searching for reviewer(jld)
aba5bfdd772bbc630ed53edb59aa43c0544484e8: Bug 1550771 - Deadlock in SharedMemoryBasic_mach triggered by AV1 playback r=jld
Haik Aftandilian <haftandilian@mozilla.com> - Wed, 22 May 2019 01:33:46 +0000 - rev 474916
Push 113176 by shindli@mozilla.com at Wed, 22 May 2019 09:18:18 +0000
Bug 1550771 - Deadlock in SharedMemoryBasic_mach triggered by AV1 playback r=jld Don't hold gMutex when calling HandleSharePortsMessage() from PortServerThread to avoid deadlock. Differential Revision: https://phabricator.services.mozilla.com/D31694
969f29e56fa1b018fe790d888e05df6f1ddbdbb1: Bug 1547085 - Discard messages sent to dead actors, r=jld
Nika Layzell <nika@thelayzells.com> - Tue, 21 May 2019 17:04:25 +0000 - rev 474793
Push 113171 by aiakab@mozilla.com at Tue, 21 May 2019 21:56:33 +0000
Bug 1547085 - Discard messages sent to dead actors, r=jld Historically we've failed very loudly when receiving a message which was destined for an actor which had already been destroyed. This had the effect of requiring manual teardown for most actors, as work would need to be done to ensure messages weren't sent when the target actor might be about to tear itself down. In addition, due to this teardown work being done outside of IPDL, this work would have to manually be checked in subactors, and involved the addition of new flags, such as `mIPCOpen`, in order to track whether IPC had begun to be shut down, and discard messages manually if it had. It is an ongoing issue that we occasionally miss places where we need to discard messages, and it is easy to not remember to perform async destruction when building a new actor, meaning that extra work is required to correctly discard messages when the actor is being torn down. Due to the correct decision, almost all of the time, being to discard the message, this patch takes the approach of transforming the crash which was previously performed into a message discard. The hope is that this will reduce the burden on actor implementors, by allowing the use of `Send__delete__` without first synchronizing with the remote actor, as well as reduce unintentional crashes. Differential Revision: https://phabricator.services.mozilla.com/D28892
a64a71ccebf79139da2288edca29186292c5a257: Bug 1513915 - LSNG: LoadItem could return more items (up to the prefill limit); r=asuth,jld
Jan Varga <jan.varga@gmail.com> - Mon, 08 Apr 2019 10:56:19 +0200 - rev 472465
Push 113013 by jvarga@mozilla.com at Fri, 03 May 2019 14:30:47 +0000
Bug 1513915 - LSNG: LoadItem could return more items (up to the prefill limit); r=asuth,jld Differential Revision: https://phabricator.services.mozilla.com/D26476
7de7d6a0be86d400ee23ca1ac806eb358555b28d: Bug 1514869 - patch 8 - Make the SetCharacterMap message async, and use the unshared gfxCharacterMap in the content process until the shared one is in place. r=jwatt,jld
Jonathan Kew <jkew@mozilla.com> - Mon, 29 Apr 2019 14:39:05 +0000 - rev 471761
Push 112948 by shindli@mozilla.com at Mon, 29 Apr 2019 21:58:23 +0000
Bug 1514869 - patch 8 - Make the SetCharacterMap message async, and use the unshared gfxCharacterMap in the content process until the shared one is in place. r=jwatt,jld Differential Revision: https://phabricator.services.mozilla.com/D24137
21ef00977ab69cd330d727d582fef291276391c4: Bug 1514869 - patch 1 - Basic implementation of a cross-process sharable font list, using shared memory to store the list of families & faces, and per-font character maps. r=jwatt,jld
Jonathan Kew <jkew@mozilla.com> - Sat, 27 Apr 2019 15:37:29 +0000 - rev 471754
Push 112948 by shindli@mozilla.com at Mon, 29 Apr 2019 21:58:23 +0000
Bug 1514869 - patch 1 - Basic implementation of a cross-process sharable font list, using shared memory to store the list of families & faces, and per-font character maps. r=jwatt,jld Differential Revision: https://phabricator.services.mozilla.com/D22937
e25e95c59e6b1b913d895183d00461175ae5ab74: Bug 1494469 - Removed duplicate IPC::MAX_MESSAGE_SIZE constant r=jld
hafsa <hijitsuhisa@gmail.com> - Fri, 12 Apr 2019 13:01:41 +0000 - rev 469524
