Up to [DragonFly] / src / sys / kern
Request diff between arbitrary revisions
Keyword substitution: kv
Default branch: MAIN
MFC numerous features from HEAD. * NFS export support for nullfs mounted filesystems, intended for nullfs mounted hammer PFSs. * Each nullfs mount constructs a unique fsid based on the underlying mount. * Each nullfs mount maintains its own netexport structure. * The mount pointer in the nch (namecache handle) is passed into FHTOVP and friends, allowing operations to occur on the underlying vnodes but still go through the nullfs mount.
* Implement the ability to export NULLFS mounts via NFS. * Enforce PFS isolation when exporting a HAMMER PFS via a NULLFS mount. NOTE: Exporting anything other then HAMMER PFS root's via nullfs does NOT protect the parent of the exported directory from being accessed via NFS. Generally speaking this feature is implemented by giving each nullfs mount a synthesized fsid based on what is being mounted and implementing the NFS export infrastructure in the nullfs code instead of just bypassing those functions to the underyling VFS.
MFC 1.117 - fix desiredvnodes calculation for machines with >2G ram. Requested-by: Francois Tigeot <email@example.com>
Adjust the desiredvnodes (kern.maxvnodes) calculation for machines with 3G+ of ram to prevent it from blowing out KVM. Reported-by: Michael Neumann <firstname.lastname@example.org>
Correct a bug in the last commit.
Add a vclean_unlocked() call that allows HAMMER to try to get rid of a vnode.
Fix a number of core kernel issues related to HAMMER operation. * The cluster code was incorrectly using the maximum IO size from the filesystem on which /dev is mounted instead of the maximum IO size of the block device. This became evident when HAMMER (with 16K blocks) tried to call cluster_read() via /dev/ad6s1h (on UFS with 8K blocks). * Change the way the VNLRU code works to avoid an infinite loop in vmntvnodescan(). The vnode LRU recycling code was cycling vnodes from the head of mp->mnt_nvnodelist to the tail. Under certain heavy load conditions this could cause a vmntvnodescan() to never finish running and eventually hit a count assertion (at 1,000,000 vnodes scanned). Instead of cycling the vnodes in the mnt_nvnodelist, use the syncer vnode (mount->mnt_syncer) as a placemarker and move *IT* within the list to represent the LRU scan. By not cycling vnodes to the end of the list, vmntvnodescan() can no longer get into an infinite loop. * Change the mount->mnt_syncer logic slightly to avoid races against a background sync while unmounting. The field is no longer cleared by the sync_reclaim() call but is instead cleared by the unmount code before vrele()ing the special vnode.
Fix a race between the namecache and the vnode recycler. A vnode cannot be recycled if it's namecache entry represents a directory with locked children. The various VOP_N*() functions require the parent dvp to be stable. The main fix is in vrecycle() (kern/vfs_subr.c). Do not vgone() the vnode if we can't clean out the children. Also create an API to assert that the parent dvp is stable, and make it vhold/vdrop the dvp. The race primarily effected HAMMER which uses the VOP_N*() API.
Have vfsync() call buf_checkwrite() on buffers with bioops to determine whether it is ok to write out a buffer or not. Used by HAMMER to prevent specfs from syncing out meta-data at the wrong time.
* Implement a mountctl() op for setting export control on a filesystem. * Adjust mountd to try to use the mountctl() op BEFORE calling a UFS-style mount() to set export ops for a filesystem. * Add a prototype for the mountctl() system call in sys/mountctl.h * Cleanup WARNS for the mountctl utility.
For kmalloc(), MALLOC() and contigmalloc(), use M_ZERO instead of explicitly bzero()ing. Reviewed-by: sephe
Add bio_ops->io_checkread and io_checkwrite - a read and write pre-check which gives HAMMER a chance to set B_LOCKED if the kernel wants to write out a passively held buffer. Change B_LOCKED semantics slightly. B_LOCKED buffers will not be written until B_LOCKED is cleared. This allows HAMMER to hold off B_DELWRI writes on passively held buffers.
Modify struct vattr: Increase va_nlink, va_fileid (the inode number), and va_gen from 32 bit to 64 bit integers. Add va_uid_uuid, va_gid_uuid, and va_fsid_uuid, and flags to indicate that these fields are valid. The original va_uid and va_gid are retained. This change has no external visibility. Modify struct statvfs: Use spare fields to add f_fsid_uuid and f_uid_uuid to the structure, and flags indicating that those fields are valid. This change has minimal external visibility. The size of the structure has not changed. Modify struct stat: Add a new file type S_IFDB. DB files are like regular files but access data on a record by record basis. The seek position is a 64 bit record key and not a byte offset. Further work in this area will be done later on to support related UIO operations. This change has minimal external visibility. The size of the structure has not changed.
Reactivate a vnode after associated it with deadfs after a forced unmount. This fixes numerous system panics that can occur due to the vnode's unexpected change in state. Submitted-by: "Nicolas Thery" <email@example.com>
Synchronize all changes made in HEAD to date with the 1.10 release branch. * usbdevs update * header file fixes * vinum root * vinum device I/O fixes * MD fixes * New PCI ids for netif rum and ural * New USB uplcom ids * linux exec memory leak * devclass ordering fixes (sound devices) * rate-limited kprintf support (filesystem full console spams) * msdosfs fixes * Manual page work
vrecycle() is typically called from a VFS's inactive function, which in turn is called when the sysref reference count transitions from 1->0xc0000000. Fix a test that was causing the vnode to not be immediately recycled. msdosfs depended on the recyclement behavior due to rename-over issues - msdosfs is not allowed to reuse an 'inode' (which is basically the position of a directory entry in a directory) until the underlying file is entirely reclaimed. This also effected ufs somewhat by preventing inodes from being reused as quickly as they could be, but ufs allocates inodes dynamically and masked the problem. Reported-by: walt <firstname.lastname@example.org>
Formalize the object sleep/wakeup code when waiting on a dead VM object and remove spurious calls to wakeup().
Give the device major / minor numbers their own separate 32 bit fields in the kernel. Change dev_ops to use a RB tree to index major device numbers and remove the 256 device major number limitation. Build a dynamic major number assignment feature into dev_ops_add() and adjust ASR (which already had a hand-rolled one), and MFS to use the feature. MFS at least does not require any filesystem visibility to access its backing device. Major devices numbers >= 256 are used for dynamic assignment. Retain filesystem compatibility for device numbers that fall within the range that can be represented in UFS or struct stat (which is a single 32 bit field supporting 8 bit major numbers and 24 bit minor numbers).
Replace NOCDEV with NULL. NOCDEV was ((void *)-1) and as inherited from *BSD a long time ago due to the device pointer / device number duality. Now that the pointer and device number has been separated, we can just use NULL to indicate no-pointer. Replace si_refs with si_sysref. Use SYSREF for ref-count cdev_t. Enable cdev_t reclamation on deletion.
