On Wednesday, July 08, 2015 10:40:00 AM Alan Stern wrote: > On Wed, 8 Jul 2015, Pavel Machek wrote: > > > > well, that depends on what the purpose of the sync is supposed to be. > > > > > > If it is there to prevent users from corrupting their filesystems as a > > > result > > > of a mistake, it is insufficient. If it's there for other reasons, I'm > > > wondering > > > what those reasons are (on systems that suspend and resume reliably, > > > because the > > > original reason to put it in there was to reduce the damage from > > > suspend/resume > > > crashes). > > > > I put it there, and there were more reasons than "crashes" to put it > > there. > > > > 1) crashes. > > > > 2) battery is quite likely to run out in suspended machine. > > > > 3) if someone pulls the stick and puts it in other machine, I wanted > > consistent filesystem at least after journal replay. > > I was going to make the same points. > > From my point of view, whether to issue a sync is a tradeoff. I can't > remember any time in the last several years where lack of a sync would > have caused a problem for my computers, but the possibility still > exists. > > So on one hand, issuing the sync can help prevent a low-probability > problem. On the other hand, issuing the sync takes a small amount of > time (negligible for my purposes but not negligible for Len and > others). > > I prefer to pay a very small cost to prevent a low-probability problem. > Others may not want to pay, because to them the cost is larger or the > probability is lower. > > _That_ is the justification for not eliminating the sync completely but > making it optional.
Agreed. Thanks, Rafael -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
