Re: [Libevent-users] [PATCH] Make evutil_read_file() close-on-exec safe.

2012-02-11 Thread Ross Lagerwall
On 12/02/2012 04:21, Nick Mathewson wrote: > On Sat, Feb 11, 2012 at 3:15 PM, Ross Lagerwall > wrote: >> If we can't trust the compile-time value and we can't detect the runtime >> behavior then I think the solution I suggested is a reasonable idea. > > Agreed. > > (Actually, we _could_ detect

Re: [Libevent-users] listen syscall and the backlog argument

2012-02-11 Thread Nick Mathewson
On Sat, Feb 11, 2012 at 1:12 AM, Ross Lagerwall wrote: > Hi, > > In listener.c, there is some duplicated code related to listen() and the > backlog argument. > > This code simplifies backlog handling by relying on the POSIX semantics > of listen(). How sure are we that every platform with listen(

Re: [Libevent-users] Assertion in evmap_io_active

2012-02-11 Thread Nick Mathewson
On Sat, Feb 11, 2012 at 12:31 PM, Gilad Benjamini wrote: > (re-sending since my previous message doesn't appear to have reached the list) > Our code uses libevent 2.0.10 and we are running into the assertion below. > We haven't seen any real problems beyond that assertion but it is alarming > nev

Re: [Libevent-users] [PATCH] Make evutil_read_file() close-on-exec safe.

2012-02-11 Thread Nick Mathewson
On Sat, Feb 11, 2012 at 3:15 PM, Ross Lagerwall wrote: > If we can't trust the compile-time value and we can't detect the runtime > behavior then I think the solution I suggested is a reasonable idea. Agreed. (Actually, we _could_ detect it at runtime by using fcntl(F_GETFD) on an fd after we op

Re: [Libevent-users] [PATCH] Make evutil_read_file() close-on-exec safe.

2012-02-11 Thread Ross Lagerwall
On 11/02/2012 18:48, Nick Mathewson wrote: > On Sat, Feb 11, 2012 at 10:34 AM, Ross Lagerwall > wrote: >> In a multi-process/threaded environment, ev_util_read_file() >> could leak fds to child processes when not using O_CLOEXEC/FD_CLOEXEC. > > Hm. I'm not sure I trust "#ifdef O_CLOEXEC" as a te

[Libevent-users] Assertion in evmap_io_active

2012-02-11 Thread Gilad Benjamini
(re-sending since my previous message doesn't appear to have reached the list) Our code uses libevent 2.0.10 and we are running into the assertion below. We haven't seen any real problems beyond that assertion but it is alarming nevertheless. Can anyone please shed some light on what this asserti

Re: [Libevent-users] [PATCH] Make evutil_read_file() close-on-exec safe.

2012-02-11 Thread Nick Mathewson
On Sat, Feb 11, 2012 at 10:34 AM, Ross Lagerwall wrote: > In a multi-process/threaded environment, ev_util_read_file() > could leak fds to child processes when not using O_CLOEXEC/FD_CLOEXEC. Hm. I'm not sure I trust "#ifdef O_CLOEXEC" as a test for whether open() supports O_CLOEXEC. Generally,

[Libevent-users] [PATCH] Make evutil_read_file() close-on-exec safe.

2012-02-11 Thread Ross Lagerwall
In a multi-process/threaded environment, ev_util_read_file() could leak fds to child processes when not using O_CLOEXEC/FD_CLOEXEC. --- evutil.c |6 ++ 1 files changed, 6 insertions(+), 0 deletions(-) diff --git a/evutil.c b/evutil.c index 306f037..2bdacaa 100644 --- a/evutil.c +++ b/evut