linux.git/drivers/accel, branch v7.1-rc7

accel/ethosu: fix OOB write in ethosu_gem_cmdstream_copy_and_validate()

2026-06-05T03:07:59+00:00

The command stream parsing loop increments the index variable a second
time when a 64-bit command word is encountered (bit 14 set), but does
not re-check the loop bound before writing the second word:

    for (i = 0; i < size / 4; i++) {
        bocmds[i] = cmds[0];
        if (cmd & 0x4000) {
            i++;
            bocmds[i] = cmds[1];   /* unchecked */
        }
    }

The buffer bocmds is backed by a DMA allocation of exactly size bytes
from drm_gem_dma_create(ddev, size), giving valid indices [0, size/4-1].

When i == size/4 - 1 on entry to an iteration and bit 14 of cmds[0] is
set, bocmds[size/4-1] is written in bounds, i is then incremented to
size/4, and bocmds[size/4] writes four bytes past the end of the
allocation.

Userspace controls both the buffer contents and the size argument via
the ioctl, making this a userspace-triggerable heap out-of-bounds write.

Fix by checking the incremented index against the buffer bound before
the second write and returning -EINVAL if the buffer is too small to
contain the extended command.

Fixes: 5a5e9c0228e6 ("accel: Add Arm Ethos-U NPU driver")
Cc: stable@vger.kernel.org
Signed-off-by: Muhammad Bilal 
Link: https://patch.msgid.link/20260523190843.33977-1-meatuni001@gmail.com
Signed-off-by: Rob Herring (Arm)

accel/ethosu: reject DMA commands with uninitialized length

2026-06-04T22:43:43+00:00

cmd_state_init() initializes the command state with memset(0xff),
leaving dma->len at U64_MAX to signal missing setup. The only setter
is NPU_SET_DMA0_LEN; if userspace omits this command and issues
NPU_OP_DMA_START, dma->len remains U64_MAX.

In dma_length(), a positive stride added to U64_MAX wraps to a small
value. With size0 == 1, check_mul_overflow() does not trigger and
dma_length() returns 0 instead of U64_MAX. The caller's U64_MAX check
then passes, region_size[] stays 0, and the bounds check in
ethosu_job.c is bypassed, allowing hardware to execute DMA with stale
physical addresses.

Fix by checking for U64_MAX at the start of dma_length() before any
arithmetic, consistent with the sentinel value used throughout the
driver to detect uninitialized fields.

Fixes: 5a5e9c0228e6 ("accel: Add Arm Ethos-U NPU driver")
Cc: stable@vger.kernel.org
Signed-off-by: Muhammad Bilal 
Link: https://patch.msgid.link/20260524130319.12747-1-meatuni001@gmail.com
Signed-off-by: Rob Herring (Arm)

accel/ethosu: fix arithmetic issues in dma_length()

2026-06-04T22:43:27+00:00

dma_length() derives DMA region usage from command stream values and
updates region_size[]:

    len = ((len + stride[0]) * size0 + stride[1]) * size1
    region_size[region] = max(..., len + dma->offset)

Several arithmetic issues can corrupt the derived region size:

- signed stride values may underflow when added to len
- intermediate multiplications may overflow
- len + dma->offset may overflow during region_size updates
- dma_length() error returns were not validated by the caller

region_size[] is later used by ethosu_job.c to validate command stream
accesses against GEM buffer sizes. Arithmetic wraparound can therefore
under-report region usage and bypass the bounds validation.

Fix by validating signed additions, using overflow helpers for
multiplications and offset updates, and propagating dma_length()
failures to the caller.

Fixes: 5a5e9c0228e6 ("accel: Add Arm Ethos-U NPU driver")
Cc: stable@vger.kernel.org
Signed-off-by: Muhammad Bilal 
Link: https://patch.msgid.link/20260524103710.47397-1-meatuni001@gmail.com
Signed-off-by: Rob Herring (Arm)

accel/ethosu: fix wrong weight index in NPU_SET_SCALE1_LENGTH on U85

2026-06-04T22:41:46+00:00

On non-U65 hardware (e.g. U85), opcode 0x4093 is NPU_SET_WEIGHT2_LENGTH.
The BASE handler for the same opcode correctly assigns to
st.weight[2].base, but the LENGTH handler mistakenly assigns cmds[1]
to st.weight[1].length instead of st.weight[2].length.

This leaves weight[2].length at its initialised sentinel value of
0xffffffff and corrupts weight[1].length with the user-supplied value,
breaking the software bounds-check state for both weight buffers on U85.

Fix the index to match the BASE handler.

Fixes: 5a5e9c0228e6 ("accel: Add Arm Ethos-U NPU driver")
Cc: stable@vger.kernel.org
Signed-off-by: Muhammad Bilal 
Link: https://patch.msgid.link/20260523210840.92039-3-meatuni001@gmail.com
Signed-off-by: Rob Herring (Arm)

accel/ethosu: reject NPU_OP_RESIZE commands from userspace

2026-06-04T22:41:46+00:00

NPU_OP_RESIZE is a U85-only command that the driver does not yet
implement. The existing WARN_ON(1) placeholder fires unconditionally
whenever userspace submits this command via DRM_IOCTL_ETHOSU_GEM_CREATE,
causing unbounded kernel log spam.

