While checking with ab, I found
commit 30cdb3ac8f
Author: Justin Chen <justinchen00@github.invalid.com>
Date: Thu Apr 14 21:40:53 2016 +0800
recv treat zero return as error
https://github.com/warmcat/libwebsockets/issues/475
turned ab performance to crap, reverting it made everything fast again.
recv manpage says there is three ways to get zero returned
1) When a stream socket peer has performed an orderly shutdown, the return value will be 0 (the traditional "end-of-file"
return).
2) Datagram sockets in various domains (e.g., the UNIX and Internet domains) permit zero-length datagrams. When such a
datagram is received, the return value is 0.
3) The value 0 may also be returned if the requested number of bytes to receive from a stream socket was 0.
we can't just assume it means the peer shut down.
If the peer shut down, then the event loop should get an event on the socket like POLLHUP and deal with it that way.
So the patch mentioned above is simply reverted here.
Signed-off-by: Andy Green <andy@warmcat.com>
This adds an info member that allows the user code to
set the library's network action timeout in seconds.
If left at the default 0, the build-time default
AWAITING_TIMEOUT continues to be used.
As suggested
https://github.com/warmcat/libwebsockets/issues/427
Signed-off-by: Andy Green <andy.green@linaro.org>
This adds support for multithreaded service to lws without adding any
threading or locking code in the library.
At context creation time you can request split the service part of the
context into n service domains, which are load-balanced so that the most
idle one gets the next listen socket accept.
There's a single listen socket on one port still.
User code may then spawn n threads doing n service loops / poll()s
simultaneously. Locking is only required (I think) in the existing
FD lock callbacks already handled by the pthreads server example,
and that locking takes place in user code. So the library remains
completely agnostic about the threading / locking scheme.
And by default, it's completely compatible with one service thread
so no changes are required by people uninterested in multithreaded
service.
However for people interested in extremely lightweight mass http[s]/
ws[s] service with minimum provisioning, the library can now do
everything out of the box.
To test it, just try
$ libwebsockets-test-server-pthreads -j 8
where -j controls the number of service threads
Signed-off-by: Andy Green <andy.green@linaro.org>
After "rerelease"
https://github.com/warmcat/libwebsockets/issues/392#issuecomment-170003294
Since we introduced partial buffering a long while ago,
user code shold never see partial sends and very few
user callbsck attempt to deal with them.
Let's just eliminate the whole concept of user callback
partial send handling under any circumstances.
Signed-off-by: Andy Green <andy.green@linaro.org>
Server side has had immediate RX flow control for quite a while.
But client side made do with RX continuing until what had been received was exhausted.
For what Autobahn tests, that's not enough.
This patch gives clientside RX flow control the same immediate effect as the server
side enjoys, re-using the same code.
Signed-off-by: Andy Green <andy.green@linaro.org>
Again we treat user code sending zero length things as a bug in user code.
But Autobahn insists to be able to do it, so now we allow it.
That buys us a pass on Autobahn test 1.1.1 (the first of a bazillion)
Reproduce with
libwebsockets-test-echo --client localhost --port 9001 -u "/runCase?case=1&agent=libwebsockets" -v -d 65535 -n 1
Signed-off-by: Andy Green <andy.green@linaro.org>
This adds an api lws_close_reason() which lets you control what will
be sent in the close frame when the connection is closed by returning
nonzero from the user callback.
The test server demo is extended to prove it works in both directions.
With this, we should have nice close support.
https://github.com/warmcat/libwebsockets/issues/196
Signed-off-by: Andy Green <andy.green@linaro.org>
The only guy who cared about this for a long while
(since I eliminated the pre-standard protocol variants)
was sending a close frame.
- Set it to 0 so old code remains happy. It only affects
user code buffer commit, if there's overcommit no harm
done so no effect directly on user ABI.
- Remove all uses inside the library. The sample apps
don't have it any more and that's the recommendation now.
Signed-off-by: Andy Green <andy.green@linaro.org>
- Mainly symbol length reduction
- Whitespace clean
- Code refactor for linear flow
- Audit @Context for API docs vs changes
Signed-off-by: Andy Green <andy.green@linaro.org>
Since struct lws (wsi) now has his own context pointer,
we were able to remove the need for passing context
almost everywhere in the apis.
In turn, that means there's no real use for context being
passed to every callback; in the rare cases context is
needed user code can get it with lws_get_ctx(wsi)
Signed-off-by: Andy Green <andy.green@linaro.org>
Extend the cleanout caused by wsi having a context pointer
into the public api.
There's no point keeping the 1.5 compatibility work,
we have changed the api in several places and
rebuilt wasn't going to be enough a while ago.