Use SYSREF to reference count struct vnode. v_usecount is now v_sysref(.refcnt). v_holdcnt is now v_auxrefs. SYSREF's termination state (using a negative reference count from -0x40000000+) now places the vnode in a VCACHED or VFREE state and deactivates it. The vnode is now assigned a 64 bit unique id via SYSREF. vhold() (which manipulates v_auxrefs) no longer reactivates a vnode and is explicitly used only to track references from auxillary structures and references to prevent premature destruction of the vnode. vdrop() will now only move a vnode from VCACHED to VFREE on the 1->0 transition of v_auxrefs if the vnode is in a termination state. vref() will now panic if used on a vnode in a termination state. vget() must now be used to explicitly reactivate a vnode. These requirements existed before but are now explicitly asserted. vlrureclaim() and allocvnode() should now interact a bit better. In particular, vlrureclaim() will do a better job of finding vnodes to flush and transition from VCACHED to VFREE, and allocvnode() will do a better job finding vnodes to reuse without getting blocked by a flush. allocvnode now uses a real VX lock to sequence vnodes into VRECLAIMED. All vnode special state processing now uses a VX lock. Vnodes are now able to be slowly returned to the memory pool when kern.maxvnodes is reduced at run time. Various initialization elements have been moved to CTOR/DTOR and are no longer in the critical path, improving performance. However, since SYSREF uses atomic_cmpset_int() (aka cmpxchgl), which reduces performance somewhat, overall performance tends to be about the same.
Introduce globals: KvaStart, KvaEnd, and KvaSize. Used by the kernel instead of the nutty VADDR and VM_*_KERNEL_ADDRESS macros. Move extern declarations for these variables as well as for virtual_start, virtual_end, and phys_avail from MD headers to MI headers. Make kernel_object a global structure instead of a pointer. Remove kmem_object and all related code (none of it is used any more).
Rename printf -> kprintf in sys/ and add some defines where necessary (files which are used in userland, too).
1:1 Userland threading stage 2.9/4: Push out p_thread a little bit more
Change the kernel dev_t, representing a pointer to a specinfo structure, to cdev_t. Change struct specinfo to struct cdev. The name 'cdev' was taken from FreeBSD. Remove the dev_t shim for the kernel. This commit generally removes the overloading of 'dev_t' between userland and the kernel. Also fix a bug in libkvm where a kernel dev_t (now cdev_t) was not being properly converted to a userland dev_t.
Rename the kernel NODEV to NOCDEV to avoid conflicts with the userland NODEV.
Rename struct specinfo into struct cdev. Add a new typedef 'cdev_t' for cdev pointers. Temporarily retain dev_t for cdev pointers until the kernel can be converted over to cdev_t.
Rename malloc->kmalloc, free->kfree, and realloc->krealloc. Pass 1
VNode sequencing and locking - part 3/4. VNode aliasing is handled by the namecache (aka nullfs), so there is no longer a need to have VOP_LOCK, VOP_UNLOCK, or VOP_ISSLOCKED as 'VOP' functions. Both NFS and DEADFS have been using standard locking functions for some time and are no longer special cases. Replace all uses with native calls to vn_lock, vn_unlock, and vn_islocked. We can't have these as VOP functions anyhow because of the introduction of the new SYSLINK transport layer, since vnode locks are primarily used to protect the local vnode structure itself.
VNode sequencing and locking - part 2/4. Control access to v_usecount and v_holdcnt with the vnode's lock's spinlock. Use the spinlock to interlock the VRECLAIMED and VINACTIVE flags during 1->0 and 0->1 transitions. N->N+1 transitions do not need to obtain the spinlock and simply use a locked bus cycle increment. Vnode operations are still not MP safe but this gets further along that road. The lockmgr can no longer fail when obtaining an exclusive lock, remove the error code return from vx_lock() and vx_get(). Add special lockmgr support routines to atomically acquire and release an exclusive lock when the caller is already holding the spinlock. The removal of vnodes from the vnode free list is now defered. Removal only occurs when allocvnode() encounters a vnode on the list which should not be on it. This improves critical code paths for vget(), vput() and vrele() by removing unnecessary manipulation of the freelist. Fix a lockmgr bug where wakeup() was being called with a spinlock held. Instead, defer the wakeup until after the spinlock is released.
VNode sequencing and locking - part 1/4. Separate vref() for the case where the ref count is already non-zero (which is nearly all uses of vref()), vs the case where it might be zero. Clean up the code in preparation for putting it under a spinlock.
Remove several layers in the vnode operations vector init code. Declare the operations vector directly instead of via a descriptor array. Remove most of the recalculation code, it stopped being needed over a year ago. This work is similar to what FreeBSD now does, but was developed along a different line. Ultimately our vop_ops will become SYSLINK ops for userland VFS and clustering support.
Disassociate the VM object after calling VOP_INACTIVE instead of before. VOP_INACTIVE may have to do some work on the vnode that requires a functional buffer cache. For example, UFS may have to truncate a removed file.
Cleanup crit_*() usage to reduce bogus warnings printed to the console when a kernel is compiled with DEBUG_CRIT_SECTIONS. NOTE: DEBUG_CRIT_SECTIONS does a direct pointer comparison rather than a strcmp in order to reduce overhead. Supply a string constant in cases where the string identifier might be (intentionally) different otherwise.
Remove an inappropriate crit_exit() in ehci.c and add a missing crit_exit() in kern/vfs_subr.c. Specify string IDs in vfsync_bp() so we don't get complaints on the console when the kernel is compiled with DEBUG_CRIT_SECTIONS. The missing crit_exit() in kern/vfs_subr.c was causing the kernel to leave threads in a critical section, causing interrupts to stop operating and cpu-bound userland programs to lock up the rest of the system. Reported-by: Sascha Wildner <email@example.com>, others.
Add some diagnostic messages to try to catch a ufs_dirbad panic before it happens. MFC: Reorder BUF_UNLOCK() - it must occur after b_flags is modified, not before. A newly created non-VMIO buffer is now marked B_INVAL. Callers of getblk() now always clear B_INVAL before issuing a READ I/O or when clearing or overwriting the buffer. Before this change, a getblk() (getnewbuf), brelse(), getblk() sequence on a non-VMIO buffer would result in a buffer with B_CACHE set yet containing uninitialized data. MFC: B_NOCACHE cannot be set on a clean VMIO-backed buffer as this will destroy the VM backing store, which might be dirty. MFC: Reorder vnode_pager_setsize() calls to close a race condition.
Fix several buffer cache issues related to B_NOCACHE. * Do not set B_NOCACHE when calling vinvalbuf(... V_SAVE). This will destroy dirty VM backing store associated with clean buffers before the VM system has a chance to check for and flush them. Taken-from: FreeBSD * Properly set B_NOCACHE when destroying buffers related to truncated data. * Fix a bug in vnode_pager_setsize() that was recently introduced. v_filesize was being set before a new/old size comparison, causing a file truncation to not destroy related VM pages past the new EOF. * Remove a bogus B_NOCACHE|B_DIRTY test in brelse(). This was originally intended to be a B_NOCACHE|B_DELWRITE test which then cleared B_NOCACHE, but now that B_NOCACHE operation has been fixed it really does indicate that the buffer, its contents, and its backing store are to be destroyed, even if the buffer is marked B_DELWRI. Instead of clearing B_NOCACHE when B_DELWRITE is found to be set, clear B_DELWRITE when B_NOCACHE is found to be set. Note that B_NOCACHE is still cleared when bdirty() is called in order to ensure that data is not lost when softupdates and other code do a 'B_NOCACHE + bwrite' sequence. Softupdates can redirty a buffer in its io completion hook and a write error can also redirty a buffer. * The VMIO buffer rundown seems to have mophed into a state where the distinction between NFS and non-NFS buffers can be removed. Remove the test.
