Aren/NetBSD
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
189,559 Commits 606 Branches 0 Tags 3.1 GiB
f95fad15ff
Commit Graph

338 Commits

Author SHA1 Message Date
pooka
5a69314b41 struct puffs_usermoint does sound more lively, but be serious and 
fix the typos to puffs_usermount
 2007-05-17 13:31:30 +00:00
pooka
4fcec1c58b remember to document puffs_framev_cb error parameter addition 2007-05-16 10:07:11 +00:00
pooka
c1f0cc6870 Add error parameter to puffs_framev_cb to signal if there is a 
result of if the function should just do resource cleanup.
 2007-05-16 10:04:08 +00:00
pooka
8be39bc354 switch the arbitrary default errno value for network i/o from 
ECONNABORTED to ECONNRESET, which is at least a little bit better
 2007-05-16 09:41:04 +00:00
pooka
7b30368988 update .Nd to reflect current reality better 2007-05-16 07:56:09 +00:00
pooka
13d0372f38 memset freshly allocated puffs_usermount to 0 2007-05-15 16:45:22 +00:00
pooka
291fe845cd Step 3 and currently final step of a more general purpose event loop: 
support removal and addition of i/o file descriptors on the fly.

* detect closed file descriptors
* automatically free waiters of a dead file descriptor
* give the file server the possibility to specify a callback which
  notifies of a dead file descriptor
* move loop function to be a property of the mainloop instead of
  framebuf (doesn't change effective behaviour)
* add the possibility to configure a timespec parameter which
  attempts to call the loop function periodically
* move the event loop functions from the puffs_framebuf namespace
  to puffs_framev to differential between pure memory management
  functions
 2007-05-15 13:44:46 +00:00
pooka
354b8e48b8 use size_t instead of int to calm over-eager lint 2007-05-12 07:44:58 +00:00
pooka
46dc457634 copyright adjustment 2007-05-11 21:44:00 +00:00
pooka
e9cad0cef9 move puffs request dispatching code into it's own module. 
no functional change
 2007-05-11 21:42:42 +00:00
pooka
e94b19cfed Step 2 of general-purpose eventloop: get rid of puffs_framebuf_eventloop(). 
puffs_mainloop() now handles that functionality as well.
 2007-05-11 21:27:13 +00:00
pooka
2b8177d9fc Step 1 in moving towards a general-purpose eventloop: allow the 
framebuf event loop to take n i/o fd's as parameters and also allow
dynamic add/remove of fd's.  (not tested except for one fd, but more
changes coming soon)
 2007-05-11 16:22:38 +00:00
pooka
7806d3dfc4 assert that yield/continue is done only with a real continuation 2007-05-10 12:36:44 +00:00
pooka
4966e1c1de Add some kludges to be able to execute requests on the regular program 
stack instead of the continuation stack.  This is for lib/36011, where
pthread gets confused since we aren't running on the regular stack.
I'm not really sure which direction to go to with this quite yet, so
make the hack hard to enable on purpose.  The whole request dispatch
code needs cleaning anyway.
 2007-05-10 12:26:28 +00:00
pooka
4862309053 Remove the puffs_usermount argument from puffs_req_handle(). It can 
be found from the other arguments and this way puffs_req_handle() is
more symmetric with the other puffs_req() routines.
 2007-05-09 18:36:52 +00:00
pooka
19db57c596 List all current ways to write a file system. Also, add a reference 
to puffs_req(3) for the manual operation case.
