public inbox for passt-dev@passt.top
 help / color / mirror / code / Atom feed
From: Laurent Vivier <lvivier@redhat.com>
To: Stefano Brivio <sbrivio@redhat.com>
Cc: passt-dev@passt.top
Subject: Re: [PATCH v5 4/5] tcp: Update TCP checksum using an iovec array
Date: Wed, 2 Oct 2024 09:39:23 +0200	[thread overview]
Message-ID: <4efad680-ccf3-45a8-8929-cc5ae2be4760@redhat.com> (raw)
In-Reply-To: <20241001202216.5c3ee274@elisabeth>

On 01/10/2024 20:22, Stefano Brivio wrote:
> On Tue, 1 Oct 2024 09:29:28 +0200
> Stefano Brivio <sbrivio@redhat.com> wrote:
> 
>> On Fri, 27 Sep 2024 15:53:48 +0200
>> Laurent Vivier <lvivier@redhat.com> wrote:
>>
>>> [...]
>>>
>>> +void tcp_update_check_tcp4(const struct iphdr *iph,
>>> +			   const struct iovec *iov, int iov_cnt,
>>> +			   size_t l4offset)
>>>   {
>>>   	uint16_t l4len = ntohs(iph->tot_len) - sizeof(struct iphdr);
>>>   	struct in_addr saddr = { .s_addr = iph->saddr };
>>>   	struct in_addr daddr = { .s_addr = iph->daddr };
>>> -	uint32_t sum = proto_ipv4_header_psum(l4len, IPPROTO_TCP, saddr, daddr);
>>> +	size_t check_ofs;
>>> +	__sum16 *check;
>>> +	int check_idx;
>>> +	uint32_t sum;
>>> +	uintptr_t ptr;
>>> +
>>> +	sum = proto_ipv4_header_psum(l4len, IPPROTO_TCP, saddr, daddr);
>>> +
>>> +	check_idx = iov_skip_bytes(iov, iov_cnt,
>>> +				   l4offset + offsetof(struct tcphdr, check),
>>> +				   &check_ofs);
>>> +
>>> +	if (check_idx >= iov_cnt) {
>>> +		err("TCP4 buffer is too small, iov size %zd, check offset %zd",
>>> +		    iov_size(iov, iov_cnt),
>>> +		    l4offset + offsetof(struct tcphdr, check));
>>> +		return;
>>> +	}
>>>   
>>> -	bp->th.check = 0;
>>> -	bp->th.check = csum(bp, l4len, sum);
>>> +	if (check_ofs + sizeof(*check) > iov[check_idx].iov_len) {
>>> +		err("TCP4 checksum field memory is not contiguous "
>>> +		    "check_ofs %zd check_idx %d iov_len %zd",
>>> +		    check_ofs, check_idx, iov[check_idx].iov_len);
>>> +		return;
>>> +	}
>>> +
>>> +	ptr = (uintptr_t)((char *)iov[check_idx].iov_base + check_ofs);
>>> +	if (ptr & (__alignof__(*check) - 1)) {
>>> +		err("TCP4 checksum field is not correctly aligned in memory");
>>> +		return;
>>> +	}
>>> +
>>> +	check = (__sum16 *)ptr;
>>
>> clang-tidy (14.0.6) says:
>>
>> /home/sbrivio/passt/tcp.c:803:10: error: integer to pointer cast pessimizes optimization opportunities [performance-no-int-to-ptr,-warnings-as-errors]
>>          check = (__sum16 *)ptr;
>>                  ^
>> /home/sbrivio/passt/tcp.c:854:10: error: integer to pointer cast pessimizes optimization opportunities [performance-no-int-to-ptr,-warnings-as-errors]
>>          check = (__sum16 *)ptr;
>>                  ^
>>
>> ...I'm still trying to figure out what that actually means. If it's
>> trivial I can fix that up on merge.
> 
> Something like this?

It's ok for me. Do you want I resend a version with it or can you update the patch on the 
merge?