Push 112801 by ccoroiu@mozilla.com at Mon, 15 Apr 2019 21:40:09 +0000
Bug 1494469 - Removed duplicate IPC::MAX_MESSAGE_SIZE constant r=jld The maximum size of an IPC message is defined twice; IPC::Channel::kMaximumMessageSize = 256 MiB and IPC::MAX_MESSAGE_SIZE = 65 KiB . The latter one is used for Windows printing and its small size could cause errors. I removed the small constant, and made changes to the files using it : - The windows printing file now uses kMaximumMessageSize / 2 - nsWebBrowserPersist uses its old constant Differential Revision: https://phabricator.services.mozilla.com/D24972
b34795658d239ba6415a9ad5e72fe10d7de38265: Bug 1541130: Add XRE_GetChildProcBinPathType to nsXULAppAPI.h; r=froydnj,jld on a CLOSED TREE
Aaron Klotz <aklotz@mozilla.com> - Tue, 09 Apr 2019 00:45:24 +0300 - rev 468442
Push 112717 by rmaries@mozilla.com at Tue, 09 Apr 2019 03:49:06 +0000
Bug 1541130: Add XRE_GetChildProcBinPathType to nsXULAppAPI.h; r=froydnj,jld on a CLOSED TREE The binary path type of a particular content process is useful information outside of IPC. Given that `XRE_EnableSameExecutableForContentProc` already exists, and given that the binary path type is closely related to `GeckoProcessType`, I've added a new function, `XRE_GetContentProcBinPathType`. The mapping of process type to binary type has been moved to the `GeckoProcessTypes` definitions. This patch also modifies `ipc::GeckoChildProcessHost` to call into the new function. Differential Revision: https://phabricator.services.mozilla.com/D25816
7ecc024df5d45ada7499e2f70d79670e8c9d4d51: Bug 1541130: Add XRE_GetChildProcBinPathType to nsXULAppAPI.h; r=froydnj,jld
Aaron Klotz <aklotz@mozilla.com> - Mon, 08 Apr 2019 20:19:51 +0000 - rev 468437
Push 112717 by rmaries@mozilla.com at Tue, 09 Apr 2019 03:49:06 +0000
Bug 1541130: Add XRE_GetChildProcBinPathType to nsXULAppAPI.h; r=froydnj,jld The binary path type of a particular content process is useful information outside of IPC. Given that `XRE_EnableSameExecutableForContentProc` already exists, and given that the binary path type is closely related to `GeckoProcessType`, I've added a new function, `XRE_GetContentProcBinPathType`. The mapping of process type to binary type has been moved to the `GeckoProcessTypes` definitions. This patch also modifies `ipc::GeckoChildProcessHost` to call into the new function. Differential Revision: https://phabricator.services.mozilla.com/D25816
656108173cde7bbc6bab7dfee15dea907b0af05a: Bug 1541130: Follow-up: Add more missing parts that had been stripped during rebase; r=froydnj, jld on a CLOSED TREE
Aaron Klotz <aklotz@mozilla.com> - Mon, 08 Apr 2019 13:28:04 -0600 - rev 468417
Push 112717 by rmaries@mozilla.com at Tue, 09 Apr 2019 03:49:06 +0000
Bug 1541130: Follow-up: Add more missing parts that had been stripped during rebase; r=froydnj, jld on a CLOSED TREE
5914a425bc67dc5131c9625052362ebd777aa841: Bug 1541130: Follow-up: Add missing parts that had been stripped out during rebase; r=froydnj, jld on a CLOSED TREE
Aaron Klotz <aklotz@mozilla.com> - Mon, 08 Apr 2019 12:55:55 -0600 - rev 468416
Push 112717 by rmaries@mozilla.com at Tue, 09 Apr 2019 03:49:06 +0000
Bug 1541130: Follow-up: Add missing parts that had been stripped out during rebase; r=froydnj, jld on a CLOSED TREE
8ef30ebad16c538e39b9fb4648a551a3257d9729: Bug 1541130: Add XRE_GetChildProcBinPathType to nsXULAppAPI.h; r=froydnj,jld
Aaron Klotz <aklotz@mozilla.com> - Mon, 08 Apr 2019 18:19:06 +0000 - rev 468406
Push 112715 by rmaries@mozilla.com at Mon, 08 Apr 2019 21:03:16 +0000
Bug 1541130: Add XRE_GetChildProcBinPathType to nsXULAppAPI.h; r=froydnj,jld The binary path type of a particular content process is useful information outside of IPC. Given that `XRE_EnableSameExecutableForContentProc` already exists, and given that the binary path type is closely related to `GeckoProcessType`, I've added a new function, `XRE_GetContentProcBinPathType`. The mapping of process type to binary type has been moved to the `GeckoProcessTypes` definitions. This patch also modifies `ipc::GeckoChildProcessHost` to call into the new function. Differential Revision: https://phabricator.services.mozilla.com/D25816
