summaryrefslogtreecommitdiffstats
path: root/sys/kern/kern_descrip.c
Commit message (Expand)AuthorAgeFilesLines
* Fix a deadlock I introduced this morning.phk2004-12-141-6/+7
* Add a new kind of reference count (fd_holdcnt) to struct filedescphk2004-12-141-16/+45
* Make fdesc_mtx private to kern_descrip.c now that the flock has come home.phk2004-12-141-1/+1
* Move the checkdirs() function from vfs_mount.c to kern_descrip.c andphk2004-12-141-0/+48
* Add new function fdunshare() which encapsulates the necessary light magicphk2004-12-141-0/+19
* Sort and wash #includes.phk2004-12-031-18/+15
* Drop ffree() as a separate function and incorporate the only place used.phk2004-12-021-17/+8
* Style polishing.phk2004-12-021-88/+37
* We already have a lock initialization function, use that for fdesc_mtxphk2004-12-011-57/+28
* Collect the stuff for the /dev/fd/{%d,std{in,out,err}} pseudo-devicephk2004-12-011-55/+52
* "nfiles" is a bad name for a global variable. Call it "openfiles" insteadphk2004-12-011-6/+6
* Style: move data to top of file.phk2004-12-011-4/+4
* Don't acquire Giant before calling closef() in close() (and elsewhere);rwatson2004-11-281-7/+7
* Fix LOR.phk2004-11-261-2/+2
* Neither of the arguments to closef() can be NULL anymore, so don'tdas2004-11-211-4/+1
* Move a FILEDESC_UNLOCK up to maintain correct nesting of FILEDESC/FILEphk2004-11-161-11/+9
* Make FILE_LOCK and FILEDESC_LOCK nest properly by postponing the thephk2004-11-151-2/+2
* Move #define up.phk2004-11-141-5/+5
* Introduce an alias for FILEDESC_{UN}LOCK() with the suffix _FAST.phk2004-11-131-15/+15
* Use more intuitive pointer for fdinit() and fdcopy().phk2004-11-081-2/+1
* Introduce fdclose() which will clean an entry in a filedesc.phk2004-11-071-1/+17
* Move fdinit() related stuff from .h to .cphk2004-11-071-0/+35
* Allow fdinit() to be called with a NULL fdp argument so we can usephk2004-11-071-15/+16
* When we open /dev/null for stdin/out/err for safety reasons, do it right:phk2004-11-061-3/+5
* Push acquisition of the accept mutex out of sofree() into the callerrwatson2004-10-181-0/+1
* Another case where we need to guard against a partiallyjulian2004-10-041-0/+2
* Remove GIANT_REQUIRED from setugidsafety() as knote_fdclose() no longerrwatson2004-08-191-2/+0
* Add the missing knote_fdclose().green2004-08-161-2/+4
* Add locking to the kqueue subsystem. This also makes the kqueue subsystemjmg2004-08-151-20/+7
* We're not yet ready to assert !Giant in kern_fcntl(), as it's calledrwatson2004-08-071-5/+4
* Avoid acquiring Giant for some common light-weight or already MPSAFErwatson2004-08-061-2/+31
* Assert Giant in the following file descriptor-related functions:rwatson2004-08-041-0/+10
* Push Giant acquisition down into fo_stat() from most callers. Acquirerwatson2004-07-221-6/+0
* Push acquisition of Giant from fdrop_closed() into fo_close() so thatrwatson2004-07-221-2/+0
* In addition to the real user ID check, do an explicit jailcsjp2004-07-141-2/+3
* Explicitly initialize f_data and f_vnode to NULL.phk2004-06-191-0/+3
* Do the dreaded s/dev_t/struct cdev */phk2004-06-161-2/+2
* Extend coverage of SOCK_LOCK(so) to include so_count, the socketrwatson2004-06-121-0/+3
* Deorbit COMPAT_SUNOS.phk2004-06-111-2/+2
* Push the VOP_ADVLOCK() call to release advisory locks on vnode filerwatson2004-06-011-11/+1
* Remove advertising clause from University of California Regent's license,imp2004-04-051-4/+0
* Conditionally assert Giant in fputsock() based on the value ofrwatson2004-03-291-0/+1
* Split the mlock() kernel code into two parts, mlock(), which unpackstruckman2004-02-261-1/+3
* Device megapatch 4/6:phk2004-02-211-0/+2
* Don't bother storing a result when all you need are the side effects.des2004-02-161-5/+5
* In fdcheckstd the descriptor table should never be shared, so justdwmalone2004-02-151-9/+7
* Locking for the per-process resource limits structure.jhb2004-02-041-7/+16
* Restore correct semantics for F_DUPFD fcntl. This should fix the errorsdes2004-01-171-4/+4
* WITNESS won't let us hold two filedesc locks at the same time, so juggledes2004-01-161-3/+9
* Remove two KASSERTs which were overly paranoid.des2004-01-161-4/+0
OpenPOWER on IntegriCloud