)]}'
{
  "log": [
    {
      "commit": "45853c975907ac9b5c4c11e9f74caa2bd01fea24",
      "tree": "74d51c4e9413b2bcae28b47e10f67bb1fca558fa",
      "parents": [
        "32820ca5f93722969129446e18d290ba81a08a5f"
      ],
      "author": {
        "name": "Mathias Agopian",
        "email": "mathias@google.com",
        "time": "Fri Feb 26 18:59:23 2010 -0800"
      },
      "committer": {
        "name": "Mathias Agopian",
        "email": "mathias@google.com",
        "time": "Fri Feb 26 18:59:23 2010 -0800"
      },
      "message": "to help debugging [2461567] Home screen redraw messed up\n\nlog SF\u0027s idea of the front buffer in dumpsys.\n"
    },
    {
      "commit": "54ba51dff21de666c5ae3bf3abd4f0634ebb0676",
      "tree": "58a69dccdbfd227b9de3c58a416de45d2a8d3513",
      "parents": [
        "36bc94173c957f33f70bcd16c369951f5b2dc3f6"
      ],
      "author": {
        "name": "Mathias Agopian",
        "email": "mathias@google.com",
        "time": "Mon Oct 26 20:12:37 2009 -0700"
      },
      "committer": {
        "name": "Mathias Agopian",
        "email": "mathias@google.com",
        "time": "Tue Oct 27 13:13:29 2009 -0700"
      },
      "message": "    fix [2143798] Need to figure out how to do video\n\n    Use EGLImageKHR instead of copybit directly.\n    We now have the basis to use streaming YUV textures (well, in fact\n    we already are). When/if we use the GPU instead of the MDP we\u0027ll\n    need to make sure it supports the appropriate YUV format.\n\n    Also make sure we compile if EGL_ANDROID_image_native_buffer is not supported\n"
    },
    {
      "commit": "57720c384af81327d327ef8b88078f11a750fd05",
      "tree": "4125091c671e9fd56bb292f54b04b3668a8fadab",
      "parents": [
        "9521fc0f17c227fa79dd4d830d4607557d060f00"
      ],
      "author": {
        "name": "Mathias Agopian",
        "email": "mathias@google.com",
        "time": "Wed Oct 21 16:27:21 2009 -0700"
      },
      "committer": {
        "name": "Mathias Agopian",
        "email": "mathias@google.com",
        "time": "Fri Oct 23 15:37:28 2009 -0700"
      },
      "message": "fix [2211532] improves sholes graphics performance\n\nInstead of using glTex{Sub}Image2D() to refresh the textures, we\u0027re using an EGLImageKHR object\nbacked up by a gralloc buffer. The data is updated using memcpy(). This is faster than\nglTex{Sub}Image2D() because the texture is not swizzled. It also uses less memory because\nEGLImageKHW is not limited to power-of-two dimensions.\n"
    },
    {
      "commit": "9ec430adaea1cb88eaa1e78c7f759cd42ab6cf7a",
      "tree": "1174b71aeaf88287d38538f8b7baa74532f6f638",
      "parents": [
        "ef8646344a7168643b50a51ebffa2e41e9717608"
      ],
      "author": {
        "name": "Mathias Agopian",
        "email": "mathias@google.com",
        "time": "Tue Oct 06 19:00:57 2009 -0700"
      },
      "committer": {
        "name": "Mathias Agopian",
        "email": "mathias@google.com",
        "time": "Tue Oct 06 19:00:57 2009 -0700"
      },
      "message": "fix [2152536] ANR in browser\n\nA window is created and the browser is about to render into it the\nvery first time, at that point it does an IPC to SF to request a new\nbuffer. Meanwhile, the window manager removes that window from the\nlist and the shared memory block it uses is marked as invalid.\nHowever, at that point, another window is created and is given the\nsame index (that just go freed), but a different identity and resets\nthe \"invalid\" bit in the shared block. When we go back to the buffer\nallocation code, we\u0027re stuck because the surface we\u0027re allocating for\nis gone and we don\u0027t detect it\u0027s invalid because the invalid bit has\nbeen reset.\n\nIt is not sufficient to check for the invalid bit, I should\nalso check that identities match.\n"