8075337e7cf44caf1b78bc41918c9e475e858a13: bug 1469694 - GeckoChildProcessHost should check the return value of LaunchApp r=jld
Khyati Agarwal <f20160385@hyderabad.bits-pilani.ac.in> - Mon, 01 Apr 2019 21:58:23 +0000 - rev 467784
Push 112658 by aciure@mozilla.com at Thu, 04 Apr 2019 04:41:45 +0000
bug 1469694 - GeckoChildProcessHost should check the return value of LaunchApp r=jld Differential Revision: https://phabricator.services.mozilla.com/D24837
80945e722421a005e54c3c37725c14992b32a23a: Bug 1535794 - Allow getrandom() in all sandbox policies. r=jld
Gian-Carlo Pascutto <gcp@mozilla.com> - Tue, 02 Apr 2019 12:37:58 +0000 - rev 467610
Push 112638 by dvarga@mozilla.com at Wed, 03 Apr 2019 06:18:49 +0000
Bug 1535794 - Allow getrandom() in all sandbox policies. r=jld Differential Revision: https://phabricator.services.mozilla.com/D25610
42b734a0a83dac48657497ba3ac904915f3e479e: Bug 1536137 - Add getrusage to RDD policy. r=jld
Gian-Carlo Pascutto <gcp@mozilla.com> - Mon, 01 Apr 2019 16:10:05 +0000 - rev 467401
Push 112625 by csabou@mozilla.com at Mon, 01 Apr 2019 22:06:22 +0000
Bug 1536137 - Add getrusage to RDD policy. r=jld Depends on D24190 Differential Revision: https://phabricator.services.mozilla.com/D25621
6de4d77e6769db34a26ef40d8558da9ddbf1ada7: Bug 1536127 - Add F_SETFD/CLOEXEC logic to RDD process. r=jld
Gian-Carlo Pascutto <gcp@mozilla.com> - Mon, 01 Apr 2019 16:10:07 +0000 - rev 467400
Push 112625 by csabou@mozilla.com at Mon, 01 Apr 2019 22:06:22 +0000
Bug 1536127 - Add F_SETFD/CLOEXEC logic to RDD process. r=jld Differential Revision: https://phabricator.services.mozilla.com/D24190
d5b5baea492ae20badf1a8d45189aeaa0ef0cc7d: Bug 1538045 - use MAP_FAILED instead of (void*)-1 in shared_memory_posix.cc. r=jld
Cameron McCormack <cam@mcc.id.au> - Tue, 26 Mar 2019 22:06:18 +0000 - rev 466267
Push 112571 by csabou@mozilla.com at Wed, 27 Mar 2019 05:03:16 +0000
Bug 1538045 - use MAP_FAILED instead of (void*)-1 in shared_memory_posix.cc. r=jld Differential Revision: https://phabricator.services.mozilla.com/D24458
9d97860471b45d4096acb7d9ed8c10c7d3c602ff: Bug 1212103 : Added assert to check if the id is exceeding bounds of signed 32 bit integer r=jld
srujana <srujana.peddinti@students.iiit.ac.in> - Fri, 22 Mar 2019 19:06:02 +0000 - rev 465780
Push 112523 by shindli@mozilla.com at Sat, 23 Mar 2019 09:54:29 +0000
Bug 1212103 : Added assert to check if the id is exceeding bounds of signed 32 bit integer r=jld Added assert to check if the id is exceeding bounds of signed 32 bit integer. ``` x < (1<<29) x+1 <= (1<<29) ((x+1)<<2) <= (1<<31) ((x+1)<<2) | (two bit tag t)<= (1<<31) ``` Differential Revision: https://phabricator.services.mozilla.com/D23808
dc8935d7c0b10afc0401049936cc9d5f9fc5b003: Bug 1375863 - fold MOZ_CONTENT_SANDBOX and MOZ_GMP_SANDBOX into MOZ_SANDBOX; r=jld,firefox-build-system-reviewers
Alex Gaynor <agaynor@mozilla.com> - Mon, 18 Mar 2019 22:31:59 +0000 - rev 465029
Push 112488 by rmaries@mozilla.com at Tue, 19 Mar 2019 22:11:19 +0000
Bug 1375863 - fold MOZ_CONTENT_SANDBOX and MOZ_GMP_SANDBOX into MOZ_SANDBOX; r=jld,firefox-build-system-reviewers Differential Revision: https://phabricator.services.mozilla.com/D22975
64abd0254f73af2d4580b5ec7188f513a3fc785a: Bug 1530245 - Launch sandbox from new remote sandbox broker process. r=jld,bobowen
Chris Pearce <cpearce@mozilla.com> - Wed, 13 Mar 2019 09:24:52 +0000 - rev 463794
Push 112413 by opoprus@mozilla.com at Wed, 13 Mar 2019 16:20:20 +0000
Bug 1530245 - Launch sandbox from new remote sandbox broker process. r=jld,bobowen Depends on D22051 Differential Revision: https://phabricator.services.mozilla.com/D22052
2302287136a3aed6a9a445377ee588ad7b05c2ab: Bug 1527135 - Remove unnecessary EI_OSABI and EI_ABIVERSION checks. r=jld
Markus Stange <mstange@themasta.com> - Tue, 19 Feb 2019 20:10:56 +0000 - rev 459992
Push 112027 by dvarga@mozilla.com at Wed, 20 Feb 2019 04:11:33 +0000