Attempt to interlock races between the buffer cache and VM backing store that might cause new buffers to be instantiated beyond the new file EOF during a truncate operation. Truncate the VM object size before attempting to flush the pages and buffers in order to prevent new VM pages from being created beyond EOF during the flush. Add an extra pass on the buffer cache after truncation to make sure the buffers have been cleaned out. Generate a warning to the console if buffers are found during the extra pass. If an old buffer were left intact during a truncate, then a re-extension of the file or directory could have resulted in granting access to the old buffer which might have had an incorrect cached block number translation (vs the new block allocated by the extension of the file or directory), causing new data to be written to the wrong disk block and resulting in file or directory corruption. The regular file truncation/extension code had other checks in it prior to this patch so if this problem could occur at all before it would have been in the directory code. There is a small chance that this race was related to reported ufs: dirbad panics. The M.O. matches but unfortunately there is still no smoking gun.
Remove vnode lock assertions that are no longer used. Remove the IS_LOCKING_VFS() macro. All VFS's are required to be locking VFSs now.
Remove the thread argument from all mount->vfs_* function vectors, replacing it with a ucred pointer when applicable. This cleans up a considerable amount of VFS function code that previously delved into the process structure to get the cred, though some code remains. Get rid of the compatibility thread argument for hpfs and nwfs. Our lockmgr calls are now mostly compatible with NetBSD (which doesn't use a thread argument either). Get rid of some complex junk in fdesc_statfs() that nobody uses. Remove the thread argument from dounmount() as well as various other filesystem specific procedures (quota calls primarily) which no longer need it due to the lockmgr, VOP, and VFS cleanups. These cleanups also have the effect of making the VFS code slightly less dependant on the calling thread's context.
The thread/proc pointer argument in the VFS subsystem originally existed for... well, I'm not sure *WHY* it originally existed when most of the time the pointer couldn't be anything other then curthread or curproc or the code wouldn't work. This is particularly true of lockmgr locks. Remove the pointer argument from all VOP_*() functions, all fileops functions, and most ioctl functions.
Simplify vn_lock(), VOP_LOCK(), and VOP_UNLOCK() by removing the thread_t argument. These calls now always use the current thread as the lockholder. Passing a thread_t to these functions has always been questionable at best.
Remove VOP_BWRITE(). This function provided a way for a VFS to override the bwrite() function and was used *only* by NFS in order to allow NFS to handle the B_NEEDCOMMIT flag as part of NFSv3's 2-phase commit operation. However, over time, the handling of this flag was moved to the strategy code. Additionally, the kernel now fully supports the redirtying of buffers during an I/O (which both softupdates and NFS need to be able to do). The override is no longer needed. All former calls to VOP_BWRITE() now simply call bwrite().
Remove b_xflags. Fold BX_VNCLEAN and BX_VNDIRTY into b_flags as B_VNCLEAN and B_VNDIRTY. Remove BX_AUTOCHAINDONE and recode the swap pager to use one of the caller data fields in the BIO instead.
Get rid of pbgetvp() and pbrelvp(). Instead fold the B_PAGING flag directly into getpbuf() (the only type of buffer that pbgetvp() could be called on anyway). Change related b_flags assignments from '=' to '|='. Get rid of remaining depdendancies on b_vp. vn_strategy() now relies solely on the vp passed to it as an argument. Remove buffer cache code that sets b_vp for anonymous pbuf's. Add a stopgap 'vp' argument to vfs_busy_pages(). This is only really needed by NFS and the clustering code do to the severely hackish nature of the NFS and clustering code. Fix a bug in the ext2fs inode code where vfs_busy_pages() was being called on B_CACHE buffers. Add an assertion to vfs_busy_pages() to panic if it encounters a B_CACHE buffer.
Get rid of the weird FSMID update path in the vnode and namecache code. Instead, mark the vnode as needing an FSMID update when the vnode is disconnected from the namecache. This fixes a bug where FSMID updates were being lost at unmount time.
vfsync() is not in the business of removing buffers beyond the file EOF. Remove the procedural argument and related code.
Remove unused code label.
MFC vfs_bio.c 1.57, vfs_subr.c 1.69 - fix race condition in vfs_bio_awrite().
Due to continuing issues with VOP_READ/VOP_WRITE ops being called without a VOP_OPEN, particularly by NFS, redo the way VM objects are associated with vnodes. * The size of the object is now passed to vinitvmio(). vinitvmio() no longer calls VOP_GETATTR(). * Instead of trying to call vinitvmio() conditionally in various places, we now call it unconditionally when a vnode is instantiated if the filesystem at any time in the future intends to use the buffer cache to access that vnode's dataspace. * Specfs 'disk' devices are an exception. Since we cannot safely do I/O on such vnodes if they have not been VOP_OPEN()'ed anyhow, the VM objects for those vnodes are still only associated on open. The performance impact is limited to the case where large numbers of vnodes are being created and destroyed. This case only occurs when a large directory topology (number of files > kernel's vnode cache) is traversed and all related inodes are cached by the system. Being a pure-cpu case the slight loss of performance due to the VM object allocations is not really a big dael.
Use the vnode v_opencount and v_writecount universally. They were previously only used by specfs. Require that VOP_OPEN and VOP_CLOSE calls match. Assert on boundary errors. Clean up umount's FORCECLOSE mode. Adjust deadfs to allow duplicate closes (which can happen due to a forced unmount or revoke). Add vop_stdopen() and vop_stdclose() and adjust the default vnode ops to call them. All VFSs except DEADFS which supply their own vop_open and vop_close now call vop_stdopen() and vop_stdclose() to handle v_opencount and v_writecount adjustments. Change the VOP_OPEN/fp specs. VOP_OPEN (aka vop_stdopen) is now responsible for filling in the file pointer information, rather than the caller of VOP_OPEN. Additionally, when supplied a file pointer, VOP_OPEN is now allowed to populate the file pointer with a different vnode then the one passed to it, which will be used later on to allow filesystems which synthesize different vnodes on open, for example so we can create a generic tty/pty pairing devices rather than scanning for an unused pty, and so we can create swap-backed generic anonymous file descriptors rather than having to use /tmp. And for other purposes as well. Fix UFS's mount/remount/unmount code to make the proper VOP_OPEN and VOP_CLOSE calls when a filesystem is remounted read-only or read-write.
A VM object is now required for vnode-based buffer cache ops. This is usually handled by VOP_OPEN but there are a few cases where UFS issues buffer cache ops on vnodes that have not been opened, such as when creating a new directory or softlink.
Remove VOP_GETVOBJECT, VOP_DESTROYVOBJECT, and VOP_CREATEVOBJECT. Rearrange the VFS code such that VOP_OPEN is now responsible for associating a VM object with a vnode. Add the vinitvmio() helper routine.
Major BUF/BIO work commit. Make I/O BIO-centric and specify the disk or file location with a 64 bit offset instead of a 32 bit block number. * All I/O is now BIO-centric instead of BUF-centric. * File/Disk addresses universally use a 64 bit bio_offset now. bio_blkno no longer exists. * Stackable BIO's hold disk offset translations. Translations are no longer overloaded onto a single structure (BUF or BIO). * bio_offset == NOOFFSET is now universally used to indicate that a translation has not been made. The old (blkno == lblkno) junk has all been removed. * There is no longer a distinction between logical I/O and physical I/O. * All driver BUFQs have been converted to BIOQs. * BMAP, FREEBLKS, getblk, bread, breadn, bwrite, inmem, cluster_*, and findblk all now take and/or return 64 bit byte offsets instead of block numbers. Note that BMAP now returns a byte range for the before and after variables.
