View Issue Details
|ID||Project||Category||View Status||Date Submitted||Last Update|
|0005484||DUMo||Bug||public||2019-05-16 01:27||2019-05-16 20:53|
|OS||Windows 10||OS Version||1809|
|Summary||0005484: wrong handling for driver Intel USB 3.0 eXtensible-Hostcontroller - 1.0 (Microsoft)|
|Description||I don't know if this issue already existed before. I observed it with the actual version of Windows 10 1809 of driver Intel USB 3.0 eXtensible-Hostcontroller - 1.0 (Microsoft). It seems to me that already within this driver there is some inconsistent use of this type of meta information and DUMo doesn't detect this inconsistency.|
There are text fields like driver name, manufacturer and driver provider. (I didn't check the English labels as my operating system with MUI has another language and region in Europe configured. I translated back the terms of this configured language and region freely to English.) The driver name lets assume that device manufacturer is Intel not AMD or any other system board manufactuerer with USB 3.0 host controllers onboard or on extension cards. The field manufacturer is filled inconsistently claiming that this driver is not only for this type of devices of Intel but also of other manufacturers claiming it as generic USB-xHCI-Hostcontroller instead. The field driver provider is filled correctly as Microsoft.
DUMo creates a report after scanning and checking. This report has a column manufacturer and no column driver publisher resp. provider. So DUMo presents this probably wrong and certainly inconsisten meta data value of manufacturer without checking with driver name nor driver publisher name. As long as DUMo doesn't report the (possibly incorrect value) of driver publisher in addition, it is an error of DUMo not to provide either the correct manufacturer name found in the driver name nor the driver publisher name. For this driver, reporting the driver publisher would not be the manufacturer and hence create another inconsistency in DUMo while the correct manufacturer is found in the as a token resp. component of the driver name.
Which field of meta data is more relevant to DUMo users, device manufacturer or driver publisher. Often the device manufacturer is part of the driver name. As long as DUMo doesn't report both fields, driver publisher is more relevant to me than device manufacturer as I don't use DUMo for deciding to exchange the device but instead for deciding to look up for available driver updates of the usually unmodified device. (There are also cases of reprogrammable hardware devices receiving corresponding firmware or microcode update or some other reprogramming of the hardware itself.)
|Additional Information||There are cases of devices where there exist different drivers of different driver publishers for the same device. I can see in the operating system that this field of driver publisher is also sometimes filled incorrectly with text like localized term of "your firm". This is sometime the case where component device manufacturers provide a device support package to original system integrators so that they adapt to their system and they not always adapting this field before release and often neither afterwards.|
My preferred solution is a feature request to report not only device manufacturer but additionally the driver publisher. This solution is generic and reflects the consistency and quality of the driver publisher not device manufacturer. As already mentioned above, this consistency and quality can be insufficient. Reflecting it, I don't consider a bug of DUMo but instead solely of the driver manufacturer. With this generic solution the DUMo user should have sufficient information to decide which data he considers trust worthy and relevant.
For me it would be a quick fix to replace the manufacturer field in the DUMo report by the driver publisher field. But this is not a protection against such inconsistencies by driver publishers, just shifting the issue of some set of drivers to another set.
|Tags||No tags attached.|