From: Hal Rosenstock <hal-LDSdmyG8hGV8YrgS2mwiifqBs+8SCbDb@public.gmane.org>
To: Alex Netes <alexne-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
Cc: "linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org"
<linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>
Subject: Re: [PATCH] opensm/osmtest/osmt_multicast.c: Use proper defines for rate
Date: Sun, 19 Jun 2011 10:46:59 -0400 [thread overview]
Message-ID: <4DFE0BE3.6060208@dev.mellanox.co.il> (raw)
In-Reply-To: <20110619120535.GA2431-iQai9MGU/dyyaiaB+Ve85laTQe2KTcn/@public.gmane.org>
Hi Alex,
On 6/19/2011 8:05 AM, Alex Netes wrote:
> Hi Hal,
>
> On 10:05 Sat 18 Jun , Hal Rosenstock wrote:
>> Rather than using width defines, use rate defines when setting rate in MCMemberRecord.
>> The conversions here are 1 for 1.
>>
>> Also, fix typo in an OSM_LOG message
>>
>> Signed-off-by: Hal Rosenstock <hal-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
>> ---
>> diff --git a/osmtest/osmt_multicast.c b/osmtest/osmt_multicast.c
>> index e4eba64..4e471ce 100644
>> --- a/osmtest/osmt_multicast.c
>> +++ b/osmtest/osmt_multicast.c
>> @@ -385,7 +385,7 @@ void osmt_init_mc_query_rec(IN osmtest_t * const p_osmt,
>> /* uint8_t mtu; - keep it zero means - anything you have please. */
>> /* uint8_t tclass; can leave as zero for now (between subnets) */
>> /* ib_net16_t pkey; leave as zero */
>> - p_mc_req->rate = IB_LINK_WIDTH_ACTIVE_4X;
>> + p_mc_req->rate = IB_PATH_RECORD_RATE_2_5_GBS;
>> /* uint8_t pkt_life; zero means greater than zero ... */
>> /* ib_net32_t sl_flow_hop; keep it all zeros */
>> /* we want to use a link local scope: 0x02 */
>> @@ -900,7 +900,7 @@ ib_api_status_t osmt_run_mcast_flow(IN osmtest_t * const p_osmt)
>>
>> /* impossible requested rate */
>> mc_req_rec.rate =
>> - IB_LINK_WIDTH_ACTIVE_12X | IB_PATH_SELECTOR_GREATER_THAN << 6;
>> + IB_PATH_RECORD_RATE_60_GBS | IB_PATH_SELECTOR_GREATER_THAN << 6;
>>
>> comp_mask = IB_MCR_COMPMASK_GID | IB_MCR_COMPMASK_PORT_GID | IB_MCR_COMPMASK_QKEY | IB_MCR_COMPMASK_PKEY | IB_MCR_COMPMASK_SL | IB_MCR_COMPMASK_FLOW | IB_MCR_COMPMASK_JOIN_STATE | IB_MCR_COMPMASK_TCLASS | /* all above are required */
>> IB_MCR_COMPMASK_RATE_SEL | IB_MCR_COMPMASK_RATE;
>> @@ -1216,14 +1216,14 @@ ib_api_status_t osmt_run_mcast_flow(IN osmtest_t * const p_osmt)
>> OSM_LOG(&p_osmt->log, OSM_LOG_INFO,
>> "Checking Create given MGID=0 valid Set several options :\n\t\t"
>> "First above min RATE, Second less than max RATE\n\t\t"
>> - "Third above min MTU, Second less than max MTU\n\t\t"
>> + "Third above min MTU, Fourth less than max MTU\n\t\t"
>> "Fifth exact MTU & RATE feasible, Sixth exact RATE feasible\n\t\t"
>> "Seventh exact MTU feasible (o15.0.1.4)...\n");
>>
>> /* Good Flow - mgid is 0 while giving all required fields for join : P_Key, Q_Key, SL, FlowLabel, Tclass */
>>
>> mc_req_rec.rate =
>> - IB_LINK_WIDTH_ACTIVE_1X | IB_PATH_SELECTOR_GREATER_THAN << 6;
