RE: [PATH for-next] RDMA/siw: Fix setting active_{speed, width} attributes

["Bernard Metzler" <BMT@zurich.ibm.com>]

-----"Kamal Heib" <kamalheib1@gmail.com> wrote: -----

>To: "Bernard Metzler" <BMT@zurich.ibm.com>
>From: "Kamal Heib" <kamalheib1@gmail.com>
>Date: 02/17/2020 03:12PM
>Cc: linux-rdma@vger.kernel.org, "Jason Gunthorpe" <jgg@ziepe.ca>,
>"Doug Ledford" <dledford@redhat.com>
>Subject: [EXTERNAL] Re: [PATH for-next] RDMA/siw: Fix setting
>active_{speed, width} attributes
>
>On Mon, Feb 17, 2020 at 10:13:21AM +0000, Bernard Metzler wrote:
>> -----"Kamal Heib" <kamalheib1@gmail.com> wrote: -----
>> 
>> >To: "Bernard Metzler" <BMT@zurich.ibm.com>
>> >From: "Kamal Heib" <kamalheib1@gmail.com>
>> >Date: 02/16/2020 02:43PM
>> >Cc: linux-rdma@vger.kernel.org, "Jason Gunthorpe" <jgg@ziepe.ca>,
>> >"Doug Ledford" <dledford@redhat.com>
>> >Subject: [EXTERNAL] Re: [PATH for-next] RDMA/siw: Fix setting
>> >active_{speed, width} attributes
>> >
>> >On Thu, Feb 13, 2020 at 01:59:30PM +0000, Bernard Metzler wrote:
>> >> -----"Kamal Heib" <kamalheib1@gmail.com> wrote: -----
>> >> 
>> >> >To: linux-rdma@vger.kernel.org
>> >> >From: "Kamal Heib" <kamalheib1@gmail.com>
>> >> >Date: 02/13/2020 02:07PM
>> >> >Cc: "Jason Gunthorpe" <jgg@ziepe.ca>, "Doug Ledford"
>> >> ><dledford@redhat.com>, "Bernard Metzler" <bmt@zurich.ibm.com>,
>> >"Kamal
>> >> >Heib" <kamalheib1@gmail.com>
>> >> >Subject: [EXTERNAL] [PATH for-next] RDMA/siw: Fix setting
>> >> >active_{speed, width} attributes
>> >> >
>> >> >Make sure to set the active_{speed, width} attributes to avoid
>> >> >reporting
>> >> >the same values regardless of the underlying device.
>> >> >
>> >> >Fixes: 303ae1cdfdf7 ("rdma/siw: application interface")
>> >> >Signed-off-by: Kamal Heib <kamalheib1@gmail.com>
>> >> >---
>> >> > drivers/infiniband/sw/siw/siw_verbs.c | 7 ++++---
>> >> > 1 file changed, 4 insertions(+), 3 deletions(-)
>> >> >
>> >> >diff --git a/drivers/infiniband/sw/siw/siw_verbs.c
>> >> >b/drivers/infiniband/sw/siw/siw_verbs.c
>> >> >index 73485d0da907..b1aaec912edb 100644
>> >> >--- a/drivers/infiniband/sw/siw/siw_verbs.c
>> >> >+++ b/drivers/infiniband/sw/siw/siw_verbs.c
>> >> >@@ -165,11 +165,12 @@ int siw_query_port(struct ib_device
>> >*base_dev,
>> >> >u8 port,
>> >> > 		   struct ib_port_attr *attr)
>> >> > {
>> >> > 	struct siw_device *sdev = to_siw_dev(base_dev);
>> >> >+	int rc;
>> >> > 
>> >> > 	memset(attr, 0, sizeof(*attr));
>> >> > 
>> >> >-	attr->active_speed = 2;
>> >> >-	attr->active_width = 2;
>> >> >+	rc = ib_get_eth_speed(base_dev, port, &attr->active_speed,
>> >> >+			 &attr->active_width);
>> >> > 	attr->gid_tbl_len = 1;
>> >> > 	attr->max_msg_sz = -1;
>> >> > 	attr->max_mtu = ib_mtu_int_to_enum(sdev->netdev->mtu);
>> >> >@@ -192,7 +193,7 @@ int siw_query_port(struct ib_device
>*base_dev,
>> >u8
>> >> >port,
>> >> > 	 * attr->subnet_timeout = 0;
>> >> > 	 * attr->init_type_repy = 0;
>> >> > 	 */
>> >> >-	return 0;
>> >> >+	return rc;
>> >> > }
>> >> > 
