Re: [PATCH 0/4] lightnvm: pblk: add support for chunk metadata on erase

From: Javier Gonzalez <javier@cnexlabs.com>
To: "Matias Bjørling" <mb@lightnvm.io>
Cc: "linux-block@vger.kernel.org" <linux-block@vger.kernel.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	Hans Holmberg <hans.holmberg@cnexlabs.com>
Subject: Re: [PATCH 0/4] lightnvm: pblk: add support for chunk metadata on erase
Date: Mon, 3 Sep 2018 09:16:53 +0000	[thread overview]
Message-ID: <7FBEB635-C943-484D-A970-D3BF870BAFFB@cnexlabs.com> (raw)
In-Reply-To: <71f5f8b3-66d5-a0e0-85ab-cb22e5ac621c@lightnvm.io>

[-- Attachment #1: Type: text/plain, Size: 1804 bytes --]

> On 31 Aug 2018, at 15.57, Matias Bjørling <mb@lightnvm.io> wrote:
> 
> On 08/31/2018 03:34 PM, Javier González wrote:
>> Matias,
>> This patchset implements support for retrieving chunk metadata when
>> submitting a reset/erase command. Patches 0 and 1 are small fixes that
>> can be directly merged into your patch:
>>      lightnvm: move bad block and chunk state logic to core
>> Also, note that these do not apply on top of your for-4.20/core due them
>> depending on patches that I sent before that you have not picked up yet.
>> You can see them though in for-4.20/pblk. I'll rebase as patches in the
>> list appear in your tree.
> 
> Thanks. It is really confusing when you guys maintains an implicit order and posts the patches separately. I will appreciate that patches that are related are posted together, such that I don't have to manually track what comes before another. That makes it less of a pain for me to keep track of and we can keep the reviews together.
> 
> This is the patches that I have in the pipeline (from before the e-mails from today):
> 
>  - This serie - Pending review
>  - Serie: pblk: support variable OOB size - Waiting on review from Igor
>  - lightnvm: pblk: recover open lines on 2.0 devices. Which doesn't apply due to the fixes to the pad distance patch.
> 

Yes, I know and I apologize - we should have a better flow. What do you
say that for windows like this, where we have a number of patches that
have dependencies that we post them in meaningful patchsets and point to
a branch where they are ordered, like in a PR? Then we can rebase and
propagate changes properly?

For this window, I'll rebase the rest of the patches in for-4.20/pblk on
top of your for-4.20/core, then we can propagate changes as they come.

Javier

[-- Attachment #2: Message signed with OpenPGP --]
[-- Type: application/pgp-signature, Size: 833 bytes --]