Re: [PATCH v4 net-next 12/12] mlx5: support BIG TCP packets

[Jakub Kicinski <kuba@kernel.org>]

On Fri, 6 May 2022 17:32:43 -0700 Eric Dumazet wrote:
> On Fri, May 6, 2022 at 3:34 PM Jakub Kicinski <kuba@kernel.org> wrote:
> >
> > On Fri,  6 May 2022 08:30:48 -0700 Eric Dumazet wrote:  
> > > From: Coco Li <lixiaoyan@google.com>
> > >
> > > mlx5 supports LSOv2.
> > >
> > > IPv6 gro/tcp stacks insert a temporary Hop-by-Hop header
> > > with JUMBO TLV for big packets.
> > >
> > > We need to ignore/skip this HBH header when populating TX descriptor.
> > >
> > > Note that ipv6_has_hopopt_jumbo() only recognizes very specific packet
> > > layout, thus mlx5e_sq_xmit_wqe() is taking care of this layout only.
> > >
> > > v2: clear hopbyhop in mlx5e_tx_get_gso_ihs()
> > > v4: fix compile error for CONFIG_MLX5_CORE_IPOIB=y  
> >
> > In file included from ../include/linux/string.h:253,
> >                  from ../arch/x86/include/asm/page_32.h:22,
> >                  from ../arch/x86/include/asm/page.h:14,
> >                  from ../arch/x86/include/asm/processor.h:19,
> >                  from ../arch/x86/include/asm/timex.h:5,
> >                  from ../include/linux/timex.h:65,
> >                  from ../include/linux/time32.h:13,
> >                  from ../include/linux/time.h:60,
> >                  from ../include/linux/skbuff.h:15,
> >                  from ../include/linux/tcp.h:17,
> >                  from ../drivers/net/ethernet/mellanox/mlx5/core/en_tx.c:33:
> > In function ‘fortify_memcpy_chk’,
> >     inlined from ‘mlx5e_insert_vlan’ at ../drivers/net/ethernet/mellanox/mlx5/core/en_tx.c:104:2,
> >     inlined from ‘mlx5e_sq_xmit_wqe’ at ../drivers/net/ethernet/mellanox/mlx5/core/en_tx.c:404:5:
> > ../include/linux/fortify-string.h:328:25: warning: call to ‘__write_overflow_field’ declared with attribute warning: detected write beyond size of field (1st parameter); maybe use struct_group()? [-Wattribute-warning]
> >   328 |                         __write_overflow_field(p_size_field, size);
> >       |                         ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> > In function ‘fortify_memcpy_chk’,
> >     inlined from ‘mlx5e_sq_xmit_wqe’ at ../drivers/net/ethernet/mellanox/mlx5/core/en_tx.c:408:5:
> > ../include/linux/fortify-string.h:328:25: warning: call to ‘__write_overflow_field’ declared with attribute warning: detected write beyond size of field (1st parameter); maybe use struct_group()? [-Wattribute-warning]
> >   328 |                         __write_overflow_field(p_size_field, size);
> >       |                         ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> > In function ‘fortify_memcpy_chk’,
> >     inlined from ‘mlx5i_sq_xmit’ at ../drivers/net/ethernet/mellanox/mlx5/core/en_tx.c:962:4:
> > ../include/linux/fortify-string.h:328:25: warning: call to ‘__write_overflow_field’ declared with attribute warning: detected write beyond size of field (1st parameter); maybe use struct_group()? [-Wattribute-warning]
> >   328 |                         __write_overflow_field(p_size_field, size);
> >       |                         ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~  
> 
> I guess these warnings show up before this BIG TCP patch ?
> 
> I do not see any struct_group() being used in mlx5
> 
> May I ask which compiler is used here, and what CONFIG_ option needs to be set ?
> 
> Thanks.

Without our patches drivers/net/ethernet/mellanox/mlx5/core/ builds
cleanly. Gotta be the new W=1 filed overflow warnings, let's bother
Kees.

I believe this is the code in question:

@@ -379,15 +393,36 @@ mlx5e_sq_xmit_wqe(struct mlx5e_txqsq *sq, struct sk_buff *skb,

+		u8 *start = eseg->inline_hdr.start;
+
+		if (unlikely(attr->hopbyhop)) {
+			/* remove the HBH header.
+			 * Layout: [Ethernet header][IPv6 header][HBH][TCP header]
+			 */
+			if (skb_vlan_tag_present(skb)) {
+				mlx5e_insert_vlan(start, skb, ETH_HLEN + sizeof(*h6));

Unhappiness #1 ^^^

Where mlx5e_insert_vlan() is:

static inline void mlx5e_insert_vlan(void *start, struct sk_buff *skb, u16 ihs)
{
	struct vlan_ethhdr *vhdr = (struct vlan_ethhdr *)start;
	int cpy1_sz = 2 * ETH_ALEN;
	int cpy2_sz = ihs - cpy1_sz;

	memcpy(&vhdr->addrs, skb->data, cpy1_sz);
	vhdr->h_vlan_proto = skb->vlan_proto;
	vhdr->h_vlan_TCI = cpu_to_be16(skb_vlan_tag_get(skb));
	memcpy(&vhdr->h_vlan_encapsulated_proto, skb->data + cpy1_sz, cpy2_sz);
}

indeed ihs == ETH_HLEN + sizeof(*h6) will make cpy2_sz come out as something
much bigger than the vhdr->h_vlan_encapsulated_proto field.

+				ihs += VLAN_HLEN;
+				h6 = (struct ipv6hdr *)(start + sizeof(struct vlan_ethhdr));
+			} else {
+				memcpy(start, skb->data, ETH_HLEN + sizeof(*h6));

Unhappiness #2 ^^^

Again, ETH_HLEN + sizeof(*h6) will be larger than eseg->inline_hdr.start
which is what start is pointing at.

+				h6 = (struct ipv6hdr *)(start + ETH_HLEN);
+			}

I didn't look where #3 is.