>> + (IB_PATH_RECORD_RATE_2_5_GBS - 1) | IB_PATH_SELECTOR_GREATER_THAN << 6;
>>
>> comp_mask = IB_MCR_COMPMASK_MGID | IB_MCR_COMPMASK_PORT_GID | IB_MCR_COMPMASK_QKEY | IB_MCR_COMPMASK_PKEY | IB_MCR_COMPMASK_SL | IB_MCR_COMPMASK_FLOW | IB_MCR_COMPMASK_JOIN_STATE | IB_MCR_COMPMASK_TCLASS | /* all above are required */
>> IB_MCR_COMPMASK_RATE_SEL | IB_MCR_COMPMASK_RATE;
>> @@ -1247,7 +1247,7 @@ ib_api_status_t osmt_run_mcast_flow(IN osmtest_t * const p_osmt)
>> /* Good Flow - mgid is 0 while giving all required fields for join : P_Key, Q_Key, SL, FlowLabel, Tclass */
>>
>> mc_req_rec.rate =
>> - IB_LINK_WIDTH_ACTIVE_12X | IB_PATH_SELECTOR_LESS_THAN << 6;
>> + IB_PATH_RECORD_RATE_60_GBS | IB_PATH_SELECTOR_LESS_THAN << 6;
>>
>> comp_mask = IB_MCR_COMPMASK_MGID | IB_MCR_COMPMASK_PORT_GID | IB_MCR_COMPMASK_QKEY | IB_MCR_COMPMASK_PKEY | IB_MCR_COMPMASK_SL | IB_MCR_COMPMASK_FLOW | IB_MCR_COMPMASK_JOIN_STATE | IB_MCR_COMPMASK_TCLASS | /* all above are required */
>> IB_MCR_COMPMASK_RATE_SEL | IB_MCR_COMPMASK_RATE;
>> @@ -1843,7 +1843,7 @@ ib_api_status_t osmt_run_mcast_flow(IN osmtest_t * const p_osmt)
>> "Checking JoinState update request should return 0x22 (o15.0.1.11)...\n");
>>
>> mc_req_rec.rate =
>> - IB_LINK_WIDTH_ACTIVE_1X | IB_PATH_SELECTOR_GREATER_THAN << 6;
>> + (IB_PATH_RECORD_RATE_2_5_GBS - 1) | IB_PATH_SELECTOR_GREATER_THAN << 6;
>
> (IB_PATH_RECORD_RATE_2_5_GBS - 1) is defined as 1 and it's a reserved value
> IBTA p917.
Yes, it's a reserved value but is equivalent to what was there.
> Moreover in the OpenSM code opensm/osm_sa_mcmember_record.c:288 the check is:
>
> if (rate_mgrp <= rate_required) { //equal or greater
Is that another bug ? It says greater than rather than greater than or
equal to. Also, other places there do the comparison without the equal.
> So I guess rate here could be defined as:
> mc_req_rec.rate =
> (IB_PATH_RECORD_RATE_2_5_GBS) | IB_PATH_SELECTOR_GREATER_THAN << 6;
AFAIT it's not greater than or equal to but just greater than.
-- Hal
>> mc_req_rec.mgid = good_mgid;
>>
>> OSM_LOG(&p_osmt->log, OSM_LOG_INFO,
>> @@ -1980,7 +1980,7 @@ ib_api_status_t osmt_run_mcast_flow(IN osmtest_t * const p_osmt)
>>
>> mc_req_rec.mgid = good_mgid;
>> mc_req_rec.rate =
>> - IB_LINK_WIDTH_ACTIVE_1X | IB_PATH_SELECTOR_LESS_THAN << 6;
>> + (IB_PATH_RECORD_RATE_2_5_GBS - 1) | IB_PATH_SELECTOR_LESS_THAN << 6;
>> comp_mask = IB_MCR_COMPMASK_GID | IB_MCR_COMPMASK_PORT_GID | IB_MCR_COMPMASK_QKEY | IB_MCR_COMPMASK_PKEY | IB_MCR_COMPMASK_SL | IB_MCR_COMPMASK_FLOW | IB_MCR_COMPMASK_JOIN_STATE | IB_MCR_COMPMASK_TCLASS | /* all above are required */
>> IB_MCR_COMPMASK_RATE_SEL | IB_MCR_COMPMASK_RATE;
>>
>> @@ -2055,7 +2055,7 @@ ib_api_status_t osmt_run_mcast_flow(IN osmtest_t * const p_osmt)
>> "Checking partial JoinState delete request - removing NonMember (o15.0.1.14)...\n");
>>
>> mc_req_rec.rate =
