one-line-patch against SCSI-Read-Error-BUG()

From: Ralf Oehler (R.Oehler@GDAmbH.com)
Date: Tue Feb 05 2002 - 09:32:10 EST


Hi, List

I think, I found a very simple solution for this annoying BUG().

Since at least kernel 2.4.16 there is a BUG() in pci.h,
that crashes the kernel on any attempt to read a SCSI-Sector
from an erased MO-Medium and on any attempt to read
a sector from a SCSI-disk, which returns "Read-Error".

There seems to be a thinko in the corresponding code, which
does not take into account the case where a SCSI-READ
does not return any data because of a "sense code: read error"
or a "sense code: blank sector".

I simply commented out this BUG() statement (see below)
and everything worked well from there on. The BUG()
seems to be inadequate.

Please could you check if I'm right and apply this change to the
current kernel? Again I want to stress that in my
oppinion it is dangerous for the stability of a production
machine if it crashes on the next SCSI-Read-Error it sees.
Most SCSI-Hardware today shows very few read errors, but
please let's not rely on that. That'd be playing vabanque.

If there are tests to do, I can offer my time and hardware
(SCSI-MO drives and media with various sector sizes) to
test and in case to provide stack traces.

Regards,
        Ralf

include/asm/pci.h:
static inline int pci_map_sg(struct pci_dev *hwdev, struct scatterlist *sg,
                             int nents, int direction)
{
        int i;

        if (direction == PCI_DMA_NONE)
                BUG();
 
        /*
         * temporary 2.4 hack
         */
        for (i = 0; i < nents; i++ ) {
                if (sg[i].address && sg[i].page)
                        BUG();
    --------> else if (!sg[i].address && !sg[i].page)
    --------> BUG();
 
                if (sg[i].address)
                        sg[i].dma_address = virt_to_bus(sg[i].address);
                else
                        sg[i].dma_address = page_to_bus(sg[i].page) + sg[i].offset;
        }
 
        flush_write_buffers();
        return nents;
}

 --------------------------------------------------------------------------
| Ralf Oehler
|
| GDA - Gesellschaft fuer Digitale _/
| Archivierungstechnik mbH & CoKG _/
| Ein Unternehmen der Bechtle AG #/_/_/_/ _/_/_/_/ _/_/_/_/
| _/ _/ _/ _/ _/
| E-Mail: R.Oehler@GDAmbH.com _/ _/ _/ _/ _/ _/
| Tel.: +49 6182-9271-23 _/_/_/_/ _/_/_/#/ _/_/_/#/
| Fax.: +49 6182-25035 _/
| Mail: GDA, Bensbruchstraße 11, _/_/_/_/
| D-63533 Mainhausen
| HTTP: www.GDAmbH.com
 --------------------------------------------------------------------------

time is a funny concept
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/



This archive was generated by hypermail 2b29 : Thu Feb 07 2002 - 21:00:42 EST