Re: r1470679, async write completion, non blocking writes, and mod_ssl

2013-08-15 Thread Stefan Fritsch
Am Donnerstag, 15. August 2013, 02:36:25 schrieb Graham Leggett: On 14 Aug 2013, at 22:43, Stefan Fritsch s...@sfritsch.de wrote: Unfortunately, I haven't been able to trigger the new code path in mod_ssl being actually used. Do you have any example setup/situation, where the

Re: r1470679, async write completion, non blocking writes, and mod_ssl

2013-08-15 Thread Stefan Fritsch
Am Donnerstag, 15. August 2013, 10:45:25 schrieb Graham Leggett: On 15 Aug 2013, at 09:56, Stefan Fritsch s...@sfritsch.de wrote: I have understood that. But I would have liked to see the sense code in action, but failed to trigger it. At least t/ssl/pr12355.t in the test suite uses

Re: r1470679, async write completion, non blocking writes, and mod_ssl

2013-08-07 Thread Jim Jagielski
On Aug 5, 2013, at 4:50 PM, Stefan Fritsch s...@sfritsch.de wrote: It gives us a new API that we need to keep. I think it's a useful API, but ymmv

Re: r1470679, async write completion, non blocking writes, and mod_ssl

2013-08-05 Thread Stefan Fritsch
On Mon, 5 Aug 2013, Graham Leggett wrote: Are you seeing a specific problem? Well, when I download a large file over a slow link, the request does not enter write completion state but rather the worker thread is still hogged for (nearly) the entire download. The way openssl's async behaviour

Re: r1470679, async write completion, non blocking writes, and mod_ssl

2013-08-05 Thread Jim Jagielski
On Aug 5, 2013, at 4:00 AM, Stefan Fritsch s...@sfritsch.de wrote: An ideal solution would put the buffering/decision for blocking/non-blocking into ap_pass_brigade(). This way other filters like deflate could also be called asynchronously. But I am not too optimistic that this can be

Re: r1470679, async write completion, non blocking writes, and mod_ssl

2013-08-05 Thread Stefan Fritsch
Am Montag, 5. August 2013, 09:57:16 schrieb Jim Jagielski: On Aug 5, 2013, at 4:00 AM, Stefan Fritsch s...@sfritsch.de wrote: An ideal solution would put the buffering/decision for blocking/non-blocking into ap_pass_brigade(). This way other filters like deflate could also be called

r1470679, async write completion, non blocking writes, and mod_ssl

2013-08-04 Thread Stefan Fritsch
Hi, I did some testing/reviewing of the ssl/event backport proposal * core, mod_ssl: Lift the restriction that prevents mod_ssl taking full advantage of the event MPM. Enable the ability for a module to reverse the sense of a poll event from a read to a write or vice versa. The

Re: r1470679, async write completion, non blocking writes, and mod_ssl

2013-08-04 Thread Graham Leggett
On 04 Aug 2013, at 8:52 PM, Stefan Fritsch s...@sfritsch.de wrote: Hi, I did some testing/reviewing of the ssl/event backport proposal * core, mod_ssl: Lift the restriction that prevents mod_ssl taking full advantage of the event MPM. Enable the ability for a module to reverse