Replace the global buffer cache hash table with a per-vnode red-black tree. Add a B_HASHED b_flags bit as a sanity check. Remove the invalhash junk and replace with assertions in several cases where the buffer must already not be hashed. Get rid of incore() and gbincore() and replace with a new function called findblk(). Merge the new RB management with bgetvp(), the two are now fully integrated. Previous work has turned reassignbuf() into a mostly degenerate call, simplify its arguments and functionality to match. Remove an unnecessary reassignbuf() call from the NFS code. Get rid of pbreassignbuf(). Adjust the code in several places where it was assumed that calling BUF_LOCK() with LK_SLEEPFAIL after previously failing with LK_NOWAIT would always fail. This code was used to sleep before a retry. Instead, if the second lock unexpectedly succeeds, simply issue an unlock and retry anyway. Testing-by: Stefan Krueger <firstname.lastname@example.org>
vfs_bio_awrite() was unconditionally locking a buffer without checking for races, potentially resulting in the wrong buffer, an invalid buffer, or a recently replaced buffer being written out. Change the call semantics to require a locked buffer to be passed into the function rather then locking the buffer in the function.
buftimespinlock is utterly useless since the spinlock is released within lockmgr(). The only real problem was with lk_prio, which no longer exists, so get rid of the spin lock and document the remaining passive races.
Pass LK_PCATCH instead of trying to store tsleep flags in the lock structure, so multiple entities competing for the same lock do not use unexpected flags when sleeping. Only NFS really uses PCATCH with lockmgr locks.
Make the entire BUF/BIO system BIO-centric instead of BUF-centric. Vnode and device strategy routines now take a BIO and must pass that BIO to biodone(). All code which previously managed a BUF undergoing I/O now manages a BIO. The new BIO-centric algorithms allow BIOs to be stacked, where each layer represents a block translation, completion callback, or caller or device private data. This information is no longer overloaded within the BUF. Translation layer linkages remain intact as a 'cache' after I/O has completed. The VOP and DEV strategy routines no longer make assumptions as to which translated block number applies to them. The use the block number in the BIO specifically passed to them. Change the 'untranslated' constant to NOOFFSET (for bio_offset), and (daddr_t)-1 (for bio_blkno). Rip out all code that previously set the translated block number to the untranslated block number to indicate that the translation had not been made. Rip out all the cluster linkage fields for clustered VFS and clustered paging operations. Clustering now occurs in a private BIO layer using private fields within the BIO. Reformulate the vn_strategy() and dev_dstrategy() abstraction(s). These routines no longer assume that bp->b_vp == the vp of the VOP operation, and the dev_t is no longer stored in the struct buf. Instead, only the vp passed to vn_strategy() (and related *_strategy() routines for VFS ops), and the dev_t passed to dev_dstrateg() (and related *_strategy() routines for device ops) is used by the VFS or DEV code. This will allow an arbitrary number of translation layers in the future. Create an independant per-BIO tracking entity, struct bio_track, which is used to determine when I/O is in-progress on the associated device or vnode. NOTE: Unlike FreeBSD's BIO work, our struct BUF is still used to hold the fields describing the data buffer, resid, and error state. Major-testing-by: Stefan Krueger
An exclusive lock on the vnode is required when running vm_object_page_clean(), otherwise a balloc may occur without the vnode/inode held locked. There is a possibility that this bug was responsible for some filesystem corrupted. Reported-by: numerous people after a sanity assertion was committed to the ffs_balloc code.
Add an argument to vfs_add_vnodeops() to specify VVF_* flags for the vop_ops structure. Add a new flag called VVF_SUPPORTS_FSMID to indicate filesystems which support persistent storage of FSMIDs. Rework the FSMID code a bit to reduce overhead. Use the spare field in the UFS inode structure to implement a persistent FSMID. The FSMID is recursively marked in the namecache but not adjusted until the next getattr() call on the related inode(s), or when the vnode is reclaimed.
Make struct dirent contain a full 64bit inode. Allow more than 255 byte filenames by increasing d_namlen to 16bit. Remove UFS specific macros from sys/dirent.h, programs which really need them should include vfs/ufs/dir.h. MAXNAMLEN should not be used, but replaced by NAME_MAX. To keep the impact for older BSD code small, d_ino and d_fileno are kept in the old meaning when __BSD_VISIBLE is defined, otherwise the POSIX version d_ino is used. This will be changed later to always define only d_ino and make d_fileno a compatiblity macro for __BSD_VISIBLE. d_name is left with hard-coded 256 byte space, this will be changed at some point in the future and doesn't affect the ABI. Programs should correctly allocate space themselve, since the maximum directory entry length can be > 256 byte. For allocating dirents (e.g. for readdir_r), _DIRENT_RECLEN and _DIRENT_DIRSIZ should be used. NetBSD has choosen the same names. Revamp the compatibility code to always use a local kernel buffer and write out the entries. This will be changed later by passing down the output function to vop_readdir, elimininating the redundant copy. Change NFS and CD9660 to use to use vop_write_dirent, for CD9660 ensure that the buffers are big enough by prepending char arrays of the right size. Tested-by & discussed-with: dillon
Add a sanity check for the length of the file name to vop_write_dirent().
Fix merge bug. d_namlen is used by GENERIC_DIRSIZ, when it isn't initialised, the argument to bzero is wrong.
Add vop_write_dirent helper functions, which isolates the caller from the layout and setup of struct dirent.
When allocating memory for the index file, query the filesystem for the maximum entry name first and use that.
Add vn_get_namelen to simplify correct emulation of statfs with maximum name length field. Discussed-with: hmp
Remove spl*() calls from kern, replacing them with critical sections. Change the meaning of safepri from a cpl mask to a thread priority. Make a minor adjustment to tests within one of the buffer cache's critical sections.
MFC 1.56. Minor kernel stack memory disclosure. Security: FreeBSD-SA-05:08.kmem
Bring in fix from FreeBSD/cperciva: Log: If we are going to 1. Copy a NULL-terminated string into a fixed-length buffer, and 2. copyout that buffer to userland, we really ought to 0. Zero the entire buffer first. Security: FreeBSD-SA-05:08.kmem Thanks to Colin Percival for notifying us!
Abstract out the routines which manipulate the mountlist. Introduce an MP-safe mountlist scanning function. This function keeps track of scans which are in-progress and properly handles ripouts that occur during the callback by advancing the matching pointers being tracked. The callback can safely block without confusing the scan. This algorithm has already been successfully used for the buffer cache and will soon be used for the vnode lists hanging off the mount point.
Implement Red-Black trees for the vnode clean/dirty buffer lists. Implement ranged fsyncs and adjust the syncer to use the new capability. This capability will also soon be used to replace the write_behind heuristic. Rewrite the fsync code for all VFSs to use the new APIs (generally simplifying them). Get rid of B_WRITEINPROG, it is no longer useful or needed. Get rid of B_SCANNED, it is no longer useful or needed. Rewrite the NFS 2-phase commit protocol to take advantage of the new Red-Black tree topology. Add RB_SCAN() for callback-scanning of Red-Black trees. Give RB_SCAN the ability to track the 'next' scan node and automatically fix it up if the callback directly or indirectly or through blocking indirectly deletes nodes in the tree while the scan is in progress. Remove most related loop restart conditions, they are no longer necessary. Disable filesystem background bitmap writes. This really needs to be solved a different way and the concept does not work well with red-black trees.
Convert the struct domain next pointer to an SLIST.
