Re: Can O_SYNC be implemented by using fsync?

From: Bill Wendling (wendling@ganymede.isdn.uiuc.edu)
Date: Mon May 15 2000 - 16:28:09 EST


Also sprach Andi Kleen:
} On Mon, May 15, 2000 at 02:16:41PM -0600, Jeff V. Merkey wrote:
} >
} >
} > Bill Wendling wrote:
} > >
} > > Also sprach Pavel Machek:
} > >
} > > }
} > > I wouldn't count on that. If my failed patches were any indication, it's
} > > heavily used by apps like Netscape...
} >
} > Why the hell is Netscape doing this? Netscape is a browser/mail
} > program, not a multi-user database, or are you refering to the server
} > side of NetScape? This doesn't sound right! Why would Netscape need to
} > write-through files all the time -- for browser failover?
}
} Netscape is a mail client. For mailers it makes sense to flush
} mailbox updates to disk. It seems to use a quite complicated
} database for its mail store.
}
} Also I guess fsync()ing bookmarks updates is useful.
}
Maybe it wants cache integrity? Dunno...I simply got VERY poor
performance when I did an ll_rw_block() call if the O_SYNC flag was set
on the file pointer.

-- 
|| Bill Wendling			wendling@ganymede.isdn.uiuc.edu

- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/



This archive was generated by hypermail 2b29 : Mon May 15 2000 - 21:00:26 EST