CVE-2026-99184

In the Linux kernel, the following vulnerability has been resolved:

accel/rocket: search every core slot when a core is removed

rocket remove() decrements rdev->num cores for each core it removes, while find core for dev() searches slots 0 to num cores - 1. Unbinding the cores in the order they were bound therefore loses the last one: by the time it is removed the search range has already shrunk past its slot, so find core for dev() returns -1 and rocket remove() gives up without doing anything.

num cores never reaches zero, rocket device fini() never runs, and the file-scoped rdev keeps pointing at a device that is going away. Binding the cores again starts from that stale count, because rocket probe() takes rdev->num cores as the slot to fill. On an RK3588, which describes three cores, the second round lands on slots 1, 2 and 3 while rdev->cores was allocated with room for three:

rocket fdab0000.npu: Rockchip NPU core 1 version: 1179210309 rocket fdac0000.npu: Rockchip NPU core 2 version: 1179210309 rocket fdad0000.npu: Rockchip NPU core 3 version: 1179210309

The write to rdev->cores[3] is past the end of the array.

Nothing in tree reads the core array often enough to notice, so the overrun is silent today. It turned up while testing a devfreq series on top of this, where a worker walks every core a few times a second, and UBSAN caught the first bool it read out of the overrun entry:

UBSAN: invalid-load in drivers/accel/rocket/rocket devfreq.c:47:10 load of value 5 is not a valid value for type ' Bool' Workqueue: devfreq wq devfreq monitor

Record how many slots were allocated and search all of them. Every core is then found on removal, num cores reaches zero, the device is torn down and a later bind starts from a clean rdev.

This does not make unbinding a single core out of several work. probe still takes num cores as the slot to fill, so rebinding one core while its siblings stay bound would write over a slot that is already in use, and rocket open() still reaches for cores[0] whether or not anything is there. Both of those want more thought than a fix should carry.

Found by unbinding and rebinding all three cores on an Orange Pi 5 Plus. With this applied, 25 unbind/rebind rounds and 5 module unload/reload rounds run clean there: the cores land in slots 0, 1 and 2 every time, whichever order they are bound in, and the shared supply goes back to a single user after each round.