Fix a cache_resolve() vs cache_inval() race which can result in a livelock. The namecache invalidation code was being a bit overzealous when asked to invalidate a subhierarchy. It was retrying until the subhierarchy was completely invalidated, re-invalidating new entries created after the initial call to cache_inval(). This can occur if the filesystem is heavily loaded and a high level directory is being recursively invalidated. It is unnecessary to retry in this case... the purpose is to invalidate as-of the call to cache_inval(), so it is acceptable to allow new entries to be resolved within the subhierarchy undergoing the invalidation. Certain higher level entities... rename, and vnode reclamation, require complete invalidation. The retry has been moved to a higher level for these entities. The basic cache_inval() code is now single-pass. Reported-by: Richard Nyberg <email@example.com>
Don't use the statfs field f_mntonname in filesystems. For the userland export code, it can synthesized from mnt_ncp. For debugging code, use f_mntfromname, it should be enough to find culprit. The vfs_unmountall doesn't use code_fullpath to avoid problems with resource allocation and to make it more likely that a call from ddb succeds. Change getfsstat and fhstatfs to not show directories outside a chroot path, with the exception of the filesystem counting the chroot root itself.
VFS messaging/interfacing work stage 10/99: Start adding the journaling, range locking, and (very slightly) cache coherency infrastructure. Continue cleaning up the VOP operations vector. Expand on past commits that gave each mount structure its own set of VOP operations vectors by adding additional vector sets for journaling or cache coherency operations. Remove the vv_jops and vv_cops fields from the vnode operations vector in favor of placing those vop_ops directly in the mount structure. Reorganize the VOP calls as a double-indirect and add a field to the mount structure which represents the current vnode operations set (which will change when e.g. journaling is turned on or off). This creates the infrastructure necessary to allow us to stack a generic journaling implementation on top of a filesystem. Introduce a hard range-locking API for vnodes. This API will be used by high level system/vfs calls in order to handle atomicy guarentees. It is a prerequisit for: (1) being able to break I/O's up into smaller pieces for the vm_page list/direct-to-DMA-without-mapping goal, (2) to support the parallel write operations on a vnode goal, (3) to support the clustered (remote) cache coherency goal, and (4) to support massive parallelism in dispatching operations for the upcoming threaded VFS work. This commit represents only infrastructure and skeleton/API work.
Clean up routing code before I parallelize it.
Cleanup some dangling issues with cache_inval(). A lot of hard work went into guarenteeing that the namecache topology would remain connected, but there were two cases (basically rmdir and rename-over-empty-target-dir) which disconnected a portion of the hierarchy. This fixes the remaining cases by having cache_inval() simply mark the namecache entry as destroyed without actually disconnecting it from the topology. The flag tells cache_nlookup() and ".." handlers that a node has been destroyed and is no longer connected to any parent directory. The new cache_inval() also now has the ability to mark an entire subhierarchy as being unresolved, which can be a useful feature to have. In-discussion-with: Richard Nyberg <firstname.lastname@example.org>
Default vfs.fastdev to 1 for wider testing, so the vnode bypass for device read and write is now the default. This is a precursor to the continued work on a kernel-managed cache layer on top of the VFS layer. That is, the intention is to eventually switch the VM page cache to be ABOVE the VFS layer rather then BELOW the VFS layer for standard read() and write() calls, with potentially major performance benefits.
VFS messaging/interfacing work stage 9/99: VFS 'NEW' API WORK. NOTE: unionfs and nullfs are temporarily broken by this commit. * Remove the old namecache API. Remove vfs_cache_lookup(), cache_lookup(), cache_enter(), namei() and lookup() are all gone. VOP_LOOKUP() and VOP_CACHEDLOOKUP() have been collapsed into a single non-caching VOP_LOOKUP(). * Complete the new VFS CACHE (namecache) API. The new API is able to supply topological guarentees and is able to reserve namespaces, including negative cache spaces (whether the target name exists or not), which the new API uses to reserve namespace for things like NRENAME and NCREATE (and others). * Complete the new namecache API. VOP_NRESOLVE, NLOOKUPDOTDOT, NCREATE, NMKDIR, NMKNOD, NLINK, NSYMLINK, NWHITEOUT, NRENAME, NRMDIR, NREMOVE. These new calls take (typicaly locked) namecache pointers rather then combinations of directory vnodes, file vnodes, and name components. The new calls are *MUCH* simpler in concept and implementation. For example, VOP_RENAME() has 8 arguments while VOP_NRENAME() has only 3 arguments. The new namecache API uses the namecache to lock namespaces without having to lock the underlying vnodes. For example, this allows the kernel to reserve the target name of a create function trivially. Namecache records are maintained BY THE KERNEL for both positive and negative hits. Generally speaking, the kernel layer is now responsible for resolving path elements. NRESOLVE is called when an unresolved namecache record needs to be resolved. Unlike the old VOP_LOOKUP, NRESOLVE is simply responsible for associating a vnode to a namecache record (positive hit) or telling the system that it's a negative hit, and not responsible for handling symlinks or other special cases or doing any of the other path lookup work, much unlike the old VOP_LOOKUP. It should be particularly noted that the new namecache topology does not allow disconnected namecache records. In rare cases where a vnode must be converted to a namecache pointer for new API operation via a file handle (i.e. NFS), the cache_fromdvp() function is provided and a new API VOP, VOP_NLOOKUPDOTDOT() is provided to allow the namecache to resolve the topology leading up to the requested vnode. These and other topological guarentees greatly reduce the complexity of the new namecache API. The new namei() is called nlookup(). This function uses a combination of cache_n*() calls, VOP_NRESOLVE(), and standard VOP calls resolve the supplied path, deal with symlinks, and so forth, in a nice small compact compartmentalized procedure. * The old VFS code is no longer responsible for maintaining namecache records, a function which was mostly adhoc cache_purge()s occuring before the VFS actually knows whether an operation will succeed or not. The new VFS code is typically responsible for adjusting the state of locked namecache records passed into it. For example, if NCREATE succeeds it must call cache_setvp() to associate the passed namecache record with the vnode representing the successfully created file. The new requirements are much less complex then the old requirements. * Most VFSs still implement the old API calls, albeit somewhat modified and in particular the VOP_LOOKUP function is now *MUCH* simpler. However, the kernel now uses the new API calls almost exclusively and relies on compatibility code installed in the default ops (vop_compat_*()) to convert the new calls to the old calls. * All kernel system calls and related support functions which used to do complex and confusing namei() operations now do far less complex and far less confusing nlookup() operations. * SPECOPS shortcutting has been implemented. User reads and writes now go directly to supporting functions which talk to the device via fileops rather then having to be routed through VOP_READ or VOP_WRITE, saving significant overhead. Note, however, that these only really effect /dev/null and /dev/zero. Implementing this was fairly easy, we now simply pass an optional struct file pointer to VOP_OPEN() and let spec_open() handle the override. SPECIAL NOTES: It should be noted that we must still lock a directory vnode LK_EXCLUSIVE before issuing a VOP_LOOKUP(), even for simple lookups, because a number of VFS's (including UFS) store active directory scanning information in the directory vnode. The legacy NAMEI_LOOKUP cases can be changed to use LK_SHARED once these VFS cases are fixed. In particular, we are now organized well enough to actually be able to do record locking within a directory for handling NCREATE, NDELETE, and NRENAME situations, but it hasn't been done yet. Many thanks to all of the testers and in particular David Rhodus for finding a large number of panics and other issues.
Remove the vfs page replacement optimization and its ENABLE_VFS_IOOPT option. This never worked properly... that is, the semantics are broken compared to a normal read or write in that the read 'buffer' will be modified out from under the caller if the underlying file is. What is really needed here is a copy-on-write feature that works in both directions, similar to how a shared buffer is copied after a fork() if either the parent or child modify it. The optimization will eventually rewritten with that in mind but not right now.
The old lookup() API is extremely complex. Even though it will be ripped out soon, I'm documenting the procedure so I don't have to keep running through it to figure out what is going on. Do a better job describing the new vgone() API (the old API required the vnode to be in a very weird state. The new API requires the vnode to be VX locked and refd and returns with the vnode in the same state).