 2007-05-09 18:24:11 +00:00
pooka
a1388b4c53 Return 0 from puffs_mainloop() if no error occurs. 
pointed out by Phil Nelson
 2007-05-09 18:23:12 +00:00
pooka
e5a894536c document PUFFS_KFLAG_WTCACHE and PUFFS_KFLAG_IAONDEMAND 2007-05-09 13:54:08 +00:00
pooka
06dd177014 document puffs_setback() 2007-05-09 13:46:16 +00:00
pooka
7faf2bccbe Add puffs_setback(), which can be used to set setbacks for nodes before 
the call returns to the kernel.  Currently it can be called only from
open(), mmap(), remove() and rmdir().  (But that might change)
 2007-05-07 17:16:07 +00:00
pooka
0a36778e8f Xr puffs_framebuf 2007-05-06 13:56:16 +00:00
pooka
bbf8b7cf94 document puffs_framebuf 2007-05-06 13:56:03 +00:00
pooka
4226bd8b40 implement the conveniently-already-prototyped puffs_framebuf_remaining(), 
which is just shorthand for tellsize() - telloff()
 2007-05-06 10:54:41 +00:00
pooka
a5f3fa93dc Remove the user-specified argument for the framebuf loop callback. 
It can be generated through other means if it is desired.
 2007-05-06 10:21:45 +00:00
pooka
bb0de0a878 add stubby version of puffs_framebuf manual page. proper one 
forthcoming to an internet near you Real Soon Now(tm)
 2007-05-05 16:09:16 +00:00
pooka
b223c403fc Add puffs "frame buffers", which are an abstraction of the buffering 
and event handling mechanisms required in file servers with blocking
I/O backends.  puffs_framebuf is built on the concept of puffs_cc
and uses those to multiplex execution where needed.

File systems are required to implement three methods:
  * read frame
  * write frame
  * compare if frame is a response to the given one

Memory management is provided by puffs_framebuf, but the file
systems must still, of course, interpret the protocol and do e.g.
byte order conversion.

As always, puffs_framebuf is work in progress.  Current users are
mount_psshfs and mount_9p.
 2007-05-05 15:48:18 +00:00
pooka
5fac7361cf Introduce PUFFS_FLAG_HASHPATH, which for PUFFS_FLAG_BUILDPATH file servers 
creates a hash of the node's path into po->po_hash to avoid a full pathcmp
every time.
 2007-05-01 15:58:00 +00:00
pooka
9601d4d304 use more accurate flag masking 2007-04-22 18:02:44 +00:00
pooka
5ad485727c assert that the file server doesn't try to return more cookies than 
it was supposed to
 2007-04-20 08:28:53 +00:00
pooka
89a0f35e6c improve descriptions and markup 2007-04-19 21:03:00 +00:00
pooka
5d86889ea0 add puffs_cc_getspecific(pcc), which is shorthand for: 
puffs_getspecific(puffs_cc_getusermount(pcc))
 2007-04-19 14:45:03 +00:00
pooka
7cfb5f2d40 fix readdir prototype 2007-04-18 20:37:05 +00:00
pooka
d17f97ea42 * describe fhtonode, nodetofh and suspend 
* update description of readdir to match reality
 2007-04-17 13:11:05 +00:00
pooka
eeff4f8efe Introduce PUFFS_STORE_DCOOKIE, which stores a directory cookie for a 
given offset iff the directory read in question wants cookies.
 2007-04-17 11:42:14 +00:00
pooka
f9d2f485d4 improve markup and descriptions 2007-04-17 10:14:27 +00:00
pooka
aaf5af07be Move file system and node callbacks into their own manual page. 
Improve markup and descriptions a bit (not completely finished yet).
 2007-04-16 19:16:02 +00:00
pooka
0d032531d1 document new interfaces 2007-04-16 16:37:02 +00:00
pooka
d1667bb48c alloc to set flags when setting fhsize 2007-04-16 13:04:49 +00:00
pooka
fca427389c some routines for setting mount parameters 2007-04-16 08:28:55 +00:00
pooka
37dd7942fc Instead of keeping on adding parameters to puffs_mount(), make it 
only take the bare essentials, which currently means removing
"maxreqlen" from the argument list (all current callers I'm aware
of set it as 0 anyway).  Introduce puffs_init(), which provides a
context for setting various parameters and puffs_domount(), which
can be used to mount the file system.  Keep puffs_mount() as a
shortcut for the above two for simple file systems.

