[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <a2338f78-8f0c-2cdb-f040-bf0040ca6522@iogearbox.net>
Date: Thu, 31 Jan 2019 23:24:19 +0100
From: Daniel Borkmann <daniel@...earbox.net>
To: Alexei Starovoitov <ast@...nel.org>, davem@...emloft.net
Cc: peterz@...radead.org, jannh@...gle.com, paulmck@...ux.ibm.com,
will.deacon@....com, mingo@...hat.com, netdev@...r.kernel.org,
kernel-team@...com
Subject: Re: [PATCH bpf] bpf: run bpf programs with preemption disabled
On 01/29/2019 03:43 AM, Alexei Starovoitov wrote:
> Disabled preemption is necessary for proper access to per-cpu maps
> from BPF programs.
>
> But the sender side of socket filters didn't have preemption disabled:
> unix_dgram_sendmsg->sk_filter->sk_filter_trim_cap->bpf_prog_run_save_cb->BPF_PROG_RUN
>
> and a combination of af_packet with tun device didn't disable either:
> tpacket_snd->packet_direct_xmit->packet_pick_tx_queue->ndo_select_queue->
> tun_select_queue->tun_ebpf_select_queue->bpf_prog_run_clear_cb->BPF_PROG_RUN
>
> Disable preemption before executing BPF programs (both classic and extended).
>
> Reported-by: Jann Horn <jannh@...gle.com>
> Signed-off-by: Alexei Starovoitov <ast@...nel.org>
Applied, thanks!
Powered by blists - more mailing lists