Renumber cluase 4 to 3, per what everybody else did when BSD granted
them permission to remove clause 3. My insistance on keeping the same
numbering for legal reasons is too pedantic, so give up on that point.
Submitted by: Jan Schaumann <jschauma@stevens.edu>
Pull Request: https://github.com/freebsd/freebsd/pull/96
-- Fix whitespace
-- Use err/errx
-- Remove superfluous braces
- Be a bit more defensive with input from the end-user
- Don't throw a floating point exception by dividing by 0 when processing a
zero-byte file
MFC after: 1 week
- Convert errx(-1, ..) to errx(1, ..)
- Move the aio(4) checks to a single function (aio_available); use modfind(2)
instead of depending on SIGSYS (doesn't work when aio(4) support is missing,
not documented in the aio syscall manpages).
- Use aio_available liberally in the testcase functions
- Use mkstemp(3) + unlink(2) instead of mktemp(3)
- Fix some -Wunused warnings
- Bump WARNS to 6
MFC after: 1 week
Submitted by: mjohnston [*]
Sponsored by: EMC / Isilon Storage Division
and finish the job. ncurses is now the only Makefile in the tree that
uses it since it wasn't a simple mechanical change, and will be
addressed in a future commit.
AIO calls.
This small program queues up a controllable number of concurrent AIO
read operations w/ controllable io size against a disk or regular file.
There are a few other things to add (notably optional write support!)
but it works well enough at the present time to stress the AIO code out
relatively harshly in the disk IO case.
http://www.ambrisko.com/doug/listio_kqueue/listio_kqueue.patch
Note: it is a good idea to run this against a physical drive to
exercise the physio fast path (ie. lio_kqueue /dev/<something safe>)
This will ensure op's counting per LIO request is correct. It is
currently broken the above patch fixes it.
Sponsored by: IronPort
against a disk as the argument. If you don't it will use a temp file.
The raw disk will use the kernel physio fast path method until the
max number of pending op's is reached then it will queue them. File
system op's are always queued. This is more important with LIO since
operation can get split across and accounting of op's is broken with LIO.
Note that this was broken when locking was added to kqueue (ie. 5.3)
My fix needs to be better integrated with FreeBSD.
Next is an LIO test and implementation.
Sponsored by: IronPort
then reads from a fairly broad range of object types: regular file,
fifo, UNIX socketpair, pty, UNIX pipe, and an md device. Not a deep
test of functionality, just a basic test that aio_write followed by
aio_read returns the correct data in a relatively timely manner.
Requested by: phk
