Граф коммитов

8035 Коммитов

Автор SHA1 Сообщение Дата
Nick Hurley 74b99cf44e bug 712914 - add some sanity checking when (de)serializing security info in the disk cache. r=michal 2012-03-30 11:21:01 +02:00
Patrick McManus 645594048f Bug 739023 - Disabling the Restart-in-progess logic temporarily since suspected of causing crashes, r=honzab 2012-03-30 00:40:42 +02:00
Michal Novotny c2005fe9a0 Bug 649194 - Abort when the cache IO thread can't be created 2012-03-27 10:20:54 +02:00
Michael Wu 03bd08a4ff Bug 737196 - Make camera code build on gonk-ics, r=cjones 2012-03-23 15:45:47 -04:00
Ms2ger 1cc3c03286 No bug - Remove always-true comparison. 2012-03-24 09:18:09 +01:00
Patrick McManus 007716fc61 bug 717759 - negative pipeline feedback on img decode problem r=bz r=jdrew r=honzab 2012-03-22 19:39:32 -04:00
Patrick McManus 6d514a6643 bug 717350 doom cache entry on last-modified mismatch r=honzab 2012-03-22 19:39:31 -04:00
Patrick McManus 844b20abf7 bug 671591 - restart partial http transaction (in progress) r=honzab 2012-03-22 19:39:31 -04:00
Patrick McManus bc1f690f50 bug 603514 - http stalled read detection r=honzab
When a connection that involves a pipelined transaction has been stalled (i.e.
idle with an open transaction, not an idle persistent connection) for a
second move any transactions that are pipelined after the current one onto
different connections (and implicitly close this connection when done with the
current transaction).

when it has been stalled for 10 seconds (pref configurable), cancel the current
transaction itself too - depending on its state it can hopefully be restarted
on a clean connection.
2012-03-22 19:39:31 -04:00
Patrick McManus 8b81d2a05f bug 603512 - large objects block pipelines r=honzab
the type and state patch tries hard not to form pipelines behind resources that
could become head of line blockers. But of course that requires the ability to
predict the future, and won't be perfect.

This patch reacts to a transaction that has a large response body (defined by
either a large content-length header or actually reading a large number of
chunked bytes) by cancelling any transactions that have been pipelined down the
same connection and rescheduling them elsewhere. It also changes the type of
the connection to "solo", which prevents new transactions from being pipelined
onto this one and provides class-specific negative feedback to the pipeline
manager so that near-future requests to the same host of the same type (e.g.
general) will not be pipelined but other types (e.g. img or js/css) can still
do that.

Content-Length is ideal, because it allows us to identify the problem so early.
But even actually reading the document for a fairly long time gives it a fairly
high probability of not ending soon. (i.e. long document sizes are spread over
a larger range than small ones. duh.)