    },
    {
      "commit": "a4b740ed89074cda898a30eb1b029b0d3a5de1a5",
      "tree": "2c7f89741b0d58e569b1392372ac344b8023267e",
      "parents": [
        "b26af23744fa73e8bc142b1eb98772fde5970c10"
      ],
      "author": {
        "name": "Mathias Agopian",
        "email": "mathias@google.com",
        "time": "Mon Oct 05 18:20:39 2009 -0700"
      },
      "committer": {
        "name": "Mathias Agopian",
        "email": "mathias@google.com",
        "time": "Tue Oct 06 17:24:26 2009 -0700"
      },
      "message": "fix [2168528] enable glTexImage2D code path in SF for software-only buffers\n"
    },
    {
      "commit": "3330b203039dea366d4981db1408a460134b2d2c",
      "tree": "f165f7801f2ee8034ad1edfd34e2b554fdaf662f",
      "parents": [
        "0ad3f9f4009c76b56f41da5d26eb7712dee938f1"
      ],
      "author": {
        "name": "Mathias Agopian",
        "email": "mathias@google.com",
        "time": "Mon Oct 05 17:07:12 2009 -0700"
      },
      "committer": {
        "name": "Mathias Agopian",
        "email": "mathias@google.com",
        "time": "Tue Oct 06 17:00:25 2009 -0700"
      },
      "message": "fix [2167050] glTexImage2D code path buggy in SurfaceFlinger\n\nWhen EGLImage extension is not available, SurfaceFlinger will fallback to using\nglTexImage2D and glTexSubImage2D instead, which requires 50% more memory and an\nextra copy. However this code path has never been exercised and had some bugs\nwhich this patch fix.\n\nMainly the scale factor wasn\u0027t computed right when falling back on glDrawElements.\nWe also fallback to this mode of operation if a buffer doesn\u0027t have the adequate\nusage bits for EGLImage usage.\n\nThis changes only code that is currently not executed. Some refactoring was needed to\nkeep the change clean. This doesn\u0027t change anything functionaly.\n"
    },
    {
      "commit": "0b3ad46a26dc3717260fa9347c77f673f3198606",
      "tree": "e101cea80b6258fa478b702cfef622747f3f9e39",
      "parents": [
        "c91af0c12e24d56262a2c100d9e46de26b33fb91"
      ],
      "author": {
        "name": "Mathias Agopian",
        "email": "mathias@google.com",
        "time": "Fri Oct 02 18:12:30 2009 -0700"
      },
      "committer": {
        "name": "Mathias Agopian",
        "email": "mathias@google.com",
        "time": "Fri Oct 02 18:12:30 2009 -0700"
      },
      "message": "Attempt to fix [2152536] ANR in browser\n\nThe ANR is caused by SurfaceFlinger waiting for buffers of a removed surface to become availlable.\nWhen it is removed from the current list, a Surface is marked as NO_INIT, which causes SF to return\nimmediately in the above case. For some reason, the surface here wasn\u0027t marked as NO_INIT.\n\nThis change makes the code more robust by always (irregadless or errors) setting the NO_INIT status\nin all code paths where a surface is removed from the list.\n\nAdditionaly added more information in the logs, should this happen again.\n"
    },
    {
      "commit": "401c257fba770a267f637184b1f532b4e03bed20",
      "tree": "5ed9e89c7436a7f738818287d7722ec6f98627b5",
      "parents": [
        "a2fe0a234bd6823f868742a3c137acb8875e1b03"
      ],
      "author": {
        "name": "Mathias Agopian",
        "email": "mathias@google.com",
        "time": "Wed Sep 23 19:16:27 2009 -0700"
      },
      "committer": {
        "name": "Mathias Agopian",
        "email": "mathias@google.com",
        "time": "Wed Sep 23 19:16:27 2009 -0700"
      },
      "message": "turn dithering off if it\u0027s not needed\n"
    },
    {