>> >> > int siw_get_port_immutable(struct ib_device *base_dev, u8
>port,
>> >> >-- 
>> >> >2.21.1
>> >> >
>> >> >
>> >> Hi Kamal, 
>> >
>> >Hi Bernard,
>> >
>> >> Many thanks for looking after this! So there definitely seem to 
>> >> be applications which are taking care of those values. So, good
>> >> to get my obvious laziness fixed.
>> >> 
>> >
>> >Sure :)
>> >
>> >> I tried your patch on a 40Gbs Ethernet link (Chelsio cxgb4
>driver).
>> >> Works in principle, but reported numbers are off. I am not
>saying
>> >> I would get right numbers when using Chelsio HW iWarp
>(iw_cxgb4),
>> >> but it's closer to reality (using ibv_devinfo <ibname> -vv)
>> >> 
>> >> iw_cxgb4 driver:
>> >> ...
>> >>    active_width:           4X (2)
>> >>    active_speed:           25.0 Gbps (32)
>> >> 
>> >> siw driver with your patch:
>> >> ...
>> >>    active_width:           4X (2)
>> >>    active_speed:           10.0 Gbps (8)
>> >> 
>> >> Any idea how we can improve that, maybe coming even
>> >> close to reality (40Gbs)?
>> >
>> >Could you please share the output of ethtool <if_name> for the
>> >underlying
>> >net device that used for both iw_cxgb4 and siw?
>> >
>> H Kamal,
>> 
>> Sure! Speed looks correct, and its also what I get
>> at maximum:
>> 
>> [bmt@spoke ~]$ ethtool enp1s0f4 
>> Settings for enp1s0f4:
>>         Supported ports: [ FIBRE ]
>>         Supported link modes:   40000baseSR4/Full 
>>         Supported pause frame use: Symmetric
>>         Supports auto-negotiation: Yes
>>         Supported FEC modes: None
>>         Advertised link modes:  40000baseSR4/Full 
>>         Advertised pause frame use: Symmetric
>>         Advertised auto-negotiation: Yes
>>         Advertised FEC modes: None
>>         Link partner advertised link modes:  40000baseSR4/Full 
>>         Link partner advertised pause frame use: Symmetric
>>         Link partner advertised auto-negotiation: Yes
>>         Link partner advertised FEC modes: None
>>         Speed: 40000Mb/s
>>         Duplex: Full
>>         Port: Direct Attach Copper
>>         PHYAD: 255
>>         Transceiver: internal
>>         Auto-negotiation: on
>> Cannot get wake-on-lan settings: Operation not permitted
>>         Current message level: 0x000000ff (255)
>>                                drv probe link timer ifdown ifup
>rx_err tx_err
>>         Link detected: yes
>>
>
>Hi Bernard,
>
>Well, this is the expected value for 40GbE, Because the reported
>value
>is 4X aggregation of FDR10. For more info please the table of speeds
>under [1].
>
>[1] -
>https://urldefense.proofpoint.com/v2/url?u=https-3A__en.wikipedia.org
>_wiki_InfiniBand&d=DwIBAg&c=jf_iaSHvJObTbx-siA1ZOg&r=2TaYXQ0T-r8ZO1PP
>1alNwU_QJcRRLfmYTAgd3QCvqSc&m=Ay2tYZ7Z-SHKRw8UZDCk76kwlZzvkXhRMrO_0jk
>YjcY&s=D4Z0CAH5UVO95WHNnUozLzrqjxRQgVe-2lc8_jwVlhw&e= 
>
>Thanks,
>Kamal
>
Hi Kamal, so I have to do the math! 4 x 10Gbs = 40Gbs.
So it is all correct as reported by ibv_devinfio
(and somehow strange what the iw_cxgb4 makes out of it).

Many thanks for the info,
Bernard.