| To: | Martins Innus <minnus@xxxxxxxxxxx> |
|---|---|
| Subject: | Re: [pcp] updated infiniband pmda |
| From: | Nathan Scott <nathans@xxxxxxxxxx> |
| Date: | Thu, 24 Jul 2014 03:12:06 -0400 (EDT) |
| Cc: | pcp@xxxxxxxxxxx |
| Delivered-to: | pcp@xxxxxxxxxxx |
| In-reply-to: | <53CFD25B.2070408@xxxxxxxxxxx> |
| References: | <53A998F1.4050003@xxxxxxxxxxx> <1097508451.33910731.1403758163030.JavaMail.zimbra@xxxxxxxxxx> <53B2EEA0.7020106@xxxxxxxxxxx> <53CFD25B.2070408@xxxxxxxxxxx> |
| Reply-to: | Nathan Scott <nathans@xxxxxxxxxx> |
| Thread-index: | JDcmeqT4wuzixt/px3lgsT9bjDO+xQ== |
| Thread-topic: | updated infiniband pmda |
Hi Martins, ----- Original Message ----- > Nathan, > I just realized that 3.9.7 missed a couple changes for the > infiniband switch metrics. Probably my fault, looking back, in my > second submission, I only included the files that changed from my first > submission, not everything. Attached patch should fix. (merged in dev branch now) > There should be no problem for existing users of 3.9.7 in terms of > previous behavior, just that switch metrics will not be reported even if > they should be available. > No worries - we should start thinking about some automated testing for this agent. The model we used for the Nvidia PMDA with a mock version of the library giving fixed results could give us basic sanity testing here too I guess? Install/Remove scripts, pmns validity - even simple valgrind-based memchecks can be done that way too, if especially keen. cheers. -- Nathan |
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | pcp updates: kenj+martins+brolley merges, dmcache PMDA, Nathan Scott |
|---|---|
| Next by Date: | Re: systemtap/pcp integration, Nathan Scott |
| Previous by Thread: | Re: [pcp] updated infiniband pmda, Martins Innus |
| Next by Thread: | Current Work, Martins Innus |
| Indexes: | [Date] [Thread] [Top] [All Lists] |