      "commit": "48d819a1315f7d1b5abfec9d4fd34fb5aed27b1d",
      "tree": "610ef98303e468061f68511380991e4b65bd547a",
      "parents": [
        "b58b5d72d6877300fe39b2a30b4742c6f962833f"
      ],
      "author": {
        "name": "Mathias Agopian",
        "email": "mathias@google.com",
        "time": "Thu Sep 10 19:41:18 2009 -0700"
      },
      "committer": {
        "name": "Mathias Agopian",
        "email": "mathias@google.com",
        "time": "Thu Sep 10 19:41:18 2009 -0700"
      },
      "message": "fix [2112575] stuck on DequeueCondition for a surface that doesn\u0027t exist anymore\n\nthis also fixes part of [2111536] Device is soft rebooted after ending the call through voice dialer\n"
    },
    {
      "commit": "cbb288bfe89f585bf48371bd31b2d4aafa32f32e",
      "tree": "73d20f36910dcab8c17c686cdd7014cee285bc39",
      "parents": [
        "f0780974fc31ae88135d1dcb67cb5fd86bb6deb6"
      ],
      "author": {
        "name": "Mathias Agopian",
        "email": "mathias@google.com",
        "time": "Mon Sep 07 16:32:45 2009 -0700"
      },
      "committer": {
        "name": "Mathias Agopian",
        "email": "mathias@google.com",
        "time": "Mon Sep 07 16:32:45 2009 -0700"
      },
      "message": "fix [2068105] implement queueBuffer/lockBuffer/dequeueBuffer properly\n\nRewrote SurfaceFlinger\u0027s buffer management from the ground-up.\nThe design now support an arbitrary number of buffers per surface, however the current implementation is limited to four. Currently only 2 buffers are used in practice.\n\nThe main new feature is to be able to dequeue all buffers at once (very important when there are only two). \n\nA client can dequeue all buffers until there are none available, it can lock all buffers except the last one that is used for composition. The client will block then, until a new buffer is enqueued.\n\nThe current implementation requires that buffers are locked in the same order they are dequeued and enqueued in the same order they are locked. Only one buffer can be locked at a time.\n\neg. Allowed sequence:   DQ, DQ, LOCK, Q, LOCK, Q\neg. Forbidden sequence: DQ, DQ, LOCK, LOCK, Q, Q\n\n"
    },
    {
      "commit": "5221271375f361b84a6eeec3d7086f223997fbb3",
      "tree": "f65cab160cf38b9ed577f1cb5bc2f9882fc8631c",
      "parents": [
        "40ef81c7fce7cd1b028aec7b8b27b77ce6bf7b90"
      ],
      "author": {
        "name": "Mathias Agopian",
        "email": "mathias@google.com",
        "time": "Tue Aug 11 22:34:02 2009 -0700"
      },
      "committer": {
        "name": "Mathias Agopian",
        "email": "mathias@google.com",
        "time": "Tue Aug 11 23:32:29 2009 -0700"
      },
      "message": "second take, hopefully this time it doesn\u0027t break one of the builds: \"SurfaceFlinger will now allocate buffers based on the usage specified by the clients. This allows to allocate the right kind of buffer automatically, without having the user to specify anything.\"\n"
    },
    {
      "commit": "b2fd4665e66fe81969f1f07fd724c6f68dd8f7f1",
      "tree": "8564d48ba85748bdb281a3bcda22b41ed17c72e8",
      "parents": [
        "01b535185805d5b22c5fffc30833decdca308b5b"
      ],
      "author": {
        "name": "Fred Quintana",
        "email": "fredq@google.com",
        "time": "Tue Aug 11 20:49:35 2009 -0700"
      },
      "committer": {
        "name": "Fred Quintana",
        "email": "fredq@google.com",
        "time": "Tue Aug 11 20:49:35 2009 -0700"
      },
      "message": "Revert \"SurfaceFlinger will now allocate buffers based on the usage specified by the clients. This allows to allocate the right kind of buffer automatically, without having the user to specify anything.\"\n\nThis reverts commit 8b76a0ac6fbf07254629ed1ea86af014d5abe050.\n"
