kernel/linux.git/drivers/accel, branch v6.12.81

accel/qaic: Handle DBC deactivation if the owner went away

2026-04-11T12:24:38+00:00

[ Upstream commit 2feec5ae5df785658924ab6bd91280dc3926507c ] When a DBC is released, the device sends a QAIC_TRANS_DEACTIVATE_FROM_DEV transaction to the host over the QAIC_CONTROL MHI channel. QAIC handles this by calling decode_deactivate() to release the resources allocated for that DBC. Since that handling is done in the qaic_manage_ioctl() context, if the user goes away before receiving and handling the deactivation, the host will be out-of-sync with the DBCs available for use, and the DBC resources will not be freed unless the device is removed. If another user loads and requests to activate a network, then the device assigns the same DBC to that network, QAIC will "indefinitely" wait for dbc->in_use = false, leading the user process to hang. As a solution to this, handle QAIC_TRANS_DEACTIVATE_FROM_DEV transactions that are received after the user has gone away. Fixes: 129776ac2e38 ("accel/qaic: Add control path") Signed-off-by: Youssef Samir Reviewed-by: Lizhi Hou Reviewed-by: Jeff Hugo Signed-off-by: Jeff Hugo Link: https://patch.msgid.link/20260205123415.3870898-1-youssef.abdulrahman@oss.qualcomm.com Signed-off-by: Sasha Levin

accel/ivpu: Fix race condition when unbinding BOs

2026-01-30T09:28:48+00:00

[ Upstream commit 00812636df370bedf4e44a0c81b86ea96bca8628 ] Fix 'Memory manager not clean during takedown' warning that occurs when ivpu_gem_bo_free() removes the BO from the BOs list before it gets unmapped. Then file_priv_unbind() triggers a warning in drm_mm_takedown() during context teardown. Protect the unmapping sequence with bo_list_lock to ensure the BO is always fully unmapped when removed from the list. This ensures the BO is either fully unmapped at context teardown time or present on the list and unmapped by file_priv_unbind(). Fixes: 48aea7f2a2ef ("accel/ivpu: Fix locking in ivpu_bo_remove_all_bos_from_context()") Signed-off-by: Tomasz Rusinowicz Reviewed-by: Jeff Hugo Signed-off-by: Karol Wachowski Link: https://patch.msgid.link/20251029071451.184243-1-karol.wachowski@linux.intel.com [ The context change is due to the commit e0c0891cd63b ("accel/ivpu: Rework bind/unbind of imported buffers") and the proper adoption is done. ] Signed-off-by: Rahul Sharma Signed-off-by: Greg Kroah-Hartman

accel/ivpu: Fix DCT active percent format

2025-12-18T12:54:40+00:00

[ Upstream commit aa1c2b073ad23847dd2e7bdc7d30009f34ed7f59 ] The pcode MAILBOX STATUS register PARAM2 field expects DCT active percent in U1.7 value format. Convert percentage value to this format before writing to the register. Fixes: a19bffb10c46 ("accel/ivpu: Implement DCT handling") Reviewed-by: Lizhi Hou Signed-off-by: Karol Wachowski Link: https://lore.kernel.org/r/20251001104322.1249896-1-karol.wachowski@linux.intel.com Signed-off-by: Sasha Levin

accel/ivpu: Make function parameter names consistent

2025-12-18T12:54:40+00:00

[ Upstream commit cf87f93847dea607e8a35983cb006ef8493f8065 ] Make ivpu_hw_btrs_dct_set_status() and ivpu_fw_boot_params_setup() declaration and definition parameter names consistent. Reviewed-by: Lizhi Hou Signed-off-by: Jacek Lawrynowicz Link: https://lore.kernel.org/r/20250808111014.328607-1-jacek.lawrynowicz@linux.intel.com Stable-dep-of: aa1c2b073ad2 ("accel/ivpu: Fix DCT active percent format") Signed-off-by: Sasha Levin

