net: fix checksum on big endian CPUs
diff mbox series

Message ID 20200710114313.7412-1-guohongzhi1@huawei.com
State Accepted
Delegated to: Thomas Monjalon
Headers show
Series
  • net: fix checksum on big endian CPUs
Related show

Checks

Context Check Description
ci/iol-intel-Performance success Performance Testing PASS
ci/iol-testing success Testing PASS
ci/Intel-compilation success Compilation OK
ci/travis-robot success Travis build: passed
ci/iol-broadcom-Performance success Performance Testing PASS
ci/checkpatch success coding style OK

Commit Message

Hongzhi Guo July 10, 2020, 11:43 a.m. UTC
With current code, the checksum of odd-length buffers is wrong on
big endian CPUs: the last byte is not properly summed to the
accumulator.

Fix this by left-shifting the remaining byte by 8. For instance,
if the last byte is 0x42, we should add 0x4200 to the accumulator
on big endian CPUs.

This change is similar to what is suggested in Errata 3133 of
RFC 1071.

Fixes: 6006818cfb26("net: new checksum functions")
Cc: stable@dpdk.org

Signed-off-by: Hongzhi Guo <guohongzhi1@huawei.com>
---
v2:
* Explain the logic in the commit log
* Fixed commit title
---
---
 lib/librte_net/rte_ip.h | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

Comments

Morten Brørup July 10, 2020, 12:20 p.m. UTC | #1
> From: dev [mailto:dev-bounces@dpdk.org] On Behalf Of Hongzhi Guo
> Sent: Friday, July 10, 2020 1:43 PM
> 
> With current code, the checksum of odd-length buffers is wrong on
> big endian CPUs: the last byte is not properly summed to the
> accumulator.
> 
> Fix this by left-shifting the remaining byte by 8. For instance,
> if the last byte is 0x42, we should add 0x4200 to the accumulator
> on big endian CPUs.
> 
> This change is similar to what is suggested in Errata 3133 of
> RFC 1071.
> 
> Fixes: 6006818cfb26("net: new checksum functions")
> Cc: stable@dpdk.org
> 
> Signed-off-by: Hongzhi Guo <guohongzhi1@huawei.com>
> ---
> v2:
> * Explain the logic in the commit log
> * Fixed commit title
> ---
> ---
>  lib/librte_net/rte_ip.h | 7 +++++--
>  1 file changed, 5 insertions(+), 2 deletions(-)
> 
> diff --git a/lib/librte_net/rte_ip.h b/lib/librte_net/rte_ip.h
> index 292f63fd7..4fb0e314a 100644
> --- a/lib/librte_net/rte_ip.h
> +++ b/lib/librte_net/rte_ip.h
> @@ -139,8 +139,11 @@ __rte_raw_cksum(const void *buf, size_t len,
> uint32_t sum)
>  	}
> 
>  	/* if length is in odd bytes */
> -	if (len == 1)
> -		sum += *((const uint8_t *)u16_buf);
> +	if (len == 1) {
> +		uint16_t left = 0;
> +		*(uint8_t *)&left = *(const uint8_t *)u16_buf;
> +		sum += left;
> +	}
> 
>  	return sum;
>  }
> --
> 2.21.0.windows.1
> 

This is correct for both big and little endian CPUs.

Reviewed-by: Morten Brørup <mb@smartsharesystems.com>
Olivier Matz July 10, 2020, 12:37 p.m. UTC | #2
On Fri, Jul 10, 2020 at 02:20:08PM +0200, Morten Brørup wrote:
> > From: dev [mailto:dev-bounces@dpdk.org] On Behalf Of Hongzhi Guo
> > Sent: Friday, July 10, 2020 1:43 PM
> > 
> > With current code, the checksum of odd-length buffers is wrong on
> > big endian CPUs: the last byte is not properly summed to the
> > accumulator.
> > 
> > Fix this by left-shifting the remaining byte by 8. For instance,
> > if the last byte is 0x42, we should add 0x4200 to the accumulator
> > on big endian CPUs.
> > 
> > This change is similar to what is suggested in Errata 3133 of
> > RFC 1071.
> > 
> > Fixes: 6006818cfb26("net: new checksum functions")
> > Cc: stable@dpdk.org
> > 
> > Signed-off-by: Hongzhi Guo <guohongzhi1@huawei.com>
> > ---
> > v2:
> > * Explain the logic in the commit log
> > * Fixed commit title
> > ---
> > ---
> >  lib/librte_net/rte_ip.h | 7 +++++--
> >  1 file changed, 5 insertions(+), 2 deletions(-)
> > 
> > diff --git a/lib/librte_net/rte_ip.h b/lib/librte_net/rte_ip.h
> > index 292f63fd7..4fb0e314a 100644
> > --- a/lib/librte_net/rte_ip.h
> > +++ b/lib/librte_net/rte_ip.h
> > @@ -139,8 +139,11 @@ __rte_raw_cksum(const void *buf, size_t len,
> > uint32_t sum)
> >  	}
> > 
> >  	/* if length is in odd bytes */
> > -	if (len == 1)
> > -		sum += *((const uint8_t *)u16_buf);
> > +	if (len == 1) {
> > +		uint16_t left = 0;
> > +		*(uint8_t *)&left = *(const uint8_t *)u16_buf;
> > +		sum += left;
> > +	}
> > 
> >  	return sum;
> >  }
> > --
> > 2.21.0.windows.1
> > 
> 
> This is correct for both big and little endian CPUs.
> 
> Reviewed-by: Morten Brørup <mb@smartsharesystems.com>

Acked-by: Olivier Matz <olivier.matz@6wind.com>

Thanks!
Thomas Monjalon July 10, 2020, 9:11 p.m. UTC | #3
10/07/2020 14:37, Olivier Matz:
> On Fri, Jul 10, 2020 at 02:20:08PM +0200, Morten Brørup wrote:
> > > From: dev [mailto:dev-bounces@dpdk.org] On Behalf Of Hongzhi Guo
> > > Sent: Friday, July 10, 2020 1:43 PM
> > > 
> > > With current code, the checksum of odd-length buffers is wrong on
> > > big endian CPUs: the last byte is not properly summed to the
> > > accumulator.
> > > 
> > > Fix this by left-shifting the remaining byte by 8. For instance,
> > > if the last byte is 0x42, we should add 0x4200 to the accumulator
> > > on big endian CPUs.
> > > 
> > > This change is similar to what is suggested in Errata 3133 of
> > > RFC 1071.
> > > 
> > > Fixes: 6006818cfb26("net: new checksum functions")
> > > Cc: stable@dpdk.org
> > > 
> > > Signed-off-by: Hongzhi Guo <guohongzhi1@huawei.com>
> > 
> > Reviewed-by: Morten Brørup <mb@smartsharesystems.com>
> 
> Acked-by: Olivier Matz <olivier.matz@6wind.com>

Applied, thank you Hongzhi Guo, we want more patches like this :-)

Patch
diff mbox series

diff --git a/lib/librte_net/rte_ip.h b/lib/librte_net/rte_ip.h
index 292f63fd7..4fb0e314a 100644
--- a/lib/librte_net/rte_ip.h
+++ b/lib/librte_net/rte_ip.h
@@ -139,8 +139,11 @@  __rte_raw_cksum(const void *buf, size_t len, uint32_t sum)
 	}
 
 	/* if length is in odd bytes */
-	if (len == 1)
-		sum += *((const uint8_t *)u16_buf);
+	if (len == 1) {
+		uint16_t left = 0;
+		*(uint8_t *)&left = *(const uint8_t *)u16_buf;
+		sum += left;
+	}
 
 	return sum;
 }