Bug 1527135 - Remove unnecessary EI_OSABI and EI_ABIVERSION checks. r=jld Differential Revision: https://phabricator.services.mozilla.com/D20069
cce258ac041bd88e0414a836de77eb86abe9753a: Bug 1500596 - pt1 - modify IPC sync message name for making remote decoders. r=jld
Michael Froman <mfroman@mozilla.com> - Thu, 14 Feb 2019 19:07:08 +0000 - rev 459602
Push 111982 by ccoroiu@mozilla.com at Sat, 16 Feb 2019 09:42:42 +0000
Bug 1500596 - pt1 - modify IPC sync message name for making remote decoders. r=jld Differential Revision: https://phabricator.services.mozilla.com/D18639
326199603b5dfa09155f0dbfd203dc477bcac768: Bug 1525813 - Ensure ipc::ShmemCreated message serializes shmem size as uint32_t. r=jld
Chris Pearce <cpearce@mozilla.com> - Thu, 07 Feb 2019 16:25:37 +1300 - rev 458070
Push 111767 by cpearce@mozilla.com at Fri, 08 Feb 2019 00:54:25 +0000
Bug 1525813 - Ensure ipc::ShmemCreated message serializes shmem size as uint32_t. r=jld Shmem sizes serialized in an ipc::ShmemCreated message should be sent as an uint32_t rather than a size_t, as size_t is defined as different sizes in 64 and 32 bit builds. If the size isn't consistent, we won't be able to reliably send this message between cross architecture processes. Also, Shmem's have a limit of 32bit for their size anyway: https://searchfox.org/mozilla-central/rev/e00ea598e52bbb35f8c45abf9c2eade17962bb5e/ipc/glue/Shmem.cpp#127 So this patch should not affect allocations of shmems. Differential Revision: https://phabricator.services.mozilla.com/D18909
4dfee3ce7d5eda9eb7ba1d0f4e91b18b11caecdb: Bug 1498079 - Use a var cache for the PreallocatedProcessManager delay pref. r=jld
Mike Conley <mconley@mozilla.com> - Tue, 15 Jan 2019 20:17:13 +0000 - rev 454015
Push 111191 by dvarga@mozilla.com at Wed, 16 Jan 2019 04:58:43 +0000
Bug 1498079 - Use a var cache for the PreallocatedProcessManager delay pref. r=jld This should allow us to avoid accessing this pref too many times during start-up. Differential Revision: https://phabricator.services.mozilla.com/D16584
c28b226ee6c1d383487fb7dd75a14f419221a143: Bug 1514874 - start RDD process on-demand r=jya,jld
Michael Froman <mfroman@mozilla.com> - Sat, 12 Jan 2019 04:51:20 +0000 - rev 453613
Push 111111 by nbeleuzu@mozilla.com at Sat, 12 Jan 2019 09:47:12 +0000
Bug 1514874 - start RDD process on-demand r=jya,jld Differential Revision: https://phabricator.services.mozilla.com/D15775
36802e2a349090305f53f7ef345bfe6a3f324243: Bug 1354200 - Exit instead of MOZ_CRASH on channel error in child process r=jld
James Willcox <snorp@snorp.net> - Thu, 10 Jan 2019 22:45:44 +0000 - rev 453389
Push 111085 by dvarga@mozilla.com at Fri, 11 Jan 2019 04:23:32 +0000
Bug 1354200 - Exit instead of MOZ_CRASH on channel error in child process r=jld Android mercilessly kills the parent in low memory situations, and we don't want that to trigger a crash when the child is abruptly disconnected. Differential Revision: https://phabricator.services.mozilla.com/D16234
8d953c25fdef67aac8b006988f072a1594516938: Bug 1514874 - start RDD process on-demand r=jya,jld
Michael Froman <mfroman@mozilla.com> - Thu, 10 Jan 2019 18:35:06 +0000 - rev 453366
Push 111085 by dvarga@mozilla.com at Fri, 11 Jan 2019 04:23:32 +0000
Bug 1514874 - start RDD process on-demand r=jya,jld Differential Revision: https://phabricator.services.mozilla.com/D15775
6a7b315a82b2ebc9402166d886fa8c93d805df10: Bug 1511099 - Make MAY_CREATE handling more consistent. r=jld
Gian-Carlo Pascutto <gcp@mozilla.com> - Mon, 07 Jan 2019 15:21:32 +0000 - rev 452746
Push 110970 by ncsoregi@mozilla.com at Mon, 07 Jan 2019 21:51:41 +0000
Bug 1511099 - Make MAY_CREATE handling more consistent. r=jld Differential Revision: https://phabricator.services.mozilla.com/D13523
a925aa11dd43eeddf065ec9f795ef4ab17050ea4: Bug 1517567 - remove unused FILETIME-related functions from chromium ipc; r=jld
Nathan Froyd <froydnj@mozilla.com> - Fri, 04 Jan 2019 09:13:05 -0500 - rev 452523
Push 110911 by nfroyd@mozilla.com at Fri, 04 Jan 2019 14:13:31 +0000
Bug 1517567 - remove unused FILETIME-related functions from chromium ipc; r=jld This removes one include of windows.h, which is nice.
