Re: [RFT 00/13] constify s390 ccw_device_id

From: Heiko Carstens
Date: Tue Aug 15 2017 - 07:30:56 EST


On Tue, Aug 15, 2017 at 04:05:31PM +0530, Arvind Yadav wrote:
> ccw_device_id are not supposed to change at runtime. All functions
> working with ccw_device_id provided by <asm/ccwdev.h> work with
> const ccw_device_id. So mark ccw_device_id member of struct ccw driver
> as const. And change all driver relatively.
>
> Arvind Yadav (13):
> [RFT 01/13] s390: ccwdev: constify ccw_device_id
> [RFT 02/13] s390: dasd_eckd: constify ccw_device_id
> [RFT 03/13] s390: dasd_fba: constify ccw_device_id
> [RFT 04/13] s390: con3215: constify ccw_device_id
> [RFT 05/13] s390: raw3270: constify ccw_device_id
> [RFT 06/13] s390: tape_34xx: constify ccw_device_id
> [RFT 07/13] s390: tape_3590: constify ccw_device_id
> [RFT 08/13] s390: vmur: constify ccw_device_id
> [RFT 09/13] s390: ctcm: constify ccw_device_id
> [RFT 10/13] s390: lcs: constify ccw_device_id
> [RFT 11/13] s390: qeth: constify ccw_device_id
> [RFT 12/13] s390: zfcp: constify ccw_device_id
> [RFT 13/13] s390: virtio: constify ccw_device_id

Did the constify work reveal a single bug until now? This causes a lot of
code churn, and I'm failing to see an added value here.
In addition this is not worth splitting it into 13(!) patches.