lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <079A7D73-697C-4CFD-97F3-7CFB741BE4C3@fb.com>
Date:   Thu, 27 Jun 2019 17:55:58 +0000
From:   Song Liu <songliubraving@...com>
To:     Andrii Nakryiko <andrii.nakryiko@...il.com>
CC:     Andrii Nakryiko <andriin@...com>, Alexei Starovoitov <ast@...com>,
        "daniel@...earbox.net" <daniel@...earbox.net>,
        Kernel Team <Kernel-team@...com>,
        "bpf@...r.kernel.org" <bpf@...r.kernel.org>,
        "netdev@...r.kernel.org" <netdev@...r.kernel.org>
Subject: Re: [PATCH bpf-next 1/3] libbpf: capture value in BTF type info for
 BTF-defined map defs



> On Jun 27, 2019, at 10:47 AM, Andrii Nakryiko <andrii.nakryiko@...il.com> wrote:
> 
> On Thu, Jun 27, 2019 at 10:27 AM Song Liu <songliubraving@...com> wrote:
>> 
>> 
>> 
>>> On Jun 26, 2019, at 4:21 PM, Andrii Nakryiko <andriin@...com> wrote:
>>> 
>>> Change BTF-defined map definitions to capture compile-time integer
>>> values as part of BTF type definition, to avoid split of key/value type
>>> information and actual type/size/flags initialization for maps.
>> 
>> If I have an old bpf program and compiled it with new llvm, will it
>> work with new libbpf?
> 
> You mean BPF programs that used previous incarnation of BTF-defined
> maps? No, they won't work. But we never released them, so I think it's
> ok to change them. Nothing should be using that except for selftests,
> which I fixed.

I see. This makes sense. 

> 
>> 
>> 
>>> 
>>> Signed-off-by: Andrii Nakryiko <andriin@...com>
>>> ---
> 
> <snip>
> 
>>> diff --git a/tools/testing/selftests/bpf/bpf_helpers.h b/tools/testing/selftests/bpf/bpf_helpers.h
>>> index 1a5b1accf091..aa5ddf58c088 100644
>>> --- a/tools/testing/selftests/bpf/bpf_helpers.h
>>> +++ b/tools/testing/selftests/bpf/bpf_helpers.h
>>> @@ -8,6 +8,9 @@
>>> */
>>> #define SEC(NAME) __attribute__((section(NAME), used))
>>> 
>>> +#define __int(name, val) int (*name)[val]
>>> +#define __type(name, val) val *name
>>> +
>> 
>> I think we need these two in libbpf.
> 
> Yes, but it's another story for another set of patches. We'll need to
> provide bpf_helpers as part of libbpf for inclusion into BPF programs,
> but there are a bunch of problems right now with existing
> bpf_heplers.h that prevents us from just copying it into libbpf. We'll
> need to resolve those first.
> 
> But then again, there is no use of __int and __type for user-space
> programs, so for now it's ok.

OK. How about we put these two lines in an separate patch?

Thanks,
Song

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