)]}'
{
  "commit": "050545fc180fc9bdb27d9c0813f9d0043a89e7f7",
  "tree": "75d9c2b4053a96f0daa0dc51511b6c85f39570a8",
  "parents": [
    "524b673421ca7802011d6d50e37b716126f3d81b"
  ],
  "author": {
    "name": "Gabe Black",
    "email": "gabeblack@google.com",
    "time": "Fri Mar 01 13:57:35 2019 -0800"
  },
  "committer": {
    "name": "Gabe Black",
    "email": "gabeblack@google.com",
    "time": "Fri Mar 01 23:41:51 2019 +0000"
  },
  "message": "util, tlm: Fix a memory error in the SCMasterPort class.\n\nIn the b_transport method of the SCMasterPort class, there is a check\nwhich determines whether the packet being sent to gem5 should be\ndeleted once the call to sendAtomic returns. This was deleting the\npacket if extension was *not* nullptr.\n\nThis check should delete the packet if the extension *is* nullptr. The\nreasoning is that the extension will equal nullptr if there was no\ngem5 packet in an extension and a new one needed to be allocated. If\nthere was an extension, ie if extension is not nullptr, then that\u0027s\nwhere the packet came from which therefore doesn\u0027t belong to us. In\nthat case, we need to leave it alone and let its owner clean it up.\n\nWith the check reversed, this method will either leak allocated packets\nit should delete, or delete packets it shouldn\u0027t that someone else will\nlikely try to use later.\n\nChange-Id: I61578d910be6e5085b9fc0ddaa82468b1ac68578\nReviewed-on: https://gem5-review.googlesource.com/c/16949\nReviewed-by: Jason Lowe-Power \u003cjason@lowepower.com\u003e\nMaintainer: Gabe Black \u003cgabeblack@google.com\u003e\n",
  "tree_diff": [
    {
      "type": "modify",
      "old_id": "ab2bc4aba31c11011b39e589fa3ba0b0f7c48cd0",
      "old_mode": 33188,
      "old_path": "util/tlm/src/sc_master_port.cc",
      "new_id": "864e16956f4a636416052ce4e6e123cd697de827",
      "new_mode": 33188,
      "new_path": "util/tlm/src/sc_master_port.cc"
    }
  ]
}