[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <2491bb34-2098-32f5-3d5b-e3d89fbce4b3@bjorling.me>
Date: Fri, 5 Jan 2018 19:33:36 +0100
From: Matias Bjørling <m@...rling.me>
To: Jens Axboe <axboe@...nel.dk>
Cc: linux-block@...r.kernel.org, linux-kernel@...r.kernel.org,
Javier González <javier@...xlabs.com>
Subject: Re: [GIT PULL 24/25] lightnvm: pblk: add iostat support
On 01/05/2018 04:42 PM, Jens Axboe wrote:
> On Fri, Jan 05 2018, Matias Bjørling wrote:
>> From: Javier González <javier@...xlabs.com>
>>
>> Since pblk registers its own block device, the iostat accounting is
>> not automatically done for us. Therefore, add the necessary
>> accounting logic to satisfy the iostat interface.
>
> Ignorant question - why is it a raw block device, not using blk-mq?
The current flow is using the raw block device, together with the blk-mq
nvme device driver. A bio is sent down to the nvme_nvm_submit_io() path
in the /drivers/nvme/host/lightnvm.c file. From there it attaches the to
NVMe blk-mq implementation.
Is there a better way to do it?
>
>> @@ -193,9 +197,9 @@ static void pblk_end_io_read(struct nvm_rq *rqd)
>> __pblk_end_io_read(pblk, rqd, true);
>> }
>>
>> -static int pblk_fill_partial_read_bio(struct pblk *pblk, struct nvm_rq *rqd,
>> - unsigned int bio_init_idx,
>> - unsigned long *read_bitmap)
>> +static int pblk_partial_read_bio(struct pblk *pblk, struct nvm_rq *rqd,
>> + unsigned int bio_init_idx,
>> + unsigned long *read_bitmap)
>> {
>> struct bio *new_bio, *bio = rqd->bio;
>> struct pblk_sec_meta *meta_list = rqd->meta_list;
>> @@ -306,6 +310,8 @@ static int pblk_fill_partial_read_bio(struct pblk *pblk, struct nvm_rq *rqd,
>> return NVM_IO_OK;
>>
>> err:
>> + pr_err("pblk: failed to perform partial read\n");
>> +
>> /* Free allocated pages in new bio */
>> pblk_bio_free_pages(pblk, bio, 0, new_bio->bi_vcnt);
>> __pblk_end_io_read(pblk, rqd, false);
>
> This seems to include unrelated changes, like the rename above and the
> addition of the error logging?
>
Grah... I missed it during review. It should have been its own patch or
part of the early refactor patches. Thanks for picking it up.
Powered by blists - more mailing lists