Thanks,
Laurent

> 
> --
> diff --git a/tcp.c b/tcp.c
> index ffd82e1..2b8edda 100644
> --- a/tcp.c
> +++ b/tcp.c
> @@ -772,7 +772,7 @@ void tcp_update_check_tcp4(const struct iphdr *iph,
>   	__sum16 *check;
>   	int check_idx;
>   	uint32_t sum;
> -	uintptr_t ptr;
> +	char *ptr;
>   
>   	sum = proto_ipv4_header_psum(l4len, IPPROTO_TCP, saddr, daddr);
>   
> @@ -794,8 +794,8 @@ void tcp_update_check_tcp4(const struct iphdr *iph,
>   		return;
>   	}
>   
> -	ptr = (uintptr_t)((char *)iov[check_idx].iov_base + check_ofs);
> -	if (ptr & (__alignof__(*check) - 1)) {
> +	ptr = (char *)iov[check_idx].iov_base + check_ofs;
> +	if ((uintptr_t)ptr & (__alignof__(*check) - 1)) {
>   		err("TCP4 checksum field is not correctly aligned in memory");
>   		return;
>   	}
> @@ -822,7 +822,7 @@ void tcp_update_check_tcp6(const struct ipv6hdr *ip6h,
>   	__sum16 *check;
>   	int check_idx;
>   	uint32_t sum;
> -	uintptr_t ptr;
> +	char *ptr;
>   
>   	sum = proto_ipv6_header_psum(l4len, IPPROTO_TCP, &ip6h->saddr,
>   				     &ip6h->daddr);
> @@ -845,8 +845,8 @@ void tcp_update_check_tcp6(const struct ipv6hdr *ip6h,
>   		return;
>   	}
>   
> -	ptr = (uintptr_t)((char *)iov[check_idx].iov_base + check_ofs);
> -	if (ptr & (__alignof__(*check) - 1)) {
> +	ptr = (char *)iov[check_idx].iov_base + check_ofs;
> +	if ((uintptr_t)ptr & (__alignof__(*check) - 1)) {
>   		err("TCP6 checksum field is not correctly aligned in memory");
>   		return;
>   	}
> 


  reply	other threads:[~2024-10-02  7:39 UTC|newest]

Thread overview: 16+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-09-27 13:53 [PATCH v5 0/5] tcp: use csum_iov() in tcp_update_check_tcp[4|6]() Laurent Vivier
2024-09-27 13:53 ` [PATCH v5 1/5] tcp: Use tcp_payload_t rather than tcphdr Laurent Vivier
2024-09-27 13:53 ` [PATCH v5 2/5] pcap: Add an offset argument in pcap_iov() Laurent Vivier
2024-09-27 13:53 ` [PATCH v5 3/5] checksum: Add an offset argument in csum_iov() Laurent Vivier
2024-09-30  2:51   ` David Gibson
2024-09-27 13:53 ` [PATCH v5 4/5] tcp: Update TCP checksum using an iovec array Laurent Vivier
2024-09-30  2:56   ` David Gibson
2024-10-01  7:29   ` Stefano Brivio
2024-10-01 18:22     ` Stefano Brivio
2024-10-02  7:39       ` Laurent Vivier [this message]
2024-10-02  9:00         ` Stefano Brivio
2024-10-03 12:58           ` Laurent Vivier
2024-09-27 13:53 ` [PATCH v5 5/5] udp: Update UDP " Laurent Vivier
2024-09-30  2:58   ` David Gibson
2024-10-02 14:32 ` [PATCH v5 0/5] tcp: use csum_iov() in tcp_update_check_tcp[4|6]() Stefano Brivio
2024-10-02 14:46   ` Stefano Brivio

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=4efad680-ccf3-45a8-8929-cc5ae2be4760@redhat.com \
    --to=lvivier@redhat.com \
    --cc=passt-dev@passt.top \
    --cc=sbrivio@redhat.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.
Code repositories for project(s) associated with this public inbox

	https://passt.top/passt

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 IMAP folder(s).