71cbbfe1d55b826bb101e1a1f8cb6abc050bb913: Bug 1511557 - Do not leak the pipe used for sending crash annotations when hitting errors during startup r=jld
Gabriele Svelto <gsvelto@mozilla.com> - Wed, 19 Dec 2018 15:29:09 +0000 - rev 451371
Push 110665 by rmaries@mozilla.com at Thu, 20 Dec 2018 05:11:56 +0000
Bug 1511557 - Do not leak the pipe used for sending crash annotations when hitting errors during startup r=jld Differential Revision: https://phabricator.services.mozilla.com/D14744
3d060df6e3258ed74b800c7ee0e89e001116f98d: Bug 1497749: Get hello message from channel ID when pipe handle passed. r=jld
Bob Owen <bobowencode@gmail.com> - Tue, 18 Dec 2018 14:26:26 +0000 - rev 451104
Push 110611 by bobowencode@gmail.com at Tue, 18 Dec 2018 14:26:46 +0000
Bug 1497749: Get hello message from channel ID when pipe handle passed. r=jld
15f8b27f13054b6c1029d2b2c47968a63cbd9c93: Bug 1511557 - Do not leak the pipe used for sending crash annotations when hitting errors during startup r=jld
Gabriele Svelto <gsvelto@mozilla.com> - Fri, 14 Dec 2018 02:47:03 +0000 - rev 450685
Push 110520 by rmaries@mozilla.com at Fri, 14 Dec 2018 16:36:04 +0000
Bug 1511557 - Do not leak the pipe used for sending crash annotations when hitting errors during startup r=jld Differential Revision: https://phabricator.services.mozilla.com/D13999
1c1f445a9b6a98cc10348c6eb1c9d8bdf72ca412: Bug 1509362 - Don't crash when constructing actor during content shutdown, r=jld
Nika Layzell <nika@thelayzells.com> - Thu, 22 Nov 2018 13:40:32 -0500 - rev 449396
Push 110362 by nika@thelayzells.com at Wed, 05 Dec 2018 15:19:54 +0000
Bug 1509362 - Don't crash when constructing actor during content shutdown, r=jld When shutting down a content process, we call `Close` on the `IToplevelProtocol`. This causes the MessageChannel to be `Close`-ed, which in turn sends a `GOODBYE_MESSAGE`: https://searchfox.org/mozilla-central/rev/876022232b15425bb9efde189caf747823b39567/ipc/glue/MessageChannel.cpp#2852 This message is intercepted on the I/O thread in the content process, before any code is informed in content, and used to set the `mChannelState` property to `ChannelClosing`: https://searchfox.org/mozilla-central/rev/876022232b15425bb9efde189caf747823b39567/ipc/glue/MessageChannel.cpp#1176 Once this state has been set, which is performed as soon as the message is received, whether or not other messages have been processed yet, no messages can be sent back to the parent process. This is usually what causes the 'Too late to send/recv' message spam in the console, as we're still trying to send messages at this time. Usually this is fine - the message send fails, but we gracefully recover, and the process begins shutting down like normal. Unfortunately, child actor constructors currently have code automatically generated in them which causes a process crash if the send fails. As it's impossible for the main thread to know that the channel has been closed ahead of time (due to this happening out-of-band), we can then cause random content process crashes during shutdown due to actor construction. Unfortunately, we can't just destroy the actor, as our caller may (and often do) depend on the actor reference they gave us still being valid after calling Send*Constructor. Fortunately, if a message send failed, it means we're in the process of being shut down. This patch handles this by ignoring ctor send errors, and treating them like messages which successfully were queued to send, but got lost due to the other side hanging up. The actor will be gracefully destroyed in DestroySubtree when its manager is destroyed. Differential Revision: https://phabricator.services.mozilla.com/D12695
7e23a1b903356b3bdfbf1844bee62513969b93a6: Bug 1509362 - Don't crash when constructing actor during content shutdown, r=jld
Nika Layzell <nika@thelayzells.com> - Thu, 22 Nov 2018 13:40:32 -0500 - rev 449285
Push 110347 by nika@thelayzells.com at Tue, 04 Dec 2018 16:25:58 +0000
