netdev - Re: [PATCH net-next 2/2] net: optimize skb_postpull

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: <20211202205806.kw4v5grupjlzober@skbuf>
Date:   Thu, 2 Dec 2021 22:58:06 +0200
From:   Vladimir Oltean <olteanv@...il.com>
To:     Eric Dumazet <edumazet@...gle.com>
Cc:     Eric Dumazet <eric.dumazet@...il.com>,
        "David S . Miller" <davem@...emloft.net>,
        Jakub Kicinski <kuba@...nel.org>,
        netdev <netdev@...r.kernel.org>,
        David Laight <David.Laight@...lab.com>
Subject: Re: [PATCH net-next 2/2] net: optimize skb_postpull_rcsum()

On Thu, Dec 02, 2021 at 10:40:36PM +0200, Vladimir Oltean wrote:
> To me it looks like the strange part is that the checksum of the removed
> block (printed by me as "csum_partial(start, len, 0)" inside
> skb_postpull_rcsum()) is the same as the skb->csum itself.
> 
> [   66.287583] fsl_enetc 0000:00:00.2 eno2: enetc_get_offloads 991: skb 0xffff4050c3671f00 csum 0x3c1d
> [   66.296716] skb csum of 20 bytes (20 to the left of skb->data) using old method: 0x0, new method: 0xffffffff, orig csum 0x3c1d, csum of removed block 0x3c1d

sorry, this line is confusing, what is printed as the "old method" is in
fact the "new method" and viceversa. I tried to rename them from
something clearer than "method 1" and "method 2" and failed.

> [   66.310786] skb len=84 headroom=98 headlen=84 tailroom=1546
> [   66.310786] mac=(84,-6) net=(78,0) trans=78
> [   66.310786] shinfo(txflags=0 nr_frags=0 gso(size=0 type=0 segs=0))
> [   66.310786] csum(0xffffffff ip_summed=2 complete_sw=0 valid=0 level=0)
> [   66.310786] hash(0x0 sw=0 l4=0) proto=0x00f8 pkttype=3 iif=7
> [   66.338997] dev name=eno2 feat=0x0x00020100001149a9
> [   66.343904] skb headroom: 00000000: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> [   66.351600] skb headroom: 00000010: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> [   66.359295] skb headroom: 00000020: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> [   66.366990] skb headroom: 00000030: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> [   66.374685] skb headroom: 00000040: 88 80 00 0a 00 38 c1 17 3d 01 01 80 00 00 08 0f
> [   66.382379] skb headroom: 00000050: 00 10 00 00 52 f3 98 af f9 8c d2 ee 27 92 2d 6c
> [   66.390073] skb headroom: 00000060: 08 00
> [   66.394105] skb linear:   00000000: 45 00 00 54 c8 59 40 00 40 01 28 fb c0 a8 64 01
> [   66.401799] skb linear:   00000010: c0 a8 64 02 08 00 ee 94 06 98 00 04 03 2e a9 61
> [   66.409493] skb linear:   00000020: 00 00 00 00 8b 6c 0c 00 00 00 00 00 10 11 12 13
> [   66.417187] skb linear:   00000030: 14 15 16 17 18 19 1a 1b 1c 1d 1e 1f 20 21 22 23
> [   66.424880] skb linear:   00000040: 24 25 26 27 28 29 2a 2b 2c 2d 2e 2f 30 31 32 33
> [   66.432574] skb linear:   00000050: 34 35 36 37
> [   66.437128] skb tailroom: 00000000: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> [   67.181735] fsl_enetc 0000:00:00.2 eno2: ocelot_rcv 131: skb 0xffff4050c3671f00 csum before skb_postpull_rcsum 0x3c1d, after 0xffffffff