linux-kernel - Re: [PATCH RFC v2 01/24] scsi: add 'nr_reserved

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Wed, 11 Mar 2020 07:55:46 +0100
From:   Hannes Reinecke <hare@...e.de>
To:     Ming Lei <ming.lei@...hat.com>, John Garry <john.garry@...wei.com>
Cc:     axboe@...nel.dk, jejb@...ux.ibm.com, martin.petersen@...cle.com,
        bvanassche@....org, hch@...radead.org, linux-block@...r.kernel.org,
        linux-kernel@...r.kernel.org, linux-scsi@...r.kernel.org,
        virtualization@...ts.linux-foundation.org,
        esc.storagedev@...rosemi.com, chenxiang66@...ilicon.com,
        Hannes Reinecke <hare@...e.com>
Subject: Re: [PATCH RFC v2 01/24] scsi: add 'nr_reserved_cmds' field to the
 SCSI host template

On 3/11/20 12:08 AM, Ming Lei wrote:
> On Wed, Mar 11, 2020 at 12:25:27AM +0800, John Garry wrote:
>> From: Hannes Reinecke <hare@...e.com>
>>
>> Add a new field 'nr_reserved_cmds' to the SCSI host template to
>> instruct the block layer to set aside a tag space for reserved
>> commands.
>>
>> Signed-off-by: Hannes Reinecke <hare@...e.com>
>> ---
>>  drivers/scsi/scsi_lib.c  | 1 +
>>  include/scsi/scsi_host.h | 6 ++++++
>>  2 files changed, 7 insertions(+)
>>
>> diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
>> index 610ee41fa54c..2967325df7a0 100644
>> --- a/drivers/scsi/scsi_lib.c
>> +++ b/drivers/scsi/scsi_lib.c
>> @@ -1896,6 +1896,7 @@ int scsi_mq_setup_tags(struct Scsi_Host *shost)
>>  		shost->tag_set.ops = &scsi_mq_ops_no_commit;
>>  	shost->tag_set.nr_hw_queues = shost->nr_hw_queues ? : 1;
>>  	shost->tag_set.queue_depth = shost->can_queue;
>> +	shost->tag_set.reserved_tags = shost->nr_reserved_cmds;
> 
> You reserve tags for special usage, meantime the whole queue depth
> isn't increased, that means the tags for IO request is decreased given
> reserved tags can't be used for IO, so IO performance may be effected.
> 
> If not the case, please explain a bit in commit log.
> 
The overall idea of this patchset is to fold _existing_ management
command handling into using the blk-mq bitmap.
Currently, quite a lot of drivers are using management commands
internally, which typically use the same hardware tag pool (ie they are
being allocated from the same hardware resources) than the 'normal' I/O
commands.
But as they are using the same tagpool these drivers already decrement
the available number of commands; check eg. hpsa:

static int hpsa_scsi_host_alloc(struct ctlr_info *h)
{
	struct Scsi_Host *sh;

	sh = scsi_host_alloc(&hpsa_driver_template, sizeof(h));
	if (sh == NULL) {
		dev_err(&h->pdev->dev, "scsi_host_alloc failed\n");
		return -ENOMEM;
	}

	sh->io_port = 0;
	sh->n_io_port = 0;
	sh->this_id = -1;
	sh->max_channel = 3;
	sh->max_cmd_len = MAX_COMMAND_SIZE;
	sh->max_lun = HPSA_MAX_LUN;
	sh->max_id = HPSA_MAX_LUN;
	sh->can_queue = h->nr_cmds - HPSA_NRESERVED_CMDS;
	sh->cmd_per_lun = sh->can_queue;

So the idea of this patchset is to convert existing use-cases; seeing
that they already reserve commands internally performance will not be
affected.

Cheers,

Hannes
-- 
Dr. Hannes Reinecke		           Kernel Storage Architect
hare@...e.de			                  +49 911 74053 688
SUSE Software Solutions Germany GmbH, Maxfeldstr. 5, 90409 Nürnberg
HRB 36809 (AG Nürnberg), GF: Felix Imendörffer