[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20190314210111.lveoorx7wfhq3cxa@ast-mbp>
Date: Thu, 14 Mar 2019 14:01:13 -0700
From: Alexei Starovoitov <alexei.starovoitov@...il.com>
To: Andrii Nakryiko <andrii.nakryiko@...il.com>
Cc: Edward Cree <ecree@...arflare.com>, Yonghong Song <yhs@...com>,
Andrii Nakryiko <andriin@...com>,
Kernel Team <Kernel-team@...com>,
Alexei Starovoitov <ast@...com>,
"acme@...nel.org" <acme@...nel.org>,
"netdev@...r.kernel.org" <netdev@...r.kernel.org>,
"bpf@...r.kernel.org" <bpf@...r.kernel.org>,
"daniel@...earbox.net" <daniel@...earbox.net>
Subject: Re: [PATCH bpf-next 1/2] btf: resolve enum fwds in btf_dedup
On Mon, Mar 11, 2019 at 10:39:37AM -0700, Andrii Nakryiko wrote:
>
> There is a GCC extension that allows to forward-declare enums:
> https://gcc.gnu.org/onlinedocs/gcc/Incomplete-Enums.html#Incomplete-Enums
>
> Kernel actually has examples of using this (see patch description
> regrading irqchip_irq_state).
we've considered several different ways of either extending or changing BTF api
to accommodate this unforeseen C langauge extension.
All of them require patches for the kernel 5.1 and 5.0 and/or llvm,
so we've decided to proceed with the existing kernel code and llvm
and only tweak btf dedup logic the way this patch does.
I've applied this set to bpf tree.
Powered by blists - more mailing lists