On Thu 8 Oct 2009 17:14, Mike Frysinger pondered: > On Thu, Oct 8, 2009 at 10:45, Robin Getz wrote: > > On Fri 17 Jul 2009 06:32, [email protected] pondered: > >> > >> Fix bug[#5145] <http://blackfin.uclinux.org/gf/tracker/5145> Walk > >> around sport rx 32-bit reading underflow error. > >> > >> Disable local interrupt to walk around the fake RX underflow error > >> when do 32-bit reading from RX fifois interrupted by other interrupt. > >> This is a possible hardware anomaly. After it is logged officially, > >> the official anomaly id should be listed bellow. > >> > >> 16-bit reading is not affected. > > > > This has been verified as a new anomaly (number will be 05-00-0473) - > > currently effects all parts.... > > > > The workaround that Sonic discovered is the recommended workaround. > > > > Details to follow after a few more tests are run. > > i'm guessing it isnt related to the interrupt controller, it's just > interrupted reads. in other words, the anomaly text seems to focus on > "when an interrupt occurs" and not "if the access is interrupted". > the latter version allows for all async events including exceptions > and NMI.
Design is looking into things, and is currently getting back to us... _______________________________________________ Linux-kernel-commits mailing list [email protected] https://blackfin.uclinux.org/mailman/listinfo/linux-kernel-commits