Bump development ABI version to 13.  After all, it's Friday the 13th.
Watch out!  Bad things can happen on Friday the 13th. --No carrier--
 2007-04-13 13:35:46 +00:00
pooka
fdd8f87db5 document accessor routines added when making struct puffs_usermount 
implementation private
 2007-04-12 21:45:29 +00:00
pooka
95a18d20b9 Hide struct puffs_usermount from the rest of the world and provide 
accessors for interesting data in it.  Namely, you can now get
pu->pu_privdata with puffs_getspecific(), pu->pu_pn_root with
puffs_set/getroot() and pu->pu_maxreqlen with puffs_getmaxreqlen().
 2007-04-12 15:09:00 +00:00
pooka
6b7023f08f * support nodetofh and fhtonode callbacks 
* use new signature of node_readdir
 2007-04-11 21:04:51 +00:00
pooka
f7d8f11391 document the flush routines 2007-04-06 17:55:10 +00:00
pooka
6cc35b910e support page cache flushing 2007-04-06 17:49:11 +00:00
pooka
1e9c4b7026 use range invalidation op also for invalidating entire page cache of a node 2007-04-06 17:06:39 +00:00
pooka
622f8591d4 document puffs_inval_pagecache_node{,_range} 2007-04-06 16:44:01 +00:00
pooka
fa10fe8f9f implement puffs_inval_pagecache_node_range() 2007-04-06 16:38:03 +00:00
pooka
dd4643d880 Remove check against null parameter along with the XXX comment wondering 
why the check was there.  CID 4408
 2007-03-29 12:54:54 +00:00
pooka
406a946463 document credentials & access control methods 2007-03-22 17:38:09 +00:00
pooka
cfd6dc233f allow kernel creds in addition to superuser creds when checking for access 2007-03-22 17:15:34 +00:00
pooka
d28a42812b unify the location of *pcr in puffs_access_* 2007-03-22 16:57:27 +00:00
pooka
3a93377a23 copy ngroups * sizeof(gid_t) groups instead of just ngroups 2007-03-22 15:48:42 +00:00
pooka
c3a2affb84 * return -1 and set errno instead of returning errno directly 
* introduce puffs_cred_isregular() which checks if the cred structure
  has regular uid/gid (i.e. "person" credentials and not kernel creds)
 2007-03-22 15:32:22 +00:00
pooka
1d029cc056 remove variable names from header 2007-03-22 14:22:08 +00:00
pooka
0f28e7c1d1 add routiens which check against ufs semantics for permission to 
chown, chmod and change a/mtime
 2007-03-21 19:55:55 +00:00
pooka
1ba4b810d1 implement counterpart of vaccess() 2007-03-20 18:28:08 +00:00
pooka
a1e98c19f2 match puffs_node_access prototype with reality 2007-03-20 18:25:56 +00:00
pooka
a4143b4298 g/c unused debug variable 2007-03-20 11:28:35 +00:00
pooka
5c0f0bcf80 initial support for cacheops 2007-03-20 10:22:22 +00:00
pooka
2df275c4cd va->va_mode doesn't contain the full argument to the mknod() system call, 
so introduce puffs_addvtype2mode() and use that in null.c to generate the
proper syscall argument
 2007-03-16 08:14:49 +00:00
pooka
b93b025715 return errno instead of -1 if file is not present in lookup 2007-03-16 07:43:14 +00:00
pooka
ec2d43895a add puffs paper presented at AsiaBSDCon 2007 to "SEE ALSO" 2007-03-13 17:06:10 +00:00
pooka
d695eff80f Create in-memory nodes only per unique inode, not unique pathname. 
Solves problem with e.g. cache coherency of hardlinked files.
