21 Jun 2006 03:12, Alan Cox wrote: > Ar Maw, 2006-06-20 am 15:19 -0700, ysgrifennodd Andrew Morton: > > hm. This seems... rather serious. I assume that in most cases, the > > firmware rev which we have in the table (eg "24.09P07") is a full-string > > match for the string which the drive returned. > > They are full matches as far as I can see so it should be ok, plus the > DMA blacklist is mostly hardware that went out ten or more years ago. > Even if it did mis-trigger we'd be blacklisting extra firmware revs of > iffy hardware so that isn't do worrying. No, they are *not* full matches. My case: CF model="TRANSCEND", revision="20050811" as reported by "hdparm -I" and bios, but id->model="TRANSCEND", id->fw_rev="20050811TRANSCEND" note the trailing in id->fw_rev, So lets look at ide_in_drive_list -- it should try to detect blacklisted revision, so it looks for substr device_table->id_firmware in id->fw_rev. NB: I don't know what causes id->fw_rev to be bogus (?), but maybe the right fix NB: is to correct its initialization. Some person with IDE background should clarify this. -- Kirill - : send the line "unsubscribe linux-ide" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