    },
    {
      "commit": "df37b62c62f8efd3b5a433f9b4d40d30cec13a31",
      "tree": "43c783e665b9e1c47210d3c5a8f929b33acda72f",
      "parents": [
        "d779f052b4180f347ed0bd725fdfedda504773e5"
      ],
      "author": {
        "name": "Mathias Agopian",
        "email": "mathias@google.com",
        "time": "Mon Aug 10 21:59:56 2009 -0700"
      },
      "committer": {
        "name": "Mathias Agopian",
        "email": "mathias@google.com",
        "time": "Tue Aug 11 16:12:56 2009 -0700"
      },
      "message": "SurfaceFlinger will now allocate buffers based on the usage specified by the clients. This allows to allocate the right kind of buffer automatically, without having the user to specify anything.\n\nThis change makes SurfaceHolder.setType(GPU) obsolete (it\u0027s now ignored).\nAdded an API to android_native_window_t to allow extending the functionality without ever breaking binary compatibility. This is used to implement the new set_usage() API. This API needs to be called by software renderers because the default is to use usage flags suitable for h/w.\n"
    },
    {
      "commit": "7303c6bf1a8b00a0e7d8165d774a1f259b4ccda9",
      "tree": "329ae25bcf03c06ddca3828a038bd499c63dbee0",
      "parents": [
        "6b5513538aedc19c624ff7a260c832dcfae630f2"
      ],
      "author": {
        "name": "Mathias Agopian",
        "email": "mathias@google.com",
        "time": "Thu Jul 02 18:11:53 2009 -0700"
      },
      "committer": {
        "name": "Mathias Agopian",
        "email": "mathias@google.com",
        "time": "Thu Jul 02 18:50:51 2009 -0700"
      },
      "message": "get rid of references to MemoryDealer in SurfaceFlinger\n"
    },
    {
      "commit": "1fed11c86a9d59d0f5282ae8ae25ceba2f802fdd",
      "tree": "940032f8fb572ae8096b79027382ddb0538404fd",
      "parents": [
        "9f8b0c909760aa5b866255fe08042a3200b736cc"
      ],
      "author": {
        "name": "Mathias Agopian",
        "email": "mathias@google.com",
        "time": "Tue Jun 23 18:08:22 2009 -0700"
      },
      "committer": {
        "name": "Mathias Agopian",
        "email": "mathias@google.com",
        "time": "Tue Jun 23 18:08:22 2009 -0700"
      },
      "message": "checkpoint. bring back video/camera\n"
    },
    {
      "commit": "f9d932774e06d5122c48b47d8cabd791783f56d2",
      "tree": "e78c8148d0e3fcc166ebb3f4bd60c931ec956344",
      "parents": [
        "cd8c5e29c245e55a5f648b7a10f8586baf64e622"
      ],
      "author": {
        "name": "Mathias Agopian",
        "email": "mathias@google.com",
        "time": "Fri Jun 19 17:00:27 2009 -0700"
      },
      "committer": {
        "name": "Mathias Agopian",
        "email": "mathias@google.com",
        "time": "Fri Jun 19 17:00:27 2009 -0700"
      },
      "message": "fix a memory corruption where a SF Client could be used after it\u0027s been destroyed\n"
    },
    {
      "commit": "0aa758d64ac530833ce9a311b164c9175a7a054c",
      "tree": "dffe53978839a5d7cf270d88a94f4c3b084a37fa",
      "parents": [
        "f1d8e87b09abf963cd5b6a026194c1940fadb7b4"
      ],
      "author": {
        "name": "Mathias Agopian",
        "email": "mathias@google.com",
        "time": "Wed Apr 22 15:23:34 2009 -0700"
      },
      "committer": {
        "name": "Mathias Agopian",
        "email": "mathias@google.com",
        "time": "Fri Apr 24 16:30:38 2009 -0700"
      },
