On 4/27/23 4:59?PM, Chaitanya Kulkarni wrote: > On 4/27/23 12:47, Christophe JAILLET wrote: >> Group some variables based on their sizes to reduce holes. >> On x86_64, this shrinks the size of 'struct nvmet_ns' from 520 to 512 >> bytes. >> > > Although this looks good, we at least need to document what > happens on other arch(s) which are not mentioned in the > commit log ? is there a possibility that someone might come > up with the contradictory data in future for the arch(s) which > arch that are not mentioned here ? The change is certainly not going to make things _worse_ for any arch, so I think that's somewhat of a pointless exercise and an unreasonable ask for something that makes sense on 64-bit arm/x86 and saves half the space. -- Jens Axboe
