The new field has one of three values - IBV_LINK_LAYER_UNSPECIFIED,
IBV_LINK_LAYER_INFINIBAND, IBV_LINK_LAYER_ETHERNET. It can be used
by applications to know the link layer used by the port, which
can be either Infiniband or Ethernet.
The addition of the new field does not change the size of stru
On Tue, Jul 19, 2011 at 12:27:17PM +0300, Or Gerlitz wrote:
> The addition of the new field does not change the size of struct
> ibv_port_attr due to alignment of the preceding fields. As such
> binary compatibility between the library to applications is kept,
What zeros the field in the current
Jason Gunthorpe wrote:
> What zeros the field in the current library?
the kernel uverbs code either zeros it (old kernels) or sets it (new kernels)
Or.
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majord...@vger.kernel.org
More majordomo
On Tue, Jul 19, 2011 at 10:36:26PM +0300, Or Gerlitz wrote:
> > What zeros the field in the current library?
>
> the kernel uverbs code either zeros it (old kernels) or sets it (new kernels)
ibv_cmd_query_port does a field by field copy of the port_attr, so
what the kernel does is not relevant to
Jason Gunthorpe wrote:
> Or Gerlitz wrote:
>>> What zeros the field in the current library?
+ port_attr->link_layer = IBV_LINK_LAYER_UNSPECIFIED;
+ port_attr->pad = 0;
Or
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majord...@v
On Tue, Jul 19, 2011 at 11:22:51PM +0300, Or Gerlitz wrote:
> Jason Gunthorpe wrote:
> > Or Gerlitz wrote:
>
> >>> What zeros the field in the current library?
>
> + port_attr->link_layer = IBV_LINK_LAYER_UNSPECIFIED;
> + port_attr->pad = 0;
Where are you getting that hunk from?
li
> More clearly: what initializes the memory you've redefined to be
> link_layer in libibverbs 1.1.5 ? I see nothing.
Why does the value in an older library cause an issue?
Or, how will this library version interact with an older kernel?
--
To unsubscribe from this list: send the line "unsubscribe
On Tue, Jul 19, 2011 at 09:20:25PM +, Hefty, Sean wrote:
> > More clearly: what initializes the memory you've redefined to be
> > link_layer in libibverbs 1.1.5 ? I see nothing.
>
> Why does the value in an older library cause an issue?
ABI compatability means if dynamic linking succeeds then
Jason Gunthorpe wrote:
> The commit comment states the goal is to have new apps see
> IBV_LINK_LAYER_UNSPECIFIED for all possible combinations of old stuff
> underneath.
> [...] The problem is with the userspace ABI out of libibverbs. Current
> libibverbs does not zero its ib_port_attr structur
On Wed, Jul 20, 2011 at 09:08:26PM +0300, Or Gerlitz wrote:
> Jason Gunthorpe wrote:
> +#define ibv_query_port(context, port_num, port_attr) \
> + ___ibv_query_port(context, port_num, port_attr)
> +
>
> So we are okay now, correct?
Looks like! Sorry for the confusion
Jason
--
To unsubscr
Jason Gunthorpe wrote:
> Or Gerlitz wrote:
>> Jason Gunthorpe wrote:
>> +#define ibv_query_port(context, port_num, port_attr) \
>> + ___ibv_query_port(context, port_num, port_attr)
>> +
>> So we are okay now, correct?
> Looks like! Sorry for the confusion
Jason, thanks for your detailed
> Sean, any comments/questions?
I was most concerned that this worked with an older kernel, but Jason answered
that.
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majord
On Wed, Jul 20, 2011 at 09:20:41PM +0300, Or Gerlitz wrote:
> Jason Gunthorpe wrote:
> > Looks like! Sorry for the confusion
>
> Jason, thanks for your detailed review && feedback, just to recap,
> the patches are okay except the missing entry for the new GID change
> event in ibv_event_type_str
Jason Gunthorpe wrote:
> Yes, that was the only important thing I saw.
> I assume Roland was happy with the #define trick for compatibility
> last time it came up?
yes, that's correct.
> Looking at the diff, maybe re-shuffle the order of the #define trick
> so it is easier to follow, and call i
14 matches
Mail list logo