)]}'
{
  "commit": "db5b8cecbdf479ad13156af750377e5b43853fab",
  "tree": "415df00daf14d4f34ea62d9cf0e2c005e4cce838",
  "parents": [
    "01c3ec507a33f30cf0992f86282d90ce9b7715c9"
  ],
  "author": {
    "name": "Francesco Lavra",
    "email": "flavra@baylibre.com",
    "time": "Mon Mar 30 18:19:14 2026 +0200"
  },
  "committer": {
    "name": "Linus Walleij",
    "email": "linusw@kernel.org",
    "time": "Tue Apr 07 11:29:57 2026 +0200"
  },
  "message": "pinctrl: mcp23s08: Disable all pin interrupts during probe\n\nA chip being probed may have the interrupt-on-change feature enabled on\nsome of its pins, for example after a reboot. This can cause the chip to\ngenerate interrupts for pins that don\u0027t have a registered nested handler,\nwhich leads to a kernel crash such as below:\n\n[    7.928897] Unable to handle kernel read from unreadable memory at virtual address 00000000000000ac\n[    7.932314] Mem abort info:\n[    7.935081]   ESR \u003d 0x0000000096000004\n[    7.938808]   EC \u003d 0x25: DABT (current EL), IL \u003d 32 bits\n[    7.944094]   SET \u003d 0, FnV \u003d 0\n[    7.947127]   EA \u003d 0, S1PTW \u003d 0\n[    7.950247]   FSC \u003d 0x04: level 0 translation fault\n[    7.955101] Data abort info:\n[    7.957961]   ISV \u003d 0, ISS \u003d 0x00000004, ISS2 \u003d 0x00000000\n[    7.963421]   CM \u003d 0, WnR \u003d 0, TnD \u003d 0, TagAccess \u003d 0\n[    7.968447]   GCS \u003d 0, Overlay \u003d 0, DirtyBit \u003d 0, Xs \u003d 0\n[    7.973734] user pgtable: 4k pages, 48-bit VAs, pgdp\u003d00000000089b7000\n[    7.980148] [00000000000000ac] pgd\u003d0000000000000000, p4d\u003d0000000000000000\n[    7.986913] Internal error: Oops: 0000000096000004 [#1]  SMP\n[    7.992545] Modules linked in:\n[    8.073678] CPU: 0 UID: 0 PID: 81 Comm: irq/18-4-0025 Not tainted 7.0.0-rc6-gd2b5a1f931c8-dirty #199\n[    8.073689] Hardware name: Khadas VIM3 (DT)\n[    8.073692] pstate: 604000c5 (nZCv daIF +PAN -UAO -TCO -DIT -SSBS BTYPE\u003d--)\n[    8.094639] pc : _raw_spin_lock_irq+0x40/0x80\n[    8.098970] lr : handle_nested_irq+0x2c/0x168\n[    8.098979] sp : ffff800082b2bd20\n[    8.106599] x29: ffff800082b2bd20 x28: ffff800080107920 x27: ffff800080104d88\n[    8.106611] x26: ffff000003298080 x25: 0000000000000001 x24: 000000000000ff00\n[    8.113707] x23: 0000000000000001 x22: 0000000000000000 x21: 000000000000000e\n[    8.120850] x20: 0000000000000000 x19: 00000000000000ac x18: 0000000000000000\n[    8.135046] x17: 0000000000000000 x16: 0000000000000000 x15: 0000000000000000\n[    8.135062] x14: ffff800081567ea8 x13: ffffffffffffffff x12: 0000000000000000\n[    8.135070] x11: 00000000000000c0 x10: 0000000000000b60 x9 : ffff800080109e0c\n[    8.135078] x8 : 1fffe0000069dbc1 x7 : 0000000000000001 x6 : ffff0000034ede00\n[    8.135086] x5 : 0000000000000000 x4 : ffff0000034ede08 x3 : 0000000000000001\n[    8.163460] x2 : 0000000000000000 x1 : 0000000000000001 x0 : 00000000000000ac\n[    8.170560] Call trace:\n[    8.180094]  _raw_spin_lock_irq+0x40/0x80 (P)\n[    8.184443]  mcp23s08_irq+0x248/0x358\n[    8.184462]  irq_thread_fn+0x34/0xb8\n[    8.184470]  irq_thread+0x1a4/0x310\n[    8.195093]  kthread+0x13c/0x150\n[    8.198309]  ret_from_fork+0x10/0x20\n[    8.201850] Code: d65f03c0 d2800002 52800023 f9800011 (885ffc01)\n[    8.207931] ---[ end trace 0000000000000000 ]---\n\nThis issue has always been present, but has been latent until commit\n\"f9f4fda15e72\" (\"pinctrl: mcp23s08: init reg_defaults from HW at probe and\nswitch cache type\"), which correctly removed reg_defaults from the regmap\nand as a side effect changed the behavior of the interrupt handler so that\nthe real value of the MCP_GPINTEN register is now being read from the chip\ninstead of using a bogus 0 default value; a non-zero value for this\nregister can trigger the invocation of a nested handler which may not exist\n(yet).\nFix this issue by disabling all pin interrupts during initialization.\n\nFixes: f9f4fda15e72 (\"pinctrl: mcp23s08: init reg_defaults from HW at probe and switch cache type\")\nSigned-off-by: Francesco Lavra \u003cflavra@baylibre.com\u003e\nSigned-off-by: Linus Walleij \u003clinusw@kernel.org\u003e\n",
  "tree_diff": [
    {
      "type": "modify",
      "old_id": "586f2f67c6177f549d869e06df9ec54f93dde369",
      "old_mode": 33188,
      "old_path": "drivers/pinctrl/pinctrl-mcp23s08.c",
      "new_id": "b89b3169e8be556542049b86ceaa45006511eb8e",
      "new_mode": 33188,
      "new_path": "drivers/pinctrl/pinctrl-mcp23s08.c"
    }
  ]
}