Re: [PATCH v2] staging/lustre: lloop depends on BLOCK

2013-08-08 Thread Christoph Hellwig
On Wed, Aug 07, 2013 at 07:45:17AM +, Dilger, Andreas wrote: > The problem still remains that the kernel loop driver eventually depends on > a local block device for the pages/bios to be written. The Lustre lloop > driver bypasses the VFS and block layer to generate RPCs from the submitted >

Re: [PATCH v2] staging/lustre: lloop depends on BLOCK

2013-08-08 Thread Christoph Hellwig
On Wed, Aug 07, 2013 at 07:45:17AM +, Dilger, Andreas wrote: The problem still remains that the kernel loop driver eventually depends on a local block device for the pages/bios to be written. The Lustre lloop driver bypasses the VFS and block layer to generate RPCs from the submitted

Re: [PATCH v2] staging/lustre: lloop depends on BLOCK

2013-08-07 Thread Dilger, Andreas
On 2013/08/02 4:49 AM, "Christoph Hellwig" wrote: >On Thu, Aug 01, 2013 at 07:57:22PM +, Dilger, Andreas wrote: >> It provides significant performance improvement for network IO on >>Lustre. >> It bypasses DLM locking in Lustre and the VFS layer on the client, >>copying >> in the loop driver,

Re: [PATCH v2] staging/lustre: lloop depends on BLOCK

2013-08-07 Thread Dilger, Andreas
On 2013/08/02 4:49 AM, Christoph Hellwig h...@infradead.org wrote: On Thu, Aug 01, 2013 at 07:57:22PM +, Dilger, Andreas wrote: It provides significant performance improvement for network IO on Lustre. It bypasses DLM locking in Lustre and the VFS layer on the client, copying in the loop

Re: [PATCH v2] staging/lustre: lloop depends on BLOCK

2013-08-02 Thread Christoph Hellwig
On Thu, Aug 01, 2013 at 07:57:22PM +, Dilger, Andreas wrote: > It provides significant performance improvement for network IO on Lustre. > It bypasses DLM locking in Lustre and the VFS layer on the client, copying > in the loop driver, and page-by-page IO submission in the normal IO path.

Re: [PATCH v2] staging/lustre: lloop depends on BLOCK

2013-08-02 Thread Christoph Hellwig
On Thu, Aug 01, 2013 at 07:57:22PM +, Dilger, Andreas wrote: It provides significant performance improvement for network IO on Lustre. It bypasses DLM locking in Lustre and the VFS layer on the client, copying in the loop driver, and page-by-page IO submission in the normal IO path. Part

Re: [PATCH v2] staging/lustre: lloop depends on BLOCK

2013-08-01 Thread Greg Kroah-Hartman
On Fri, Aug 02, 2013 at 10:20:28AM +0800, Xiong Zhou wrote: > > > On Wed, 31 Jul 2013, Greg Kroah-Hartman wrote: > > > On Wed, Jul 31, 2013 at 10:30:41AM +0800, Xiong Zhou wrote: > > > From: Xiong Zhou > > > > > > First version of this patch makes LUSTRE_FS depends on BLOCK. Second > > >

Re: [PATCH v2] staging/lustre: lloop depends on BLOCK

2013-08-01 Thread Xiong Zhou
On Wed, 31 Jul 2013, Greg Kroah-Hartman wrote: > On Wed, Jul 31, 2013 at 10:30:41AM +0800, Xiong Zhou wrote: > > From: Xiong Zhou > > > > First version of this patch makes LUSTRE_FS depends on BLOCK. Second > > version makes only lloop depends on BLOCK with a config option for this > >

Re: [PATCH v2] staging/lustre: lloop depends on BLOCK

2013-08-01 Thread Dilger, Andreas
On 2013/08/01 2:45 AM, "Christoph Hellwig" wrote: >On Tue, Jul 30, 2013 at 08:29:51AM +0800, Xiong Zhou wrote: >> From: Xiong Zhou >> >> In the lustre client driver, lloop depends on BLOCK. Add an >> option for this dependence. Last version of this patch makes >> LUSTRE_FS depends on BLOCK. >>

Re: [PATCH v2] staging/lustre: lloop depends on BLOCK

2013-08-01 Thread Christoph Hellwig
On Tue, Jul 30, 2013 at 08:29:51AM +0800, Xiong Zhou wrote: > From: Xiong Zhou > > In the lustre client driver, lloop depends on BLOCK. Add an > option for this dependence. Last version of this patch makes > LUSTRE_FS depends on BLOCK. > Remove unnecessary jdb head files which depends on BLOCK.