If panic_on_warn is set the kernel panics, giving any unprivileged user
with access to the DRM device a trivial denial-of-service primitive.

Replace the WARN_ON(1) with an explicit -EINVAL return so the ioctl
rejects the command before it reaches hardware.

Fixes: 5a5e9c0228e6 ("accel: Add Arm Ethos-U NPU driver")
Cc: stable@vger.kernel.org
Signed-off-by: Muhammad Bilal 
Link: https://patch.msgid.link/20260523210840.92039-2-meatuni001@gmail.com
Signed-off-by: Rob Herring (Arm)

accel/ethosu: fix IFM region index out-of-bounds in command stream parser

2026-06-04T22:31:41+00:00

NPU_SET_IFM_REGION extracts the region index with param & 0x7f, giving
a maximum value of 127. However region_size[] and output_region[] in
struct ethosu_validated_cmdstream_info are both sized to
NPU_BASEP_REGION_MAX (8), giving valid indices [0..7].

Every other region assignment in the same switch uses param & 0x7:
  NPU_SET_OFM_REGION:  st.ofm.region  = param & 0x7;
  NPU_SET_IFM2_REGION: st.ifm2.region = param & 0x7;
  NPU_SET_WEIGHT_REGION: st.weight[0].region = param & 0x7;
  NPU_SET_SCALE_REGION:  st.scale[0].region  = param & 0x7;

The 0x7f mask on IFM is inconsistent and appears to be a typo.

feat_matrix_length() and calc_sizes() use the region index directly
as an array subscript into the kzalloc'd info struct:
  info->region_size[fm->region] = max(...);

A userspace caller supplying NPU_SET_IFM_REGION with param > 7 causes
a write up to 127*8 = 1016 bytes past the start of region_size[],
corrupting adjacent kernel heap data.

Fix by applying the same & 0x7 mask used by all other region
assignments.

Fixes: 5a5e9c0228e6 ("accel: Add Arm Ethos-U NPU driver")
Cc: stable@vger.kernel.org
Signed-off-by: Muhammad Bilal 
Link: https://patch.msgid.link/20260523195159.55801-1-meatuni001@gmail.com
Signed-off-by: Rob Herring (Arm)

accel/ivpu: Add buffer overflow check in MS get_info_ioctl

2026-06-02T05:47:50+00:00

Add validation that the info size returned from the metric stream info
query is not exceeded when checked against the allocated buffer size.
If the firmware returns a size larger than the buffer, reject the
operation with -EOVERFLOW instead of proceeding with an incorrect
buffer copy.

Fixes: cdfad4db7756 ("accel/ivpu: Add NPU profiling support")
Cc: stable@vger.kernel.org # v6.18+
Signed-off-by: Andrzej Kacprowski 
Reviewed-by: Karol Wachowski 
Signed-off-by: Karol Wachowski 
Link: https://patch.msgid.link/20260529120841.135852-1-andrzej.kacprowski@linux.intel.com

accel/ivpu: Add bounds checks for firmware log indices

2026-06-02T05:47:50+00:00

Add validation that read and write indices in the firmware log buffer
are within valid bounds (< data_size) before using them. If
out-of-bounds indices are encountered (from firmware), clamp them to
safe values instead of proceeding with invalid offsets.

This prevents potential out-of-bounds buffer access when firmware
supplies invalid log indices.

Fixes: 1fc1251149a7 ("accel/ivpu: Refactor functions in ivpu_fw_log.c")
Cc: stable@vger.kernel.org # v6.18+
Signed-off-by: Andrzej Kacprowski 
Reviewed-by: Karol Wachowski 
Signed-off-by: Karol Wachowski 
Link: https://patch.msgid.link/20260529115842.135378-1-andrzej.kacprowski@linux.intel.com

accel/ivpu: Add bounds check for firmware runtime memory

2026-06-02T05:45:18+00:00

Validate that the firmware runtime memory specified in the image
header is properly aligned and sized to hold the firmware image.
This prevents errors during memory allocation and image transfer.

Fixes: 2007e210b6a1 ("accel/ivpu: Split FW runtime and global memory buffers")
Cc: stable@vger.kernel.org # v7.0+
Signed-off-by: Andrzej Kacprowski 
Reviewed-by: Karol Wachowski 
Signed-off-by: Karol Wachowski 
Link: https://patch.msgid.link/20260529120853.135876-1-andrzej.kacprowski@linux.intel.com

accel/ivpu: prevent uninitialized data bug in debugfs

2026-05-26T06:04:07+00:00

The simple_write_to_buffer() will only initialize data starting from
the *pos offset so if it's non-zero then the first part of the buffer
uninitialized.  Really, if *pos is non-zero then this code won't work
so just check for that at the start of the function.

Fixes: 320323d2e545 ("accel/ivpu: Add debugfs interface for setting HWS priority bands")
Signed-off-by: Dan Carpenter 
Reviewed-by: Karol Wachowski 
Signed-off-by: Karol Wachowski 
Link: https://patch.msgid.link/ahP24m6Mii9EDL7Q@stanley.mountain