From mboxrd@z Thu Jan 1 00:00:00 1970 From: wyung@micron.com (Winson Yung (wyung)) Date: Mon, 26 May 2014 22:35:16 -0700 Subject: [PATCH] NVMe: Add support to receive NVMe asynchronous events In-Reply-To: References: <53839E67.1010006@micron.com> <20140526235156.GS6121@linux.intel.com> Message-ID: <53842414.7020207@micron.com> Keith, I do see a need to enable async event in the NVMe kernel driver. For example, when there is a temperature above threshold, driver can take an action (by telling firmware) to lower down operating frequency, or throttle IO request to protect drive from premature over heat damage. /Winson On 5/26/2014 5:56 PM, Keith Busch wrote: > On Mon, 26 May 2014, Matthew Wilcox wrote: >> You should probably take a look at Keith's work last year to allow userspace >> to send async event requests: >> >> http://lists.infradead.org/pipermail/linux-nvme/2013-August/000350.html > > Yeah, that proposal just lets userspace consume the results of an async > request from a FIFO read through the character device. The driver still > has to do the sending since there's no mechanism for requests from user > space with infinite timeout. > > I've no idea if the proposed method for consuming results of this kind > of information was typical. I assumed it wasn't since it didn't get much > traction. :) >