[PATCH] Workaround for 745x data corruption bug

Mark A. Greer mgreer at mvista.com
Tue Aug 3 04:20:30 EST 2004

Adrian Cox wrote:

>Recently released errata documents show a new bug in all 745x family
>processors. This can cause data corruption when memory is mapped
>non-coherent and one of these conditions is true:
>1) L2 hardware prefetch is enabled (as it is in Linux)
>2) instructions and data are fetched from the same or adjacent cache
>The attached patch adds a workaround, by setting CPU_FTR_NEED_COHERENT
>on all 745x processors.

Well that sucks (the bug, not the patch :).  Many people like to turn
off coherency when using Marvell host bridges b/c they struggle
performance-wise with coherency on (at least on some versions).

One change to the patch, though.  According to the 7447/7457 errata doc,
rev 1.2 doesn't have the bug.  The attached patch accounts for that.

-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: coherent2.patch
URL: <http://lists.ozlabs.org/pipermail/linuxppc-dev/attachments/20040802/2d11a7a6/attachment.txt>

More information about the Linuxppc-dev mailing list