Re: [PATCH v7 00/19] Add Command Duration Limits support

From: Damien Le Moal <dlemoal@kernel.org>
To: Niklas Cassel <Niklas.Cassel@wdc.com>,
	"Martin K. Petersen" <martin.petersen@oracle.com>
Cc: Niklas Cassel <nks@flawful.org>, Jens Axboe <axboe@kernel.dk>,
	"James E.J. Bottomley" <jejb@linux.ibm.com>,
	Bart Van Assche <bvanassche@acm.org>,
	Christoph Hellwig <hch@lst.de>, Hannes Reinecke <hare@suse.de>,
	"linux-scsi@vger.kernel.org" <linux-scsi@vger.kernel.org>,
	"linux-ide@vger.kernel.org" <linux-ide@vger.kernel.org>,
	"linux-block@vger.kernel.org" <linux-block@vger.kernel.org>
Subject: Re: [PATCH v7 00/19] Add Command Duration Limits support
Date: Tue, 23 May 2023 19:08:27 +0900	[thread overview]
Message-ID: <09f5d62b-1bd4-3a25-e178-2225f1c7b603@kernel.org> (raw)
In-Reply-To: <ZGyN1KkCXsTo8ZwG@x1-carbon>

On 5/23/23 18:56, Niklas Cassel wrote:
> On Mon, May 22, 2023 at 05:41:19PM -0400, Martin K. Petersen wrote:
>>
>> Niklas,
>>
>>> This series adds support for Command Duration Limits.
>>
>> Applied to 6.5/scsi-staging, thanks!
> 
> Thank you Martin!
> 
> 
> Damien, Martin,
> considering that the libata changes depend on the scsi changes,
> and considering that further libata EH cleanups are planned for
> 6.5 now when the IPR driver is gone, I think that the best move
> is to follow the advice of:
> https://docs.kernel.org/maintainer/rebasing-and-merging.html#merging-from-sibling-or-upstream-trees

Hannes cleanup of EH will create a conflict with the scsi tree but can go in
through the ata tree independently so I was not planning on doing a rebase,
especially not on the scsi tree. I will notify Stephen about the conflict send
him a resolution to apply and carry for linux-next. When the 6.5 merge window
open, I will wait for the James to send the scsi PR and send my PR to Linus
after that with the conflict resolution, as usual.

So far, I do not see any big issue with that.

> 
> Specifically:
> "Merging another subsystem tree to resolve a dependency risks bringing in
> other bugs and should almost never be done. If that subsystem tree fails
> to be pulled upstream, whatever problems it had will block the merging of
> your tree as well.
> Preferable alternatives include agreeing with the maintainer to carry both
> sets of changes in one of the trees or creating a topic branch dedicated
> to the prerequisite commits that can be merged into both trees."
> 
> 
> 
> Martin created a topic branch/SHA1 for the CDL series:
> 18bd7718b5c489b3161b6c2ab4685d57c1e2da3b
> in order for him to be able to have a nice merge commit:
> https://git.kernel.org/pub/scm/linux/kernel/git/mkp/scsi.git/commit/?h=6.5/scsi-staging&id=8b60e2189fcd8b10b592608256eb97aebfcff147
> 
> So, I suggest that, after this has been applied to
> 6.5/scsi-queue (right now it is only in 6.5/scsi-staging),
> that Damien merges the same topic branch/SHA1:
> 18bd7718b5c489b3161b6c2ab4685d57c1e2da3b
> to libata/for-6.5.
> 
> Perhaps the fix:
> https://lore.kernel.org/linux-scsi/20230523074701.293502-1-dlemoal@kernel.org/T/#u
> could be applied on top of that SHA1, or folded in,
> the important thing is that libata merges the exact same SHA1
> for the CDL series as scsi-queue.
> (Especially since I noticed that Martin did some minor changes to
> the ioprio hints patch, namely changed IO to I/O in the comments
> describing the macros, so Damien can't just take the patches from
> the list as is, as that would create conflicts for Linus when he
> merges the two different subsystem trees.)
> 
> 
> Kind regards,
> Niklas

-- 
Damien Le Moal
Western Digital Research