[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <9da1c8f9-7ca5-e10b-8931-6871fdbffb23@intel.com>
Date: Mon, 13 Jan 2020 13:31:38 +0100
From: Björn Töpel <bjorn.topel@...el.com>
To: Jiri Olsa <jolsa@...hat.com>
Cc: Alexei Starovoitov <alexei.starovoitov@...il.com>,
Jiri Olsa <jolsa@...nel.org>,
Alexei Starovoitov <ast@...nel.org>,
Daniel Borkmann <daniel@...earbox.net>, netdev@...r.kernel.org,
bpf@...r.kernel.org, Andrii Nakryiko <andriin@...com>,
Yonghong Song <yhs@...com>, Martin KaFai Lau <kafai@...com>,
Jakub Kicinski <jakub.kicinski@...ronome.com>,
David Miller <davem@...hat.com>
Subject: Re: [PATCH 5/5] bpf: Allow to resolve bpf trampoline in unwind
On 2020-01-13 13:21, Björn Töpel wrote:
>
> On 2020-01-13 10:43, Jiri Olsa wrote:
>> hi,
>> attached patch seems to work for me (trampoline usecase), but I don't
>> know
>> how to test it for dispatcher.. also I need to check if we need to
>> decrease
>> BPF_TRAMP_MAX or BPF_DISPATCHER_MAX, it might take more time;-)
>>
>
> Thanks for working on it! I'll take the patch for a spin.
>
> To test the dispatcher, just run XDP!
>
> With your change, the BPF_DISPATCHER_MAX is still valid. 48 entries =>
> 1890B which is < (BPF_IMAGE_SIZE / 2).
>
...and FWIW, it would be nice with bpf_dispatcher_<...> entries in
kallsyms as well. If that code could be shared with the trampoline code
as well (bpf_trampoline_<btf_id>), that'd be great!
Powered by blists - more mailing lists