| Commit message (Expand) | Author | Age | Files | Lines |
* | Fix a deadlock I introduced this morning. | phk | 2004-12-14 | 1 | -6/+7 |
* | Add a new kind of reference count (fd_holdcnt) to struct filedesc | phk | 2004-12-14 | 1 | -16/+45 |
* | Make fdesc_mtx private to kern_descrip.c now that the flock has come home. | phk | 2004-12-14 | 1 | -1/+1 |
* | Move the checkdirs() function from vfs_mount.c to kern_descrip.c and | phk | 2004-12-14 | 1 | -0/+48 |
* | Add new function fdunshare() which encapsulates the necessary light magic | phk | 2004-12-14 | 1 | -0/+19 |
* | Sort and wash #includes. | phk | 2004-12-03 | 1 | -18/+15 |
* | Drop ffree() as a separate function and incorporate the only place used. | phk | 2004-12-02 | 1 | -17/+8 |
* | Style polishing. | phk | 2004-12-02 | 1 | -88/+37 |
* | We already have a lock initialization function, use that for fdesc_mtx | phk | 2004-12-01 | 1 | -57/+28 |
* | Collect the stuff for the /dev/fd/{%d,std{in,out,err}} pseudo-device | phk | 2004-12-01 | 1 | -55/+52 |
* | "nfiles" is a bad name for a global variable. Call it "openfiles" instead | phk | 2004-12-01 | 1 | -6/+6 |
* | Style: move data to top of file. | phk | 2004-12-01 | 1 | -4/+4 |
* | Don't acquire Giant before calling closef() in close() (and elsewhere); | rwatson | 2004-11-28 | 1 | -7/+7 |
* | Fix LOR. | phk | 2004-11-26 | 1 | -2/+2 |
* | Neither of the arguments to closef() can be NULL anymore, so don't | das | 2004-11-21 | 1 | -4/+1 |
* | Move a FILEDESC_UNLOCK up to maintain correct nesting of FILEDESC/FILE | phk | 2004-11-16 | 1 | -11/+9 |
* | Make FILE_LOCK and FILEDESC_LOCK nest properly by postponing the the | phk | 2004-11-15 | 1 | -2/+2 |
* | Move #define up. | phk | 2004-11-14 | 1 | -5/+5 |
* | Introduce an alias for FILEDESC_{UN}LOCK() with the suffix _FAST. | phk | 2004-11-13 | 1 | -15/+15 |
* | Use more intuitive pointer for fdinit() and fdcopy(). | phk | 2004-11-08 | 1 | -2/+1 |
* | Introduce fdclose() which will clean an entry in a filedesc. | phk | 2004-11-07 | 1 | -1/+17 |
* | Move fdinit() related stuff from .h to .c | phk | 2004-11-07 | 1 | -0/+35 |
* | Allow fdinit() to be called with a NULL fdp argument so we can use | phk | 2004-11-07 | 1 | -15/+16 |
* | When we open /dev/null for stdin/out/err for safety reasons, do it right: | phk | 2004-11-06 | 1 | -3/+5 |
* | Push acquisition of the accept mutex out of sofree() into the caller | rwatson | 2004-10-18 | 1 | -0/+1 |
* | Another case where we need to guard against a partially | julian | 2004-10-04 | 1 | -0/+2 |
* | Remove GIANT_REQUIRED from setugidsafety() as knote_fdclose() no longer | rwatson | 2004-08-19 | 1 | -2/+0 |
* | Add the missing knote_fdclose(). | green | 2004-08-16 | 1 | -2/+4 |
* | Add locking to the kqueue subsystem. This also makes the kqueue subsystem | jmg | 2004-08-15 | 1 | -20/+7 |
* | We're not yet ready to assert !Giant in kern_fcntl(), as it's called | rwatson | 2004-08-07 | 1 | -5/+4 |
* | Avoid acquiring Giant for some common light-weight or already MPSAFE | rwatson | 2004-08-06 | 1 | -2/+31 |
* | Assert Giant in the following file descriptor-related functions: | rwatson | 2004-08-04 | 1 | -0/+10 |
* | Push Giant acquisition down into fo_stat() from most callers. Acquire | rwatson | 2004-07-22 | 1 | -6/+0 |
* | Push acquisition of Giant from fdrop_closed() into fo_close() so that | rwatson | 2004-07-22 | 1 | -2/+0 |
* | In addition to the real user ID check, do an explicit jail | csjp | 2004-07-14 | 1 | -2/+3 |
* | Explicitly initialize f_data and f_vnode to NULL. | phk | 2004-06-19 | 1 | -0/+3 |
* | Do the dreaded s/dev_t/struct cdev */ | phk | 2004-06-16 | 1 | -2/+2 |
* | Extend coverage of SOCK_LOCK(so) to include so_count, the socket | rwatson | 2004-06-12 | 1 | -0/+3 |
* | Deorbit COMPAT_SUNOS. | phk | 2004-06-11 | 1 | -2/+2 |
* | Push the VOP_ADVLOCK() call to release advisory locks on vnode file | rwatson | 2004-06-01 | 1 | -11/+1 |
* | Remove advertising clause from University of California Regent's license, | imp | 2004-04-05 | 1 | -4/+0 |
* | Conditionally assert Giant in fputsock() based on the value of | rwatson | 2004-03-29 | 1 | -0/+1 |
* | Split the mlock() kernel code into two parts, mlock(), which unpacks | truckman | 2004-02-26 | 1 | -1/+3 |
* | Device megapatch 4/6: | phk | 2004-02-21 | 1 | -0/+2 |
* | Don't bother storing a result when all you need are the side effects. | des | 2004-02-16 | 1 | -5/+5 |
* | In fdcheckstd the descriptor table should never be shared, so just | dwmalone | 2004-02-15 | 1 | -9/+7 |
* | Locking for the per-process resource limits structure. | jhb | 2004-02-04 | 1 | -7/+16 |
* | Restore correct semantics for F_DUPFD fcntl. This should fix the errors | des | 2004-01-17 | 1 | -4/+4 |
* | WITNESS won't let us hold two filedesc locks at the same time, so juggle | des | 2004-01-16 | 1 | -3/+9 |
* | Remove two KASSERTs which were overly paranoid. | des | 2004-01-16 | 1 | -4/+0 |