Re: [PATCH v2] staging/lustre: lloop depends on BLOCK

2013-08-01 Thread Greg Kroah-Hartman
On Wed, Jul 31, 2013 at 10:30:41AM +0800, Xiong Zhou wrote: > From: Xiong Zhou > > First version of this patch makes LUSTRE_FS depends on BLOCK. Second > version makes only lloop depends on BLOCK with a config option for this > dependence, and remove unnecessary jdb header files which depends

Re: [PATCH v2] staging/lustre: lloop depends on BLOCK

2013-08-01 Thread Greg Kroah-Hartman
On Wed, Jul 31, 2013 at 10:30:41AM +0800, Xiong Zhou wrote: From: Xiong Zhou jencce.ker...@gmail.com First version of this patch makes LUSTRE_FS depends on BLOCK. Second version makes only lloop depends on BLOCK with a config option for this dependence, and remove unnecessary jdb header

Re: [PATCH v2] staging/lustre: lloop depends on BLOCK

2013-08-01 Thread Christoph Hellwig
On Tue, Jul 30, 2013 at 08:29:51AM +0800, Xiong Zhou wrote: From: Xiong Zhou jencce.ker...@gmail.com In the lustre client driver, lloop depends on BLOCK. Add an option for this dependence. Last version of this patch makes LUSTRE_FS depends on BLOCK. Remove unnecessary jdb head files which

Re: [PATCH v2] staging/lustre: lloop depends on BLOCK

2013-08-01 Thread Dilger, Andreas
On 2013/08/01 2:45 AM, Christoph Hellwig h...@infradead.org wrote: On Tue, Jul 30, 2013 at 08:29:51AM +0800, Xiong Zhou wrote: From: Xiong Zhou jencce.ker...@gmail.com In the lustre client driver, lloop depends on BLOCK. Add an option for this dependence. Last version of this patch makes

Re: [PATCH v2] staging/lustre: lloop depends on BLOCK

2013-08-01 Thread Xiong Zhou
On Wed, 31 Jul 2013, Greg Kroah-Hartman wrote: On Wed, Jul 31, 2013 at 10:30:41AM +0800, Xiong Zhou wrote: From: Xiong Zhou jencce.ker...@gmail.com First version of this patch makes LUSTRE_FS depends on BLOCK. Second version makes only lloop depends on BLOCK with a config option for

Re: [PATCH v2] staging/lustre: lloop depends on BLOCK

2013-08-01 Thread Greg Kroah-Hartman
On Fri, Aug 02, 2013 at 10:20:28AM +0800, Xiong Zhou wrote: On Wed, 31 Jul 2013, Greg Kroah-Hartman wrote: On Wed, Jul 31, 2013 at 10:30:41AM +0800, Xiong Zhou wrote: From: Xiong Zhou jencce.ker...@gmail.com First version of this patch makes LUSTRE_FS depends on BLOCK. Second

Re: [PATCH v2] staging/lustre: lloop depends on BLOCK

2013-07-30 Thread Xiong Zhou
On Tue, 30 Jul 2013, Peng Tao wrote: > On Tue, Jul 30, 2013 at 8:29 AM, Xiong Zhou wrote: > > From: Xiong Zhou > > > > In the lustre client driver, lloop depends on BLOCK. Add an > > option for this dependence. Last version of this patch makes > > LUSTRE_FS depends on BLOCK. > > Remove

Re: [PATCH v2] staging/lustre: lloop depends on BLOCK

2013-07-30 Thread Peng Tao
On Tue, Jul 30, 2013 at 8:29 AM, Xiong Zhou wrote: > From: Xiong Zhou > > In the lustre client driver, lloop depends on BLOCK. Add an > option for this dependence. Last version of this patch makes > LUSTRE_FS depends on BLOCK. > Remove unnecessary jdb head files which depends on BLOCK. > Thanks

Re: [PATCH v2] staging/lustre: lloop depends on BLOCK

2013-07-30 Thread Peng Tao
On Tue, Jul 30, 2013 at 8:29 AM, Xiong Zhou jencce.ker...@gmail.com wrote: From: Xiong Zhou jencce.ker...@gmail.com In the lustre client driver, lloop depends on BLOCK. Add an option for this dependence. Last version of this patch makes LUSTRE_FS depends on BLOCK. Remove unnecessary jdb head

Re: [PATCH v2] staging/lustre: lloop depends on BLOCK

2013-07-30 Thread Xiong Zhou
On Tue, 30 Jul 2013, Peng Tao wrote: On Tue, Jul 30, 2013 at 8:29 AM, Xiong Zhou jencce.ker...@gmail.com wrote: From: Xiong Zhou jencce.ker...@gmail.com In the lustre client driver, lloop depends on BLOCK. Add an option for this dependence. Last version of this patch makes LUSTRE_FS

[PATCH v2] staging/lustre: lloop depends on BLOCK

2013-07-29 Thread Xiong Zhou
From: Xiong Zhou In the lustre client driver, lloop depends on BLOCK. Add an option for this dependence. Last version of this patch makes LUSTRE_FS depends on BLOCK. Remove unnecessary jdb head files which depends on BLOCK. Signed-off-by: Xiong Zhou --- drivers/staging/lustre/lustre/Kconfig

[PATCH v2] staging/lustre: lloop depends on BLOCK

2013-07-29 Thread Xiong Zhou
From: Xiong Zhou jencce.ker...@gmail.com In the lustre client driver, lloop depends on BLOCK. Add an option for this dependence. Last version of this patch makes LUSTRE_FS depends on BLOCK. Remove unnecessary jdb head files which depends on BLOCK. Signed-off-by: Xiong Zhou