| Message ID | 1471011841-4126-1-git-send-email-xose.vazquez@gmail.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
Could you please rebase this one ? Thanks. On Fri, Aug 12, 2016 at 4:24 PM, Xose Vazquez Perez <xose.vazquez@gmail.com> wrote: > New undocumented feature, added in e83068a5f: > > dm mpath: add optional "queue_mode" feature > > Allow a user to specify an optional feature 'queue_mode <mode>' where > <mode> may be "bio", "rq" or "mq" -- which corresponds to bio-based, > request_fn rq-based, and blk-mq rq-based respectively. > > If the queue_mode feature isn't specified the default for the > "multipath" target is still "rq" but if dm_mod.use_blk_mq is set to Y > it'll default to mode "mq". > > This new queue_mode feature introduces the ability for each multipath > device to have its own queue_mode (whereas before this feature all > multipath devices effectively had to have the same queue_mode). > > > Also add more info to other features. > > Cc: Mike Snitzer <snitzer@redhat.com> > Cc: Christophe Varoqui <christophe.varoqui@opensvc.com> > Cc: device-mapper development <dm-devel@redhat.com> > Signed-off-by: Xose Vazquez Perez <xose.vazquez@gmail.com> > --- > multipath/multipath.conf.5 | 36 ++++++++++++++++++++++++------------ > 1 file changed, 24 insertions(+), 12 deletions(-) > > diff --git a/multipath/multipath.conf.5 b/multipath/multipath.conf.5 > index fd4b1f2..40df315 100644 > --- a/multipath/multipath.conf.5 > +++ b/multipath/multipath.conf.5 > @@ -3,11 +3,11 @@ > .\" Make sure there are no errors with: > .\" groff -z -wall -b -e -t multipath/multipath.conf.5 > .\" > -.\" TODO: Look for XXX > +.\" TODO: Look for XXX and ??? > .\" > .\" ------------------------------------------------------------ > ---------------- > . > -.TH MULTIPATH.CONF 5 2016-08-09 "Linux" > +.TH MULTIPATH.CONF 5 2016-08-12 "Linux" > . > . > .\" ------------------------------------------------------------ > ---------------- > @@ -166,12 +166,15 @@ Loop through every path in the path group, sending > the same amount of IO to > each. Some aspects of behavior can be controlled with the attributes: > \fIrr_min_io\fR, \fIrr_min_io_rq\fR and \fIrr_weight\fR. > .TP > +.\" XXX > .I "queue-length 0" > -Send the next bunch of IO down the path with the least amount of > outstanding IO. > +(Since ??? kernel) Choose the path for the next bunch of IO based on the > amount > +of outstanding IO to the path. > .TP > +.\" XXX > .I "service-time 0" > -Choose the path for the next bunch of IO based on the amount of > outstanding IO > -to the path and its relative throughput. > +(Since ??? kernel) Choose the path for the next bunch of IO based on the > amount > +of outstanding IO to the path and its relative throughput. > .TP > Default value is: \fBservice-time 0\fR > .RE > @@ -333,22 +336,31 @@ Default value is: \fB<unset>\fR > .TP > .B features > Specify any device-mapper features to be used. Syntax is \fInum list\fR > -where \fInum\fR is the number, between 0 and 6, of features in \fIlist\fR. > +where \fInum\fR is the number, between 0 and 8, of features in \fIlist\fR. > Possible values for the feature list are: > .RS > .TP 12 > +.\" XXX > .I queue_if_no_path > -(Superseded by \fIno_path_retry\fR) Queue IO if no path is active. > Identical to > -the \fIno_path_retry\fR with \fIqueue\fR value. See KNOWN ISSUES. > +(Superseded by \fIno_path_retry\fR) (Since ??? kernel) Queue IO if no > path is active. > +Identical to the \fIno_path_retry\fR with \fIqueue\fR value. See KNOWN > ISSUES. > .TP > .I no_partitions > Disable automatic partitions generation via kpartx. > .TP > -.I pg_init_retries > -Number of times to retry pg_init, it must be between 1 and 50. > +.\" XXX > +.I pg_init_retries <times> > +(Since ??? kernel) Number of times to retry pg_init, it must be between 1 > and 50. > .TP > -.I pg_init_delay_msecs > -Number of msecs before pg_init retry, it must be between 0 and 60000. > +.\" XXX > +.I pg_init_delay_msecs <msecs> > +(Since ??? kernel) Number of msecs before pg_init retry, it must be > between 0 and 60000. > +.TP > +.\" XXX > +.I queue_mode <mode> > +(Since ??? kernel) Select the the queue_mode per multipath device. > +Where <mode> can be \fIbio\fR, \fIrq\fR or \fImq\fR. Which corresponds to > +bio-based, request_fn rq-based, and blk-mq rq-based respectively. > .TP > Default value is: \fB0\fR > .RE > -- > 2.7.4 > > -- dm-devel mailing list dm-devel@redhat.com https://www.redhat.com/mailman/listinfo/dm-devel
On 08/16/2016 09:49 AM, Christophe Varoqui wrote:
> Could you please rebase this one ?
