All of lore.kernel.org
 help / color / mirror / Atom feed
From: Bart Van Assche <bvanassche@acm.org>
To: James Bottomley <James.Bottomley@HansenPartnership.com>,
	"Martin K . Petersen" <martin.petersen@oracle.com>
Cc: linux-scsi@vger.kernel.org
Subject: Re: [PATCH v2 00/50] Remove the request pointer from struct scsi_cmnd
Date: Tue, 18 May 2021 14:21:28 -0700	[thread overview]
Message-ID: <f3dae885-5826-11da-0eec-04f0a3e04cd9@acm.org> (raw)
In-Reply-To: <af39cb904e0f0450549f9fdcee3c256e61bfab93.camel@HansenPartnership.com>

On 5/18/21 10:55 AM, James Bottomley wrote:
> On Tue, 2021-05-18 at 10:44 -0700, Bart Van Assche wrote:
>> Hi Martin,
>>
>> This patch series implements the following two changes for all SCSI
>> drivers:
>> - Use blk_mq_rq_from_pdu() instead of the request member of struct
>> scsi_cmnd
>>   since adding an offset to a pointer is faster than pointer
>> indirection.
> 
> Are there any performance results to back up this assertion?  It's
> quite a lot of churn so it would be nice to know it's worth it.

I have not yet run any performance measurements because I expect that it
will be challenging to measure the performance impact of a change like
this one accurately. The performance measurement tool itself (e.g. fio)
might introduce more variation between runs than the performance
improvement of this patch series. Another reason I have not yet run any
performance measurements is because I was assuming that everyone would
be happy with a patch series that makes code faster and that reduces the
size of a key SCSI data structure.

Anyway, I have run 'make drivers/scsi/scsi_lib.lst' with and without
this patch series applied. What I see is that without this patch series
the assembly code for converting a SCSI command pointer into a request
pointer looks like this:

48 8b bb 10 01 00 00    mov    0x110(%rbx),%rdi

With this patch series applied that conversion code changes into the
following:

48 8d bb f0 fe ff ff    lea    -0x110(%rbx),%rdi

The above shows that struct request has a size of 0x110 = 272 bytes with
my kernel configuration.

This illustrates that this patch series realizes an improvement since
"mov" instructions used for converting SCSI command pointers into struct
request pointers are converted into "lea" instructions. "mov" fetches
data from memory while "lea" does not.

Bart.

      reply	other threads:[~2021-05-18 21:21 UTC|newest]

