ide-scsi problem in 2.4.6?!?

From: Nils Holland (
Date: Wed Jul 04 2001 - 15:03:05 EST

Hi folks,

there seems to be a problem in the 2.4.6. kernel, but unluckily, I have no
idea what it is. Seriously: Let me tell you the whole story:

I tend to burn a lot of CDs using my IDE CD-writer. In order to accomplish
that, I'm using cdrecord 1.11a04, of course in conjunction with the ide-scsi
emulation in the kernel. Now, up to 2.4.5 this has always worked fine - with
2.4.5 alone, I have probably burned a few dozen CDs without any problems.

Today I updated to 2.4.6. I did not update any other software on my computer,
only the kernel. When I now try to burn a CD with cdrecord, it fails in about
80% of all cases (yes, I've already wasted about 5 CDs trying to figure out
what's wrong).

What happens is the following: The burn process starts fine, but then,
suddenly, after about 24 MB have been written, the process suddenly stops,
giving me a bunch of SCSI error messages on the screen, including one that
tells me something about "Streaming lost..."

Now, I'm not a hard-core programmer, but I've still tried to figure out where
the problem lies. Obviously, since everything worked fine under 2.4.5, and
since the problem only occured after updating to 2.4.6, while everything else
on my machine remained the same, it seems as if the problem has something to
do with the kernel. What comes to my mine first is, of course, the ide-scsi
stuff. However, since I'm using a VIA-based Athlon system, I also wouldn't
exclude the possibility of this being another manifestation of the VIA-bug.

Currently, I seem to be able to work around the problem by turning the
BURN-Proof feature of my CD-burner on. I've never used it before, and what I
can see from the information cdrecord shows on the screen, the buffer is not
nearly empty when the burn process suddenly stops, but still BURN-Proof seems
to help. Just note that I never had any problems with kernels < 2.4.6 that
required me to use BURN-Proof.

I guess that if this is really a problem with the kernel (and it really seems
to be), then others will also see this problem. If so, I'd be glad to hear
about it. Again: I'm not a kernel-hacker, but I hope that if we're really
facing a serious problem here, enough information can be gathered so that
someone can fix it.


Nils Holland -
NightCastle Productions - Linux in Tiddische, Germany
"Give the heavens above more than just a passing glance
And when you get the choice to sit it out or dance - 
I hope you dance!"

- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to More majordomo info at Please read the FAQ at

This archive was generated by hypermail 2b29 : Sat Jul 07 2001 - 21:00:14 EST