kernel.git/drivers/media/pci/intel, branch linux-6.16.y

media: ivsc: Fix crash at shutdown due to missing mei_cldev_disable() calls

2025-08-28T14:34:27Z

commit 0c92c49fc688cfadacc47ae99b06a31237702e9e upstream. Both the ACE and CSI driver are missing a mei_cldev_disable() call in their remove() function. This causes the mei_cl client to stay part of the mei_device->file_list list even though its memory is freed by mei_cl_bus_dev_release() calling kfree(cldev->cl). This leads to a use-after-free when mei_vsc_remove() runs mei_stop() which first removes all mei bus devices calling mei_ace_remove() and mei_csi_remove() followed by mei_cl_bus_dev_release() and then calls mei_cl_all_disconnect() which walks over mei_device->file_list dereferecing the just freed cldev->cl. And mei_vsc_remove() it self is run at shutdown because of the platform_device_unregister(tp->pdev) in vsc_tp_shutdown() When building a kernel with KASAN this leads to the following KASAN report: [ 106.634504] ================================================================== [ 106.634623] BUG: KASAN: slab-use-after-free in mei_cl_set_disconnected (drivers/misc/mei/client.c:783) mei [ 106.634683] Read of size 4 at addr ffff88819cb62018 by task systemd-shutdow/1 [ 106.634729] [ 106.634767] Tainted: [E]=UNSIGNED_MODULE [ 106.634770] Hardware name: Dell Inc. XPS 16 9640/09CK4V, BIOS 1.12.0 02/10/2025 [ 106.634773] Call Trace: [ 106.634777] ... [ 106.634871] kasan_report (mm/kasan/report.c:221 mm/kasan/report.c:636) [ 106.634901] mei_cl_set_disconnected (drivers/misc/mei/client.c:783) mei [ 106.634921] mei_cl_all_disconnect (drivers/misc/mei/client.c:2165 (discriminator 4)) mei [ 106.634941] mei_reset (drivers/misc/mei/init.c:163) mei ... [ 106.635042] mei_stop (drivers/misc/mei/init.c:348) mei [ 106.635062] mei_vsc_remove (drivers/misc/mei/mei_dev.h:784 drivers/misc/mei/platform-vsc.c:393) mei_vsc [ 106.635066] platform_remove (drivers/base/platform.c:1424) Add the missing mei_cldev_disable() calls so that the mei_cl gets removed from mei_device->file_list before it is freed to fix this. Fixes: 78876f71b3e9 ("media: pci: intel: ivsc: Add ACE submodule") Fixes: 29006e196a56 ("media: pci: intel: ivsc: Add CSI submodule") Cc: stable@vger.kernel.org Signed-off-by: Hans de Goede Signed-off-by: Sakari Ailus Signed-off-by: Hans Verkuil Signed-off-by: Greg Kroah-Hartman

media: ipu6: isys: Use correct pads for xlate_streams()

2025-08-28T14:34:27Z

commit ff49672a28f3a856717f09d61380e524e243121f upstream. The pad argument to v4l2_subdev_state_xlate_streams() is incorrect, static pad number is used for the source pad even though the pad number is dependent on the stream. Fix it. Fixes: 3a5c59ad926b ("media: ipu6: Rework CSI-2 sub-device streaming control") Cc: stable@vger.kernel.org Signed-off-by: Sakari Ailus Signed-off-by: Hans Verkuil Signed-off-by: Greg Kroah-Hartman

media: ipu-bridge: Add _HID for OV5670

2025-08-20T16:41:24Z

[ Upstream commit 484f8bec3ddb453321ef0b8621c25de6ce3d0302 ] The OV5670 is found on Dell 7212 tablets paired with an IPU3 ISP and needs to be connected by the ipu-bridge. Add it to the list of supported devices. Signed-off-by: Daniel Scally Signed-off-by: Sakari Ailus Signed-off-by: Hans Verkuil Signed-off-by: Sasha Levin

media: intel/ipu6: Change deprecated lock comment

2025-05-06T11:06:31Z

pre_streamon_queued is no longer used. The lock now is protecting active and incoming lists. Signed-off-by: Stanislaw Gruszka Signed-off-by: Sakari Ailus Signed-off-by: Hans Verkuil

media: intel/ipu6: Remove line_align

2025-05-06T11:06:31Z

isys->line_align value is only used in one place and we can just use the proper value directly there. Signed-off-by: Stanislaw Gruszka Signed-off-by: Sakari Ailus Signed-off-by: Hans Verkuil

media: intel/ipu6: Remove pin_ready function pointer

2025-05-06T11:06:31Z

We can call ipu6_isys_queue_buf_ready() directly. The only current usage is pin_ready pointer is to check if pin was prepared before IPU6_FW_ISYS_RESP_TYPE_PIN_DATA_READY interrupt, we can use queue pointer for that purpose. Signed-off-by: Stanislaw Gruszka Signed-off-by: Sakari Ailus Signed-off-by: Hans Verkuil

media: intel/ipu6: Remove unused ipu6_isys_csi2_pdata

2025-05-06T11:06:29Z

The pointer to ipu6_isys_csi2_pdata is not used. Signed-off-by: Stanislaw Gruszka Signed-off-by: Sakari Ailus Signed-off-by: Hans Verkuil

media: intel/ipu6: Fix dma mask for non-secure mode

2025-04-25T08:15:35Z

We use dma_get_mask() of auxdev device for calculate iova pfn limit. This is always 32 bit mask as we do not initialize the mask (and we can not do so, since dev->dev_mask is NULL anyways for auxdev). Since we need 31 bit mask for non-secure mode use mmu_info->aperture_end which is properly initialized to correct mask for both modes. Fixes: daabc5c64703 ("media: ipu6: not override the dma_ops of device in driver") Cc: stable@vger.kernel.org Signed-off-by: Stanislaw Gruszka Signed-off-by: Sakari Ailus Signed-off-by: Hans Verkuil

media: intel/ipu6: Minor dma_mask clenaup

2025-04-25T08:15:35Z

Remove unused dma_mask field and ipu-dma.h includes. Signed-off-by: Stanislaw Gruszka Signed-off-by: Sakari Ailus Signed-off-by: Hans Verkuil

media: intel/ipu6: Replace deprecated PCI functions

2025-04-25T08:15:29Z

pcim_iomap_table() and pcim_iomap_regions() have been deprecated. Furthermore, the "name" parameter in pcim_iomap_regions() and its successor, pcim_iomap_region(), should always reflect the driver name, whereas currently it is the device's name. Replace the deprecated functions with pcim_iomap_region() and pass the actual driver name. Signed-off-by: Philipp Stanner Signed-off-by: Sakari Ailus Signed-off-by: Hans Verkuil