Thread overview: 60+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-05-18 17:44 [PATCH v2 00/50] Remove the request pointer from struct scsi_cmnd Bart Van Assche
2021-05-18 17:44 ` [PATCH v2 01/50] core: Introduce the scsi_cmd_to_rq() function Bart Van Assche
2021-05-18 17:44 ` [PATCH v2 02/50] core: Use scsi_cmd_to_rq() instead of scsi_cmnd.request Bart Van Assche
2021-05-18 20:01   ` Douglas Gilbert
2021-05-18 21:59     ` Bart Van Assche
2021-05-18 17:44 ` [PATCH v2 03/50] sd: " Bart Van Assche
2021-05-18 17:44 ` [PATCH v2 04/50] sr: " Bart Van Assche
2021-05-18 17:44 ` [PATCH v2 05/50] scsi_transport_fc: " Bart Van Assche
2021-05-18 17:44 ` [PATCH v2 06/50] scsi_transport_spi: " Bart Van Assche
2021-05-18 17:44 ` [PATCH v2 07/50] ata: " Bart Van Assche
2021-05-18 17:44 ` [PATCH v2 08/50] RDMA/iser: " Bart Van Assche
2021-05-18 17:58   ` Sagi Grimberg
2021-05-18 17:44 ` [PATCH v2 09/50] RDMA/srp: " Bart Van Assche
2021-05-18 17:44 ` [PATCH v2 10/50] zfcp: " Bart Van Assche
2021-05-18 17:44 ` [PATCH v2 11/50] 53c700: " Bart Van Assche
2021-05-18 17:44 ` [PATCH v2 12/50] NCR5380: " Bart Van Assche
2021-05-18 17:44 ` [PATCH v2 13/50] aacraid: " Bart Van Assche
2021-05-18 17:44 ` [PATCH v2 14/50] advansys: " Bart Van Assche
2021-05-18 17:44 ` [PATCH v2 15/50] bnx2i: " Bart Van Assche
2021-05-18 17:44 ` [PATCH v2 16/50] csiostor: " Bart Van Assche
2021-05-18 17:44 ` [PATCH v2 17/50] cxlflash: " Bart Van Assche
2021-05-18 17:44 ` [PATCH v2 18/50] dpt_i2o: " Bart Van Assche
2021-05-18 17:44 ` [PATCH v2 19/50] fnic: " Bart Van Assche
2021-05-18 17:44 ` [PATCH v2 20/50] hisi_sas: " Bart Van Assche
2021-05-18 17:44 ` [PATCH v2 21/50] hpsa: " Bart Van Assche
2021-05-18 17:44 ` [PATCH v2 22/50] ibmvfc: " Bart Van Assche
2021-05-18 17:44 ` [PATCH v2 23/50] ibmvscsi: " Bart Van Assche
2021-05-18 17:44 ` [PATCH v2 24/50] ips: " Bart Van Assche
2021-05-18 17:44 ` [PATCH v2 25/50] libsas: " Bart Van Assche
2021-05-19 10:12   ` John Garry
2021-05-18 17:44 ` [PATCH v2 26/50] lpfc: " Bart Van Assche
2021-05-18 17:44 ` [PATCH v2 27/50] megaraid: " Bart Van Assche
2021-05-19  7:43   ` Sumit Saxena
2021-05-18 17:44 ` [PATCH v2 28/50] mpt3sas: " Bart Van Assche
2021-05-18 17:44 ` [PATCH v2 29/50] mvumi: " Bart Van Assche
2021-05-18 17:44 ` [PATCH v2 30/50] myrb: " Bart Van Assche
2021-05-18 17:44 ` [PATCH v2 31/50] myrs: " Bart Van Assche
2021-05-18 17:44 ` [PATCH v2 32/50] ncr53c8xx: " Bart Van Assche
2021-05-18 17:44 ` [PATCH v2 33/50] qedf: " Bart Van Assche
2021-05-18 17:44 ` [PATCH v2 34/50] qedi: " Bart Van Assche
2021-05-18 17:44 ` [PATCH v2 35/50] qla1280: " Bart Van Assche
2021-05-18 17:44 ` [PATCH v2 36/50] qla2xxx: " Bart Van Assche
2021-05-18 17:44 ` [PATCH v2 37/50] qla4xxx: " Bart Van Assche
2021-05-18 17:44 ` [PATCH v2 38/50] qlogicpti: " Bart Van Assche
2021-05-18 17:44 ` [PATCH v2 39/50] scsi_debug: " Bart Van Assche
2021-05-18 17:44 ` [PATCH v2 40/50] smartpqi: " Bart Van Assche
2021-05-18 17:44 ` [PATCH v2 41/50] snic: " Bart Van Assche
2021-05-18 17:44 ` [PATCH v2 42/50] stex: " Bart Van Assche
2021-05-18 17:44 ` [PATCH v2 43/50] sun3_scsi: " Bart Van Assche
2021-05-18 17:44 ` [PATCH v2 44/50] sym53c8xx: " Bart Van Assche
2021-05-18 17:44 ` [PATCH v2 45/50] ufs: " Bart Van Assche
2021-05-18 17:44 ` [PATCH v2 46/50] virtio_scsi: " Bart Van Assche
2021-05-18 17:44 ` [PATCH v2 47/50] xen-scsifront: " Bart Van Assche
2021-05-19  9:20   ` Juergen Gross
2021-05-18 17:44 ` [PATCH v2 48/50] tcm_loop: " Bart Van Assche
2021-05-19 18:14   ` Bodo Stroesser
2021-05-18 17:44 ` [PATCH v2 49/50] usb-storage: " Bart Van Assche
2021-05-18 17:44 ` [PATCH v2 50/50] core: Remove the request member from struct scsi_cmnd Bart Van Assche
2021-05-18 17:55 ` [PATCH v2 00/50] Remove the request pointer " James Bottomley
2021-05-18 21:21   ` Bart Van Assche [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=f3dae885-5826-11da-0eec-04f0a3e04cd9@acm.org \
    --to=bvanassche@acm.org \
    --cc=James.Bottomley@HansenPartnership.com \
    --cc=linux-scsi@vger.kernel.org \
    --cc=martin.petersen@oracle.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body. 
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.