On 14/08/2013 12:06, Or Gerlitz wrote:
On 13/08/2013 22:18, Roland Dreier wrote:
>+struct ib_uverbs_cmd_hdr_ex {
>+ __u32 command;
>+ __u16 in_words;
>+ __u16 out_words;
>+ __u16 provider_in_words;
>+ __u16 provider_out_words;
>+ __u32 cmd_hdr_reserved;
>+};
>+
If I understand the vague explanations and the analogy to the
presentation about userspace, then cmd_hdr_reserved is supposed to be
used as a mask field. Should a kernel that doesn't understand any
mask components make sure that this reserved field is 0, and return an
error if it isn't? I don't see any code to do that, and it seems to
risk new userspace silently getting wrong answers on an old kernel.
Is there any reason not to name it as a mask field from the start?
NO.
The last field in the extended uverbs command header isn't for
extending the command for which this instance serves as the header.
Its role is to make sure we are aligned to intergral multiplicative of
64bit as the non extended header.
--
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/majordomo-info.html
