Re: cache modes in LIO

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Hi Manuel & Valambal,

Apologies for the delayed response, comments are below.

On Thu, 2014-07-17 at 12:42 +0000, Manuel Buil wrote:
> Hi,
> 
> One newby question: I create a iSCSI target using my local disk
> device /dev/sdb (iBLOCK type of backstore). Is there a way in the CLI
> to specify the cache mode that LIO uses when opening the /dev/sdb
> device (O_SYNC or O_DIRECT or...)?
> 

So IBLOCK backstores reference the underlying struct block_device
directly and submit struct bio, which means there is no struct file
involved to signal O_SYNC or O_DIRECT operation.

FILEIO backstores default to O_DSYNC usage for safety reasons, and
enabling 'buffered' mode at FILEIO backstore creation time will
explicitly disable O_DSYNC, making all WRITEs hit buffer cache and
acknowledging back to the initiator.  Buffered mode obviously has
implications for data integrity during target power loss, so for real
production systems you'll want to considering using something like
bcache in write-back mode with IBLOCK backstores, to ensure that all
writes hit persistent storage (flash SSD) before acknowledging back to
the initiator.

Also note that O_DIRECT is not supported by FILEIO in the kernel
backstore, as there is a long standing limitation of fs/direct-io.c only
supporting user pages.  Eventually a userspace backend will be supported
for this, and/or the limitation in fs/direct-io.c will be addressed.

--nab

--
To unsubscribe from this list: send the line "unsubscribe target-devel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html




[Index of Archives]     [Linux SCSI]     [Kernel Newbies]     [Linux SCSI Target Infrastructure]     [Share Photos]     [IDE]     [Security]     [Git]     [Netfilter]     [Bugtraq]     [Yosemite News]     [MIPS Linux]     [ARM Linux]     [Linux Security]     [Linux RAID]     [Linux ATA RAID]     [Linux IIO]     [Device Mapper]

  Powered by Linux