From: Ferruh Yigit <ferruh.yigit@intel.com>
To: Michael Pfeiffer <michael.pfeiffer@tu-ilmenau.de>,
Keith Wiles <keith.wiles@intel.com>,
Olivier Matz <olivier.matz@6wind.com>
Cc: dev@dpdk.org
Subject: Re: [dpdk-dev] [PATCH] net/tap: Allow all-zero checksum for UDP over IPv4
Date: Fri, 13 Nov 2020 13:02:41 +0000 [thread overview]
Message-ID: <036c2d5f-107b-bb5b-1893-9c3cdd26a1da@intel.com> (raw)
In-Reply-To: <40d63e07-f179-b3a8-ee7a-ff37e87c1008@intel.com>
On 11/11/2020 9:31 AM, Ferruh Yigit wrote:
> On 11/11/2020 7:23 AM, Michael Pfeiffer wrote:
>> Hi,
>>
>> On Tue, 2020-11-10 at 15:59 +0000, Ferruh Yigit wrote:
>>> On 11/9/2020 2:22 PM, Michael Pfeiffer wrote:
>>>> Unlike TCP, UDP checksums are optional and may be zero to indicate "not
>>>> set" [RFC 768] (except for IPv6, where this prohibited [RFC 8200]). Add
>>>> this special case to the checksum offload emulation in net/tap.
>>>>
>>>> Signed-off-by: Michael Pfeiffer <michael.pfeiffer@tu-ilmenau.de>
>>>> ---
>>>> drivers/net/tap/rte_eth_tap.c | 13 +++++++++++--
>>>> 1 file changed, 11 insertions(+), 2 deletions(-)
>>>>
>>>> diff --git a/drivers/net/tap/rte_eth_tap.c b/drivers/net/tap/rte_eth_tap.c
>>>> index 2f8abb12c..e486b41c5 100644
>>>> --- a/drivers/net/tap/rte_eth_tap.c
>>>> +++ b/drivers/net/tap/rte_eth_tap.c
>>>> @@ -303,6 +303,7 @@ tap_verify_csum(struct rte_mbuf *mbuf)
>>>> uint16_t cksum = 0;
>>>> void *l3_hdr;
>>>> void *l4_hdr;
>>>> + struct rte_udp_hdr *udp_hdr;
>>>> if (l2 == RTE_PTYPE_L2_ETHER_VLAN)
>>>> l2_len += 4;
>>>> @@ -349,10 +350,18 @@ tap_verify_csum(struct rte_mbuf *mbuf)
>>>> /* Don't verify checksum for multi-segment packets. */
>>>> if (mbuf->nb_segs > 1)
>>>> return;
>>>> - if (l3 == RTE_PTYPE_L3_IPV4)
>>>> + if (l3 == RTE_PTYPE_L3_IPV4) {
>>>> + if (l4 == RTE_PTYPE_L4_UDP) {
>>>> + udp_hdr = (struct rte_udp_hdr *)l4_hdr;
>>>> + if (udp_hdr->dgram_cksum == 0) {
>>>
>>> Overall patch looks good to me, but can you please add a comment on top of
>>> above
>>> check to describe why checksum can be zero, as done in the commit log.
>>
>> Sure, I will update the patch. I am also not completely sure whether
>> PKT_RX_L4_CKSUM_NONE is the right flag for this case (rather than _UNKNOWN).
>> From rte_core_mbuf.h:
>>
>> * - PKT_RX_L4_CKSUM_UNKNOWN: no information about the RX L4 checksum
>> * - PKT_RX_L4_CKSUM_NONE: the L4 checksum is not correct in the packet
>> * data, but the integrity of the L4 data is verified.
>>
>> The second part after the "but" is not really the case here. I don't know how
>> relevant the distinction is, as most application side code will probably only
>> do something like
>>
>> if ((mbuf->ol_flags & PKT_RX_L4_CKSUM_MASK) == PKT_RX_L4_CKSUM_BAD)
>> rte_pktmbuf_free(mbuf);
>>
>> anyway. Do you have any opinions on that?
>>
>
> I also checked for that and wasn't sure about it :) cc'ed Olivier too for comment.
>
> I think it is NOT 'PKT_RX_L4_CKSUM_UNKNOWN', since we know that checksum value
> is 0x0000 which means it is not provided.
>
> 'PKT_RX_L4_CKSUM_NONE' suits better but not sure about the expectation on
> "integrity of the L4 data is verified" part, I assume that explanation is just
> to differentiate between 'CKSUM_BAD'.
I suggest to continue with 'PKT_RX_L4_CKSUM_NONE'.
Can it be possible to get the new version today, so we can include this to the -rc4?
Thanks
next prev parent reply other threads:[~2020-11-13 13:02 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-11-09 14:22 Michael Pfeiffer
2020-11-10 14:46 ` Ferruh Yigit
2020-11-10 15:56 ` Ferruh Yigit
2020-11-10 16:01 ` Morten Brørup
2020-11-10 17:42 ` Ferruh Yigit
2020-11-11 7:06 ` Morten Brørup
2020-11-10 15:59 ` Ferruh Yigit
2020-11-11 7:23 ` Michael Pfeiffer
2020-11-11 9:31 ` Ferruh Yigit
2020-11-13 13:02 ` Ferruh Yigit [this message]
2020-11-13 14:03 ` [dpdk-dev] [PATCH v2] " Michael Pfeiffer
2020-11-13 14:49 ` Ferruh Yigit
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=036c2d5f-107b-bb5b-1893-9c3cdd26a1da@intel.com \
--to=ferruh.yigit@intel.com \
--cc=dev@dpdk.org \
--cc=keith.wiles@intel.com \
--cc=michael.pfeiffer@tu-ilmenau.de \
--cc=olivier.matz@6wind.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).