On 5 June 2017 at 21:55, Rafael J. Wysocki <rafael@xxxxxxxxxx> wrote: > On Mon, Jun 5, 2017 at 10:51 PM, Linda Knippers <linda.knippers@xxxxxxx> wrote: >> On 6/5/2017 4:42 PM, Rafael J. Wysocki wrote: >> <snip> >> >>>> I talked to our FW team and we do generate checksums and not a zero for >>>> at >>>> least some >>>> of the AML. Please revert this change until you can also validate >>>> a checksum. Or shall I post a patch to remove the check? >>> >>> >>> I'll skip this patch, no need to do anything else. Thanks for your >>> report! >> >> >> Thanks. This patch is already in as of 4.12rc1 and not part of the most >> recent >> ACPICA drop. > > Ah, OK. > > I should have checked I guess. :-) > > Anyway, I'll revert it, then. > >>> Bob, can you revert this upstream, please? It looks like the >>> assumption it is based on doesn't hold. >> >> Bob, I'm happy to test something if there is a new patch that looks for zero >> or a valid checksum. TBH, I'm not 100% certain that our checksums are >> correct because nothing has ever tried to verify them. > > Well, that's part of the problem here I guess. If they have never > been tested, they cannot be trusted. > > Still, the commit in question clearly assumed that value to always be > 0 and it clearly is not the case here. > Just stumbled across this on the moonshot m400 and breaks the parsing of the PNP0A08 _CRS on that platform. Graeme -- To unsubscribe from this list: send the line "unsubscribe linux-acpi" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
