> -----Original Message-----
> From: Intel-wired-lan <intel-wired-lan-boun...@osuosl.org> On Behalf Of 
> Corinna Vinschen
> Sent: Thursday, July 18, 2024 2:27 PM
> To: net...@vger.kernel.org; intel-wired-...@lists.osuosl.org; Eric Dumazet 
> <eduma...@google.com>
> Cc: Jason Xing <kerneljasonx...@gmail.com>; Nikolay Aleksandrov 
> <ra...@blackwall.org>; linux-ker...@vger.kernel.org; Nguyen, Anthony L 
> <anthony.l.ngu...@intel.com>; Jakub Kicinski <k...@kernel.org>; Paolo Abeni 
> <pab...@redhat.com>; David S . Miller <da...@davemloft.net>
> Subject: [Intel-wired-lan] [PATCH net v3] igb: cope with large MAX_SKB_FRAGS.
>
> From: Paolo Abeni <pab...@redhat.com>
>
> Sabrina reports that the igb driver does not cope well with large 
> MAX_SKB_FRAG values: setting MAX_SKB_FRAG to 45 causes payload corruption on 
> TX.
>
> An easy reproducer is to run ssh to connect to the machine.  With
> MAX_SKB_FRAGS=17 it works, with MAX_SKB_FRAGS=45 it fails.
>
> The root cause of the issue is that the driver does not take into account 
> properly the (possibly large) shared info size when selecting the ring 
> layout, and will try to fit two packets inside the same 4K page even when the 
> 1st fraglist will trump over the 2nd head.
>
> Address the issue forcing the driver to fit a single packet per page, leaving 
> there enough room to store the (currently) largest possible skb_shared_info.
>
> Fixes: 3948b05950fd ("net: introduce a config option to tweak MAX_SKB_FRAGS")
> Reported-by: Jan Tluka <jtl...@redhat.com>
> Reported-by: Jirka Hladky <jhla...@redhat.com>
> Reported-by: Sabrina Dubroca <s...@queasysnail.net>
> Tested-by: Sabrina Dubroca <s...@queasysnail.net>
> Tested-by: Corinna Vinschen <vinsc...@redhat.com>
> Signed-off-by: Paolo Abeni <pab...@redhat.com>
> ---
> v2: fix subject, add a simple reproducer
> v3: fix Fixes, tested with all MTUs from 1200 to 1280 per Eric's suggestion
>
>  drivers/net/ethernet/intel/igb/igb_main.c | 1 +
>  1 file changed, 1 insertion(+)
>

Tested-by: Pucha Himasekhar Reddy <himasekharx.reddy.pu...@intel.com> (A 
Contingent worker at Intel)


Reply via email to