summaryrefslogtreecommitdiff
path: root/usr.sbin/puffs/mount_psshfs/node.c
AgeCommit message (Expand)Author
2021-09-16fix typos in word "successfully", mainly s/succesfully/successfully/.andvar
2016-10-07PR/514612: Mateusz Paszwa: When no more data, bail out (reading from emptychristos
2012-01-20Ensure psshfs_node_read() completely reads all data before return.jakllsch
2011-08-12Cache vattr in psshfs's setattr.riastradh
2010-10-29Make error message more userfriendly in cases where server doespooka
2010-04-01In case we create a file, reclaim the vnode, and lookup the filepooka
2010-01-07Add -u and -g, which allow to remap one (local,remote) uidgid, i.e.pooka
2009-11-05more lp64 it's-a-sign! fixespooka
2009-05-20Copyright year management. no functional change (unless sarah disagrees)pooka
2009-05-20Introduce -c [12], which can be used to open 1 or 2 ssh connectionspooka
2009-03-29Save errno before descending to cleanup gehennom.pooka
2009-01-01Track parent directory in cross-directory renames so that mv & rmpooka
2008-08-12void * -> puffs_cookie_tpooka
2007-12-09When creating a node, do the server-dance before creating the localpooka
2007-12-07Don't use puffs_cc_getusermount()pooka
2007-12-07Get rid of a lot of extra complexity due to doing separate getattrpooka
2007-12-07In case a node is created in lookup in the readdir path, initializepooka
2007-12-04When scheduling waiters for re-run, remove them from the waitqueuepooka
2007-11-30Rototill.pooka
2007-11-30Introduce the concept of lazyopens. This means that when a filepooka
2007-11-27Remove "puffs_cid" from the puffs interface following l-removalpooka
2007-11-18Don't bother tracking opencount and try to close file handles inpooka
2007-11-11Don't overload "attributes last read" for the symlink cache,pooka
2007-11-10Cache readlink results similarly to attributes. Makes "cd" fastpooka
2007-10-21Use puffs_cc_schedule() (which is an exported interface) instead ofpooka
2007-10-20Add option -r to specify maximum number of active read requestspooka
2007-10-16Close file handles if opencount drops to zero instead of closingpooka
2007-08-24If a node is forcibly reclaimed, close can be called more timespooka
2007-08-23Prevent crash in case there is more than one reader for the samepooka
2007-07-29make closehandles() void, because that's what it ispooka
2007-07-27Track memory mappings. Close file handles already in close() if therepooka
2007-07-19Oops, the cookies for nfs of course represent the *next* directorypooka
2007-07-16If the target node exists in rename, do not try to issue a setbackpooka
2007-07-02adapt: inactive no longer comes with int *refcount (not that it waspooka
2007-07-01adapt: **newnode etc. pointers -> struct puffs_newinfopooka
2007-07-01adapt: pid -> const struct puffs_cid *pooka
2007-06-06Move puffs to a two clause license where it already isn't so. Andpooka
2007-05-18remove obvious incorrect assert() from doreclaim() and remove thepooka
2007-05-18* use NOREF setback to prompt immediate (well, once the kernelpooka
2007-05-15adapt to "step 3" of puffs_framebuf changespooka
2007-05-11adapt to puffs_framebuf changespooka
2007-05-07Use inactive-on-demand, set inactive request in open() to closepooka
2007-05-06Open files in open and close them in inactive. This avoids thepooka
2007-05-06Fire off FXP_CLOSE in create() and return immediately instead ofpooka
2007-05-05Adapt to use puffs_framebuf. This gets rid of most of the duplicatedpooka
2007-05-01g/c forgotten printfpooka
2007-04-19use puffs_cc_getspecific() where puffs_usermount is not requiredpooka
2007-04-17use PUFFS_STORE_DCOOKIEpooka
2007-04-12Actually, we can't reclaim our memory image for nodes which havepooka
2007-04-12Support nfs exporting.pooka