From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <Allen.Hubbe@dell.com>
Received: from esa3.dell-outbound.iphmx.com (esa3.dell-outbound.iphmx.com. [68.232.153.94])
        by gmr-mx.google.com with ESMTPS id z137si3819613ywd.7.2016.11.29.08.04.24
        for <linux-ntb@googlegroups.com>
        (version=TLS1_2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128);
        Tue, 29 Nov 2016 08:04:24 -0800 (PST)
From: "Allen Hubbe" <Allen.Hubbe@dell.com>
References: <c44465a3-780b-3dbd-22dc-4be241009e3c@amd.com> <000001d23e84$5720d3a0$05627ae0$@emc.com> <ca26d2ef-f3ab-3713-eacb-cb2d05d61547@amd.com> <CAPoiz9xo8TJeLpS6Ft77CHhM=6Yi+ukSek7qfQTxBORnyZDupQ@mail.gmail.com> <550c9e3a-c499-ecee-310f-1bd53f381650@amd.com> <000001d244cb$7a767880$6f636980$@dell.com> <631ba7a7-3c3e-3c7c-d164-b53c52d81753@amd.com>
In-Reply-To: <631ba7a7-3c3e-3c7c-d164-b53c52d81753@amd.com>
Subject: RE: [PATCH] NTB: Register and offset values fix for memory window
Date: Tue, 29 Nov 2016 11:03:46 -0500
Message-ID: <000001d24a5a$2dcbaac0$89630040$@dell.com>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="utf-8"
Content-Transfer-Encoding: 7bit
Content-Language: en-us
To: 'Shyam Sundar S K' <ssundark@amd.com>, 'Jon Mason' <jdmason@kudzu.us>
Cc: "'Yu, Xiangliang'" <Xiangliang.Yu@amd.com>, 'Dave Jiang' <dave.jiang@intel.com>, linux-ntb@googlegroups.com, "'Sen, Pankaj'" <Pankaj.Sen@amd.com>, "'Shah, Nehal-bakulchandra'" <Nehal-bakulchandra.Shah@amd.com>, "'Agrawal, Nitesh-kumar'" <Nitesh-kumar.Agrawal@amd.com>, "'Subramaniyan, Ramkumar'" <ramkumar.ks@amd.com>, Richard1.Su@amd.com
List-ID: <linux-ntb.googlegroups.com>

From: Shyam Sundar S K
> On 11/22/2016 7:49 PM, Allen Hubbe wrote:
> > Instead, limit the number of memory windows actually used by ntb_transport, such that
> ntb_transport can configure those memory windows using only the available scratchpads.  If
> there are not enough scratchpads to exchange values for all of the memory windows, then
> ntb_transport should limit itself to use fewer memory windows.
> >
> 
> Allen, I get your point. So, you are asking to limit the number of memory windows we
> advertise in our driver since there are less number of scratchpads available when we have
> to configure the 3rd
> window. I believe you are asking me to do
> 
> -       ndev->mw_count = AMD_MW_CNT;
> +       ndev->mw_count = AMD_MW_CNT - 1;
> 
> in amd_init_ntb(). Is this understanding correct ?

That would work, but it would limit other transport drivers from using all the memory windows of your device.
 
This is the change I meant to describe, in ntb_transport:

static int ntb_transport_probe(struct ntb_client *self, struct ntb_dev *ndev)
{
...
	mw_count = ntb_mw_count(ndev);

Instead of this:

	nt->mw_count = mw_count;

Limit the mw count like this:

	spad_count = ntb_spad_count(ndev);

	max_mw_count_for_spads = some_arithmetic(spad_count);

	nt->mw_count = min(mw_count, max_mw_count_for_spads);