All of lore.kernel.org
 help / color / mirror / Atom feed
From: Sagi Grimberg <sagig@dev.mellanox.co.il>
To: Steve Wise <swise@opengridcomputing.com>, dledford@redhat.com
Cc: infinipath@intel.com, sagig@mellanox.com, ogerlitz@mellanox.com,
	roid@mellanox.com, linux-rdma@vger.kernel.org, eli@mellanox.com,
	target-devel@vger.kernel.org
Subject: Re: [PATCH V5 3/5] RDMA/iser: Limit sg tablesize and max_sectors to device fastreg max depth
Date: Tue, 7 Jul 2015 09:26:53 +0300	[thread overview]
Message-ID: <559B712D.7020105@dev.mellanox.co.il> (raw)
In-Reply-To: <001401d0b7f9$09c21b40$1d4651c0$@opengridcomputing.com>

On 7/6/2015 5:35 PM, Steve Wise wrote:
>
>
>> -----Original Message-----
>> From: Sagi Grimberg [mailto:sagig@dev.mellanox.co.il]
>> Sent: Monday, July 06, 2015 2:51 AM
>> To: Steve Wise; dledford@redhat.com
>> Cc: infinipath@intel.com; sagig@mellanox.com; ogerlitz@mellanox.com; roid@mellanox.com; linux-rdma@vger.kernel.org;
>> eli@mellanox.com; target-devel@vger.kernel.org
>> Subject: Re: [PATCH V5 3/5] RDMA/iser: Limit sg tablesize and max_sectors to device fastreg max depth
>>
>> On 7/5/2015 8:44 PM, Steve Wise wrote:
>>> Currently the sg tablesize, which dictates fast register page list
>>> depth to use, does not take into account the limits of the rdma device.
>>> So adjust it once we discover the device fastreg max depth limit.  Also
>>> adjust the max_sectors based on the resulting sg tablesize.
>>>
>>> Signed-off-by: Steve Wise <swise@opengridcomputing.com>
>>> ---
>>>
>>>    drivers/infiniband/ulp/iser/iscsi_iser.c |    9 +++++++++
>>>    1 files changed, 9 insertions(+), 0 deletions(-)
>>>
>>> diff --git a/drivers/infiniband/ulp/iser/iscsi_iser.c b/drivers/infiniband/ulp/iser/iscsi_iser.c
>>> index 6a594aa..de8730d 100644
>>> --- a/drivers/infiniband/ulp/iser/iscsi_iser.c
>>> +++ b/drivers/infiniband/ulp/iser/iscsi_iser.c
>>> @@ -640,6 +640,15 @@ iscsi_iser_session_create(struct iscsi_endpoint *ep,
>>>    						   SHOST_DIX_GUARD_CRC);
>>>    		}
>>>
>>> +		/*
>>> +		 * Limit the sg_tablesize and max_sectors based on the device
>>> +		 * max fastreg page list length.
>>> +		 */
>>> +		shost->sg_tablesize = min_t(unsigned short, shost->sg_tablesize,
>>> +			ib_conn->device->dev_attr.max_fast_reg_page_list_len);
>>> +		shost->max_sectors = min_t(unsigned int,
>>> +			1024, (shost->sg_tablesize * PAGE_SIZE) >> 9);
>>> +
>>
>> The min statement is meaningless for max_sectors - you do a min between
>> default sg_tablesize and frpl length - so the maximum sg_tablesize is
>> 128 which is 1024 max_sectors.
>
> I'm not following.  What if ib_conn->device->dev_attr.max_fast_reg_page_list_len is say, 32?
> Then shost->sg_tablesize is set to 32, and max_sectors is set to (32*4K) >> 9 == 256 512B sectors.

Correct - but it cannot exceed 1024 (as it is derived from sg_tablesize
which is maximum 128).

  reply	other threads:[~2015-07-07  6:26 UTC|newest]

Thread overview: 23+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-07-05 17:44 [PATCH V5 0/5] iSER support for iWARP Steve Wise
2015-07-05 17:44 ` [PATCH V5 1/5] mlx4, mlx5, mthca: Expose max_sge_rd correctly Steve Wise
2015-07-05 17:44 ` [PATCH V5 2/5] ipath,qib: " Steve Wise
2015-07-05 17:44 ` [PATCH V5 3/5] RDMA/iser: Limit sg tablesize and max_sectors to device fastreg max depth Steve Wise
     [not found]   ` <20150705174452.10042.44695.stgit-PBeJgSbIpn97NCTnQtmixQ@public.gmane.org>
2015-07-06  7:50     ` Sagi Grimberg
     [not found]       ` <559A335A.90506-LDSdmyG8hGV8YrgS2mwiifqBs+8SCbDb@public.gmane.org>
2015-07-06 14:35         ` Steve Wise
2015-07-07  6:26           ` Sagi Grimberg [this message]
2015-07-07 13:59             ` Steve Wise
2015-07-07 14:32               ` Or Gerlitz
2015-07-07 15:41                 ` Steve Wise
2015-07-07 16:30                   ` Sagi Grimberg
2015-07-05 17:45 ` [PATCH V5 4/5] RDMA/isert: Set REMOTE_WRITE on DMA MRs to support iWARP devices Steve Wise
     [not found]   ` <20150705174459.10042.37298.stgit-PBeJgSbIpn97NCTnQtmixQ@public.gmane.org>
2015-07-06  7:52     ` Sagi Grimberg
2015-07-05 17:45 ` [PATCH V5 5/5] RDMA/isert: Limit read depth based on the device max_sge_rd capability Steve Wise
     [not found]   ` <20150705174505.10042.28442.stgit-PBeJgSbIpn97NCTnQtmixQ@public.gmane.org>
2015-07-06  7:52     ` Sagi Grimberg
2015-07-14  8:27     ` Christoph Hellwig
     [not found]       ` <20150714082721.GA30072-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>
2015-07-14 14:41         ` Steve Wise
2015-07-14 15:42           ` 'Christoph Hellwig'
     [not found]             ` <20150714154200.GE11026-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>
2015-07-14 15:49               ` Steve Wise
2015-07-14 18:47                 ` Chuck Lever
2015-07-14 19:11                   ` Steve Wise
2015-07-14 19:25                     ` Chuck Lever
     [not found] ` <20150705174353.10042.39648.stgit-PBeJgSbIpn97NCTnQtmixQ@public.gmane.org>
2015-07-07  0:38   ` [PATCH V5 0/5] iSER support for iWARP Nicholas A. Bellinger

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=559B712D.7020105@dev.mellanox.co.il \
    --to=sagig@dev.mellanox.co.il \
    --cc=dledford@redhat.com \
    --cc=eli@mellanox.com \
    --cc=infinipath@intel.com \
    --cc=linux-rdma@vger.kernel.org \
    --cc=ogerlitz@mellanox.com \
    --cc=roid@mellanox.com \
    --cc=sagig@mellanox.com \
    --cc=swise@opengridcomputing.com \
    --cc=target-devel@vger.kernel.org \
    /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.