[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20180116150002.77orcgphqcv52dhh@linux-x5ow.site>
Date: Tue, 16 Jan 2018 16:00:02 +0100
From: Johannes Thumshirn <jthumshirn@...e.de>
To: Keith Busch <keith.busch@...el.com>
Cc: Christoph Hellwig <hch@....de>, Sagi Grimberg <sagi@...mberg.me>,
Linux Kernel Mailinglist <linux-kernel@...r.kernel.org>,
Hannes Reinecke <hare@...e.de>,
Linux NVMe Mailinglist <linux-nvme@...ts.infradead.org>
Subject: Re: [PATCH v2 0/2] add tracepoints for nvme command submission and
completion
On Tue, Jan 16, 2018 at 07:55:55AM -0700, Keith Busch wrote:
> On Tue, Jan 16, 2018 at 03:28:19PM +0100, Johannes Thumshirn wrote:
> > Add tracepoints for nvme command submission and completion. The tracepoints
> > are modeled after SCSI's trace_scsi_dispatch_cmd_start() and
> > trace_scsi_dispatch_cmd_done() tracepoints and fulfil a similar purpose,
> > namely a fast way to check which command is going to be queued into the HW or
> > Fabric driver and which command is completed again.
>
> I like this very much, thanks for doing this. I think you could make the
> submission trace point tighter for PCI as Hannes was suggesting since
> an MMIO write can't fail, but doesn't look as doable for FC and RDMA.
That's what I concluded as well and I wanted to have some sort of smallest
common divisor when tracing customer problems on nvme.
I think I can come up with additional tracepoints in pci/rdma/fc but these can
be added later as well and should not hold up this series.
Byte,
Johannes
--
Johannes Thumshirn Storage
jthumshirn@...e.de +49 911 74053 689
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: Felix Imendörffer, Jane Smithard, Graham Norton
HRB 21284 (AG Nürnberg)
Key fingerprint = EC38 9CAB C2C4 F25D 8600 D0D0 0393 969D 2D76 0850
Powered by blists - more mailing lists