Bug 1509362 - Don't crash when constructing actor during content shutdown, r=jld When shutting down a content process, we call `Close` on the `IToplevelProtocol`. This causes the MessageChannel to be `Close`-ed, which in turn sends a `GOODBYE_MESSAGE`: https://searchfox.org/mozilla-central/rev/876022232b15425bb9efde189caf747823b39567/ipc/glue/MessageChannel.cpp#2852 This message is intercepted on the I/O thread in the content process, before any code is informed in content, and used to set the `mChannelState` property to `ChannelClosing`: https://searchfox.org/mozilla-central/rev/876022232b15425bb9efde189caf747823b39567/ipc/glue/MessageChannel.cpp#1176 Once this state has been set, which is performed as soon as the message is received, whether or not other messages have been processed yet, no messages can be sent back to the parent process. This is usually what causes the 'Too late to send/recv' message spam in the console, as we're still trying to send messages at this time. Usually this is fine - the message send fails, but we gracefully recover, and the process begins shutting down like normal. Unfortunately, child actor constructors currently have code automatically generated in them which causes a process crash if the send fails. As it's impossible for the main thread to know that the channel has been closed ahead of time (due to this happening out-of-band), we can then cause random content process crashes during shutdown due to actor construction. Unfortunately, we can't just destroy the actor, as our caller may (and often do) depend on the actor reference they gave us still being valid after calling Send*Constructor. Fortunately, if a message send failed, it means we're in the process of being shut down. This patch handles this by ignoring ctor send errors, and treating them like messages which successfully were queued to send, but got lost due to the other side hanging up. The actor will be gracefully destroyed in DestroySubtree when its manager is destroyed. Differential Revision: https://phabricator.services.mozilla.com/D12695
7725c317d72320510a3c792a60185179a4183632: Bug 1507830 - When using Wayland, don't allow connections to the X server. r=jld
Gian-Carlo Pascutto <gcp@mozilla.com> - Fri, 30 Nov 2018 22:23:31 +0000 - rev 449046
Push 110294 by ccoroiu@mozilla.com at Sat, 01 Dec 2018 05:58:04 +0000
Bug 1507830 - When using Wayland, don't allow connections to the X server. r=jld Differential Revision: https://phabricator.services.mozilla.com/D13560
a06e4cff09e6fe6ca41626310d6494a3d276772a: Bug 1509930 - Simplify CycleCollectWithLogsChild's lifetime, r=jld
Nika Layzell <nika@thelayzells.com> - Fri, 23 Nov 2018 17:07:41 -0500 - rev 448512
Push 110205 by nika@thelayzells.com at Wed, 28 Nov 2018 17:59:22 +0000
Bug 1509930 - Simplify CycleCollectWithLogsChild's lifetime, r=jld Previously this actor would call `Send__delete__` in its destructor, meaning that the lifetime of the actor is a bit hard to follow. This changes the actor to instead use a separate object for XPCOM, and use refcounting universally. This should also help with the transition to universally refcounted actors. Differential Revision: https://phabricator.services.mozilla.com/D12955
c3fe435e473a463fbc22d4afa531bdedb757079c: Bug 1509362 - Don't crash when constructing actor during content shutdown, r=jld
Nika Layzell <nika@thelayzells.com> - Thu, 22 Nov 2018 13:40:32 -0500 - rev 448407
Push 110168 by nika@thelayzells.com at Tue, 27 Nov 2018 23:11:59 +0000
Bug 1509362 - Don't crash when constructing actor during content shutdown, r=jld When shutting down a content process, we call `Close` on the `IToplevelProtocol`. This causes the MessageChannel to be `Close`-ed, which in turn sends a `GOODBYE_MESSAGE`: https://searchfox.org/mozilla-central/rev/876022232b15425bb9efde189caf747823b39567/ipc/glue/MessageChannel.cpp#2852 This message is intercepted on the I/O thread in the content process, before any code is informed in content, and used to set the `mChannelState` property to `ChannelClosing`: https://searchfox.org/mozilla-central/rev/876022232b15425bb9efde189caf747823b39567/ipc/glue/MessageChannel.cpp#1176 Once this state has been set, which is performed as soon as the message is received, whether or not other messages have been processed yet, no messages can be sent back to the parent process. This is usually what causes the 'Too late to