)]}'
{
  "log": [
    {
      "commit": "68aeec07de4a1894a5cefde405abcb5a49801df6",
      "tree": "a61f11ed0d1f343f982f8a76556b04d4d0269cae",
      "parents": [
        "c8098329300b4c6421c541d16442ba13dad57c33"
      ],
      "author": {
        "name": "Brian Salomon",
        "email": "bsalomon@google.com",
        "time": "Tue Apr 16 11:01:13 2019 -0400"
      },
      "committer": {
        "name": "Skia Commit-Bot",
        "email": "skia-commit-bot@chromium.org",
        "time": "Tue Apr 16 16:52:44 2019 +0000"
      },
      "message": "Clarify that buffers of type GrGpuBufferType::kXferGpuToCpu are mapped for reading.\n\nDon\u0027t allow GrGpuBuffer::updateData on kXferGpuToCpu (in GL this can resize the buffer\nto a smaller size).\n\nDon\u0027t call glBufferData with NULL prior to mapping for reading.\n\nDon\u0027t set GL_MAP_WRITE_BIT when using glMapBufferRange for reading only.\n\nBug: skia:8962\nChange-Id: I605d83fa40c7b170082c48a456436d97cf3b70a5\nReviewed-on: https://skia-review.googlesource.com/c/skia/+/206707\nReviewed-by: Jim Van Verth \u003cjvanverth@google.com\u003e\nCommit-Queue: Brian Salomon \u003cbsalomon@google.com\u003e\n"
    },
    {
      "commit": "c8098329300b4c6421c541d16442ba13dad57c33",
      "tree": "5481152d4353d2db4e7e56af1b0627eea4110394",
      "parents": [
        "8c4cafd297fc4cc7bb566896bfa295b567c3c22a"
      ],
      "author": {
        "name": "Jim Van Verth",
        "email": "jvanverth@google.com",
        "time": "Tue Apr 16 10:50:01 2019 -0400"
      },
      "committer": {
        "name": "Skia Commit-Bot",
        "email": "skia-commit-bot@chromium.org",
        "time": "Tue Apr 16 15:29:14 2019 +0000"
      },
      "message": "Convert 565 and 4444 bitmaps to 8888 for Mac Metal.\n\nFixes a large number of GMs.\n\nBug: skia:8243\nChange-Id: I8053ac218d894205d36a3a3cdc8b7a7ea2e27e71\nReviewed-on: https://skia-review.googlesource.com/c/skia/+/208320\nReviewed-by: Brian Osman \u003cbrianosman@google.com\u003e\nCommit-Queue: Jim Van Verth \u003cjvanverth@google.com\u003e\n"
    },
    {
      "commit": "8c4cafd297fc4cc7bb566896bfa295b567c3c22a",
      "tree": "914201a91cca2a76722048fbf3ab8452f123f9e0",
      "parents": [
        "c48bf3a34b765f5beabe8d79656df8f2d6cca800"
      ],
      "author": {
        "name": "Chris Dalton",
        "email": "csmartdalton@google.com",
        "time": "Mon Apr 15 19:14:36 2019 -0600"
      },
      "committer": {
        "name": "Skia Commit-Bot",
        "email": "skia-commit-bot@chromium.org",
        "time": "Tue Apr 16 15:21:33 2019 +0000"
      },
      "message": "Don\u0027t flush pipeline state when querying sample locations in GL\n\nThis was a relic from back when we could use EXT_raster_multisample.\nSince MSAA is now always baked into the framebuffer attachments, this\nis no longer necessary.\n\nBug: skia:\nChange-Id: I47c66fcdbd008e057dab5eae457bb70ec8e6bb9f\nReviewed-on: https://skia-review.googlesource.com/c/skia/+/208460\nReviewed-by: Brian Salomon \u003cbsalomon@google.com\u003e\nCommit-Queue: Chris Dalton \u003ccsmartdalton@google.com\u003e\n"
    },
    {
      "commit": "c48bf3a34b765f5beabe8d79656df8f2d6cca800",
      "tree": "605af0eb6d0b6e5348abd59d15aac3ddebb5d93c",
      "parents": [
        "4aa13e7da372363b626b2d9ddc50891de25fb043"
      ],
      "author": {
        "name": "Mike Klein",
        "email": "mtklein@google.com",
        "time": "Mon Apr 15 15:59:11 2019 -0500"
      },
      "committer": {
        "name": "Mike Klein",
        "email": "mtklein@google.com",
        "time": "Tue Apr 16 14:41:51 2019 +0000"
      },
      "message": "just some refactoring for fm_bot.go\n\nA couple little spots starting to look unwieldy.\n\nChange-Id: If2971b71ae202b152f54ec3df6896d906c34a081\nReviewed-on: https://skia-review.googlesource.com/c/skia/+/208276\nReviewed-by: Brian Osman \u003cbrianosman@google.com\u003e\n"
