)]}'
{
  "commit": "e729aa16b168fb202d1a20f936028cb7c2a0278d",
  "tree": "d0267225ebfadbaa4f55f55c3025597d86ff1c5f",
  "parents": [
    "428e6ce023c5890cfecc8ad10335da3f28dbf893"
  ],
  "author": {
    "name": "Ravikiran G Thirumalai",
    "email": "kiran@scalex86.org",
    "time": "Tue May 08 00:29:13 2007 -0700"
  },
  "committer": {
    "name": "Linus Torvalds",
    "email": "torvalds@woody.linux-foundation.org",
    "time": "Tue May 08 11:15:09 2007 -0700"
  },
  "message": "Pad irq_desc to internode cacheline size\n\nWe noticed a drop in n/w performance due to the irq_desc being cacheline\naligned rather than internode aligned.  We see 50% of expected performance\nwhen two e1000 nics local to two different nodes have consecutive irq\ndescriptors allocated, due to false sharing.\n\nNote that this patch does away with cacheline padding for the UP case, as\nit does not seem useful for UP configurations.\n\nSigned-off-by: Ravikiran Thirumalai \u003ckiran@scalex86.org\u003e\nSigned-off-by: Shai Fultheim \u003cshai@scalex86.org\u003e\nCc: \"Siddha, Suresh B\" \u003csuresh.b.siddha@intel.com\u003e\nCc: Ingo Molnar \u003cmingo@elte.hu\u003e\nCc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n",
  "tree_diff": [
    {
      "type": "modify",
      "old_id": "a6899402b5220233c53fe926655d788b2dce7f1c",
      "old_mode": 33188,
      "old_path": "include/linux/irq.h",
      "new_id": "1695054e8c63909a62b8e12d16fe5ca1c8cc625c",
      "new_mode": 33188,
      "new_path": "include/linux/irq.h"
    },
    {
      "type": "modify",
      "old_id": "aff1f0fabb0df0b462c0d83ea4eb35d0ecc24fe0",
      "old_mode": 33188,
      "old_path": "kernel/irq/handle.c",
      "new_id": "515ad40bde15a874ea7471ceef32f8978a317cfd",
      "new_mode": 33188,
      "new_path": "kernel/irq/handle.c"
    }
  ]
}