> On Sep 17, 2019, at 11:21 AM, Damjan Marion via Lists.Fd.Io
> wrote:
>
>
>
>> On 17 Sep 2019, at 17:12, Christian Hopps wrote:
>>
>>
>>
>>> On Sep 17, 2019, at 3:56 AM, Benoit Ganne (bganne) via Lists.Fd.Io
>>> wrote:
>>>
> AFAIK the default for x86_64 is to omit frame pointer
>> AFAIK the default for x86_64 is to omit frame pointer
>> for optimized builds so you probably do not have it in most cases in VPP
>> builds (but I did not check).
>> So when perf tries to interpret random data as frame pointer, this is
>> what you get :)
>> The most reliable (but slowest)
> On Sep 16, 2019, at 10:31 AM, Benoit Ganne (bganne) via Lists.Fd.Io
> wrote:
>
>> Could these odd symbols I'm seeing be the result of link-time
>> optimization?
>
> We do not do LTO. AFAIK the default for x86_64 is to omit frame pointer for
> optimized builds so you probably do not have
> Could these odd symbols I'm seeing be the result of link-time
> optimization?
We do not do LTO. AFAIK the default for x86_64 is to omit frame pointer for
optimized builds so you probably do not have it in most cases in VPP builds
(but I did not check).
So when perf tries to interpret random
> On Sep 16, 2019, at 8:49 AM, Damjan Marion wrote:
>
>
>
>> On 13 Sep 2019, at 18:58, Christian Hopps wrote:
>>
>>
>>
>>> On Sep 13, 2019, at 12:41 PM, Christian Hopps wrote:
>>>
>>> Signed PGP part
>>>
>>>
On Sep 13, 2019, at 12:23 PM, Damjan Marion wrote:
> On Sep 16, 2019, at 7:22 AM, Benoit Ganne (bganne) wrote:
>
>> I tried "perf record --call-graph fp" but the results seemed odd (e.g., I
>> see a tcp function symbol deep within a call-branch rooted in the dpdk
>> crypto device enqueue routine for the marvell mvsam cryptodev).
>
> Just in
> On 13 Sep 2019, at 18:58, Christian Hopps wrote:
>
>
>
>> On Sep 13, 2019, at 12:41 PM, Christian Hopps wrote:
>>
>> Signed PGP part
>>
>>
>>> On Sep 13, 2019, at 12:23 PM, Damjan Marion wrote:
>>>
>>>
>>>
On 13 Sep 2019, at 18:15, Christian Hopps wrote:
Is it
> I tried "perf record --call-graph fp" but the results seemed odd (e.g., I
> see a tcp function symbol deep within a call-branch rooted in the dpdk
> crypto device enqueue routine for the marvell mvsam cryptodev).
Just in case, did you try "--call-graph lbr" (on Intel CPU older than Haswell I
> On Sep 13, 2019, at 12:41 PM, Christian Hopps wrote:
>
> Signed PGP part
>
>
>> On Sep 13, 2019, at 12:23 PM, Damjan Marion wrote:
>>
>>
>>
>>> On 13 Sep 2019, at 18:15, Christian Hopps wrote:
>>>
>>> Is it possible to profile vpp with gprof?
>>
>> What exactly do you want to
> On Sep 13, 2019, at 12:23 PM, Damjan Marion wrote:
>
>
>
>> On 13 Sep 2019, at 18:15, Christian Hopps wrote:
>>
>> Is it possible to profile vpp with gprof?
>>
>> I google a bit but I'm not turning anything up. I have made some build
>> changes to compile/link with -pg (starting with
> On 13 Sep 2019, at 18:15, Christian Hopps wrote:
>
> Is it possible to profile vpp with gprof?
>
> I google a bit but I'm not turning anything up. I have made some build
> changes to compile/link with -pg (starting with the gcov target), and a
> change to not handle SIGPROF. This produced
Is it possible to profile vpp with gprof?
I google a bit but I'm not turning anything up. I have made some build changes
to compile/link with -pg (starting with the gcov target), and a change to not
handle SIGPROF. This produced a binary that created a gmon.out file. That file
only has some
12 matches
Mail list logo