powerpc: fix csum_ipv6_magic() on little endian platforms
diff mbox series

Message ID 3c4cddf185438cc7200908b4f971a8e6ed7472fd.1536558233.git.christophe.leroy@c-s.fr
State New, archived
Headers show
Series
  • powerpc: fix csum_ipv6_magic() on little endian platforms
Related show

Commit Message

Christophe Leroy Sept. 10, 2018, 6:09 a.m. UTC
On little endian platforms, csum_ipv6_magic() keeps len and proto in
CPU byte order. This generates a bad results leading to ICMPv6 packets
from other hosts being dropped by powerpc64le platforms.

In order to fix this, len and proto should be converted to network
byte order ie bigendian byte order. However checksumming 0x12345678
and 0x56341278 provide the exact same result so it is enough to
rotate the sum of len and proto by 1 byte.

PPC32 only support bigendian so the fix is needed for PPC64 only

Fixes: e9c4943a107b ("powerpc: Implement csum_ipv6_magic in assembly")
Reported-by: Jianlin Shi <jishi@redhat.com>
Reported-by: Xin Long <lucien.xin@gmail.com>
Cc: <stable@vger.kernel.org> # 4.18+
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
---
 arch/powerpc/lib/checksum_64.S | 3 +++
 1 file changed, 3 insertions(+)

Comments

Xin Long Sept. 10, 2018, 2:28 p.m. UTC | #1
On Mon, Sep 10, 2018 at 2:09 PM Christophe Leroy
<christophe.leroy@c-s.fr> wrote:
>
> On little endian platforms, csum_ipv6_magic() keeps len and proto in
> CPU byte order. This generates a bad results leading to ICMPv6 packets
> from other hosts being dropped by powerpc64le platforms.
>
> In order to fix this, len and proto should be converted to network
> byte order ie bigendian byte order. However checksumming 0x12345678
> and 0x56341278 provide the exact same result so it is enough to
> rotate the sum of len and proto by 1 byte.
>
> PPC32 only support bigendian so the fix is needed for PPC64 only
>
> Fixes: e9c4943a107b ("powerpc: Implement csum_ipv6_magic in assembly")
> Reported-by: Jianlin Shi <jishi@redhat.com>
> Reported-by: Xin Long <lucien.xin@gmail.com>
> Cc: <stable@vger.kernel.org> # 4.18+
> Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
> ---
>  arch/powerpc/lib/checksum_64.S | 3 +++
>  1 file changed, 3 insertions(+)
>
> diff --git a/arch/powerpc/lib/checksum_64.S b/arch/powerpc/lib/checksum_64.S
> index 886ed94b9c13..2a68c43e13f5 100644
> --- a/arch/powerpc/lib/checksum_64.S
> +++ b/arch/powerpc/lib/checksum_64.S
> @@ -443,6 +443,9 @@ _GLOBAL(csum_ipv6_magic)
>         addc    r0, r8, r9
>         ld      r10, 0(r4)
>         ld      r11, 8(r4)
> +#ifndef CONFIG_CPU_BIG_ENDIAN
> +       rotldi  r5, r5, 8
> +#endif
>         adde    r0, r0, r10
>         add     r5, r5, r7
>         adde    r0, r0, r11
> --
> 2.13.3
>
Tested-by: Xin Long <lucien.xin@gmail.com>
Christophe Leroy Sept. 17, 2018, 12:27 p.m. UTC | #2
Hi Michael,

Le 10/09/2018 à 16:28, Xin Long a écrit :
> On Mon, Sep 10, 2018 at 2:09 PM Christophe Leroy
> <christophe.leroy@c-s.fr> wrote:
>>
>> On little endian platforms, csum_ipv6_magic() keeps len and proto in
>> CPU byte order. This generates a bad results leading to ICMPv6 packets
>> from other hosts being dropped by powerpc64le platforms.
>>
>> In order to fix this, len and proto should be converted to network
>> byte order ie bigendian byte order. However checksumming 0x12345678
>> and 0x56341278 provide the exact same result so it is enough to
>> rotate the sum of len and proto by 1 byte.
>>
>> PPC32 only support bigendian so the fix is needed for PPC64 only
>>
>> Fixes: e9c4943a107b ("powerpc: Implement csum_ipv6_magic in assembly")
>> Reported-by: Jianlin Shi <jishi@redhat.com>
>> Reported-by: Xin Long <lucien.xin@gmail.com>
>> Cc: <stable@vger.kernel.org> # 4.18+
>> Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
>> ---
>>   arch/powerpc/lib/checksum_64.S | 3 +++
>>   1 file changed, 3 insertions(+)
>>
>> diff --git a/arch/powerpc/lib/checksum_64.S b/arch/powerpc/lib/checksum_64.S
>> index 886ed94b9c13..2a68c43e13f5 100644
>> --- a/arch/powerpc/lib/checksum_64.S
>> +++ b/arch/powerpc/lib/checksum_64.S
>> @@ -443,6 +443,9 @@ _GLOBAL(csum_ipv6_magic)
>>          addc    r0, r8, r9
>>          ld      r10, 0(r4)
>>          ld      r11, 8(r4)
>> +#ifndef CONFIG_CPU_BIG_ENDIAN
>> +       rotldi  r5, r5, 8
>> +#endif
>>          adde    r0, r0, r10
>>          add     r5, r5, r7
>>          adde    r0, r0, r11
>> --
>> 2.13.3
>>
> Tested-by: Xin Long <lucien.xin@gmail.com>
> 

Could you take this fix for 4.19 ?

Unless someone takes it through the netdev tree ?

Thanks
Christophe
Michael Ellerman Sept. 21, 2018, 11:59 a.m. UTC | #3
On Mon, 2018-09-10 at 06:09:04 UTC, Christophe Leroy wrote:
> On little endian platforms, csum_ipv6_magic() keeps len and proto in
> CPU byte order. This generates a bad results leading to ICMPv6 packets
> from other hosts being dropped by powerpc64le platforms.
> 
> In order to fix this, len and proto should be converted to network
> byte order ie bigendian byte order. However checksumming 0x12345678
> and 0x56341278 provide the exact same result so it is enough to
> rotate the sum of len and proto by 1 byte.
> 
> PPC32 only support bigendian so the fix is needed for PPC64 only
> 
> Fixes: e9c4943a107b ("powerpc: Implement csum_ipv6_magic in assembly")
> Reported-by: Jianlin Shi <jishi@redhat.com>
> Reported-by: Xin Long <lucien.xin@gmail.com>
> Cc: <stable@vger.kernel.org> # 4.18+
> Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
> Tested-by: Xin Long <lucien.xin@gmail.com>

Applied to powerpc fixes, thanks.

https://git.kernel.org/powerpc/c/85682a7e3b9c664995ad477520f917

cheers

Patch
diff mbox series

diff --git a/arch/powerpc/lib/checksum_64.S b/arch/powerpc/lib/checksum_64.S
index 886ed94b9c13..2a68c43e13f5 100644
--- a/arch/powerpc/lib/checksum_64.S
+++ b/arch/powerpc/lib/checksum_64.S
@@ -443,6 +443,9 @@  _GLOBAL(csum_ipv6_magic)
 	addc	r0, r8, r9
 	ld	r10, 0(r4)
 	ld	r11, 8(r4)
+#ifndef CONFIG_CPU_BIG_ENDIAN
+	rotldi	r5, r5, 8
+#endif
 	adde	r0, r0, r10
 	add	r5, r5, r7
 	adde	r0, r0, r11