Mercurial > hg > nginx-quic
view src/core/ngx_spinlock.c @ 4113:a28ba1cdec27
Buffers reuse in chunked filter.
There were 2 buffers allocated on each buffer chain sent through chunked
filter (one buffer for chunk size, another one for trailing CRLF, about
120 bytes in total on 32-bit platforms). This resulted in large memory
consumption with long-lived requests sending many buffer chains. Usual
example of problematic scenario is streaming though proxy with
proxy_buffering set to off.
Introduced buffers reuse reduces memory consumption in the above problematic
scenario.
See here for initial report:
http://mailman.nginx.org/pipermail/nginx/2010-April/019814.html
author | Maxim Dounin <mdounin@mdounin.ru> |
---|---|
date | Thu, 15 Sep 2011 15:25:42 +0000 |
parents | 3f8a2132b93d |
children | d620f497c50f |
line wrap: on
line source
/* * Copyright (C) Igor Sysoev */ #include <ngx_config.h> #include <ngx_core.h> void ngx_spinlock(ngx_atomic_t *lock, ngx_atomic_int_t value, ngx_uint_t spin) { #if (NGX_HAVE_ATOMIC_OPS) ngx_uint_t i, n; for ( ;; ) { if (*lock == 0 && ngx_atomic_cmp_set(lock, 0, value)) { return; } if (ngx_ncpu > 1) { for (n = 1; n < spin; n <<= 1) { for (i = 0; i < n; i++) { ngx_cpu_pause(); } if (*lock == 0 && ngx_atomic_cmp_set(lock, 0, value)) { return; } } } ngx_sched_yield(); } #else #if (NGX_THREADS) #error ngx_spinlock() or ngx_atomic_cmp_set() are not defined ! #endif #endif }