search for: aio

On Tue, Jul 19, 2016 at 12:42:53PM -0700, Jeremy Allison wrote: > On Tue, Jul 19, 2016 at 12:49:09PM -0500, Russell R Poyner wrote: > > I'm tuning a samba 4.3 install on freebsd and I'm confused about aio > > settings. > > > > I've loaded the freebsd aio kernel module and tried various values > > or aio read size and aio write size, but it seems to make no > > difference in the speed. > > > > Using MS diskspd against a samba share from a fast zfs pool...

I'm tuning a samba 4.3 install on freebsd and I'm confused about aio settings. I've loaded the freebsd aio kernel module and tried various values or aio read size and aio write size, but it seems to make no difference in the speed. Using MS diskspd against a samba share from a fast zfs pool I get something like 25MB/s tops. That's well below the capaci...

Hello, I'm wondering if anyone out there is doing tap:aio disk devices on a Xen HVM vm with any success. Please let me know if you are, and perhaps your version info and a config file snippet. I've posted this to the xen-users list, and what I'm trying to do seems to be correct, works on Ubuntu, Debian, and XenSource; but it doesn't work on...

...c at evilplan.org> V3 changes: - add Cc: stable at vger.kernel.org in the patch header to align with stable rules - add Acked-by: Jeff Moyer <jmoyer at redhat.com> V2 changes: - update the patch header of the first patch to make it more clear. This patch list fixes an issue about ocfs2 aio/dio write process hang. The call trace is like this: @ cat /proc/15913/stack @ [<ffffffffa06e1085>] ocfs2_aiodio_wait+0x85/0xc0 [ocfs2] @ [<ffffffffa06e3e39>] ocfs2_file_aio_write+0x99/0xa0 [ocfs2] @ [<ffffffff811ace97>] aio_rw_vect_retry+0x87/0x150 @ [<ffffffff811aebd1>] a...

I'm looking into trying out AIO on one of our file servers in the office but I can't find any info on what the values for "aio read size" and "aio write size" correspond to. Is it a buffer size of some sort or a tuning type parameter were it allows the socket to read/write x blocks? Thanks, David Miller

hi, i just have some questions about specific (uncommon) compile options and aio-settings in smb.conf: Asynchronous IO Support ======================= "Experimental support for async IO has been added to smbd for certain platforms. To enable this new feature, Samba must be compiled to include the --with-aio-support configure option. In addition, the "aio read size&...

...lo, I''m using Xen (installed from source - "hg clone http://xenbits.xen.org/xen-4.0-testing.hg") on Debian Lenny. Dom0 kernel is 2.6.32-bpo.5-xen-686. Until now I have successfully ran the Lucid Ubuntu as domU guest, defining disks as (and also as file:): disk = [''tap2:aio:/home/xen/domains/lucid/disk.img,xvda1,w'', ''tap2:aio:/home/xen/domains/lucid/swap.img,xvda2,w''] However, when I try "disk = [''tap2:remus:backup_ip:9000|aio:/home/xen/domains/lucid/disk.img,xvda1,w'', ''tap2:remus:backup_ip:9000|aio:/home/xen...

Hi folks, This patchset adds vhost-blk support. vhost-blk is a in kernel virito-blk device accelerator. Compared to userspace virtio-blk implementation, vhost-blk gives about 5% to 15% performance improvement. Asias He (5): aio: Export symbols and struct kiocb_batch for in kernel aio usage eventfd: Export symbol eventfd_file_create() vhost: Make vhost a separate module vhost-net: Use VHOST_NET_FEATURES for vhost-net vhost-blk: Add vhost-blk support drivers/vhost/Kconfig | 20 +- drivers/vhost/Makefile | 6...

Hi folks, This patchset adds vhost-blk support. vhost-blk is a in kernel virito-blk device accelerator. Compared to userspace virtio-blk implementation, vhost-blk gives about 5% to 15% performance improvement. Asias He (5): aio: Export symbols and struct kiocb_batch for in kernel aio usage eventfd: Export symbol eventfd_file_create() vhost: Make vhost a separate module vhost-net: Use VHOST_NET_FEATURES for vhost-net vhost-blk: Add vhost-blk support drivers/vhost/Kconfig | 20 +- drivers/vhost/Makefile | 6...