accel/ivpu: Ensure rpm_runtime_put in case of engine reset/resume fail

2025-12-18T12:54:40+00:00

[ Upstream commit 9f6c63285737b141ca25a619add80a96111b8b96 ] Previously, aborting work could return early after engine reset or resume failure, skipping the necessary runtime_put cleanup leaving the device with incorrect reference count breaking runtime power management state. Replace early returns with goto statements to ensure runtime_put is always executed. Fixes: a47e36dc5d90 ("accel/ivpu: Trigger device recovery on engine reset/resume failure") Reviewed-by: Lizhi Hou Signed-off-by: Karol Wachowski Link: https://lore.kernel.org/r/20250916084809.850073-1-karol.wachowski@linux.intel.com Signed-off-by: Sasha Levin

accel/ivpu: Prevent runtime suspend during context abort work

2025-12-18T12:54:40+00:00

[ Upstream commit 7806bad76ac397a767f0c369534133c71c73b157 ] Increment the runtime PM counter when entering ivpu_context_abort_work_fn() to prevent the device from suspending while the function is executing. Reviewed-by: Jacek Lawrynowicz Signed-off-by: Andrzej Kacprowski Signed-off-by: Jacek Lawrynowicz Link: https://patchwork.freedesktop.org/patch/msgid/20250204084622.2422544-3-jacek.lawrynowicz@linux.intel.com Stable-dep-of: 9f6c63285737 ("accel/ivpu: Ensure rpm_runtime_put in case of engine reset/resume fail") Signed-off-by: Sasha Levin

accel/habanalabs: support mapping cb with vmalloc-backed coherent memory

2025-11-13T20:34:30+00:00

[ Upstream commit 513024d5a0e34fd34247043f1876b6138ca52847 ] When IOMMU is enabled, dma_alloc_coherent() with GFP_USER may return addresses from the vmalloc range. If such an address is mapped without VM_MIXEDMAP, vm_insert_page() will trigger a BUG_ON due to the VM_PFNMAP restriction. Fix this by checking for vmalloc addresses and setting VM_MIXEDMAP in the VMA before mapping. This ensures safe mapping and avoids kernel crashes. The memory is still driver-allocated and cannot be accessed directly by userspace. Signed-off-by: Moti Haimovski Reviewed-by: Koby Elbaz Signed-off-by: Koby Elbaz Signed-off-by: Sasha Levin

accel/habanalabs/gaudi2: read preboot status after recovering from dirty state

2025-11-13T20:34:30+00:00

[ Upstream commit a0d866bab184161ba155b352650083bf6695e50e ] Dirty state can occur when the host VM undergoes a reset while the device does not. In such a case, the driver must reset the device before it can be used again. As part of this reset, the device capabilities are zeroed. Therefore, the driver must read the Preboot status again to learn the Preboot state, capabilities, and security configuration. Signed-off-by: Konstantin Sinyuk Reviewed-by: Koby Elbaz Signed-off-by: Koby Elbaz Signed-off-by: Sasha Levin

accel/habanalabs: return ENOMEM if less than requested pages were pinned

2025-11-13T20:34:30+00:00

[ Upstream commit 9f5067531c9b79318c4e48a933cb2694f53f3de2 ] EFAULT is currently returned if less than requested user pages are pinned. This value means a "bad address" which might be confusing to the user, as the address of the given user memory is not necessarily "bad". Modify the return value to ENOMEM, as "out of memory" is more suitable in this case. Signed-off-by: Tomer Tayar Reviewed-by: Koby Elbaz Signed-off-by: Koby Elbaz Signed-off-by: Sasha Levin

accel/habanalabs/gaudi2: fix BMON disable configuration

2025-11-13T20:34:29+00:00

[ Upstream commit b4fd8e56c9a3b614370fde2d45aec1032eb67ddd ] Change the BMON_CR register value back to its original state before enabling, so that BMON does not continue to collect information after being disabled. Signed-off-by: Vered Yavniely Reviewed-by: Koby Elbaz Signed-off-by: Koby Elbaz Signed-off-by: Sasha Levin