Signed-off-by: Andy Green <andy.green@linaro.org>
Now we bit the bullet and gave each wsi an lws_context *, many
internal apis that take both a context and wsi parameter only
need the wsi.
Also simplify parser code by making a temp var for
allocated_headers * instead of the longwinded
dereference chain everywhere.
Signed-off-by: Andy Green <andy.green@linaro.org>
Having the lws_context alone doesn't let us track state or act different
by wsi, which is the most interesting usecase. Eg not only simply track
file position / decompression state per wsi but also act differently
according to wsi authentication state / associated cookies.
Signed-off-by: Andy Green <andy.green@linaro.org>
This is a rewrite of the patch from Soapyman here
https://github.com/warmcat/libwebsockets/pull/363
The main changes compared to Soapyman's original patch are
- There's no new stuff in the info struct user code does any overrides
it may want to do explicitly after lws_context_create returns
- User overrides for file ops can call through (subclass) to the original
platform implementation using lws_get_fops_plat()
- A typedef is provided for plat-specific fd type
- Public helpers are provided to allow user code to be platform-independent
about file access, using the lws platform file operations underneath:
static inline lws_filefd_type
lws_plat_file_open(struct lws_plat_file_ops *fops, const char *filename,
unsigned long *filelen, int flags)
static inline int
lws_plat_file_close(struct lws_plat_file_ops *fops, lws_filefd_type fd)
static inline unsigned long
lws_plat_file_seek_cur(struct lws_plat_file_ops *fops, lws_filefd_type fd,
long offset_from_cur_pos)
static inline int
lws_plat_file_read(struct lws_plat_file_ops *fops, lws_filefd_type fd,
unsigned long *amount, unsigned char *buf, unsigned long len)
static inline int
lws_plat_file_write(struct lws_plat_file_ops *fops, lws_filefd_type fd,
unsigned long *amount, unsigned char *buf, unsigned long len)
There's example documentation and implementation in the test server.
Signed-off-by: Andy Green <andy.green@linaro.org>
This nukes all the oldstyle prefixes except in the compatibility code.
struct libwebsockets becomes struct lws too.
The api docs are updated accordingly as are the READMEs that mention
those apis.
Signed-off-by: Andy Green <andy.green@linaro.org>
Between changing to lws_ a few years ago and the previous two
patches migrating the public apis, there are only a few
internal functions left using libwebsocket_*.
Change those to also use lws_ without regard to compatibility
since they were never visible outside the library.
Signed-off-by: Andy Green <andy.green@linaro.org>
Change all internal uses of rationalized public apis to reflect the
new names.
Theer are a few things that got changed as side effect of search/replace
matches, but these are almost all internal. I added a compatibility define
for the public enum that got renamed.
Theoretically existing code should not notice the difference from these
two patches. And new code will find the new names.
https://github.com/warmcat/libwebsockets/issues/357
Signed-off-by: Andy Green <andy.green@linaro.org>
It was forgotten in two places that pending close ack should be
processed when wsi state is WSI_STATE_RETURNED_CLOSE_ALREADY, but
not WSI_STATE_ESTABLISHED. As a result, close ack wasn't sent out
to the peer.
Improvemed patches to address travis and appveyor build errors
Reduced WINVER and _WIN32_WINNT to 0x0501 to be less restrictive
Refined CMakeLists.txt to allow for normal Windows and MinGW-specific OpenSSL certificate generation
Simplified include path to gettimeofday.h
Removed unnecessary list(APPEND LWS_LIBRARIES zlib_internal) export
Added back #include <windows.h> to gettimeofday.c to fix build for normal Windows
Made sure that pollfd gets defined on libwebsockets side when _WIN32_WINNT < 0x0600
Made sure that WINVER and _WIN32_WINNT don't get overridden by libwebsockets headers when already set to something greater than 0x0501
Added missing declaration of WSAPoll function for WINVER < 0x0600 in libwebsockets.h, eliminated invalid usages of pollfd instead of libwebsocket_pollfd in test-server.c
Cleaned up duplicate content in gettimeofday.c, removed header inclusions from gettimeofday.h and fixed include order in test-echo.c, test-ping.c and test-server.c to enable build with normal Windows and MinGW
Re-enabled debug_level in test-echo.c and made sure that the call to lws_set_log_level() is also active under Windows (just like in test-server.c); replaced all WIN32 occurrences by _WIN32 in test-echo.c, test-ping.c, and test-server.c
Removed build-msys.sh and added new section about how to build libwebsockets using MinGW to README.build.md
Read the full incoming TLS/SSL record at once in libwebsocket_service_fd().
SSL_read() is called until no more pending data for the current record is buffered in SSL.
SSL_read() is never requested more than the pending data size for the current record
to ensure that the fd is not read again for new data, which would be copied in the SSL buffer otherwise.
