On 2/24/22 11:53 PM, Péter Ujfalusi wrote:
On 24/02/2022 16:51, Ammar Faizi wrote:From: Ammar Faizi <ammarfaizi2@xxxxxxxxxxx> Do not call snd_dma_free_pages() when snd_dma_alloc_pages() returns -ENOMEM because it leads to a NULL pointer dereference bug.Reviewed-by: Peter Ujfalusi <peter.ujfalusi@xxxxxxxxxxxxxxx>
Thanks for the review, I will append this tag in the v2.
The dmesg says: <6>[109482.497835][T138537] usb 1-2: Manufacturer: SIGMACHIP <6>[109482.502506][T138537] input: SIGMACHIP USB Keyboard as /devices/pci0000:00/0000:00:14.0/usb1/1-2/1-2:1.0/0003:1C4F:0002.000D/input/input34 <6>[109482.558976][T138537] hid-generic 0003:1C4F:0002.000D: input,hidraw1: USB HID v1.10 Keyboard [SIGMACHIP USB Keyboard] on usb-0000:00:14.0-2/input0 <6>[109482.561653][T138537] input: SIGMACHIP USB Keyboard Consumer Control as /devices/pci0000:00/0000:00:14.0/usb1/1-2/1-2:1.1/0003:1C4F:0002.000E/input/input35 <6>[109482.615490][T138537] input: SIGMACHIP USB Keyboard System Control as /devices/pci0000:00/0000:00:14.0/usb1/1-2/1-2:1.1/0003:1C4F:0002.000E/input/input36 <6>[109482.615643][T138537] hid-generic 0003:1C4F:0002.000E: input,hidraw2: USB HID v1.10 Device [SIGMACHIP USB Keyboard] on usb-0000:00:14.0-2/input1 <4>[110102.335460][T140985] rtw_8822ce 0000:01:00.0: timed out to flush queue 1 <3>[118575.730928][ T1387] sof-audio-pci-intel-tgl 0000:00:1f.3: error: memory alloc failed: -12Wow, I wonder what is going on that allocation fails...
Can't really tell the reason behind the -ENOMEM. It's hard to reproduce, randomly happens. I have hit this bug at least 4 times on my laptop with 16 GB of RAM. All happened randomly. I tried to torture the memory with mmap() + mlock() + memset() + sleep(), then playing some audio while almost running out of memory, but still can't reproduce the same -ENOMEM. -- Ammar Faizi
Attachment:
OpenPGP_0x364FBA34FF170A4B.asc
Description: OpenPGP public key
Attachment:
OpenPGP_signature
Description: OpenPGP digital signature