 2007-03-13 17:05:23 +00:00
pooka
8022394bca Until someone can sort non-priviledged mounts and kauth, force 
MNT_NOSUID | MNT_NODEV for mounts with geteuid() != 0.  I'm tired
of typing them on the command line every time I test a file system.
 2007-02-18 17:38:10 +00:00
pooka
b681c2611b dump return values for operations 2007-02-18 17:36:48 +00:00
pooka
29d5922025 #include <ucontext.h> 2007-02-17 23:00:16 +00:00
pooka
70247c3530 in case rootpath is "/", make dotdot lookup in "/foo" build 
a path of "/", not ""
 2007-02-15 21:26:50 +00:00
pooka
76fead3a3e Readlink should place the length of the link (without terminating 
nul) in the length field.  Make it so.
 2007-02-15 19:33:51 +00:00
pooka
09156a1e18 use puffs_path_walkcmp() instead of a homegrown comparison routine 2007-02-15 17:05:25 +00:00
pooka
7fea503eb4 Introduce puffs_path_walkcmp(), which is to be called from nodewalk 
and compares the path of the node against the given pathobject.
Also make comparison method take a flag to indicate if it should
check if the second path is a true prefix of the first.

plus some namespace cleanup
 2007-02-15 17:04:46 +00:00
pooka
a32e39e624 * get rid of the holy kernel-given ISDOTDOT, and instead define a 
macro which does strcmp against ".." and (the untranslated)
  componentname
* make PUFFS_FLAG_BUILDPATH build paths also if dotdot is the case,
  and adapt the regular path objects to this
* make nullfs lookup readable because we can now get rid of dotdot
  processing there
 2007-02-15 12:51:45 +00:00
pooka
1427bda45d generic node reclaim, which just puts a puffs_node 2007-02-15 12:51:24 +00:00
pooka
f7a084b3b0 Nm the manual puffs instead of libpuffs 2007-02-08 05:24:36 +00:00
pooka
6c42c843ee make it possible for the root path to be "/" and the next path to 
be "/foo" instead of "//foo"
 2007-02-06 01:46:41 +00:00
wiz
e6a160bd78 Fix spellos. 2007-01-28 07:58:05 +00:00
agc
9715dfff81 If we can't set the attributes when making the directory, use rmdir to 
remove the directory, not unlink.
 2007-01-27 11:49:44 +00:00
pooka
f4e2aba8ce document puffs_fs_suspend() 2007-01-26 23:55:27 +00:00
pooka
d83ed90176 add interfaces for suspending the file system 2007-01-26 23:00:33 +00:00
wiz
6d53f7319f New sentence, new line. 2007-01-20 16:49:36 +00:00
pooka
1118cadca7 document puffs_docc 2007-01-20 15:11:02 +00:00
pooka
5581d777a3 xref puffs_req 2007-01-20 15:10:38 +00:00
pooka
2fbe1b2261 document request handling routines 2007-01-20 15:09:34 +00:00
pooka
288823a738 Be consistent with the rest of the framework about the argument 
order of puffs_docc and puffs_dopreq
 2007-01-20 14:37:06 +00:00
pooka
84e83df1bb be consistent in naming request handlers 2007-01-20 13:52:14 +00:00
pooka
937097aa47 document puffs_getstate() and puffs_setstacksize() 2007-01-20 13:34:35 +00:00
wiz
2b5809387d Sort SEE ALSO. 2007-01-20 13:23:59 +00:00
pooka
809ae5f803 Bring the documentation slightly more into sync with the current 
reality.  It's still really a far cry from proper documentation,
though.
 2007-01-19 21:10:55 +00:00
pooka
c068d370f2 revoke revoke 2007-01-16 22:37:17 +00:00
pooka
05861b929d Checkpoint some more work in progress: for the benefit those file 
system backends which operate purely based on paths, push out more
path management into the library and make path management more
abstract: enable a file system to define a bunch of path management
callbacks, which are used by the framework.  Management of normal
/this/is/a/path type paths is provided by the library.
