mfp

On Tue, Oct 16, 2018 at 08:43:56PM -0700, Vincent Bernardoff wrote: > Great, thanks. > @mfp Could you make a new OPAM release please? New release coming shortly since it...

On Sat, Jul 08, 2017 at 01:59:21PM -0700, ygrek wrote: > It appears extprot accepts message alias to message, but generates bad code. This possibility is not documented afaik, not...

On Sun, Jul 09, 2017 at 12:13:14AM +0000, ygrek wrote: > I am confused because there is Message_alias in Ptypes.message_expr which is only produced by `+.` rule, but then generates...

The issue is very easily reproduced once you know what you're looking for, all it takes is a consumer that's slower than the sender. Here's a self-contained testcase: https://gist.github.com/mfp/1250890c2e7fa790365369116855aa0f I...

I settled with a simpler expression for the lower bound (I originally had `max sz (Int32.of_int (((3 * t.tx_mss + 1 lsl t.rx_wnd_scale) lsr t.rx_wnd_scale) lsl t.rx_wnd_scale))`, in an attempt...

> Code that uses the Unix module anywhere other than the socket stack unfortunately won't be merged, because tcpip needs to be buildable and usable without Unix, Sys, and other...

I've taken a quick look at NodeJS' API and many things in Lwt_unix would be implementable straightforwardly atop it (minus the type equalities, etc., with `Unix`, which would have to...

On Sun, Mar 26, 2017 at 12:24:09PM -0700, Anton Bachin wrote: > However, libuv is maintained by more people, and has a larger user base, so one can expect it...

I'm not especially knowledgeable on this, but for sure serving files is what sendfile(2) was made for. It would only help for plain (non-SSL) connections, which is an important limitation....

On Wed, Nov 26, 2014 at 09:54:52AM -0800, David Kaloper wrote: > Another alternative would be to use `Cstruct.t`. `sendfile(2)` avoids > copying the file into OCaml memory altogether, sure,...