On Sunday 15 January 2006 18:10, Bob Hynes wrote: >Kent West wrote: >>(Bob; let me encourage you to keep conversations on the list rather >> than private replies; this way, others who might be helpful can >> help; and others who are looking for the same answers can find the >> answers both on the list and later in the archives.) >> >>Bob Hynes wrote: >>>Kent West wrote: >>>>Bob Hynes reports a slow Debian setup. >>>> >>>>My first suspicion would be that DMA is not turned on. What's the >>>>results of running "hdparm /dev/hda" (assuming your setup is >>>>installed on /dev/hda)? >>> >>>Well, I ran that command on konsole, but got "no such file or >>>directory. I did install it earlier, so I know it's there. When I >>> run "hdparm" I get a bunch of switch options, so it's installed. >> >>The only explanation I might have is that when you got the "no such >> file or directory", you did it as a non-root user, whereas you ran >> it as root when you got the switch options. Or perhaps you >> misspelled hdparm the first time. Or /sbin is not in your path. Or >> you were in the /sbin directory the first time, in which case you >> would have wanted to type "./hdparm /dev/hda". >> >>>Can't say I know the file structure or the KDE interface well enough >>>right now to know where something gets install by default, or where >>>hdparm got installed, but if the command responds with a list of >>>switches, I'm assuming there is something I can do without running >>> it with the /dev/hda path. >> >>"locate hdparm" (or "locate [whatever]") is a handy tool to use for >>locating a file. >> >>"which hdparm" should find the first executable instance of hdparm. >> (If you have more than one copy in different locations, the command >> travels down your PATH setting until it finds an executable copy. >> The file must be executable (permissions, path settings, etc) by the >> same user that is running the "which" command, or you'll get >> different results.) >> >>>(Interestingly enough, the file "hda" exists at that directory level >>>as a "0 B" file...not sure how big the file is supposed to be.) >> >>Like so: >>nesa[westk]:/home/westk/Desktop> ls -l /dev/hda >>brw-rw---- 1 root disk 3, 0 Jul 30 2004 /dev/hda >> >>The "/dev/hda" file you've found is a "device file", reffering to a >>"block" device; that's what the "b" means at the start of the line. >> This type of device file refers to something like a hard drive or a >> cdrom drive or something else that moves data in blocks. Items that >> move data in single "characters" at a time, like a modem, would have >> a "c" at the start of the line. "d" means directory; "-" means >> normal file; "l" means link; etc. >> >>In *nix, "everything is a file". What this means in this context is >> that when your system accesses your hard drive, it's really doing so >> through this device file. >> >>But back to your issue. You can try "/sbin/hdparm /dev/hda" instead >> of "hdparm /dev/hda"; this will specify the path to the hdparm >> utility. >> >>The "mount" command should result in something like this: >> >>nesa[westk]:/home/westk/Desktop> mount >>/dev/hda8 on / type ext3 (rw,errors=remount-ro) >>proc on /proc type proc (rw) >>sysfs on /sys type sysfs (rw) >>devpts on /dev/pts type devpts (rw,gid=5,mode=620) >>tmpfs on /dev/shm type tmpfs (rw) >>/dev/hda9 on /usr type ext3 (rw) >>/dev/hda6 on /var type ext3 (rw) >>/dev/hda7 on /home type ext3 (rw) >>/dev/hda1 on /tmp type ext3 (rw) >>/dev/hda5 on /spare type ext3 (rw) >>usbfs on /proc/bus/usb type usbfs (rw) >> >> >>This indicates that my system is installed to /dev/hda. If you had >> your system spread over several drives (which you'd have a hard time >> doing in Windows - wow, that's a lame OS in many ways), you might >> see references to /dev/hdb or /dev/hdc, etc. In fact, your CDROM is >> likely on /dev/hdc. If you have a SCSI system instead of an IDE >> system (not likely for your average Dell or Gateway or locally-built >> whitebox computer), you might see things like /dev/sda or /dev/sdb. >> This might also be the case if you mounted a USB keyfob or camera, >> etc, as they often are mounted as a "simulated" SCSI device. >> >>hda refers to the master drive on the first IDE port of your mother >> board; hdb refers to the slave on the first IDE port. >>hdc refers to the master on the second IDE port; >>hdd refers to the slave on the second IDE port. >> >>A number after these designations refers to the partition. Most >> newbies probably only have one big partition instead of multiple >> partitions like I have above, so you might only see something like >> "/dev/hda1 on /". >> >>If your system says something other than /dev/hda for most of your >>mounts, then your "hdparm" command needs to refer to that (those?) >>drive(s) instead of /dev/hda. And remember, hdparm works on drives >>(hda), not partitions (hda1). > >Thanks for your help, this is what I got from that command: > > >/dev/hdc: > multcount = 0 (off) > IO_support = 0 (default 16-bit) > unmaskirq = 0 (off) > using_dma = 1 (on) > keepsettings = 0 (off) > readonly = 0 (off) > readahead = 256 (on) > geometry = 26548/16/63, sectors = 26760384, start = 0 > >So I'm running in default 16-bit, I take it. I tried using the -c > switch to change it to 32-bit, but it just displays > >/dev/hdc: > IO_support = 0 (default 16-bit) > >how do I "set" the parameter to 32, rather than just displaying the >current setting of 16-bit? Sorry, I'm super Linux dumb.
The -c takes an argument to change it, otherwise its just a query operation. 0 = 16 bit, 1, 2, & 3 set other variations of the 32 bit transfer modes. Man hdparm is your friend. -- Cheers, Gene People having trouble with vz bouncing email to me should add the word 'online' between the 'verizon', and the dot which bypasses vz's stupid bounce rules. I do use spamassassin too. :-) Yahoo.com and AOL/TW attorneys please note, additions to the above message by Gene Heskett are: Copyright 2005 by Maurice Eugene Heskett, all rights reserved. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
