Re: [PATCH 08/27] staging: lustre: avoid to use bio->bi_vcnt directly

From	James Simmons <jsimmons@infradead.org>
Newsgroups	linux.kernel
Subject	Re: [PATCH 08/27] staging: lustre: avoid to use bio->bi_vcnt directly
Date	2016-04-10 18:10 +0200
Message-ID	<rmq5Y-7Fu-9@gated-at.bofh.it> (permalink)
References	<rkxOi-2K6-15@gated-at.bofh.it> <rkxXY-36a-31@gated-at.bofh.it> <rkyU1-3Ze-1@gated-at.bofh.it> <rmoGS-6sx-13@gated-at.bofh.it> <rmoQy-6wz-15@gated-at.bofh.it>
Organization	linux.* mail to news gateway

Show all headers | View raw

> On Sun, Apr 10, 2016 at 03:37:42PM +0100, James Simmons wrote:
> > 
> > > The lloop driver should be removed entirely - use the loop driver
> > > instead.
> > 
> > I talked with Andreas last week at our annual Lustre users group meeting 
> > about this. The reason I was told for existance is that some users were
> > using files on a Lustre file system with the loop back device. The 
> > performance was really bad at the time so a lloop was developed to 
> > overcome those limitations. Its been a long time so perhaps its time
> > to look at the default loop driver again to see if can perform now. If
> > it doesn't we will go the route of reworking the lloop driver in the
> > spirit of the cryptoloop device.
> 
> The loop driver now supports using AIO/DIO on any file systems that
> implements ->read_iter and ->write_iter. If lustre doesn't support
> those or doesn't have proper performance using them it should be
> addressed in the file system.
> 
> Note that the dio mode in the loop device is not the default and you
> need to manually enabled it, keep that in mind when testing.

This is excellent news. The only sad thing is that most lustre users
are running distros that use kernels before the AIO/DIO enhancements
were landed :-( We will have to keep a copy around for those guys. But
first I need to test the performance of the loop back driver this
week before this can be dropped.

Thread

[PATCH 00/27] block: cleanup direct access on .bi_vcnt & .bi_io_vec Ming Lei <tom.leiming@gmail.com> - 2016-04-05 14:00 +0200
  [PATCH 02/27] block: drbd: use bio_get_base_vec() to retrieve the 1st bvec Ming Lei <tom.leiming@gmail.com> - 2016-04-05 14:00 +0200
  [PATCH 05/27] block: pktcdvd: use bio_get_base_vec() to retrive bvec table Ming Lei <tom.leiming@gmail.com> - 2016-04-05 14:00 +0200
  [PATCH 04/27] block: loop: use bio_get_base_vec() to retrive bvec table Ming Lei <tom.leiming@gmail.com> - 2016-04-05 14:00 +0200
  [PATCH 03/27] block: drbd: remove impossible failure handling Ming Lei <tom.leiming@gmail.com> - 2016-04-05 14:00 +0200
    Re: [PATCH 03/27] block: drbd: remove impossible failure handling Lars Ellenberg <lars.ellenberg@linbit.com> - 2016-04-05 14:50 +0200
  [PATCH 11/27] bcache: io.c: use bio_set_vec_table Ming Lei <tom.leiming@gmail.com> - 2016-04-05 14:10 +0200
    Re: [PATCH 11/27] bcache: io.c: use bio_set_vec_table Christoph Hellwig <hch@infradead.org> - 2016-04-05 14:50 +0200
      Re: [PATCH 11/27] bcache: io.c: use bio_set_vec_table Ming Lei <tom.leiming@gmail.com> - 2016-04-05 17:30 +0200
        Re: [PATCH 11/27] bcache: io.c: use bio_set_vec_table Christoph Hellwig <hch@infradead.org> - 2016-04-05 19:40 +0200
      Re: [PATCH 11/27] bcache: io.c: use bio_set_vec_table Kent Overstreet <kent.overstreet@gmail.com> - 2016-04-06 02:40 +0200
  [PATCH 12/27] bcache: journal.c: use bio_set_vec_table() Ming Lei <tom.leiming@gmail.com> - 2016-04-05 14:10 +0200
  [PATCH 08/27] staging: lustre: avoid to use bio->bi_vcnt directly Ming Lei <tom.leiming@gmail.com> - 2016-04-05 14:10 +0200
    Re: [PATCH 08/27] staging: lustre: avoid to use bio->bi_vcnt directly Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-04-05 15:00 +0200
    Re: [PATCH 08/27] staging: lustre: avoid to use bio->bi_vcnt directly Christoph Hellwig <hch@infradead.org> - 2016-04-05 15:10 +0200
      Re: [PATCH 08/27] staging: lustre: avoid to use bio->bi_vcnt  directly James Simmons <jsimmons@infradead.org> - 2016-04-10 16:40 +0200
        Re: [PATCH 08/27] staging: lustre: avoid to use bio->bi_vcnt directly Christoph Hellwig <hch@infradead.org> - 2016-04-10 16:50 +0200
          Re: [PATCH 08/27] staging: lustre: avoid to use bio->bi_vcnt  directly James Simmons <jsimmons@infradead.org> - 2016-04-10 18:10 +0200
            Re: [PATCH 08/27] staging: lustre: avoid to use bio->bi_vcnt directly Ming Lei <tom.leiming@gmail.com> - 2016-04-11 05:40 +0200
  [PATCH 10/27] bcache: debug: avoid to access .bi_io_vec directly Ming Lei <tom.leiming@gmail.com> - 2016-04-05 14:10 +0200
  [PATCH 07/27] block: floppy: use bio_add_page() Ming Lei <tom.leiming@gmail.com> - 2016-04-05 14:10 +0200
  [PATCH 09/27] target: use bio_is_full() Ming Lei <tom.leiming@gmail.com> - 2016-04-05 14:10 +0200
    Re: [PATCH 09/27] target: use bio_is_full() Christoph Hellwig <hch@infradead.org> - 2016-04-05 15:10 +0200
      Re: [PATCH 09/27] target: use bio_is_full() Ming Lei <tom.leiming@gmail.com> - 2016-04-07 06:10 +0200
  [PATCH 14/27] bcache: writeback: use bio_set_vec_table() Ming Lei <tom.leiming@gmail.com> - 2016-04-05 14:10 +0200
  [PATCH 13/27] bcache: movinggc: use bio_set_vec_table() Ming Lei <tom.leiming@gmail.com> - 2016-04-05 14:10 +0200
  [PATCH 06/27] block: floppy: use bio_set_vec_table() Ming Lei <tom.leiming@gmail.com> - 2016-04-05 14:20 +0200
    Re: [PATCH 06/27] block: floppy: use bio_set_vec_table() Christoph Hellwig <hch@infradead.org> - 2016-04-05 15:10 +0200

csiph-web