Re: [dm-devel] [PATCH] kpartx: support device names with spaces

2016-01-19 Thread Christophe Varoqui
Hello, shouldn't you either define the MAJOR and MINOR macros, or include the appropriate kernel header ? As is the build emits : kpartx.c:481:8: warning: implicit declaration of function ‘MAJOR’ [-Wimplicit-function-declaration] MAJOR(buf.st_rdev), MINOR(buf.st_rdev), slices[j].start)) {

Re: [dm-devel] Question about multipath device compiled defaults

2016-01-19 Thread Christophe Varoqui
Ok, merged. On Tue, Jan 19, 2016 at 11:17 PM, Brian Bunker wrote: > There are just two files I touch, so the patch is simple: > > diff -Naur 3/multipath-tools/libmultipath/hwtable.c > 4/multipath-tools/libmultipath/hwtable.c > --- 3/multipath-tools/libmultipath/hwtable.c2016-01-14 > 17:47:06

[dm-devel] How can I get my sector size (via queue_limits?) from a dm_target?

2016-01-19 Thread Eric Wheeler
Hello all, dm_target->len gives me sector count, but I need to know the default sector size discovered by the blk stacking code without assuming 512b. Now that I have the dm .iterate_devices callback defined and my queue_limits are sane by default, how can I get a pointer to the queue_limits s

Re: [dm-devel] How can dm use limits.max_hw_sectors from the bdev?

2016-01-19 Thread Eric Wheeler
On Mon, 18 Jan 2016, Mike Snitzer wrote: > On Sun, Jan 17 2016 at 1:34am -0500, > Eric Wheeler wrote: > > > Hello all, > > > > I'm writing a trivial dm target and hitting errors like this: > >io too big device loop0 (2048 > 255) > > > > which looks like the problem described here: > >

Re: [dm-devel] dm-multipath low performance with blk-mq

2016-01-19 Thread Mike Snitzer
On Mon, Jan 18 2016 at 7:04am -0500, Sagi Grimberg wrote: > Hi All, > > I've recently tried out dm-multipath over a "super-fast" nvme device > and noticed a serious lock contention in dm-multipath that requires some > extra attention. The nvme device is a simple loopback device emulation > back

Re: [dm-devel] Question about multipath device compiled defaults

2016-01-19 Thread Brian Bunker
There are just two files I touch, so the patch is simple: diff -Naur 3/multipath-tools/libmultipath/hwtable.c 4/multipath-tools/libmultipath/hwtable.c --- 3/multipath-tools/libmultipath/hwtable.c2016-01-14 17:47:06.0 -0800 +++ 4/multipath-tools/libmultipath/hwtable.c2016-01-14 18:3

Re: [dm-devel] Question about multipath device compiled defaults

2016-01-19 Thread Christophe Varoqui
Hi, and when you are ready, you can post the patch here so I can merge it upstream. Thanks, Christophe Varoqui OpenSVC On Tue, Jan 19, 2016 at 5:35 PM, Mauricio Faria de Oliveira < mauri...@linux.vnet.ibm.com> wrote: > On 11/05/2015 04:59 AM, Doron Tal1 wrote: > >> I would like to know how and

Re: [dm-devel] Question about multipath device compiled defaults

2016-01-19 Thread Mauricio Faria de Oliveira
On 11/05/2015 04:59 AM, Doron Tal1 wrote: I would like to know how and what procedure i should follow if i want to modify or add new settings to the device mapper defaults. The defaults are in hwtable.c [1]. So for example how can i modify existing compiled settings for the 2810XIV device? in

Re: [dm-devel] dm-multipath low performance with blk-mq

2016-01-19 Thread Sagi Grimberg
This time with the correct dm-devel... Hi All, I've recently tried out dm-multipath over a "super-fast" nvme device and noticed a serious lock contention in dm-multipath that requires some extra attention. The nvme device is a simple loopback device emulation backed by null_blk device. With th