    },
    {
      "commit": "4aa13e7da372363b626b2d9ddc50891de25fb043",
      "tree": "4906aa46ac087d62548f8d1f157d78180fde0465",
      "parents": [
        "26de56ef099f7cc4b5e3e02481d43a3bd10ccb42"
      ],
      "author": {
        "name": "Greg Daniel",
        "email": "egdaniel@google.com",
        "time": "Mon Apr 15 14:42:20 2019 -0400"
      },
      "committer": {
        "name": "Skia Commit-Bot",
        "email": "skia-commit-bot@chromium.org",
        "time": "Tue Apr 16 14:08:09 2019 +0000"
      },
      "message": "Remove deprecated SkSurface::prepareForExternalIO and related uses of the name.\n\nIn a future CL I intend to use prepareForExternalIO to be a specific operation\nto do to an SkSurface/SkImage so this is a pre CL that removes our use of that\nphrase when related to flushing.\n\nBug: skia:8802\nChange-Id: I3efe4c4061fe962e2fc1339185feb5ded6e23005\nReviewed-on: https://skia-review.googlesource.com/c/skia/+/208223\nCommit-Queue: Brian Salomon \u003cbsalomon@google.com\u003e\nCommit-Queue: Greg Daniel \u003cegdaniel@google.com\u003e\nReviewed-by: Brian Salomon \u003cbsalomon@google.com\u003e\n"
    },
    {
      "commit": "26de56ef099f7cc4b5e3e02481d43a3bd10ccb42",
      "tree": "80b2d601c94bf4083855fbf0cd4dbbeabd5b54ac",
      "parents": [
        "c0d79e525fbc7a57da366afc6f662c7d288a0774"
      ],
      "author": {
        "name": "Brian Salomon",
        "email": "bsalomon@google.com",
        "time": "Wed Apr 10 12:14:26 2019 -0400"
      },
      "committer": {
        "name": "Skia Commit-Bot",
        "email": "skia-commit-bot@chromium.org",
        "time": "Tue Apr 16 14:05:34 2019 +0000"
      },
      "message": "GrSurface -\u003e GrGpuBuffer transfer improvements:\n\nGrCaps now only provides the offset alignment requirement. The row bytes\nis always bpp * width.\n\nGrGpu::transferPixelsFrom now just returns bool since row bytes value is\nimplicit. It now asserts offset is aligned with GrCap\u0027s provided value\nin base class.\n\nImplement caps for GL.\n\nBug: skia:8962\nChange-Id: I3299b62efe9fe05bfe02f2a6a4c2704f647d0f8a\nReviewed-on: https://skia-review.googlesource.com/c/skia/+/206686\nReviewed-by: Greg Daniel \u003cegdaniel@google.com\u003e\nCommit-Queue: Brian Salomon \u003cbsalomon@google.com\u003e\n"
    },
    {
      "commit": "c0d79e525fbc7a57da366afc6f662c7d288a0774",
      "tree": "4a6b03062a1eaf5b28498d34ea930c88f15089df",
      "parents": [
        "6c431d52020c6ed8d48681c9443ec3131fd0ed34"
      ],
      "author": {
        "name": "Brian Salomon",
        "email": "bsalomon@google.com",
        "time": "Wed Apr 10 15:02:11 2019 -0400"
      },
      "committer": {
        "name": "Skia Commit-Bot",
        "email": "skia-commit-bot@chromium.org",
        "time": "Tue Apr 16 14:03:04 2019 +0000"
      },
      "message": "Avoid multiplication by alpha in fragment shader when known to be 1.\n\nImplemented for image shaders, image draws, and gradient shaders.\n\nReimplement GrFragmentProcessor::OverrideInput as GrOverrideInputFragmentProcessor.fp.\nIt allows specification of whether the replacement input color should be\na literal in the shader code or a uniform. For above use case use with literal white.\n\nMake key in variables in fp files work for 4f colors.\n\nFix issue in CPP code gen from .fp where when + key vars that pushed multiple values\ninto the shader key only skipped the first key value when the when condition is not\ntrue.\n\nBug: skia:7722\n\nChange-Id: Id7c865132d620e8cdea8b00f2a627103eef171ac\nReviewed-on: https://skia-review.googlesource.com/c/skia/+/201985\nCommit-Queue: Brian Salomon \u003cbsalomon@google.com\u003e\nReviewed-by: Brian Osman \u003cbrianosman@google.com\u003e\nReviewed-by: Ethan Nicholas \u003cethannicholas@google.com\u003e\n"
    }
  ]
}