| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
Fixed AIO error handling on FreeBSD.
The aio_return() must be called regardless of the error returned by
aio_error(). Not calling it resulted in various problems up to segmentation
faults (as AIO events are level-triggered and were reported again and again).
Additionally, in "aio sendfile" case r->blocked was incremented in case of
error returned from ngx_file_aio_read(), thus causing request hangs.
|
|
|
|
Fixed sched_setaffinity(2) to correctly pass size.
Second argument (cpusetsize) is size in bytes, not in bits. Previously
used constant 32 resulted in reading of uninitialized memory and caused
EINVAL to be returned on some Linux kernels.
|
|
Microoptimization of sendfile(2) usage under FreeBSD.
FreeBSD kernel checks headers/trailers pointer against NULL, not
corresponding count. Passing NULL if there are no headers/trailers
helps to avoid unneeded work in kernel, as well as unexpected 0 bytes
GIO in traces.
|
|
Introduction of simple ngx_write_stderr() instead of ngx_log_stderr()
for output of ./configure options, etc., since ngx_log_stderr() output
length is limited by 2048 characters defined as NGX_MAX_ERROR_STR.
|
|
Configure/build changes and fixes:
*) Revamped "configure --help" text.
*) FreeBSD 10-current has recently gotten POSIX_FADV_* macros.
A fix for the broken build applied.
*) AIX 7 defines sys_nerr in errno.h, therefore <errno.h> included
in the sys_nerr test.
When sys_nerr and _sys_nerr are missed on a particular platform
and our euristic for a maximum errno detection applied always
print the maximum errno number we reached instead of printing void.
This makes possible to build nginx on AIX 7.1.
*) Made it possible to build/install from the SVN checkout.
|
|
Fixed range checking for the "somaxconn" sysctl. Fixed port range checking.
|
|
Fixed unix ngx_write_chain_to_file() to return total bytes written.
Previously result of last iteration's writev() was returned. This was
unnoticed as return value was only used if chain contained only one or
two buffers.
|
|
Various fixes related to error messages:
*) Removed old warning that suggested to use "server_name_in_redirect off"
(now the default) in place of no longer supported "server_name *".
*) Fixing directive name in error message if types hash size is not enough.
*) Replaced "can not" with "cannot" and "could not" in a bunch of places.
Fixed nearby grammar errors.
*) Overhauled error messages.
|
|
SIGWINCH/NOACCEPT signal is disabled now in non-daemon mode.
Non-daemon mode is currently used by supervisord, daemontools and so on
or during debugging. The NOACCEPT signal is only used for online upgrade
which is not supported when nginx is run under supervisord, etc.,
so this change should not break existant setups.
|
|
Move SO_ACCEPTFILTER and TCP_DEFER_ACCEPT checks into configure.
NetBSD 5.0+ has SO_ACCEPTFILTER support merged from FreeBSD, and having
accept filter check in FreeBSD-specific ngx_freebsd_config.h prevents it
from being used on NetBSD. Therefore move the check into configure (and
do the same for Linux-specific TCP_DEFER_ACCEPT, just to be in line).
|
|
Linux AIO related fixes:
*) Fixing Linux AIO syscalls return value handling:
syscall(2) uses usual libc convention, it returns -1 on error and
sets errno. Obsolete _syscall(2) returns negative value of error.
*) Fixing Linux AIO initiatialization: AIO operations are disabled
if kernel does not support them. Previously worker just exited.
*) The "worker_aio_requests" directive.
The default value is 32 AIO simultaneous requests per worker. Previously
they were hardcoded to 1024, and it was too large, since Linux allocated
them early on io_setup(), but not on request itself. So with default value
of /proc/sys/fs/aio-max-nr equal to 65536 only 64 worker processes could
be run simultaneously. 32 AIO requests are enough for modern disks even if
server runs only 1 worker.
|
|
In Mac OS X Lion, Apple deprecated the use of the system OpenSSL library
in favour of their CommonCrypto library. This change adds a work-around
that allows nginx to still be built on Lion with OpenSSL.
|
|
open_file_cache related fixes:
*) Bugfix: open_file_cache lost is_directio flag.
On file retest open_file_cache lost is_directio if file wasn't changed.
This caused unaligned operations under Linux to fail with EINVAL.
It wasn't noticeable with AIO though, as errors wasn't properly logged.
*) Bugfix: open_file_cache did not update file info on retest.
If file inode was not changed, cached file information was not updated
on retest. As a result stale information might be cached forever if file
attributes was changed and/or file was extended.
|
|
Cache loader improvements:
The cache loader performs two tasks: inserting cache objects in inactivity
list and evaluating total cache size. Reading just directory is enough for
this purpose. Elimination of reading cache files saves at least one disk I/O
operation per file.
|
|
patch by Maxim Dounin
|
|
number of spinlock spins are increased twice
|
|
|
|
instead of file length rounded to a file system block size.
There is no similar way on Windows, so rounding to a cache->bsize is kept.
|
|
thanks to Andrei Nigmatulin
|
|
patch by Andrei Nigmatulin
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
*) change ngx_time_update() interface
|
|
*) change ngx_time_update() interface since there are no notification methods
those return time
|
|
of non Async-Signal-Safe strerror_r()
|
|
dereferencing type-punned pointer will break strict-aliasing rules
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
