Hi Luca, On Tue, Oct 20, 2015 at 08:32:50AM +0000, Luca Porzio (lporzio) wrote: > +Alex Lemberg > > > -----Original Message----- > > From: Luca Porzio (lporzio) > > Sent: Tuesday, 20 October, 2015 10:05 > > To: linux-f2fs-devel@lists.sourceforge.net > > Cc: Ulf Hansson (ulf.hans...@linaro.org); (alexey.skida...@sandisk.com); > > an...@tuxera.com; jaeg...@kernel.org > > Subject: REQ_META and Reliable write in eMMC > > > > Hi, > > > > By discussing in the MELS (Mobile & Embedded Linux Storage Forum) we are > > deciding whether it is ok to remove the bonding between REQ_META and > > "Reliable Write" in eMMC devices. > > We already got a first positive ack from FS developers here > > http://comments.gmane.org/gmane.linux.file-systems/97219 > > > > >This performance impact comes from the existing MMC SW driver solution > > where each IO request marked with > > >REQ_META is handled as a "Reliable Write" operation. > > >Pretty much as a as a FUA write access, a "Reliable write" means that the > > data will be written to the > > >non-volatile memory. > > >Due to the fact that every REQ_META is immediately bypassing the internal > > cache of eMMC device, write > > >performance is affected significantly. > > > > Wanted also to check for side effects on F2FS. > > Please advise how critical is to send REQ_META as "Reliable Write"?
As Chao mentioned, it's non-critical. F2FS sets REQ_META to give a hint where a bio contains FS metadata blocks to storage, which is not related to FS consistency. > > Can REQ_META be sent as a regular Write operation? Yup. Thanks, > > > > Thanks, > > Luca Porzio > > Micron ------------------------------------------------------------------------------ _______________________________________________ Linux-f2fs-devel mailing list Linux-f2fs-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
