The description was rather vague, citing a "list of I/O contexts"
and stating that it "can" cancel outstanding requests. This
update makes things more concrete so that the reader knows exactly
what's going on.
Signed-off-by: Jeff Moyer <jmoyer@redhat.com>
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>
nr_events is technically the number of completion events that can
be stored in the completion ring. The wording of the man page:
"capable of receiving at least nr_events" seems dubious to me,
only because I worry that folks might interpret that to mean
'nr_events' total, instead of 'nr_events' concurrently.
Further, I've added information on where to find the per-user
limit on 'nr_events', /proc/sys/fs/aio-max-nr. Let me know if
you think that is not relevant.
Signed-off-by: Jeff Moyer <jmoyer@redhat.com>
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>
I looked back through the kernel code, and the timeout was
never updated in any case. I've submitted a patch upstream
to change the comment above io_getevents.
Signed-off-by: Jeff Moyer <jmoyer@redhat.com>
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>
When describing e_ident, most of the EI_xxx defines mention the
exact byte number. This is useful when manually hacking an ELF
with a hex editor. However, the last few fields don't do this,
which means you have to count things up yourself.
Add a single word to each so you don't have to do that.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>
Split into tables so that the information does not render wider
than 80 columns. Add some explanation of tables and table columns.
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>
Since Linux 3.9, /proc/sys/kernel/sched_rr_timeslice_ms can
be used to change the SCHED_RR quantum.
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>
The units(7) man page uses an ASCII u in place of the actual Greek
letter mu. Since we're in the twenty-first century, with
UTF-8-compatible terminals and terminal emulators, we should use
the actual letter µ instead of an ASCII approximation.
See http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=704787
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>
Also: in sync_file_range.2 and posix_fadvise.2 remove description
of conventional calling signature as flawed, and in
posix_fadvise.2, de-emphasize focus on ARM, and rather phrase
as a more general discussion of certain architectures.
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>
For example, passing 'long long' on ARM-32 requires special
treatment.
Signed-off-by: Changhee Han <ch0.han@lge.com>
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>
The binutils package contains a very handy utility to
print out null-byte delimited strings from a file. This
can replace a rather complex expression with cat(1)
provided as an example for printing out /proc/[pid]/environ.
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>
The directory position dealt with by the readdir() and
friends is not a simple file offset in modern file systems.
Typically, it is some kind of cookie value. Add text and
make other changes to these pages to eliminate the
implication that this is an offset, and warn the reader
that directory positions should be treated strictly as
opaque values.
In the process, rename the 'offset' argument of seekdir(3)
to 'loc', and add some text to readdir(3) to note that
the 'd_off' field is the same value returned by telldir(3)
at the current directory position.
See also https://lwn.net/Articles/544298/
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>