[BACK]Return to TODO.kqueue CVS log [TXT][DIR] Up to [cvs.NetBSD.org] / src / doc

File: [cvs.NetBSD.org] / src / doc / Attic / TODO.kqueue (download)

Revision 1.8, Wed Nov 12 12:35:52 2008 UTC (12 years, 3 months ago) by ad
Branch: MAIN
CVS Tags: yamt-pagecache-tag8, yamt-pagecache-base8, yamt-pagecache-base7, yamt-pagecache-base6, yamt-pagecache-base5, yamt-pagecache-base4, yamt-pagecache-base3, yamt-pagecache-base2, yamt-pagecache-base, netbsd-6-base, netbsd-6-1-RELEASE, netbsd-6-1-RC4, netbsd-6-1-RC3, netbsd-6-1-RC2, netbsd-6-1-RC1, netbsd-6-1-5-RELEASE, netbsd-6-1-4-RELEASE, netbsd-6-1-3-RELEASE, netbsd-6-1-2-RELEASE, netbsd-6-1-1-RELEASE, netbsd-6-1, netbsd-6-0-RELEASE, netbsd-6-0-RC2, netbsd-6-0-RC1, netbsd-6-0-6-RELEASE, netbsd-6-0-5-RELEASE, netbsd-6-0-4-RELEASE, netbsd-6-0-3-RELEASE, netbsd-6-0-2-RELEASE, netbsd-6-0-1-RELEASE, netbsd-6-0, netbsd-6, matt-premerge-20091211, matt-nb6-plus-nbase, matt-nb6-plus-base, matt-nb6-plus, matt-mips64-premerge-20101231, jym-xensuspend-nbase, jym-xensuspend-base, jym-xensuspend, cherry-xenmp-base, cherry-xenmp, bouyer-quota2-nbase, bouyer-quota2-base, bouyer-quota2
Branch point for: yamt-pagecache, tls-maxphys
Changes since 1.7: +1 -4 lines

Remove LKMs and switch to the module framework, pass 1.

Proposed on tech-kern@.

#	$NetBSD: TODO.kqueue,v 1.8 2008/11/12 12:35:52 ad Exp $

items todo:
	- dir watched - what happens if something is mounted over it?
	  should get write event
	- multiple entry points
	- events should remain across fork()
	- rewrite sowakeup()
	- high-level docco overview of kqueues [c.f jlemon's various papers]
	- make sure there are no selwakeup() calls left (they should
	  all be selnotify() now).
	- kqfilter entry points for:
		- smb_trantcp
		- dev/pci/bktr/bktr_os.c		??? doesn't poll?
	- Test-compile the following ports:
		- amigappc		XXX not compilable even on trunk

merge in from FreeBSD:
	sys/netinet/tcp_subr.c	1.93 (?)
	libc/libc_r/uthread (?)
	pkgsrc/security/l0pht-watch (?)
	sys/fs/fifofs/fifo_vnops.c rev. 1.54
	  - jdolecek: this change seems to be bogus, but perhaps I just
	    miss something?
	usr.bin/make/job.c rev 1.44 - check if it causes any problems
	    on NetBSD (shouldn't)

to fix:

	- look for XXXLUKEM in the comments. these usually indicate
	  areas to either fix or sanity check they're ok (and remove
	  the XXXLUKEM comment)
	- update sys/dev/audio.c:audio_kqfilter() filters for
	  audio_poll() fix in rev. 1.157

not implemented:
	- aio support
		not supported on netbsd; a noop for now

	  EVFILT_AIO	FAIL: not implemented, returns EINVAL