VFS messaging/interfacing work stage 8/99: Major reworking of the vnode interlock and other miscellanious things. This patch also fixes FS corruption due to prior vfs work in head. In particular, prior to this patch the namecache locking could introduce blocking conditions that confuse the old vnode deactivation and reclamation code paths. With this patch there appear to be no serious problems even after two days of continuous testing. * VX lock all VOP_CLOSE operations. * Fix two NFS issues. There was an incorrect assertion (found by David Rhodus), and the nfs_rename() code was not properly purging the target file from the cache, resulting in Stale file handle errors during, e.g. a buildworld with an NFS-mounted /usr/obj. * Fix a TTY session issue. Programs which open("/dev/tty" ,...) and then run the TIOCNOTTY ioctl were causing the system to lose track of the open count, preventing the tty from properly detaching. This is actually a very old BSD bug, but it came out of the woodwork in DragonFly because I am now attempting to track device opens explicitly. * Gets rid of the vnode interlock. The lockmgr interlock remains. * Introduced VX locks, which are mandatory vp->v_lock based locks. * Rewrites the locking semantics for deactivation and reclamation. (A ref'd VX lock'd vnode is now required for vgone(), VOP_INACTIVE, and VOP_RECLAIM). New guarentees emplaced with regard to vnode ripouts. * Recodes the mountlist scanning routines to close timing races. * Recodes getnewvnode to close timing races (it now returns a VX locked and refd vnode rather then a refd but unlocked vnode). * Recodes VOP_REVOKE- a locked vnode is now mandatory. * Recodes all VFS inode hash routines to close timing holes. * Removes cache_leaf_test() - vnodes representing intermediate directories are now held so the leaf test should no longer be necessary. * Splits the over-large vfs_subr.c into three additional source files, broken down by major function (locking, mount related, filesystem syncer). * Changes splvm() protection to a critical-section in a number of places (bleedover from another patch set which is also about to be committed). Known issues not yet resolved: * Possible vnode/namecache deadlocks. * While most filesystems now use vp->v_lock, I haven't done a final pass to make vp->v_lock mandatory and to clean up the few remaining inode based locks (nwfs I think and other obscure filesystems). * NullFS gets confused when you hit a mount point in the underlying filesystem. * Only UFS and NFS have been well tested * NFS is not properly timing out namecache entries, causing changes made on the server to not be properly detected on the client if the client already has a negative-cache hit for the filename in question. Testing-by: David Rhodus <email@example.com>, Peter Kadau <firstname.lastname@example.org>, walt <email@example.com>, others
VFS messaging/interfacing work stage 7d/99: More firming up of stage 7. Additional work to deal with old-api/new-api issues. Cut more stuff out of the old-api's cache_enter() routine to deal with deadlocks, at the cost of some performance loss (temporary until the VFS's start using the new APIs). Change UFS and NFS to not purge whole directories in *_rename() and *_rmdir(). Add some minor breakage to the API which will not be fixed until the VFS's get new rename implementations - renaming a directory in which a process has chdir'd will create problems for that process. This doesn't happen normally anyway so this temporary breakage should not cause any significant problems. Bug-reports-by: walt, Sascha Wildner, others
VFS messaging/interfacing work stage 5b/99. More cleanups, remove the (unused) ni_ncp and ni_dncp from struct nameidata. A new structure will be used for the new API.
Remove unused variable.
Fix a bug in sillyrename handling in nfs_inactive(). The code was improperly ignoring the lock state of the passed vp and recursing nfs_inactive() by calling vrele() from within nfs_inactive(). Since NFS uses real vnode locking now, this resulted in a panic. KDE startup problems reported by: Emiel Kollof <firstname.lastname@example.org>
VFS messaging/interfacing work stage 4/99. This stage goes a long ways towards allowing us to move the vnode locking into a kernel layer. It gets rid of a lot of cruft from FreeBSD-4. FreeBSD-5 has done some of this stuff too (such as changing the default locking to stdlock from nolock), but DragonFly is going further. * Consolidate vnode locks into the vnode structure, add an embedded v_lock, and getting rid of both v_vnlock and v_data based head-of-structure locks. * Change the default vops to use a standard vnode lock rather then a fake non-lock. * Get rid of vop_nolock() and friends, we no longer support non-locking vnodes. * Get rid of vop_sharedlock(), we no longer support non standard shared-only locks (only NFS was using it and the mount-crossing lookup code should now prevent races to root from dead NFS volumes). * Integrate lock initialization into getnewvnode(). We do not yet incorporate automatically locking into getnewvnode(). getnewvnode() now has two additional arguments, lktimeout and lkflags, for lock structure initialization. * Change the sync vnode lock from nolock to stdlock. This may require more tuning down the line. Fix various sync_inactive() to properly unlock the lock as per the VOP API. * Properly flag the 'rename' vop operation regarding required tdvp and tvp unlocks (the flags are only used by nullfs). * Get rid of all inode-embedded vnode locks * Remove manual lockinit and use new getnewvnode() args instead. Lock the vnode prior to doing anything that might block in order to avoid synclist access before the vnode has been properly initialize. * Generally change inode hash insertion to also check for a hash collision and return failure if it occurs, rather then doing (often non-atomic) relookups and other checks. These sorts of collisions can occur if a vnode is being destroyed at the same time a new vnode is being created from an inode. A new vnode is not generally accessible, except by the sync code (from the mountlist) until it's underlying inode has been hashed so dealing with a hash collision should be as simple as throwing away the vnode with a vput(). * Do not initialize a new vnode's v_data until after the associated inode has been successfully added to the hash, and make the xxx_inactive() and xxx_reclaim() code friendly towards vnodes with a NULL v_data. * NFS now uses standard locks rather then shared-only locks. * PROCFS now uses standard locks rather then non-locks, and PROCFS's lookup code now understands VOP lookup semantics. PROCFS now uses a real hash table for its node search rather then a single singly-linked list (which should better scale to systems with thousands of processes). * NULLFS should now properly handle lookup() and rename() locks. NULLFS's node handling code has been rewritten. NULLFS's bypass code now understands vnode unlocks (rename case). * UFS no longer needs the ffs_inode_hash_lock hacks. It now uses the new collision-on-hash-add methodology. This will speed up UFS when operating on lots of small files (reported by David Rhodus).
VFS messaging/interfacing work stage 2/99. This stage retools the vnode ops vector dispatch, making the vop_ops a per-mount structure rather then a per-filesystem structure. Filesystem mount code, typically in blah_vfsops.c, must now register various vop_ops pointers in the struct mount to compile its VOP operations set. This change will allow us to begin adding per-mount hooks to VFSes to support things like kernel-level journaling, various forms of cache coherency management, and so forth. In addition, the vop_*() calls now require a struct vop_ops pointer as the first argument instead of a vnode pointer (note: in this commit the VOP_*() macros currently just pull the vop_ops pointer from the vnode in order to call the vop_*() procedures). This change is intended to allow us to divorce ourselves from the requirement that a vnode pointer always be part of a VOP call. In particular, this will allow namespace based routines such as remove(), mkdir(), stat(), and so forth to pass namecache pointers rather then locked vnodes and is a very important precursor to the goal of using the namecache for namespace locking.