>> - IB_LINK_WIDTH_ACTIVE_1X | IB_PATH_SELECTOR_GREATER_THAN << 6;
>> + (IB_PATH_RECORD_RATE_2_5_GBS - 1) | IB_PATH_SELECTOR_GREATER_THAN << 6;
>> mc_req_rec.mgid = good_mgid;
>> comp_mask = IB_MCR_COMPMASK_GID | IB_MCR_COMPMASK_PORT_GID | IB_MCR_COMPMASK_QKEY | IB_MCR_COMPMASK_PKEY | IB_MCR_COMPMASK_SL | IB_MCR_COMPMASK_FLOW | IB_MCR_COMPMASK_JOIN_STATE | IB_MCR_COMPMASK_TCLASS | /* all above are required */
>> IB_MCR_COMPMASK_RATE_SEL | IB_MCR_COMPMASK_RATE;
>> @@ -2136,7 +2136,7 @@ ib_api_status_t osmt_run_mcast_flow(IN osmtest_t * const p_osmt)
>>
>> mc_req_rec.mgid = osm_ipoib_mgid;
>> mc_req_rec.rate =
>> - IB_LINK_WIDTH_ACTIVE_1X | IB_PATH_SELECTOR_GREATER_THAN << 6;
>> + (IB_PATH_RECORD_RATE_2_5_GBS - 1) | IB_PATH_SELECTOR_GREATER_THAN << 6;
>> mc_req_rec.scope_state = 0x21; /* delete full member */
>>
>> status = osmt_send_mcast_request(p_osmt, 0, /* delete flag */
>> @@ -2194,7 +2194,7 @@ ib_api_status_t osmt_run_mcast_flow(IN osmtest_t * const p_osmt)
>>
>> mc_req_rec.mgid = osm_ipoib_mgid;
>> mc_req_rec.rate =
>> - IB_LINK_WIDTH_ACTIVE_1X | IB_PATH_SELECTOR_GREATER_THAN << 6;
>> + (IB_PATH_RECORD_RATE_2_5_GBS - 1) | IB_PATH_SELECTOR_GREATER_THAN << 6;
>> mc_req_rec.scope_state = 0x21; /* delete full member */
>>
>> status = osmt_send_mcast_request(p_osmt, 0, /* delete flag */
>> @@ -2411,7 +2411,7 @@ ib_api_status_t osmt_run_mcast_flow(IN osmtest_t * const p_osmt)
>> join : P_Key, Q_Key, SL, FlowLabel, Tclass */
>>
>> mc_req_rec.rate =
>> - IB_LINK_WIDTH_ACTIVE_1X |
>> + (IB_PATH_RECORD_RATE_2_5_GBS - 1) |
>> IB_PATH_SELECTOR_GREATER_THAN << 6;
>> mc_req_rec.mlid = max_mlid;
>> memset(&mc_req_rec.mgid, 0, sizeof(ib_gid_t));
>> --
>> To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
>> the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
>> More majordomo info at http://vger.kernel.org/majordomo-info.html
>
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
next prev parent reply other threads:[~2011-06-19 14:46 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-06-18 14:05 [PATCH] opensm/osmtest/osmt_multicast.c: Use proper defines for rate Hal Rosenstock
[not found] ` <4DFCB0AE.80400-LDSdmyG8hGV8YrgS2mwiifqBs+8SCbDb@public.gmane.org>
2011-06-19 12:05 ` Alex Netes
[not found] ` <20110619120535.GA2431-iQai9MGU/dyyaiaB+Ve85laTQe2KTcn/@public.gmane.org>
2011-06-19 14:46 ` Hal Rosenstock [this message]
-- strict thread matches above, loose matches on Subject: below --
2011-07-05 17:24 Hal Rosenstock
[not found] ` <4E1348D2.8080109-LDSdmyG8hGV8YrgS2mwiifqBs+8SCbDb@public.gmane.org>
2011-07-08 11:12 ` Alex Netes
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=4DFE0BE3.6060208@dev.mellanox.co.il \
--to=hal-ldsdmyg8hgv8yrgs2mwiifqbs+8scbdb@public.gmane.org \
--cc=alexne-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org \
--cc=linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox