[PATCH 01/14] nvmet: for pt I/O cmds annotate req->sg_cnt likely

[Chaitanya Kulkarni]

On 8/10/20 14:54, Logan Gunthorpe wrote:
> Hey,
> 
> You forgot to CC me again....
> 
Ugghhh..
> On 2020-08-10 12:54 p.m., Chaitanya Kulkarni wrote:
>> I/O Commands (nvme_cmd_read, nvme_cmd_write) are most command commands
>> when accessing passthru controller. For I/O commands data is represented
>> in SG list, so sg_cnt != 0 in most cases.
>>
>> Annotate req->sg_cnt check with likely.
> Does adding these likely annotations actually produce any benefit? Why
> can't we rely on the branch predictor in these instances?
> 

They are present in the target code. I don't see any problem having it 
in a passthru code unless it shows severe performance degradation.

> I vaguely remember somebody complaining about all the extra likelys and
> unlikelys which is why I removed them.
> 
> Logan
>