Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1717561

Possible memory leak in cafe_ccic.ko

From Anton Volkov <avolkov@ispras.ru>
Newsgroups linux.kernel
Subject Possible memory leak in cafe_ccic.ko
Date 2017-08-22 18:30 +0200
Message-ID <uhkdZ-4RO-31@gated-at.bofh.it> (permalink)
Organization linux.* mail to news gateway

Show all headers | View raw


Hello.

While searching for races in the Linux kernel I've come across 
"drivers/media/platform/marvell-ccic/cafe_ccic.ko" module. Here are 
questions that I came up with while analyzing results. Lines are given 
using the info from Linux v4.12.

Consider the following case:

Thread 1:                  Thread 2:
                            mcam_v4l_release
                            ->mcam_free_dma_bufs
                                cam->dma_bufs[i] = NULL
                                cam->nbufs = 0
cafe_pci_resume                (mcam-core.c: line 413)
->mccic_resume
   ->mcam_read_setup
     ->mcam_alloc_dma_bufs
       cam->dma_bufs[i] =
          dma_alloc_coherent()
          (mcam-core.c: line 381)

It looks like mcam_v4l_release() doesn't really shut the device down. In 
this case cafe_pci_resume() leaks memory for cam->dma_bufs[i] after 
mcam_v4l_release() freed and poisoned them. Is this feasible from your 
point of view?

Thank you for your time.

-- Anton Volkov
Linux Verification Center, ISPRAS
web: http://linuxtesting.org
e-mail: avolkov@ispras.ru

Back to linux.kernel | Previous | Next | Find similar | Unroll thread


Thread

Possible memory leak in cafe_ccic.ko Anton Volkov <avolkov@ispras.ru> - 2017-08-22 18:30 +0200

csiph-web