Maybe it''s something I''m doing wrong, but I have a 15gb image that I use with disk = [ "tap:aio:/root/images/vista32.img,hda,w" ] It was working fine on Friday, but today it failed to boot with an error message. The same works fine when I replace "tap:aio:" with "file:". It seems like qemu-dm tries to open the file with the "aio:" still attached - judging...

One more data point for comparison I installed the stock samba 4.2 rpm on a centos 7 machine and ran the same diskspd tests against a share configured with: vfs objects = aio_pthread aio read size = 1024 aio read size = 1024 smb2 leases = yes I get 27MB/s with 4k blocks and 145MB/s with 64k blocks. Disabling cacheing by passing the -h switch to diskspd lowered these to 72MB/s and 11MB/s. Which I view as 'close enough' to wire speed. Thus it seems...

hi, i'm trying to get the asynchronous file transfer to work on a FC4 for the latest samba verion (3.0.20b). The things i've done: ./configure --with-aio-support ... included "aio read size = 1" and "aio write size = 1" in the smb.conf file What did i forget?? michiel ========================= uname -a Linux wired11 2.6.12-1.1398_FC4 #1 Fri Jul 15 00:52:32 EDT 2005 i686 athlon i386 GNU/Linux here's my smb.conf: [global...

Hello list, I setup Iometer to test AIO for 100% random read. If "Transfer Request Size" is more than or equal to 256 kilobytes,in the beginning the transmission is good. But 3~5 seconds later,the throughput will drop to zero. Server OS: Ubuntu Server 14.04.1 LTS Samba: Version 4.1.6-Ubuntu Dialect: SMB 2.0 AIO settings : a...

I have never been able to get tap:aio devices running. When I try it in a Linux PV domain, it says: XENBUS: Waiting for devices to initialise: 295s... And they don''t work in my GPLPV drivers, which is the problem I am trying to solve, but I''d like to see them work under Linux first. Is there anything extra I need t...

Create a class for AIO buffers. This is a mostly neutral code refactoring, but we add a convenient function for getting the size of the buffer (same as previous commit for OCaml). --- generator/generator | 111 +++++++++++++++++------------ python/handle.c | 26 +++++-- python/t/500-a...

This doesn't yet work. However it does make me more convinced than ever that we really need to sort out persistent buffer lifetimes in the library (similar to what we did for closures). Rich.

I'm trying to install samba-3.0.20 on aix 5.2, and I don't want to use aio, but the build process is compiling it in anyway. By default, it isn't supposed to build in aio support, but it does. When I pass --with-aio-support=no to configure, it also builds in support. It builds in aio support whether I use gcc or xlc. Here's what ldd shows: # ldd bin/smbd b...

Hi list, I have been trying to enable AIO on ARMv5 Samba built with kernel headers 2.6.31.8, the configure indicated that AIO is supported via kernel headers and all went smooth (regarding the build process). Once I added the smb.conf 'aio write size=1' and tried to perform a file copy to the Samba share I noticed that 32K of the...

Hi. I've been recently looking into asynchronous feature and currently stuck with few stuff that I would appreciate to get some help with. 1. I tried turning on the feature by setting 'server multi channel support = yes', 'aio read size = 1' and 'aio write size = 1' and I couldn't figure out whether it was really working or not. Is there any way I can verify this? 2. Is there anything else I have to do except configuring smb.conf to turn on aio? To what I have googled, it says that you have to compile wi...

Anyone who wants to do copy to/from user from a kernel thread, needs use_mm (like what fs/aio has). Move that into mm/, to make reusing and exporting easier down the line, and make aio use it. Next intended user, besides aio, will be vhost-net. Acked-by: Andrea Arcangeli <aarcange at redhat.com> Signed-off-by: Michael S. Tsirkin <mst at redhat.com> --- fs/aio.c...