* struct nbd_reply not packed => trouble?
@ 2013-01-10 16:09 Sebastian Pipping
2013-01-10 16:36 ` Yann Droneaud
0 siblings, 1 reply; 5+ messages in thread
From: Sebastian Pipping @ 2013-01-10 16:09 UTC (permalink / raw)
To: kernelnewbies
Hello kernel hackers,
I noticed that in [linux]/include/uapi/linux/nbd.h the structure
struct nbd_request {
__be32 magic;
__be32 type; /* == READ || == WRITE */
char handle[8];
__be64 from;
__be32 len;
} __attribute__((packed));
is packed but its reply counter part
struct nbd_reply {
__be32 magic;
__be32 error; /* 0 = ok, else error */
char handle[8]; /* handle you got from request */
};
is not. Since Linux seems to read sizeof(nbd_reply) bytes from the
network (see [1]), the number of bytes read varies with the number of
bytes of the structure.
So my understanding is that if the size of struct nbd_reply varies from
platform/compiler to another that means trouble. I wonder:
- Is there anything about struct nbd_reply that would keep its size
equal everywhere or a reason why variance in size is no problem here?
- Any ideas for a platform where you would expect struct nbd_reply to
be other than 4 + 4 + 8 = 16 bytes in size?
Best,
Sebastian
[1] http://lxr.linux.no/#linux+v3.7.1/drivers/block/nbd.c#L336
^ permalink raw reply [flat|nested] 5+ messages in thread* struct nbd_reply not packed => trouble?
2013-01-10 16:09 struct nbd_reply not packed => trouble? Sebastian Pipping
@ 2013-01-10 16:36 ` Yann Droneaud
2013-01-13 16:45 ` Sebastian Pipping
2013-01-14 23:59 ` 卜弋天
0 siblings, 2 replies; 5+ messages in thread
From: Yann Droneaud @ 2013-01-10 16:36 UTC (permalink / raw)
To: kernelnewbies
> Hello kernel hackers,
>
>
> I noticed that in [linux]/include/uapi/linux/nbd.h the structure
>
> struct nbd_request {
> __be32 magic;
> __be32 type; /* == READ || == WRITE */
> char handle[8];
> __be64 from;
> __be32 len;
> } __attribute__((packed));
>
> is packed
Given a target where ABI requires types to be aligned on their size (1 =>
1, 2 => 2, 4 => 4, 8 => 8, etc.)
4 + 4 + 8 + 8 + 4 = 28 bytes
To satisfy __be64 alignement (the strictest), it must be aligned on a 8
bytes boundary, but 28 is not multiple of 8. 32 is. Padding is required.
[The padding is required so that in a array of struct nbd_request[],
all members of all structures are aligned to their requirement alignement]
> but its reply counter part
>
> struct nbd_reply {
> __be32 magic;
> __be32 error; /* 0 = ok, else error */
> char handle[8]; /* handle you got from request */
> };
>
4 + 4 + 8 = 16 bytes.
To satisfy __be32 alignement (the strictest), it must be aligned on a 4
bytes boundary and 16 is a multiple of 4: no need for padding.
> is not. Since Linux seems to read sizeof(nbd_reply) bytes from the
> network (see [1]), the number of bytes read varies with the number of
> bytes of the structure.
It won't.
>
> So my understanding is that if the size of struct nbd_reply varies from
> platform/compiler to another that means trouble. I wonder:
>
> - Is there anything about struct nbd_reply that would keep its size
> equal everywhere or a reason why variance in size is no problem here?
>
> - Any ideas for a platform where you would expect struct nbd_reply to
> be other than 4 + 4 + 8 = 16 bytes in size?
>
Look for an ABI that require 'char' with either property:
- size > 1 byte
- alignment > 8 (16 bytes !)
Regards.
--
Yann Droneaud
OPTEYA
^ permalink raw reply [flat|nested] 5+ messages in thread* struct nbd_reply not packed => trouble?
2013-01-10 16:36 ` Yann Droneaud
@ 2013-01-13 16:45 ` Sebastian Pipping
2013-01-14 23:59 ` 卜弋天
1 sibling, 0 replies; 5+ messages in thread
From: Sebastian Pipping @ 2013-01-13 16:45 UTC (permalink / raw)
To: kernelnewbies
I see, thank you!
Best,
Sebastian
^ permalink raw reply [flat|nested] 5+ messages in thread
* struct nbd_reply not packed => trouble?
