lkml.org 
[lkml]   [2021]   [Jan]   [29]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
SubjectRe: [PATCH v9 3/4] kasan: Add report for async mode
From
Date


On 1/29/21 6:09 PM, Andrey Konovalov wrote:
> On Fri, Jan 29, 2021 at 6:56 PM Vincenzo Frascino
> <vincenzo.frascino@arm.com> wrote:
>>
>> Hi Andrey,
>>
>> On 1/29/21 5:40 PM, Andrey Konovalov wrote:
>>> I suggest to call end_report(&flags, 0) here and check addr !=0 in
>>> end_report() before calling trace_error_report_end().
>>>
>>
>> Probably this is better as:
>>
>> if (!IS_ENABLED(CONFIG_KASAN_HW_TAGS))
>>
>> Because that condition passes always addr == 0.
>
> Not sure I understand. Call report_end(&flags, 0) and then there do:
>
> if (addr) trace_error_report_end(...);
>
> Although maybe it makes sense to still trace all async bugs to address
> 0. Or to some magic address.
>
> Alex, WDYT?
>

What I meant is instead of:

if (addr) trace_error_report_end(...);

you might want to do:

if (!IS_ENABLED(CONFIG_KASAN_HW_TAGS)) trace_error_report_end(...);

because, could make sense to trace 0 in other cases?

I could not find the implementation of trace_error_report_end() hence I am not
really sure on what it does.

--
Regards,
Vincenzo

\
 
 \ /
  Last update: 2021-01-29 19:14    [W:0.069 / U:0.232 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site