| [Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] |
> -----Original Message----- > From: Guochun Shi [mailto:gshi@ncsa.uiuc.edu] > Sent: Monday, August 18, 2003 3:06 PM > To: opengfs-devel@lists.sourceforge.net > Subject: [ogfs-dev]scheduing ogfs_glockd > > > On docs/ogfs-locking:626 reads: > > "In addition, the cache handling code in glock.c can schedule > ogfs_glockd on an > as-needed basis to attempt to free up some G-Lock cache space. Note: > there are comments about this in new_glock(), but I can't > find where it > is actually done!" > > Do a simple grep will find glockd_process is waken up in > ogfs_clear_inode() src/fs/arch_linux_2_4/super_linux.c:811 : > wake_up_process(sdp->sd_glockd_process). > > It is also waken up on other two place when mount fails or > umount happens, super_linux.c:529, 900 Yes, I think my comments above were meant to be strictly within the scope of new_glock(), which does not seem to do what its comments (previously) said. Since writing this, I've changed the comments in new_glock(), and changed the ogfs-locking doc. I should probably add something about ogfs_clear_inode() ... thanks. BTW, it looks like you're reading an old copy of ogfs-locking. I've made many changes since then (based on new understandings), so you might want to grab a new copy, or look through the diffs in CVS. THANK YOU for reading and commenting! -- Ben -- Opinions are mine, not Intel's > > Let me know if I understand it wrongly. > > Thanks > Guochun > ------------------------------------------------------- This SF.Net email sponsored by: Free pre-built ASP.NET sites including Data Reports, E-commerce, Portals, and Forums are available now. Download today and enter to win an XBOX or Visual Studio .NET. http://aspnet.click-url.com/go/psa00100003ave/direct;at.aspnet_072303_01/01 _______________________________________________ Opengfs-devel mailing list Opengfs-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/opengfs-devel