VFS messaging/interfacing work stage 1/99. This stage replaces the old dynamic VFS descriptor and inlined wrapper mess with a fixed structure and fixed procedural wrappers. Most of the work is straightforward except for vfs_init, which was basically rewritten (and greatly simplified). It is my intention to make the vop_*() call wrappers eventually handle range locking and cache coherency issues as well as implementing the direct call -> messaging interface layer. The call wrappers will also API translation as we shift the APIs over to new, more powerful mechanisms in order to allow the work to be incrementally committed. This is the first stage of what is likely to be a huge number of stages to modernize the VFS subsystem.
There was a mountlist race in getnewvnode() whereby the system could block obtaining the mountlist token while adding a vnode to the mountlist prior to initializing the vnode's v_usecount and v_data fields. This bug is possibly responsible for or related to occassional reports of duplicate inodes in the system. Fix the potential problem by more completely initializing the vnode prior to adding it to the mountlist. Note that FreeBSD-5 also rearranged thei r code along the same lines (though this change is independant of their work).
When booting from CD, check cd1a and acd1a after cd0a and acd0a, allowing a CD to be booted off of a second CDRom drive. When checking for the same rootdev selection as already has been set, only the major number of the existing rootdev was being checked. This prevent other unit numbers from being tried properly (e.g. trying cd1 after having already tried cd0). Completely parse the rootdev's (unit, slice, partition) tuple before trying to look up the device, rather then assuming that devname(0,0,0) will exist (this is no longer necessarily true).
Fix a race with the clearing of p->p_session->s_ttyvp. NULL the pointer out before calling vrele() rather then after. Fix a bug with v_opencount accounting on revoke(). The underlying device was being closed properly but v_opencount was being decremented which causes it to go negative when close() is called on the descriptor later on. To fix the bug we zero out v_opencount() when the underlying vnode's device is disassociated and spec_close() now only decrements it when the device is associated. Reported-by: GeekGod Testing-by: GeekGod, Hiten, David Rhodus.
ANSIfication and general cleanup. No operational changes.
Cleanup pass. Removed code that is not needed anymore. Cleanup VOP_LEASE() uses and document. Add in a debug function for buffer pool statistical information which can be toggled via debug.syncprt.
Device layer rollup commit. * cdevsw_add() is now required. cdevsw_add() and cdevsw_remove() may specify a mask/match indicating the range of supported minor numbers. Multiple cdevsw_add()'s using the same major number, but distinctly different ranges, may be issued. All devices that failed to call cdevsw_add() before now do. * cdevsw_remove() now automatically marks all devices within its supported range as being destroyed. * vnode->v_rdev is no longer resolved when the vnode is created. Instead, only v_udev (a newly added field) is resolved. v_rdev is resolved when the vnode is opened and cleared on the last close. * A great deal of code was making rather dubious assumptions with regards to the validity of devices associated with vnodes, primarily due to the persistence of a device structure due to being indexed by (major, minor) instead of by (cdevsw, major, minor). In particular, if you run a program which connects to a USB device and then you pull the USB device and plug it back in, the vnode subsystem will continue to believe that the device is open when, in fact, it isn't (because it was destroyed and recreated). In particular, note that all the VFS mount procedures now check devices via v_udev instead of v_rdev prior to calling VOP_OPEN(), since v_rdev is NULL prior to the first open. * The disk layer's device interaction has been rewritten. The disk layer (i.e. the slice and disklabel management layer) no longer overloads its data onto the device structure representing the underlying physical disk. Instead, the disk layer uses the new cdevsw_add() functionality to register its own cdevsw using the underlying device's major number, and simply does NOT register the underlying device's cdevsw. No confusion is created because the device hash is now based on (cdevsw,major,minor) rather then (major,minor). NOTE: This also means that underlying raw disk devices may use the entire device minor number instead of having to reserve the bits used by the disk layer, and also means that can we (theoretically) stack a fully disklabel-supported 'disk' on top of any block device. * The new reference counting scheme prevents this by associating a device with a cdevsw and disconnecting the device from its cdevsw when the cdevsw is removed. Additionally, all udev2dev() lookups run through the cdevsw mask/match and only successfully find devices still associated with an active cdevsw. * Major work on MFS: MFS no longer shortcuts vnode and device creation. It now creates a real vnode and a real device and implements real open and close VOPs. Additionally, due to the disk layer changes, MFS is no longer limited to 255 mounts. The new limit is 16 million. Since MFS creates a real device node, mount_mfs will now create a real /dev/mfs<PID> device that can be read from userland (e.g. so you can dump an MFS filesystem). * BUF AND DEVICE STRATEGY changes. The struct buf contains a b_dev field. In order to properly handle stacked devices we now require that the b_dev field be initialized before the device strategy routine is called. This required some additional work in various VFS implementations. To enforce this requirement, biodone() now sets b_dev to NODEV. The new disk layer will adjust b_dev before forwarding a request to the actual physical device. * A bug in the ISO CD boot sequence which resulted in a panic has been fixed. Testing by: lots of people, but David Rhodus found the most aggregious bugs.
namecache work stage 4: (1) Remove vnode->v_dd, vnode->v_ddid, namecache->nc_dvp_data, and namecache->nc_dvp_id. These identifiers were being used to detect stale parent directory linkages in the namecache and were leftovers from the original FreeBSD-4.x namecache topology. The new namecache topology actively discards such linkages and does not require them. (2) Cleanup kern/vfs_cache.c, abstracting out allocation and parent link/unlink operations into their own procedures. (3) Formally allow a disjoint topology. That is, allow the case where nc_parent is NULL. When constructing namecache entries (dvp,vp), require that that dvp be associated with a namecache record so we can create the proper parent->child linkage. Since no naming information is known for dbp, formally allow unnamed namecache records to be created in order to create the association. (4) Properly relink parent namecache entries when ".." is entered into the cache. This is what relinks a disjoint namecache topology after it has been partially purged or when the namecache is instantiated in the middle of the logical topology (and thus disjoint). Note that the original plan was to not allow a disjoint topology, but after much hair pulling I've come to the conclusion that it is impossible to do this. So the work now formally allows a disjoint topology but also, unlike the original FreeBSD code, takes pains to try to keep the topology intact by only recycling 'leaf' vnodes. This is accomplished by vref()ing a vnode when its namecache records have children.
Protect v_usecount with a critical section for now (we depend on the BGL), and assert that it does not drop below 0. Suggested-by: David Rhodus <email@example.com>
Move the ASSERT_VOP_LOCKED and ASSERT_VOP_UNLOCKED macros into its own functions. Idea taken from: FreeBSD
Newtoken commit. Change the token implementation as follows: (1) Obtaining a token no longer enters a critical section. (2) tokens can be held through schedular switches and blocking conditions and are effectively released and reacquired on resume. Thus tokens serialize access only while the thread is actually running. Serialization is not broken by preemptive interrupts. That is, interrupt threads which preempt do no release the preempted thread's tokens. (3) Unlike spl's, tokens will interlock w/ interrupt threads on the same or on a different cpu. The vnode interlock code has been rewritten and the API has changed. The mountlist vnode scanning code has been consolidated and all known races have been fixed. The vnode interlock is now a pool token. The code that frees unreferenced vnodes whos last VM page has been freed has been moved out of the low level vm_page_free() code and moved to the periodic filesystem sycer code in vfs_msycn(). The SMP startup code and the IPI code has been cleaned up considerably. Certain early token interactions on AP cpus have been moved to the BSP. The LWKT rwlock API has been cleaned up and turned on. Major testing by: David Rhodus
Use a globaldata_t instead of a cpuid in the lwkt_token structure. The LWKT subsystem already uses globaldata_t instead of cpuid for its thread td_gd reference, and the IPI messaging code will soon be converted to take a globaldata_t instead of a cpuid as well. This reduces the number of memory indirections we have to make to access the per-cpu globaldata space in various procedures.