      "message": "Surfaces are now destroyed properly in SurfaceFlinger.\n\nFirst, the window manager tells us when a surface is no longer needed. At this point, several things happen:\n- the surface is removed from the active/visible list\n- it is added to a purgatory list, where it waits for all clients to release their reference\n- it destroys all data/state that can be spared\n\nLater, when all clients are done, the remains of the Surface are disposed off: it is removed from the purgatory and destroyed.\nIn particular its gralloc buffers are destroyed at that point (when we\u0027re sure nobody is using them anymore).\n"
    },
    {
      "commit": "9a11206fe793363c0e8897b478cbe6ef8c52b543",
      "tree": "0b7fc29027a6803e1ab992f47f0dd66c2a3e3843",
      "parents": [
        "9f88afb013a7560bf1362d7999a4609e38d0ea77"
      ],
      "author": {
        "name": "Mathias Agopian",
        "email": "mathias@google.com",
        "time": "Fri Apr 17 19:36:26 2009 -0700"
      },
      "committer": {
        "name": "Mathias Agopian",
        "email": "mathias@google.com",
        "time": "Fri Apr 24 15:00:41 2009 -0700"
      },
      "message": "more Surface lifetime management\n\nSurfaces are now destroyed once all references from the clients are gone, but they go through a partial destruction as soon as the window manager requests it.\nThis last part is still buggy. see comments in SurfaceFlinger::destroySurface()\n"
    },
    {
      "commit": "076b1cc3a9b90aa5b381a1ed268ca0b548444c9b",
      "tree": "b87f7863623b37fa9a6cc4ef48934aab46f79986",
      "parents": [
        "4ec21527096d0dcaa99634e6058aca3a415461af"
      ],
      "author": {
        "name": "Mathias Agopian",
        "email": "mathias@google.com",
        "time": "Fri Apr 10 14:24:30 2009 -0700"
      },
      "committer": {
        "name": "Mathias Agopian",
        "email": "mathias@google.com",
        "time": "Fri Apr 10 14:24:30 2009 -0700"
      },
      "message": "Integrate from //sandbox/mathias/donut/...@145728\n\nSurfaceFlinger rework for new EGL driver model support.\n"
    },
    {
      "commit": "edbf3b6af777b721cd2a1ef461947e51e88241e1",
      "tree": "f09427b843b192cccf8c3b5328cb81dddf6489fa",
      "parents": [
        "d5193d9394c5e58176d7bcdf50ef017f8a3b9e1e"
      ],
      "author": {
        "name": "The Android Open Source Project",
        "email": "initial-contribution@android.com",
        "time": "Tue Mar 03 19:31:44 2009 -0800"
      },
      "committer": {
        "name": "The Android Open Source Project",
        "email": "initial-contribution@android.com",
        "time": "Tue Mar 03 19:31:44 2009 -0800"
      },
      "message": "auto import from //depot/cupcake/@135843\n"
    },
    {
      "commit": "d5193d9394c5e58176d7bcdf50ef017f8a3b9e1e",
      "tree": "4b825dc642cb6eb9a060e54bf8d69288fbee4904",
      "parents": [
        "43aa2b1cbf7a03e248e10f4d0fec0463257cd52d"
      ],
      "author": {
        "name": "The Android Open Source Project",
        "email": "initial-contribution@android.com",
        "time": "Tue Mar 03 18:28:45 2009 -0800"
      },
      "committer": {
        "name": "The Android Open Source Project",
        "email": "initial-contribution@android.com",
        "time": "Tue Mar 03 18:28:45 2009 -0800"
      },
      "message": "auto import from //depot/cupcake/@135843\n"
    },
    {
      "commit": "7c1b96a165f970a09ed239bb4fb3f1b0d8f2a407",
      "tree": "df5a6539447324de36e95b057d6b9f0361b7a250",
      "parents": [],
      "author": {
        "name": "The Android Open Source Project",
        "email": "initial-contribution@android.com",
        "time": "Tue Oct 21 07:00:00 2008 -0700"
      },
      "committer": {
        "name": "The Android Open Source Project",
        "email": "initial-contribution@android.com",
        "time": "Tue Oct 21 07:00:00 2008 -0700"
      },
      "message": "Initial Contribution\n"
    }
  ]
}