send/recv' message spam in the console, as we're still trying to send messages at this time. Usually this is fine - the message send fails, but we gracefully recover, and the process begins shutting down like normal. Unfortunately, child actor constructors currently have code automatically generated in them which causes a process crash if the send fails. As it's impossible for the main thread to know that the channel has been closed ahead of time (due to this happening out-of-band), we can then cause random content process crashes during shutdown due to actor construction. Unfortunately, we can't just destroy the actor, as our caller may (and often do) depend on the actor reference they gave us still being valid after calling Send*Constructor. Fortunately, if a message send failed, it means we're in the process of being shut down. This patch handles this by ignoring ctor send errors, and treating them like messages which successfully were queued to send, but got lost due to the other side hanging up. The actor will be gracefully destroyed in DestroySubtree when its manager is destroyed. Differential Revision: https://phabricator.services.mozilla.com/D12695
bce195f98895f82bd5ace5fc5ed5a5ca3b9f1afd: Bug 1509930 - Simplify CycleCollectWithLogsChild's lifetime, r=jld
Nika Layzell <nika@thelayzells.com> - Fri, 23 Nov 2018 17:07:41 -0500 - rev 448404
Push 110168 by nika@thelayzells.com at Tue, 27 Nov 2018 23:11:59 +0000
Bug 1509930 - Simplify CycleCollectWithLogsChild's lifetime, r=jld Previously this actor would call `Send__delete__` in its destructor, meaning that the lifetime of the actor is a bit hard to follow. This changes the actor to instead use a separate object for XPCOM, and use refcounting universally. This should also help with the transition to universally refcounted actors. Differential Revision: https://phabricator.services.mozilla.com/D12955
b4754021fc2dbeb9a96fc85cacc52ec1c63d1453: Bug 1499465 - Don't collect KillHard crashes on Beta or Release. r=jld,gsvelto
Mike Conley <mconley@mozilla.com> - Tue, 20 Nov 2018 21:34:36 +0000 - rev 447431
Push 110040 by btara@mozilla.com at Wed, 21 Nov 2018 09:51:14 +0000
Bug 1499465 - Don't collect KillHard crashes on Beta or Release. r=jld,gsvelto Differential Revision: https://phabricator.services.mozilla.com/D10968
82b7ea01b869a9fc0731e4f4f07c214312d1f6aa: Bug 1471535 - pt8 - Add RDD process code. r=jld
Michael Froman <mfroman@mozilla.com> - Wed, 14 Nov 2018 18:06:24 +0000 - rev 446405
Push 109870 by rmaries@mozilla.com at Wed, 14 Nov 2018 22:38:51 +0000
Bug 1471535 - pt8 - Add RDD process code. r=jld Depends on D8489 Differential Revision: https://phabricator.services.mozilla.com/D8490
52bfaa088f57fd69fde36b54ffc6c1b4017ea1cc: Bug 1471535 - pt7 - Add remote decoding in dom/media/ipc alongside gpu decoding. r=jya,mattwoodrow,jld
Michael Froman <mfroman@mozilla.com> - Wed, 14 Nov 2018 18:06:11 +0000 - rev 446404
Push 109870 by rmaries@mozilla.com at Wed, 14 Nov 2018 22:38:51 +0000
Bug 1471535 - pt7 - Add remote decoding in dom/media/ipc alongside gpu decoding. r=jya,mattwoodrow,jld Depends on D8488 Differential Revision: https://phabricator.services.mozilla.com/D8489
404d760a9e82041d15736ce99d46b9f943641f24: Bug 1471535 - pt8 - Add RDD process code. r=jld
Michael Froman <mfroman@mozilla.com> - Tue, 13 Nov 2018 17:35:44 +0000 - rev 446096
Push 109839 by ccoroiu@mozilla.com at Wed, 14 Nov 2018 10:01:06 +0000
Bug 1471535 - pt8 - Add RDD process code. r=jld Depends on D8489 Differential Revision: https://phabricator.services.mozilla.com/D8490
28ae4b6fab688d07caaa083f0c0975f05bcfdd86: Bug 1471535 - pt7 - Add remote decoding in dom/media/ipc alongside gpu decoding. r=jya,mattwoodrow,jld
Michael Froman <mfroman@mozilla.com> - Tue, 13 Nov 2018 17:35:31 +0000 - rev 446095
Push 109839 by ccoroiu@mozilla.com at Wed, 14 Nov 2018 10:01:06 +0000
Bug 1471535 - pt7 - Add remote decoding in dom/media/ipc alongside gpu decoding. r=jya,mattwoodrow,jld Depends on D8488 Differential Revision: https://phabricator.services.mozilla.com/D8489
1a991ac2e1f8db3b9410d1cea09686ebfc38ad1b: Bug 1471535 - pt8 - Add RDD process code. r=jld
Michael Froman <mfroman@mozilla.com> - Thu, 08 Nov 2018 03:47:22 +0000 - rev 445398