Try to work-around a DFly-specific crash that can occur in ufs_ihashget() if the underlying vnode is being reclaimed at the same time. Bump the vnodes ref count to interlock against vget's VXLOCK test.
64 bit address space cleanups which are a prerequisit for future 64 bit address space work and PAE. Note: this is not PAE. This patch basically adds vm_paddr_t, which represents a 'physical address'. Physical addresses may be larger then virtual addresses and on IA32 we make vm_paddr_t a 64 bit quantity. Submitted-by: Hiten Pandya <firstname.lastname@example.org>
namecache work stage 3a: Adjust the VFS APIs to include a namecache pointer where necessary. For the moment we pass NULL for these parameters (the old 'dvp' vnode pointer's cannot be ripped out quite yet).
namecache work stage 2: move struct namecache to its own header file and have vnode.h include it for now. Re-engineer the namecache topology to make it possible to track different parent directories and to make it possible to namei/lookup paths using the namecache structure as the primary placeholder rather then a directory vnode. Add a few minor hacks to stabilize the system that will be removed (no longer be necessary) in stage 3. Get rid of the leafonly sysctl and make its effect the default, but in order to avoid doing too much in this stage it is still possible to disassociate a vnode from its namecache entry, which a lot of filesystems (e.g. NFS) depend on as a poor-man's way of invalidating entries. The namecache topology itself, however, will be left intact even if a vnode is disassociated in the middle of a path.
namecache work stage 1: namespace cleanups. Add a NAMEI_ prefix to CREATE, LOOKUP, DELETE, and RENAME. Add a CNP_ prefix too all the name lookup flags (nd_flags) e.g. ISDOTDOT->CNP_ISDOTDOT.
1) Add new tunable, kern.syncdelay: kern.syncdelay can be used to change the delay time between file system data synchronization. This is useful when you have notebooks. 2) Document the following sysctls: kern.dirdelay, kern.metadelay and kern.filedelay
Properly handle an error return from udev2dev(). Reviewed by: dillon
The make release process tries to stat/open a non-existant device, which causes addalias() to be called with a NULL dev. Add code to addaliasu() to check that the device actually exists before trying to add the vnode to its alias list. Found by: ROBERT GARRETT <email@example.com>, Jeffrey Hsu <hsu@FreeBSD.org>
Register keyword removal Approved by: Matt Dillon
DEV messaging stage 2/4: In this stage all DEV commands are now being funneled through the message port for action by the port's beginmsg function. CONSOLE and DISK device shims replace the port with their own and then forward to the original. FB (Frame Buffer) shims supposedly do the same thing but I haven't been able to test it. I don't expect instability in mainline code but there might be easy-to-fix, and some drivers still need to be converted. See primarily: kern/kern_device.c (new dev_*() functions and inherits cdevsw code from kern/kern_conf.c), sys/device.h, and kern/subr_disk.c for the high points. In this stage all DEV messages are still acted upon synchronously in the context of the caller. We cannot create a separate handler thread until the copyin's (primarily in ioctl functions) are made thread-aware. Note that the messaging shims are going to look rather messy in these early days but as more subsystems are converted over we will begin to use pre-initialized messages and message forwarding to avoid having to constantly rebuild messages prior to use. Note that DEV itself is a mess oweing to its 4.x roots and will be cleaned up in subsequent passes. e.g. the way sub-devices inherit the main device's cdevsw was always a bad hack and it still is, and several functions (mmap, kqfilter, psize, poll) return results rather then error codes, which will be fixed since now we have a message to store the result in :-)
Throw better sanity checks into vfs_hang_addrlist() for argp->ex_addrlen and argp->ex_masklen which are otherwise totally unchecked from userland.
Remove the priority part of the priority|flags argument to tsleep(). Only flags are passed now. The priority was a user scheduler thingy that is not used by the LWKT subsystem. For process statistics assume sleeps without P_SINTR set to be disk-waits, and sleeps with it set to be normal sleeps. This commit should not contain any operational changes.
The syncer is not a process any more, deal with it as a thread.
MP Implementation 1/2: Get the APIC code working again, sweetly integrate the MP lock into the LWKT scheduler, replace the old simplelock code with tokens or spin locks as appropriate. In particular, the vnode interlock (and most other interlocks) are now tokens. Also clean up a few curproc/cred sequences that are no longer needed. The APs are left in degenerate state with non IPI interrupts disabled as additional LWKT work must be done before we can really make use of them, and FAST interrupts are not managed by the MP lock yet. The main thing for this stage was to get the system working with an APIC again. buildworld tested on UP and 2xCPU/MP (Dell 2550)
Split the struct vmmeter cnt structure into a global vmstats structure and a per-cpu cnt structure. Adjust the sysctls to accumulate statistics over all cpus.
proc->thread stage 6: kernel threads now create processless LWKT threads. A number of obvious curproc cases were removed, tsleep/wakeup was made to work with threads (wmesg, ident, and timeout features moved to threads). There are probably a few curproc cases left to fix.
proc->thread stage 5: BUF/VFS clearance! Remove the ucred argument from vop_close, vop_getattr, vop_fsync, and vop_createvobject. These VOPs can be called from multiple contexts so the cred is fairly useless, and UFS ignorse it anyway. For filesystems (like NFS) that sometimes need a cred we use proc0.p_ucred for now. This removal also removed the need for a 'proc' reference in the related VFS procedures, which greatly helps our proc->thread conversion. bp->b_wcred and bp->b_rcred have also been removed, and for the same reason. It makes no sense to have a particular cred when multiple users can access a file. This may create issues with certain types of NFS mounts but if it does we will solve them in a way that doesn't pollute the struct buf.
proc->thread stage 4: rework the VFS and DEVICE subsystems to take thread pointers instead of process pointers as arguments, similar to what FreeBSD-5 did. Note however that ultimately both APIs are going to be message-passing which means the current thread context will not be useable for creds and descriptor access.
proc->thread stage 2: MAJOR revamping of system calls, ucred, jail API, and some work on the low level device interface (proc arg -> thread arg). As -current did, I have removed p_cred and incorporated its functions into p_ucred. p_prison has also been moved into p_ucred and adjusted accordingly. The jail interface tests now uses ucreds rather then processes. The syscall(p,uap) interface has been changed to just (uap). This is inclusive of the emulation code. It makes little sense to pass a proc pointer around which confuses the MP readability of the code, because most system call code will only work with the current process anyway. Note that eventually *ALL* syscall emulation code will be moved to a kernel-protected userland layer because it really makes no sense whatsoever to implement these emulations in the kernel. suser() now takes no arguments and only operates with the current process. The process argument has been removed from suser_xxx() so it now just takes a ucred and flags. The sysctl interface was adjusted somewhat.
proc->thread stage 1: change kproc_*() API to take and return threads. Note: we won't be able to turn off the underlying proc until we have a clean thread path all the way through, which aint now.
thread stage 5: Separate the inline functions out of sys/buf.h, creating sys/buf2.h (A methodology that will continue as time passes). This solves inline vs struct ordering problems. Do a major cleanup of the globaldata access methodology. Create a gcc-cacheable 'mycpu' macro & inline to access per-cpu data. Atomicy is not required because we will never change cpus out from under a thread, even if it gets preempted by an interrupt thread, because we want to be able to implement per-cpu caches that do not require locked bus cycles or special instructions.
Add the DragonFly cvs id and perform general cleanups on cvs/rcs/sccs ids. Most ids have been removed from !lint sections and moved into comment sections.
import from FreeBSD RELENG_4 220.127.116.11