 2007-01-15 00:39:02 +00:00
pooka
5014a4c9c2 For directory renames we need to rename all the child nodes for 
the given directory if the file system wants paths (PUFFS_FLAG_BUILDPATH).
Do this by walking the nodelist and adjusting the path prefix of
each matching node.
 2007-01-11 18:18:36 +00:00
pooka
a3d07fb3e9 implement fsync 2007-01-11 17:48:21 +00:00
pooka
b599bb507d * do the l-variations of syscalls, since some nodes are symlinks 
* truncate only regular files to set size
* do the chmod()-dance for cache flush to now write-protected files
  until I can think of a nicer way to solve this
 2007-01-11 14:59:35 +00:00
pooka
e639bec8c2 Add preliminary code for a nullfs layer, which can be used to mount 
a directory hierarchy to another point, just like with the kernel
nullfs.  This is not really a layering scheme yet, but it should
evolve into one.  Currently it can just be used to do 1:1 mapping.
 2007-01-11 01:01:55 +00:00
pooka
14b3edf81b * don't chdir in possible daemon() call for the benefit of file systems 
which specify a relative path as the root
* if (buildpath), build link source full pathname for the duration of
  the operation
 2007-01-10 23:02:50 +00:00
pooka
2811d45fdd Be less shy about the default stack size, it's a only virtual memory space. 
TODO: put a guard page at the end
 2007-01-10 20:11:04 +00:00
pooka
71ddd8df08 rename name cache invalidation interface to be in sync with the 
kernel operation names, i.e. contain "namecache" instead of just "name"
 2007-01-09 18:19:01 +00:00
pooka
03a190dc22 lib interface for invalidation routines 2007-01-09 18:15:08 +00:00
pooka
7cc096d8ec * get rid of the mount callback; it's no great surprise to the 
server that it needs to mount the file system backend if it wants
  to call mount
* provide some options for getmntopts(), assume that callers will parse
  command line (or fstab) args
* reorganize the puffs_cc interface just a bit, preparing for a bigger
  revamp later
 2007-01-06 18:22:09 +00:00
pooka
f70a3c8c19 * inform kernel of our version 
* homegrown lib ABI check (to avoid bumping major a ludicrillion times
  before reaching stability)
 2007-01-02 15:53:05 +00:00
pooka
184a7d787c checkpoint some experimental work-in-progress, namely: 
Add support for having multiple outstanding operations.  This is done
by exposing enough interfaces so that it is convenient to have the
main event loop in the implementation itself and by providing a
continuation framework for convinient blocking and rescheduling.

works fine, but will undergo further cleanup & development
 2006-12-29 15:28:11 +00:00
alc
1a0e3f4a48 CID4360: don't leak `buf' upon return 
ok pooka@
 2006-12-14 18:15:59 +00:00
pooka
49df5bfc7b outline & export code for interfacing with kernel GET/PUT requests 2006-12-07 23:15:20 +00:00
pooka
24fb4d5d8c g/c out-of-date comment 2006-12-07 17:39:54 +00:00
pooka
43d8340ec6 support mmap 2006-12-07 16:59:14 +00:00
pooka
79fa3dd2c2 cleanup generated stuff from comments 2006-12-07 16:13:51 +00:00
pooka
55544f739a rototill the operation callbacks: unify the ops under struct puffs_ops 
and namespace them
 2006-12-07 10:53:21 +00:00
pooka
a0eb335104 adapt to kernel change: 
Allow multiple requests to be transferred in each GET/PUTOP
 2006-12-05 23:04:21 +00:00
pooka
b98361ea36 sauce catchup 2006-12-01 15:31:25 +00:00
pooka
