comparison src/http/ngx_http_request.c @ 6536:f7849bfb6d21

Improved EPOLLRDHUP handling. When it's known that the kernel supports EPOLLRDHUP, there is no need in additional recv() call to get EOF or error when the flag is absent in the event generated by the kernel. A special runtime test is done at startup to detect if EPOLLRDHUP is actually supported by the kernel because epoll_ctl() silently ignores unknown flags. With this knowledge it's now possible to drop the "ready" flag for partial read. Previously, the "ready" flag was kept until the recv() returned EOF or error. In particular, this change allows the lingering close heuristics (which relies on the "ready" flag state) to actually work on Linux, and not wait for more data in most cases. The "available" flag is now used in the read event with the semantics similar to the corresponding counter in kqueue.
author Valentin Bartenev <vbart@nginx.com>
date Fri, 13 May 2016 17:19:23 +0300
parents f01ab2dbcfdc
children 055cbb52ac1d
comparison
equal deleted inserted replaced
6535:db699978a33f 6536:f7849bfb6d21
2750 2750
2751 #endif 2751 #endif
2752 2752
2753 #if (NGX_HAVE_EPOLLRDHUP) 2753 #if (NGX_HAVE_EPOLLRDHUP)
2754 2754
2755 if ((ngx_event_flags & NGX_USE_EPOLL_EVENT) && rev->pending_eof) { 2755 if ((ngx_event_flags & NGX_USE_EPOLL_EVENT) && ngx_use_epoll_rdhup) {
2756 socklen_t len; 2756 socklen_t len;
2757
2758 if (!rev->pending_eof) {
2759 return;
2760 }
2757 2761
2758 rev->eof = 1; 2762 rev->eof = 1;
2759 c->error = 1; 2763 c->error = 1;
2760 2764
2761 err = 0; 2765 err = 0;