On Thu, Aug 31, 2017 at 12:46 PM, Kostya Serebryany <[email protected]> wrote:

>
>
> On Thu, Aug 31, 2017 at 12:41 PM, Ben Pfaff <[email protected]> wrote:
>
>> What's the issue with file I/O?  If you can help me understand why you
>> want to get rid of it, maybe I can suggest what to do instead.
>>
>
> The file IO takes more time than direct access to memory.
> Even if the actual IO happens on tmpfs, this is still slower, because the
> accesses to the data require syscalls.
> using fmemopen instead of fopen in the fuzz target
> (LLVMFuzzerTestOneInput()) should reduce the overhead.
>

I take my words about fmemopen back -- it's nearly as slow as the real file
IO. (


>
> --kcc
>
>
>>
>> On Thu, Aug 31, 2017 at 09:24:29PM +0200, Bhargava Shastry wrote:
>> > An update from my side. I have written a small test case for catching
>> > CVE-2016-2074 here [1]. KCC strongly encourages me to get rid of file
>> > I/O based APIs such as ovs_pcap_read() and so on. So, my question to
>> > dev@OVS is: Any suggestions how I can do this? Right now, the test runs
>> > but is relatively slow. I haven't really benchmarked it so I can't
>> > provide hard numbers.
>> >
>> > [1]:
>> > https://github.com/bshastry/fuzzer-test-suite/blob/master/op
>> envswitch-2.3.2/target.c
>>
>
>
_______________________________________________
dev mailing list
[email protected]
https://mail.openvswitch.org/mailman/listinfo/ovs-dev

Reply via email to