Please drop my pending patches from your queue.
I am going to send, again, five missed patches.
Thank you.
--
dm-devel mailing list
dm-devel@redhat.com
https://www.redhat.com/mailman/listinfo/dm-devel
diff --git a/multipath/multipath.conf.5 b/multipath/multipath.conf.5 index fd4b1f2..40df315 100644 --- a/multipath/multipath.conf.5 +++ b/multipath/multipath.conf.5 @@ -3,11 +3,11 @@ .\" Make sure there are no errors with: .\" groff -z -wall -b -e -t multipath/multipath.conf.5 .\" -.\" TODO: Look for XXX +.\" TODO: Look for XXX and ??? .\" .\" ---------------------------------------------------------------------------- . -.TH MULTIPATH.CONF 5 2016-08-09 "Linux" +.TH MULTIPATH.CONF 5 2016-08-12 "Linux" . . .\" ---------------------------------------------------------------------------- @@ -166,12 +166,15 @@ Loop through every path in the path group, sending the same amount of IO to each. Some aspects of behavior can be controlled with the attributes: \fIrr_min_io\fR, \fIrr_min_io_rq\fR and \fIrr_weight\fR. .TP +.\" XXX .I "queue-length 0" -Send the next bunch of IO down the path with the least amount of outstanding IO. +(Since ??? kernel) Choose the path for the next bunch of IO based on the amount +of outstanding IO to the path. .TP +.\" XXX .I "service-time 0" -Choose the path for the next bunch of IO based on the amount of outstanding IO -to the path and its relative throughput. +(Since ??? kernel) Choose the path for the next bunch of IO based on the amount +of outstanding IO to the path and its relative throughput. .TP Default value is: \fBservice-time 0\fR .RE @@ -333,22 +336,31 @@ Default value is: \fB<unset>\fR .TP .B features Specify any device-mapper features to be used. Syntax is \fInum list\fR -where \fInum\fR is the number, between 0 and 6, of features in \fIlist\fR. +where \fInum\fR is the number, between 0 and 8, of features in \fIlist\fR. Possible values for the feature list are: .RS .TP 12 +.\" XXX .I queue_if_no_path -(Superseded by \fIno_path_retry\fR) Queue IO if no path is active. Identical to -the \fIno_path_retry\fR with \fIqueue\fR value. See KNOWN ISSUES. +(Superseded by \fIno_path_retry\fR) (Since ??? kernel) Queue IO if no path is active. +Identical to the \fIno_path_retry\fR with \fIqueue\fR value. See KNOWN ISSUES. .TP .I no_partitions Disable automatic partitions generation via kpartx. .TP -.I pg_init_retries -Number of times to retry pg_init, it must be between 1 and 50. +.\" XXX +.I pg_init_retries <times> +(Since ??? kernel) Number of times to retry pg_init, it must be between 1 and 50. .TP -.I pg_init_delay_msecs -Number of msecs before pg_init retry, it must be between 0 and 60000. +.\" XXX +.I pg_init_delay_msecs <msecs> +(Since ??? kernel) Number of msecs before pg_init retry, it must be between 0 and 60000. +.TP +.\" XXX +.I queue_mode <mode> +(Since ??? kernel) Select the the queue_mode per multipath device. +Where <mode> can be \fIbio\fR, \fIrq\fR or \fImq\fR. Which corresponds to +bio-based, request_fn rq-based, and blk-mq rq-based respectively. .TP Default value is: \fB0\fR .RE
New undocumented feature, added in e83068a5f: dm mpath: add optional "queue_mode" feature Allow a user to specify an optional feature 'queue_mode <mode>' where <mode> may be "bio", "rq" or "mq" -- which corresponds to bio-based, request_fn rq-based, and blk-mq rq-based respectively. If the queue_mode feature isn't specified the default for the "multipath" target is still "rq" but if dm_mod.use_blk_mq is set to Y it'll default to mode "mq". This new queue_mode feature introduces the ability for each multipath device to have its own queue_mode (whereas before this feature all multipath devices effectively had to have the same queue_mode). Also add more info to other features. Cc: Mike Snitzer <snitzer@redhat.com> Cc: Christophe Varoqui <christophe.varoqui@opensvc.com> Cc: device-mapper development <dm-devel@redhat.com> Signed-off-by: Xose Vazquez Perez <xose.vazquez@gmail.com> --- multipath/multipath.conf.5 | 36 ++++++++++++++++++++++++------------ 1 file changed, 24 insertions(+), 12 deletions(-)