 21:10:04 +00:00
pooka
d4069e2c5e Provide an interface for reboot. 2009-10-09 14:41:36 +00:00
pooka
d3da4377ca regen 2009-10-09 14:32:07 +00:00
pooka
4dd39169ff namespace RB_ from reboot.h 2009-10-09 14:31:47 +00:00
pooka
0111b794da Seems like module loading code is tasked with calling secmodel_register() 
these days ...
 2009-10-08 00:47:47 +00:00
pooka
b15be8fcc2 remove extraneous call to secmodel_suser_start() 2009-10-08 00:36:56 +00:00
pooka
4ecb17968a dlclose() in error branch 2009-10-08 00:34:54 +00:00
pooka
366ae82927 Use uvm_readahead.c instead of null stubs. 2009-10-07 10:23:50 +00:00