Push 109742 by dluca@mozilla.com at Fri, 09 Nov 2018 17:47:42 +0000
Bug 1471535 - pt8 - Add RDD process code. r=jld Depends on D8489 Differential Revision: https://phabricator.services.mozilla.com/D8490
a32288737e57ffba27e3e7dffcbc0c263298104d: Bug 1471535 - pt7 - Add remote decoding in dom/media/ipc alongside gpu decoding. r=jya,mattwoodrow,jld
Michael Froman <mfroman@mozilla.com> - Fri, 09 Nov 2018 00:12:46 +0000 - rev 445397
Push 109742 by dluca@mozilla.com at Fri, 09 Nov 2018 17:47:42 +0000
Bug 1471535 - pt7 - Add remote decoding in dom/media/ipc alongside gpu decoding. r=jya,mattwoodrow,jld Depends on D8488 Differential Revision: https://phabricator.services.mozilla.com/D8489
4755e3693c1ff7d9e1729a51f6d9a82df367cdb7: Bug 1500219: Part 2 - Test UniquePtr integration with IPDL (r=jld)
David Parks <dparks@mozilla.com> - Wed, 24 Oct 2018 21:04:44 +0000 - rev 443219
Push 109323 by ncsoregi@mozilla.com at Sat, 27 Oct 2018 10:01:40 +0000
Bug 1500219: Part 2 - Test UniquePtr integration with IPDL (r=jld) Tests that pass UniquePtrs in IPDL messages. Depends on D9143 Differential Revision: https://phabricator.services.mozilla.com/D9144
a887c9b3dad3dd4d37cb0fb14d7aa15fc9d24118: Bug 1500219: Part 1 - Add ability to send and receive UniquePtr<T> with IPDL (r=jld)
David Parks <dparks@mozilla.com> - Fri, 26 Oct 2018 17:09:19 +0000 - rev 443217
Push 109323 by ncsoregi@mozilla.com at Sat, 27 Oct 2018 10:01:40 +0000
Bug 1500219: Part 1 - Add ability to send and receive UniquePtr<T> with IPDL (r=jld) This patch adds the ability to use UniquePtr<T> in IPDL messages if T is serializable. The behavior works as expected -- a UniquePtr in a Send is cleared and is passed by move. Some design points: * The UniquePtr identification is done in the parser. This is because the parser immediately normalizes CxxTemplateInst -- an old version of the patch did hacky string-parsing to pull it apart. I, instead, created CxxUniquePtrInst. * The implementation allows passing to Send... by move or by reference. This is valid UniquePtr API behavior but passing by reference is not really useful in this case (but not harmful). This could probably piggy-back on the "moveonly" IPDL work but that is newer than this work and will require some refactoring. Differential Revision: https://phabricator.services.mozilla.com/D9143
ee8274a9b6c09241f0da9b6ed3b83a95b8091574: Bug 1445994: Rename variables to kungFuDeathGrip to make it clear it's keeping a reference alive. r=jld
Diego Pino Garcia <dpino@igalia.com> - Thu, 18 Oct 2018 00:26:00 -0400 - rev 441802
Push 109052 by spohl@mozilla.com at Thu, 18 Oct 2018 04:27:13 +0000
Bug 1445994: Rename variables to kungFuDeathGrip to make it clear it's keeping a reference alive. r=jld
766131a8c0a9a8361c313b45f28ae8262af3dd8a: Bug 1475139 part 2 - Add 'moveonly' qualifier to IPDL using statements. r=jld
Ryan Hunt <rhunt@eqrion.net> - Mon, 24 Sep 2018 21:24:24 -0500 - rev 439953
Push 108717 by rhunt@eqrion.net at Mon, 08 Oct 2018 14:09:25 +0000
Bug 1475139 part 2 - Add 'moveonly' qualifier to IPDL using statements. r=jld I'd like to use a struct over IPDL which is move only. This patch adds a modifier to usings statements to indicate the C++ type should always be moved, similar to refcounted. Differential Revision: https://phabricator.services.mozilla.com/D6781
36356f99018a7bd5a7ef1fc57ed74139421076f1: Bug 1495871 - remove chromium atomics code; r=jld
Nathan Froyd <froydnj@mozilla.com> - Fri, 05 Oct 2018 13:43:46 -0400 - rev 439818
Push 108681 by nfroyd@mozilla.com at Fri, 05 Oct 2018 17:44:10 +0000
Bug 1495871 - remove chromium atomics code; r=jld The only consumer of this code was Singleton, which we previously removed, and everything that this code accomplished can be done more simply and more foolproof-y by standard constructs these days.
323979f32e2de8a896e97b4471daa56c20b9ee70: Bug 1495871 - remove singleton headers from ipc/; r=jld
Nathan Froyd <froydnj@mozilla.com> - Fri, 05 Oct 2018 13:43:46 -0400 - rev 439817
Push 108681 by nfroyd@mozilla.com at Fri, 05 Oct 2018 17:44:10 +0000
Bug 1495871 - remove singleton headers from ipc/; r=jld These headers are no longer used.