netdev - Re: [PATCH bpf-next 0/8] bpf: Follow up to RCU enforcement in the verifier.

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]

Date:   Thu, 6 Apr 2023 18:17:57 -0700
From:   Alexei Starovoitov <alexei.starovoitov@...il.com>
To:     Jakub Kicinski <kuba@...nel.org>
Cc:     Andrii Nakryiko <andrii.nakryiko@...il.com>,
        David Vernet <void@...ifault.com>,
        "David S. Miller" <davem@...emloft.net>,
        Daniel Borkmann <daniel@...earbox.net>,
        Andrii Nakryiko <andrii@...nel.org>,
        Martin KaFai Lau <martin.lau@...nel.org>,
        Dave Marchevsky <davemarchevsky@...a.com>,
        Tejun Heo <tj@...nel.org>,
        Kumar Kartikeya Dwivedi <memxor@...il.com>,
        Network Development <netdev@...r.kernel.org>,
        bpf <bpf@...r.kernel.org>, Kernel Team <kernel-team@...com>,
        Yonghong Song <yhs@...a.com>, Song Liu <song@...nel.org>
Subject: Re: [PATCH bpf-next 0/8] bpf: Follow up to RCU enforcement in the verifier.

On Thu, Apr 6, 2023 at 8:42 AM Jakub Kicinski <kuba@...nel.org> wrote:
>
> On Wed, 5 Apr 2023 22:13:26 -0700 Alexei Starovoitov wrote:
> > On Wed, Apr 5, 2023 at 11:19 AM Jakub Kicinski <kuba@...nel.org> wrote:
> > >
> > > On Wed, 5 Apr 2023 10:22:16 -0700 Andrii Nakryiko wrote:
> > > > So I'm exclusively using `pw-apply -c <patchworks-url>` to apply
> > > > everything locally.
> > >
> > > I think you can throw -M after -c $url? It can only help... :)
> >
> > Yeah. If only...
> > I'm exclusively using -c.
> > -M only works with -s, but I couldn't make -s -M work either.
> > Do you pass the series as a number?
>
> Yes, it copy just the numerical ID into the terminal.
>
> > but then series_json=$(curl -s $srv/series/$1/) line
> > doesn't look right, since it's missing "/mbox/" ?
>
> That's loading JSON from the patchwork's REST API.

This line still doesn't work for me.
curl -s https://patchwork.kernel.org/series/736654/
returns:
The page URL requested (<code>/series/736654/</code>) does not exist.

while
curl -s https://patchwork.kernel.org/series/736654/mbox/
returns proper mbox format.

> > User error on my side, I guess.
> > My bash skills were too weak to make -c and -M work,
> > but .git/hooks tip is great!
> > Thank you.
>
> FWIW I think the below may work for using -c instead of -s.
> But it is mixing "Daniel paths" and "Jakub paths" in the script.
> The output is still a bit different than when using -s.
>
> diff --git a/pw-apply b/pw-apply
> index 5fc37a24b027..c9cec94a4a8c 100755
> --- a/pw-apply
> +++ b/pw-apply
> @@ -81,17 +81,15 @@ accept_series()
>  }
>
>  cover_from_url()
>  {
>    curl -s $1 | gunzip -f -c > tmp.i
> -  series_num=`grep "href=\"/series" tmp.i|cut -d/ -f3|head -1`
> +  series=`grep "href=\"/series" tmp.i|cut -d/ -f3|head -1`
>    cover_url=`grep "href=\"/project/netdevbpf/cover" tmp.i|cut -d\" -f2`
>    if [ ! -z "$cover_url" ]; then
> -    curl -s https://patchwork.kernel.org${cover_url}mbox/ | gunzip -f -c > cover.i
>      merge="1"
>    fi
> -  curl -s https://patchwork.kernel.org/series/$series_num/mbox/ | gunzip -f -c > mbox.i
>  }
>
>  edits=""
>  am_flags=""
>  branch="mbox"
> @@ -118,18 +116,20 @@ while true; do
>      -h | --help ) usage; break ;;
>      -- ) shift; break ;;
>      * )  break ;;
>    esac
>  done
> +# Load the info from cover first, it may will populate $series and $merge
> +[ ! -z "$cover" ]  && cover_from_url $cover
> +
>  [ ! -z "$auto_branch" ] && [ -z "$series" ] && usage
>  [ ! -z "$mbox" ]   && [ ! -z "$series" ] && usage
>  [   -z "$mbox" ]   && [   -z "$series" ] && [ -z "$cover" ] && usage
>  [ ! -z "$accept" ] && [ ! -z "$mbox" ]   && usage
>  [ ! -z "$series" ] && mbox_from_series $series
>  [ ! -z "$mbox" ]   && mbox_from_url $mbox
>  [ ! -z "$accept" ] && accept_series $series
> -[ ! -z "$cover" ]  && cover_from_url $cover

This part completely makes sense! Would be great to converge
on common usage, but json fetch doesn't work for me for some reason.
While mbox via original -c works fine.