The pref network.http.pipelining.maxsize controls the threshold. I set the
default at 300KB, which is roughly the bandwidth delay product of a 2mbit 120ms
rtt connection and 1 rtt is mostly what you are giving up by canceling it on
one connection and sending it on another. (modulo maybe needing a handshake).
2012-03-22 19:39:31 -04:00
Patrick McManus d58afda113 bug 665885 respect keepalive: "max=" attribute r=honzab 2012-03-22 19:39:31 -04:00
Patrick McManus ff250f550b bug 737155 - httpconnectionmgr::processPendingQ allow more than 1 with pipeline r=honzab 2012-03-22 19:39:31 -04:00
Patrick McManus c1ef6c81a2 bug 599164 pipeline with type and state r=honzab 2012-03-22 19:39:31 -04:00
Patrick McManus 6acbbecb41 bug 597684 Implement HTTP Assoc-req and Banned Pipelines on nsHttpConnectionInfo r=honzab 2012-03-22 19:39:31 -04:00
Patrick McManus e600a6343f bug 447866 http pipelining is bursty r=honzab
--HG--
extra : rebase_source : 87370228074eaff2615b8dcef9b3c0c9b919b3c3
2012-03-22 19:39:31 -04:00
Patrick McManus 3c85a1c5cf bug 447866 - patch pre.1 for empty nsHttpPipeline::TakeSubTransactions r=honzab
--HG--
extra : rebase_source : 5c21007525883fc74dc70332cf74d8d0bacd9147
2012-03-22 19:39:28 -04:00
Michal Novotny c8537395fd Bug 722033 - Invalidate cache entry in nsHttpChannel::DoInvalidateCacheEntry() asynchronously 2012-03-22 23:54:20 +01:00
Michal Novotny 2df9700ac8 Bug 722033 - Use asyncOpenCacheEntry() in nsHttpChannel::OpenOfflineCacheEntryForWriting() 2012-03-22 23:53:55 +01:00
Michal Novotny d4e6a87f9b Bug 722033 - Use asyncOpenCacheEntry() in nsHttpChannel when flag LOAD_BYPASS_LOCAL_CACHE_IF_BUSY is set 2012-03-22 23:53:10 +01:00
Henri Sivonen 2783a8258e Bug 737417 part 1 - Split charset source constants out of nsIParser.h. r=smaug. 2012-03-22 16:42:42 +02:00
Jim Mathies c958ac3180 Bug 737986 - win8 dev kit fixup for win wifi includes, r=dougt 2012-03-22 09:18:43 -05:00
julian.reschke@gmx.de 77c36bf727 Bug 714302: add suport for HTTP status code 308 r=mcmanus
--HG--
extra : rebase_source : 2e4a46ac5ebe89a36fb4b12b1985d07d674db5e8
2012-03-21 16:13:42 -04:00
julian.reschke@gmx.de adff062582 Bug 384571: RFC 2231 parameters not decoded when appearing in reversed order. r=jduell 2012-03-20 14:08:30 -07:00
Matt Brubeck f60d72f679 Back out 1bbe7ea3a01e through 672cc4ee2dd9 because of mochitest-browser-chrome leaks 2012-03-20 12:53:10 -07:00
Patrick McManus dc1aed3edd bug 717759 - negative pipeline feedback on img decode problem r=bz r=jdrew r=honzab
--HG--
extra : rebase_source : cb9795b5442102a777602f59c225ab9572e06cea
2012-03-20 13:11:32 -04:00
Patrick McManus f2f489fbb1 bug 717350 doom cache entry on last-modified mismatch r=honzab
--HG--
extra : rebase_source : 5c8692c94bf7e5dcb2d381a85e06bddbabffe22c
2012-03-20 13:11:32 -04:00
Patrick McManus 1c33f42938 bug 671591 - restart partial http transaction (in progress) r=honzab
--HG--
extra : rebase_source : 9f71e1803e25c10f6be5f48f324eb47fa8d707c3
2012-03-20 13:11:32 -04:00
Patrick McManus 031058589a bug 603514 - http stalled read detection r=honzab
When a connection that involves a pipelined transaction has been stalled (i.e.
idle with an open transaction, not an idle persistent connection) for a
second move any transactions that are pipelined after the current one onto
different connections (and implicitly close this connection when done with the
current transaction).

when it has been stalled for 10 seconds (pref configurable), cancel the current
transaction itself too - depending on its state it can hopefully be restarted
on a clean connection.

--HG--
extra : rebase_source : 58377c9eab2f8a81b7a2d3b9a5f12e1134da4031
2012-03-20 13:11:32 -04:00
Patrick McManus 2809e23a08 bug 603512 - large objects block pipelines r=honzab
the type and state patch tries hard not to form pipelines behind resources that
could become head of line blockers. But of course that requires the ability to
predict the future, and won't be perfect.

This patch reacts to a transaction that has a large response body (defined by
either a large content-length header or actually reading a large number of
chunked bytes) by cancelling any transactions that have been pipelined down the
same connection and rescheduling them elsewhere. It also changes the type of
the connection to "solo", which prevents new transactions from being pipelined
onto this one and provides class-specific negative feedback to the pipeline
manager so that near-future requests to the same host of the same type (e.g.
general) will not be pipelined but other types (e.g. img or js/css) can still
do that.

Content-Length is ideal, because it allows us to identify the problem so early.
But even actually reading the document for a fairly long time gives it a fairly
high probability of not ending soon. (i.e. long document sizes are spread over
a larger range than small ones. duh.)

