[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20180627163706.GA9361@localhost.localdomain>
Date: Wed, 27 Jun 2018 10:37:06 -0600
From: Keith Busch <keith.busch@...ux.intel.com>
To: Johannes Thumshirn <jthumshirn@...e.de>
Cc: Christoph Hellwig <hch@....de>, Sagi Grimberg <sagi@...mberg.me>,
Keith Busch <keith.busch@...el.com>,
James Smart <james.smart@...adcom.com>,
Linux Kernel Mailinglist <linux-kernel@...r.kernel.org>,
Linux NVMe Mailinglist <linux-nvme@...ts.infradead.org>
Subject: Re: [PATCH v5 2/2] nvme: trace: add disk name to tracepoints
On Wed, Jun 27, 2018 at 02:53:24PM +0200, Johannes Thumshirn wrote:
> TP_fast_assign(
> - __entry->qid = qid;
> + __assign_str(name, disk_name);
> + __entry->ctrl_id = nvme_req(req)->ctrl->cntlid;
> + __entry->qid = req->q->id;
> __entry->opcode = cmd->common.opcode;
> __entry->flags = cmd->common.flags;
> __entry->cid = cmd->common.command_id;
> @@ -122,10 +130,10 @@ TRACE_EVENT(nvme_setup_nvm_cmd,
> memcpy(__entry->cdw10, cmd->common.cdw10,
> sizeof(__entry->cdw10));
> ),
> - TP_printk("qid=%d, nsid=%u, cmdid=%u, flags=0x%x, meta=0x%llx, cmd=(%s %s)",
> - __entry->qid, __entry->nsid, __entry->cid,
> - __entry->flags, __entry->metadata,
> - show_opcode_name(__entry->opcode),
> + TP_printk("nvme%d: disk=%s, qid=%d, nsid=%u, cmdid=%u, flags=0x%x, meta=0x%llx, cmd=(%s %s)",
> + __entry->ctrl_id, __get_str(name), __entry->qid,
> + __entry->nsid, __entry->cid, __entry->flags,
> + __entry->metadata, show_opcode_name(__entry->opcode),
> __parse_nvme_cmd(__entry->opcode, __entry->cdw10))
> );
This looks a bit confusing to me. The ctrl->cntlid you're using in
__entry->ctrl_id isn't a unique value across subsystems, and it will
typically be 0 for all controllers that are the only controller in
their subsystem.
It looks like you want the # assigned to /dev/nvme<#>. Do you want to
use ctrl->instance here instead?
Powered by blists - more mailing lists