lkml.org 
[lkml]   [1999]   [Mar]   [4]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: Maybe Disk Problem.


On Tue, 2 Mar 1999, Pavel Ravits wrote:

> "wish to be personally CC'ed the answers/comments posted to the list
> in response to my posting".
>
>
> +++++++++ Why nobody answers me - I'm not :|nice:| enough ? !!!!!
> ++++++++++++++++
>

Perhaps the list was slow.. My messages sometimes appear after quite a
while, occasionally even days.

>
> Hi,
>
> Please answer,
> (3nd message);
>
>
> Steve Dodd wrote:
> >
> > Hi,
> >
> > On Tue, Feb 23, 1999 at 09:59:54AM +0200, Pavel Ravits wrote:
> >
> > > Hello, ( I'm sorry for the length )
> >
> > One of the shorter posts to l-k, actually :)
> >
> > > sig11 with kernel compiling,
> > > crc error on opening archives
> > > error on casual copying of files.
> >
> > The knee-jerk reaction here is 'memory problems', but maybe not. If you restart
> > make after a sig11 kernel compile problem, does it blow up again straight away
> > at the same place, or does it run on for a bit then blow up somewhere else?
>
> It doesn't blow up straight away, but more chances that on the next make
> will be sig11.
>
> > > My guess all the time was that the new kernels support UDMA and my
> > > disk don't like it, and today I proved it
> > > when installed the 35 kernel on old disk and everything was fine.
> >
> > Was the old disk in the same machine? (If so, I guess that eliminates the idea
> > of memory problems).
>
> Not only the old disk was on the same computer, but I worked until now
> on
> 2.0.29 flawlessly but in slow speed of the disk
>
> > 2.2.x allow you to specify if (U)DMA should be used in the kernel config; I
> > don't know off-hand how 2.0.x decides whether to enable DMA on boot up or not.
> > You're not forcing DMA on with hdparm, are you?
>
> I'm not forcing DMA with hdparm and I don't know how to do it, but as
> you said here
> I indeed found option in 2.2.1 to turn off DMA ( but not UDMA ; What
> mode
> is it working
> now anyway :)) I turned it off, compiled the kernel 80 times with no
> problems.
>

This means your disk isn't compatible with UDMA and/or your motherboard.
Actually noting you have P2 your motherboard probably uses a standard
Intel chipset so it must be due to the disk. I have also heard in the past
some stuff about (some) Quantum disks not doing UDMA well.

UDMA behaves in much the same manner as DMA - at least from kernel point
of view, it's just faster.

You should also check for the following:
IDE cables - is it long ? Is your drive plugged in the middle or in the
back ? Are there other devices on this cable ? Try reconfiguring them.
For example if you have a cd-rom or zip disk on the same cable try to see
what happens if you disconnect them.

Also your video problems are indicative of some noise going around
inside computer case - you might cure it by repositioning your video card
and checking that no cables are touching it. Also check grounding and see
if it(problem) dissappears if you remove one of the devices (usually
cd-rom) - just in case you have a power shortage.

I know all this above sounds a bit strange - but I have seen a _lot_ of
computers made by well-known manufacturers that had above problems
(though power problems usually relate to computers assembled by users
themselves).

Vladimir Dergachev


> > > What do you think - how can I prove that this is a disk problem
> > > so I can exchange it?
> >
> > Well, I reckon if toggling DMA on using hdparm brings the problems straight
> > back, that's a pretty clear indicator. More important than getting your disk
> > exchanged though, is reporting the model type here (hdparm -i, hdparm -I) so
> > it can be added to the DMA black list if necessary.
>
> Outputs of hdparm running 2.2.1 are attached.
> The question being raised is How is it possible that this is a known
> model,
> known manufacturer, the disk is ok - EXCEPT the DMA thing.
> And from here- Is buying the same thing safe or with change of size
> from 5.1GB to 6GB or with QUANTUM --> WD, and generally - is there a way
> to prove the disk being bad with some win95 ( i hate to say this word)
> program.
>
> > Oh, what motherboard do you have? (VIA?)
>
> My configuration is:
> P2 266 + 64MB + 5.1 Quantum
> Motherboard: SP-P2LXC ( i don't have the company name).
>
> As I thinking now It may be problem with the Motherboard controller,
> so I don't know what to do, help please I can't goon with speed anymore,
> (and the `funny` thing I suspect in not a hardware problem.)
>
> > S.
> >
> > --
> > "His mind is like a steel trap -- full of mice"
>
> by the way - I have this distortion of the screen every time I logining
> one of consoles
> or starting mc for example - minor problem but annoying. -- what can be
> done ? --
>
> Thanks for your help,
> Pavel.
>
>
>
>
>
> /dev/hda:
>
> Model=UQNAUT MIFERABLLE 5L1. A , FwRev=0A.81100,
> SerialNo=43856117
> Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs }
> RawCHS=10602/15/63, TrkSize=32256, SectSize=21298, ECCbytes=4
> BuffType=3(DualPortCache), BuffSize=418kB, MaxMultSect=16, MultSect=off
> DblWordIO=no, maxPIO=2(fast), DMA=yes, maxDMA=2(fast)
> CurCHS=10602/15/63, CurSects=10018890, LBA=yes, LBAsects=10018890
> tDMA={min:120,rec:120}, DMA modes: sword0 sword1 sword2 mword0 mword1
> mword2
> IORDY=on/off, tPIO={min:120,w/IORDY:120}, PIO modes: mode3 mode4
>
>
>
>
>
>
> /dev/hda:
>
> Model=QUANTUM FIREBALL EL5.1A, FwRev=A08.1100, SerialNo=34581671
> Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs }
> RawCHS=10602/15/63, TrkSize=32256, SectSize=21298, ECCbytes=4
> BuffType=3(DualPortCache), BuffSize=418kB, MaxMultSect=16, MultSect=off
> DblWordIO=no, maxPIO=2(fast), DMA=yes, maxDMA=2(fast)
> CurCHS=623/255/63, CurSects=10018890, LBA=yes, LBAsects=10018890
> tDMA={min:120,rec:120}, DMA modes: sword0 sword1 sword2 mword0 mword1
> mword2
> IORDY=on/off, tPIO={min:120,w/IORDY:120}, PIO modes: mode3 mode4
>
> -
> 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/
>


-
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/

\
 
 \ /
  Last update: 2005-03-22 13:50    [W:0.704 / U:0.052 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site