The pref network.http.pipelining.maxsize controls the threshold. I set the
default at 300KB, which is roughly the bandwidth delay product of a 2mbit 120ms
rtt connection and 1 rtt is mostly what you are giving up by canceling it on
one connection and sending it on another. (modulo maybe needing a handshake).

--HG--
extra : rebase_source : ebc44f8dfc3fa7bcee613ff3b74269d908ceac51
2012-03-20 13:11:32 -04:00
Patrick McManus 57a87f3770 bug 665885 respect keepalive: "max=" attribute r=honzab
--HG--
extra : rebase_source : ce4009c19bb786388338778d594ca8038a4398c1
2012-03-20 13:11:32 -04:00
Patrick McManus 90405ceda1 bug 737155 - httpconnectionmgr::processPendingQ allow more than 1 with pipeline r=honzab
--HG--
extra : rebase_source : 94c50c125fc66c9d5b4389e20b0187ee40c2e315
2012-03-20 13:11:32 -04:00
Patrick McManus ffa953ec74 bug 599164 pipeline with type and state r=honzab
--HG--
extra : rebase_source : b060665a1726eab3e3ef0ab0d94c34cfea5c1460
2012-03-20 13:11:32 -04:00
Patrick McManus 2602c52e7b bug 597684 Implement HTTP Assoc-req and Banned Pipelines on nsHttpConnectionInfo r=honzab
--HG--
extra : rebase_source : 26243e1af9e0554553520dae435cc4dcfd551746
2012-03-20 13:11:32 -04:00
Patrick McManus 7abd00c179 bug 447866 http pipelining is bursty r=honzab
--HG--
extra : rebase_source : e927c5961b2f6c79c1aa73897900e95ed3f286c5
2012-03-20 13:11:32 -04:00
Gervase Markham b30378e740 Bug 716607 - Add .pm, .wf, .yt to PSL. r=pkasting. 2012-03-20 15:04:40 +00:00
Phil Ringnalda bdcda40739 Merge last PGO-green changeset of mozilla-inbound to mozilla-central 2012-03-17 16:57:05 -07:00
Chris Jones fc3225b548 Bug 736778: Work around probable compiler bug breaking DEBUG builds of GonkCaptureProvider. r=fabrice (npotb) 2012-03-17 14:42:24 -07:00
Jason Duell 6569e9493a Backed out changeset cd5ac132e763 2012-03-15 23:36:01 -07:00
Jason Duell a8eaf5a43f Backed out changeset cd5ac132e763 2012-03-15 23:35:51 -07:00
Jason Duell f8e3cd2793 Backed out changeset 5aaec7f808b6 2012-03-15 20:56:23 -07:00
Jason Duell 515075a988 Backed out changeset 5aaec7f808b6 2012-03-15 20:55:44 -07:00
John Daggett aab65b1694 Bug 512566. Changes based on review comments. r=bz,jfkthame 2012-03-16 12:31:01 +09:00
John Daggett e080a64cde Bug 512566. Load font data url's synchronously, netwerk/style changes. r=bz 2012-03-16 12:31:01 +09:00
Jason Duell 9ba0c759c5 Bug 664163: I will back this out when xpcshell test_traceable.js fails eventually. r=jduell 2012-03-15 18:54:58 -07:00
julian.reschke@gmx.de 6e8cb9e4a8 Bug 384571: RFC 2231 parameters not decoded when appearing in reversed order r=jduell 2012-03-04 14:12:00 -08:00
Jason Duell e3a3c49b9b Bug 648878 - I will backout: investigating inbound-only failure r=jduell 2012-03-15 13:00:20 -07:00
Jason Duell 613c1c4358 Backed out c9265ada6f40 2012-03-15 15:05:41 -07:00
Jason Duell 079fc95539 Backed out changeset c9265ada6f40 2012-03-15 15:05:22 -07:00
Derrick Rice 028c0b7534 Bug 732363 - Avoid running test_net_addr on windows; it is failing for unknown reasons (race?) r=jdm 2012-03-15 07:10:00 -04:00
Patrick McManus ef5538d062 bug 735552 - wrong assert on shutdown in nsHttpConnectionMgr::ReportSpdyConnection r=honzab
--HG--
extra : rebase_source : 8baf5fbae8d9ad8196fd699191c0013f7ea8f424
2012-03-15 00:03:04 -04:00