2013-01-10 16:36 ` Yann Droneaud
2013-01-13 16:45 ` Sebastian Pipping
@ 2013-01-14 23:59 ` 卜弋天
2013-01-15 9:59 ` Yann Droneaud
1 sibling, 1 reply; 5+ messages in thread
From: 卜弋天 @ 2013-01-14 23:59 UTC (permalink / raw)
To: kernelnewbies
? 2013-1-11?0:42?"Yann Droneaud" <ydroneaud@opteya.com> ???
>> Hello kernel hackers,
>>
>>
>> I noticed that in [linux]/include/uapi/linux/nbd.h the structure
>>
>> struct nbd_request {
>> __be32 magic;
>> __be32 type; /* == READ || == WRITE */
>> char handle[8];
>> __be64 from;
>> __be32 len;
>> } __attribute__((packed));
>>
>> is packed
>
> Given a target where ABI requires types to be aligned on their size (1 =>
> 1, 2 => 2, 4 => 4, 8 => 8, etc.)
>
>
> 4 + 4 + 8 + 8 + 4 = 28 bytes
>
> To satisfy __be64 alignement (the strictest), it must be aligned on a 8
> bytes boundary, but 28 is not multiple of 8. 32 is. Padding is required.
>
> [The padding is required so that in a array of struct nbd_request[],
> all members of all structures are aligned to their requirement alignement]
>
>
>> but its reply counter part
>>
>> struct nbd_reply {
>> __be32 magic;
>> __be32 error; /* 0 = ok, else error */
>> char handle[8]; /* handle you got from request */
>> };
>
> 4 + 4 + 8 = 16 bytes.
> To satisfy __be32 alignement (the strictest), it must be aligned on a 4
> bytes boundary and 16 is a multiple of 4: no need for padding.
>
>> is not. Since Linux seems to read sizeof(nbd_reply) bytes from the
>> network (see [1]), the number of bytes read varies with the number of
>> bytes of the structure.
>
>
> It won't.
>
>>
>> So my understanding is that if the size of struct nbd_reply varies from
>> platform/compiler to another that means trouble. I wonder:
>>
>> - Is there anything about struct nbd_reply that would keep its size
>> equal everywhere or a reason why variance in size is no problem here?
>>
>> - Any ideas for a platform where you would expect struct nbd_reply to
>> be other than 4 + 4 + 8 = 16 bytes in size?
>
> Look for an ABI that require 'char' with either property:
> - size > 1 byte
> - alignment > 8 (16 bytes !)
>
>
Do you mean this structure size will vary on different platform? If yes, why Linux does not packed here? Thanks
> Regards.
>
> --
> Yann Droneaud
> OPTEYA
>
>
>
> _______________________________________________
> Kernelnewbies mailing list
> Kernelnewbies at kernelnewbies.org
> http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies
^ permalink raw reply [flat|nested] 5+ messages in thread* struct nbd_reply not packed => trouble?
2013-01-14 23:59 ` 卜弋天
@ 2013-01-15 9:59 ` Yann Droneaud
0 siblings, 0 replies; 5+ messages in thread
From: Yann Droneaud @ 2013-01-15 9:59 UTC (permalink / raw)
To: kernelnewbies
Hi,
Le mardi 15 janvier 2013 ? 07:59 +0800, ??? a ?crit :
>
> ? 2013-1-11?0:42?"Yann Droneaud" <ydroneaud@opteya.com> ???
> >> - Any ideas for a platform where you would expect struct nbd_reply to
> >> be other than 4 + 4 + 8 = 16 bytes in size?
> >
> > Look for an ABI that require 'char' with either property:
> > - size > 1 byte
> > - alignment > 8 (16 bytes !)
> >
> >
>
> Do you mean this structure size will vary on different platform? If yes, why Linux does not packed here? Thanks
>
There's probably no such architecture for a general purpose processor
with the ABI properties I've mentionned. eg. (sizeof(char) > 1 ||
alignof(char) > 8).
And if one day Linux is going to be ported on such target, the packing
of the nbd_reply struct will be the smallest problem to fix.
Regards
--
Yann Droneaud
OPTEYA
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2013-01-15 9:59 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2013-01-10 16:09 struct nbd_reply not packed => trouble? Sebastian Pipping
2013-01-10 16:36 ` Yann Droneaud
2013-01-13 16:45 ` Sebastian Pipping
2013-01-14 23:59 ` 卜弋天
2013-01-15 9:59 ` Yann Droneaud
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).