Adds a std::map for hostname (or urls) --> PerHostRequestQueue
objects. The latter keeps track of the number of added curl easy
requests and decides if a new request should be throttled or
not, as well as provides the queue to queue throttled requests.
At the moment CurlConcurrentConnectionsPerHost is set to 16,
because things really don't work without LL supporting connection
reuse if we limit it to 2. CurlConcurrentConnectionsPerHost is
also set to non-persistent so that we can easily change it in the future
(once we decide on it's final value it can be set to persistent).
Moved CURLOPT_ENCODING from CurlEasyRequest::setPost_raw, and
CURLOPT_SSL_VERIFYPEER and CURLOPT_SSL_VERIFYHOST from
CurlResponderBuffer::prepRequest, to LLURLRequest::configure,
enabling the debug setting NoVerifySSLCert for the latter
two to work as follows: old behavior if "NoVerifySSLCert"
is not set, and check neither if it is set. However, if
the (new) bool mIsAuth is set the behavior of LLXMLRPCTransaction::Impl::init
is used. This is so in a next commit we can replace
LLXMLRPCTransaction with LLURLRequest: LLXMLRPCTransaction::Impl::init
will be removed. For the same reason, when the new boolean
mNoCompression is set then CURLOPT_ENCODING is set to "identity",
otherwise the old behavior (of clearing it) is used.
Introduces AIHTTPTimeoutPolicy objects which do not just
specify a single "timeout" in seconds, but a plethora of
timings related to the life cycle of the average HTTP
transaction.
This knowledge is that moved to the Responder being
used instead of floating constants hardcoded in the
callers of http requests. This assumes that the same
timeout policy is wanted for each transaction that
uses the same Responder, which can be enforced is needed.
I added a AIHTTPTimeoutPolicy for EVERY responder,
only to make it easier later to tune timeout values
and/or to get feedback about which responder runs
into HTTP errors in debug output (especially time outs),
so that they can be tuned later. If we already understood
exactly what we were doing then most responders could
have been left alone and just return the default timeout
policy: by far most timeout policies are just a copy
of the default policy, currently.
This commit is not finished... It's a work in progress
(viewer runs fine with it though).
Conflicts:
indra/newview/statemachine/aistatemachine.cpp
Huh - you re-INDENTED a file that you didn't merge yet?
How about merging curlthreading2 and curlthreading3
before doing THAT? :/
Obviously I did an "use ours" here.
Updated AntiSpam with documentation, safety measures, and better presentations and clarifications for the end-user from FS.
If it looks like something was removed from an xml, it was just moved to fall into place better with v-d.
Corrected log end message being log start message.
Conflicts:
indra/llcommon/CMakeLists.txt
indra/llmessage/llcurl.cpp
indra/llmessage/llcurl.h
indra/llmessage/llhttpclient.cpp
indra/llmessage/llhttpclient.h
indra/llmessage/llpumpio.cpp
indra/llmessage/llpumpio.h
indra/llmessage/llurlrequest.cpp
indra/llmessage/llurlrequest.h
indra/newview/hipporestrequest.cpp
indra/newview/llappviewer.cpp
indra/newview/llspatialpartition.cpp
indra/newview/llviewermedia.cpp
indra/newview/llxmlrpctransaction.cpp
Conflicts resolved by choosing curlthreading2
for any llmessage file regardless (which looks
correct upon investigation); the rest also
turned out to need to use curlthreading2, except
in one line where I added a semi-colon after
an assert(), and the assert was changed in
singu/master.
Tweaked LLRenderTarget to support depth textures if FBO support is lacking.
Prefer LLRenderTarget::getFBO() over LLRenderTarget::sUseFBO when determining how to handle a specific LLRenderTarget object. (Decoupling to simplify logic without having to track a global)
AntiSpamNotify setting to turn off notifications, which seemed to become a new form of spam, otherwise.
Add tool_tips to UI explaining how to make the system less sensitive.
Made defaults for antispam a lot less sensitive, so users used to the old antispam defaults won't likely need to configure the new panel.
Pull in license from NaCl... (Why wasn't it already there, Ruby?!)
Fixed spaces being where tabs should be...
Added in selective Dialog disabling, instead of just all, nice little feature in itself.
Potentially fixed a bug where sounds would be blocked when they shouldn't be... but, if this still happens, I'll put in a debug setting to disable sound antispam.
SHFMODExStreamBufferSize added. Determines stream buffer size in ms. (stream restart required)
SHFMODExDecodeBufferSize added. Determines decode buffer size in ms. (stream restart required)
Streams will mute themselves if they are starving, until they are free of starvation for 5 full seconds.
Streams that fail to accumulate any buffer progress while starving for 10 full updates will be stopped.
Stream buffer progress(buffer percent) is llinfos spewed every update. (temporary)
Doubled default stream buffer size
Increased default decode buffer size to 1000ms (from 400)
Temporarily using FMOD::Memory_Initialize to display raw stream/decode buffer sizes via llinfos.
Added llwarns messages for SigmaTel hardware or bad audio acceleration configuration.
