Re: [PATCH net] net: validate SO_TXTIME clockid coming from userspace

From: Willem de Bruijn
Date: Wed May 29 2024 - 12:00:26 EST


Abhishek Chauhan (ABC) wrote:
>
>
> On 5/29/2024 6:58 AM, Willem de Bruijn wrote:
> > minor: double space before userspace
> >
> > Abhishek Chauhan wrote:
> >> Currently there are no strict checks while setting SO_TXTIME
> >> from userspace. With the recent development in skb->tstamp_type
> >> clockid with unsupported clocks results in warn_on_once, which causes
> >> unnecessary aborts in some systems which enables panic on warns.
> >>
> >> Add validation in setsockopt to support only CLOCK_REALTIME,
> >> CLOCK_MONOTONIC and CLOCK_TAI to be set from userspace.
> >>
> >> Link: https://lore.kernel.org/netdev/bc037db4-58bb-4861-ac31-a361a93841d3@xxxxxxxxx/
> >> Link: https://lore.kernel.org/lkml/20240509211834.3235191-1-quic_abchauha@xxxxxxxxxxx/
> >
> > These discussions can be found directly from the referenced commit?
> > If any, I'd like to the conversation we had that arrived at this
> > approach.
> >
> Not Directly but from the patch series.
> 1. First link is for why we introduced skb->tstamp_type
> 2. Second link points to the series were we discussed on two approach to solve the problem
> one being limit the skclockid to just TAI,MONO and REALTIME.

Ah, I missed that.
Perhaps point directly to the start of that follow-up conversation?

https://lore.kernel.org/lkml/6bdba7b6-fd22-4ea5-a356-12268674def1@xxxxxxxxxxx/

>
>
> >> Fixes: 1693c5db6ab8 ("net: Add additional bit to support clockid_t timestamp type")
> >> Reported-by: syzbot+d7b227731ec589e7f4f0@xxxxxxxxxxxxxxxxxxxxxxxxx
> >> Closes: https://syzkaller.appspot.com/bug?extid=d7b227731ec589e7f4f0
> >> Reported-by: syzbot+30a35a2e9c5067cc43fa@xxxxxxxxxxxxxxxxxxxxxxxxx
> >> Closes: https://syzkaller.appspot.com/bug?extid=30a35a2e9c5067cc43fa
> >> Signed-off-by: Abhishek Chauhan <quic_abchauha@xxxxxxxxxxx>
> >> ---
> >> net/core/sock.c | 16 ++++++++++++++++
> >> 1 file changed, 16 insertions(+)
> >>
> >> diff --git a/net/core/sock.c b/net/core/sock.c
> >> index 8629f9aecf91..f8374be9d8c9 100644
> >> --- a/net/core/sock.c
> >> +++ b/net/core/sock.c
> >> @@ -1083,6 +1083,17 @@ bool sockopt_capable(int cap)
> >> }
> >> EXPORT_SYMBOL(sockopt_capable);
> >>
> >> +static int sockopt_validate_clockid(int value)
> >
> > sock_txtime.clockid has type __kernel_clockid_t.
> >
>
> __kernel_clockid_t is typedef of int.

It is now, but the stricter type definition exists for a reason.
Try to keep the strict types where possible. Besides aiding
syntactic checks, it also helps self document code.