0d5746a9ac PUFFSFLAG -> PUFFS_FLAG to be consistent with the kernel 2006-12-01 12:50:06 +00:00
pooka
465f49d71b build & send implemented operations list to kernel 2006-12-01 12:38:11 +00:00
pooka
d4280e60eb update with reality 2006-11-30 05:53:34 +00:00
pooka
c145fb5a17 summon daemon(3) in mainloop unless the nodaemon flag is given 2006-11-30 05:37:48 +00:00
pooka
c9aa28f463 * document open, close and access 
* nomenclature improvements
 2006-11-23 17:45:11 +00:00
pooka
4b5bf45935 const coherence 2006-11-23 16:44:28 +00:00
pooka
194d17d824 mode_t to enum vtype conversion 2006-11-21 23:11:09 +00:00
wiz
30ef64a953 Some mdoc and grammar nits. 2006-11-19 00:11:21 +00:00
pooka
d69d91cf3e * call statvfs() at mount() time, pass that info along 
* start is gone
 2006-11-18 12:40:35 +00:00
pooka
a0e314dc77 shuffle flags a bit 2006-11-17 17:48:32 +00:00
pooka
a40549faca namespace previous change under puffs_vfsnop to avoid confusions 2006-11-14 11:45:03 +00:00
pooka
2d9b770ec5 implement unmount, sync and statvfs as dummies for file system which 
couldn't care less about implementing them
 2006-11-14 11:23:44 +00:00
pooka
c025c6ae8f add note to BUGS about the questionable nature of puffs_node 2006-11-13 21:09:16 +00:00
pooka
c3f04e4e04 sync with change to lookup 2006-11-13 21:06:16 +00:00
pooka
80ce20607e * catch up with kernel changes 
* better error handling when mounting
 2006-11-09 13:11:01 +00:00
pooka
b8a1eae019 initial documentation for the puffs library interface 2006-11-09 01:30:15 +00:00
pooka
010234739c * support FAF (Fire-And-Forget) class operations 
* adapt to new lookup
 2006-11-07 22:10:53 +00:00
pooka
1fc270ecf0 pass rdev for specfs support 2006-10-26 22:53:01 +00:00
pooka
bda8cd7266 handle PUFFS_VN_INACTIVE 2006-10-25 18:15:50 +00:00
pooka
c643be45eb dump correct name for vfs operations 2006-10-23 16:53:17 +00:00
mrg
6bb0c03185 use casts to properly truncate some assignments. (sparc64 gcc complained.) 2006-10-23 03:18:30 +00:00
christos
d6a795f6a8 put back the comparison in a lint friendly way to appease pooka. 2006-10-23 03:13:13 +00:00
dogcow
acba3dfe4b comment out comparison of unsigned expression to >= 0 2006-10-23 03:11:53 +00:00
christos
7ef97c92be use portable prints 2006-10-23 03:10:09 +00:00
pooka
718426155f still check for array bounds, fix off-by-one 2006-10-23 01:36:13 +00:00
christos
bc7a8f1f3d fix a lint warning. 2006-10-23 01:23:42 +00:00
christos
b6f6a56052 pass lint; fix install. 2006-10-23 00:22:24 +00:00
dogcow
8f7efdd62a add the .include <bsd.own.mk>, so .d and .o files don't end up in the CWD. 
remove INCDIR=/usr/include, as my leenooks system doesn't have puffs .h files.
 2006-10-22 23:51:37 +00:00
pooka
1635cf06bc add libpuffs - the userspace programming interface for puffs 
This work was initially started and completed for Google SoC 2005
and tweaked to work a bit better in the past few weeks.  While
being far from complete, it is functional enough to be able and
stable to host a fairly general-purpose in-memory file system in
userspace.  Even so, puffs should be considered experimental and
no binary compatibility for interfaces or crash-freedom or zero
security implications should be relied upon just yet.

The GSoC project was mentored by William Studenmund and the final
review for the code was done by Christos.
 2006-10-22 22:52:21 +00:00