[lkml]   [2018]   [Sep]   [3]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: [PATCH] isdn: mISDN: tei: Fix a sleep-in-atomic-context bug in create_teimgr()

On 2018/9/3 0:31, wrote:
> Hi,
> I do not understand the analysis and do not see that the spinlock is a
> problem here.
> I think your DSAC analyzer assumes that the FUNC_PTR mgr_ctrl call calls
> the mgr_ctrl in tei.c, but in real it calls l2->ch.ctrl() which is the
> function in layer2.c, not tei.c. And the function in layer2.c should not
> do any GFP_KERNEL allocation.
> Same for your 2. reported issue.

Okay, thanks for your reply.
My analysis handles the function pointer using the function type and
structure field, but it cannot distinguish the different variables of
the same type and field now.
I will try to improve my tool, thanks for your explanation.

Best wishes,
Jia-Ju Bai

 \ /
  Last update: 2018-09-03 03:41    [W:0.039 / U:1.676 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site