summaryrefslogtreecommitdiff
path: root/sys/kern/sys_pipe.c
AgeCommit message (Expand)Author
2021-10-11Setting EV_EOF requires modifying kn->kn_flags. However, that relies onthorpej
2021-10-02Fix a deadlock where one thread writes to a pipe, has more datahannken
2021-09-27Tweak filt_piperead() and filt_pipewrite() so that:thorpej
2021-09-26The pipe kq filter ops are MPSAFE.thorpej
2021-09-26Change the kqueue filterops::f_isfd field to filterops::f_flags, andthorpej
2021-09-07s/aquire/acquire/ in comments, also one typo fix acqure->acquire.andvar
2021-01-25Fix a thundering herd problem in pipes.dholland
2020-12-11Use sel{record,remove}_knote().thorpej
2020-06-25Fix NULL deref. The original code before Jaromir's cleanup had an #ifndefmaxv
2020-06-25remove experimental direct pipe code (using uvm_loan()) I added in 2001 - it'sjdolecek
2019-04-26Handle half-closed pipes in FIONWRITE and FIONSPACE.mlelstv
2019-04-26Clean up pipe structure before recycling it.mlelstv
2018-06-10convert the (still disabled) 'direct write' for pipes to use thejdolecek
2018-05-19Remove emap support. Unfortunately it never got to state where it would bejdolecek
2018-04-20add prot parameter for uvm_emap_enter(), so that it's possible tojdolecek
2017-12-26Refactor pipe1() and correct a bug in sys_pipe2() (SYS_pipe2)kamil
2017-11-30add fo_name so we can identify the fileops in a simple way.christos
2017-10-25Use C99 initializer for filteropsmaya
2014-09-05Try not to use f_data, use f_{vnode,socket,pipe,mqueue,kqueue,ksem} to getmatt
2014-09-05Don't next structure and enum definitions.matt
2014-02-25Ensure that the top level sysctl nodes (kern, vfs, net, ...) exist beforepooka
2013-06-28Make page loaning in pipes color aware.matt
2012-05-16Make sure we can deliver two file descriptors for pipe2() before we setmartin
2012-01-25As discussed in tech-kern, provide the means to prevent delivery of SIGPIPEchristos
2011-10-20Do call fd_set_exclose() on both file descriptors, to set thenjoly
2011-10-05When pipe1() calls pipe_create() and it fails, use the errorapb
2011-07-15fail with EINVAL if flags not are not O_CLOEXEC|O_NONBLOCK in pipe2(2) andchristos
2011-06-26* Arrange for interfaces that create new file descriptors to be able tochristos
2011-04-10- Add O_CLOEXEC to open(2)christos
2011-01-17Include internal definitions (uvm/uvm.h) only where necessary.uebayasi
2010-08-11Keep condvar wmesg within 8-char limitpgoyette
2009-12-20If a multithreaded app closes an fd while another thread is blocked indsl
2009-12-15Don't ERESTART write() calls for now.dsl
2009-12-13Another, better, fix for PR/26567.dsl
2009-12-13Revert most of the previous change.dsl
2009-12-12Add support for unblocking read/write when close called.dsl
2009-12-10Avoid leaking a mutex_obj when pipe_create() fails for the read pipe.dsl
2009-12-09Rename fo_drain() to fo_abort(), 'drain' is used to mean 'wait for outputdsl
2009-12-06Correct comment, pipelock() no longer releases the mutex.dsl
2009-08-31Turn off pipe's direct I/O again, it corrupts the data (although build andrmind
2009-08-29- Re-enable direct I/O with emap for pipe.rmind
2009-07-15Revert previous: disable direct I/O on pipe, it cought a problem with emap.rmind
2009-07-13Re-enable direct I/O for pipe:rmind
2009-06-28Ephemeral mapping (emap) implementation. Concept is based on the idea thatrmind
2009-06-28Amend previous.rmind
2009-06-28- Convert some #ifdefs to KASSERT()s.rmind
2009-04-11Fix locking as Andy explained. Also fill in uid and gid like sys_pipe did.christos
2009-04-11rename ctime to btime for consistency.christos
2009-04-11- maintain timespec internally.christos
2009-04-04Add fileops::fo_drain(), to be called from fd_close() when there is moread