Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1408663 > unrolled thread

Re: WMI driver no longer load after switching to generic UUID library

Started byLinus Torvalds <torvalds@linux-foundation.org>
First post2016-05-29 22:30 +0200
Last post2016-05-31 06:30 +0200
Articles 5 — 4 participants

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.


Contents

  Re: WMI driver no longer load after switching to generic UUID library Linus Torvalds <torvalds@linux-foundation.org> - 2016-05-29 22:30 +0200
    Re: WMI driver no longer load after switching to generic UUID  library Andy Shevchenko <andriy.shevchenko@linux.intel.com> - 2016-05-30 10:00 +0200
      Re: WMI driver no longer load after switching to generic UUID library Bjørn Mork <bjorn@mork.no> - 2016-05-30 13:00 +0200
        Re: WMI driver no longer load after switching to generic UUID  library Andy Shevchenko <andriy.shevchenko@linux.intel.com> - 2016-05-30 16:40 +0200
          Re: WMI driver no longer load after switching to generic UUID library Kui Zhang <kuizhang@gmail.com> - 2016-05-31 06:30 +0200

#1408663 — Re: WMI driver no longer load after switching to generic UUID library

FromLinus Torvalds <torvalds@linux-foundation.org>
Date2016-05-29 22:30 +0200
SubjectRe: WMI driver no longer load after switching to generic UUID library
Message-ID<rEfvr-2wx-11@gated-at.bofh.it>
Andy?

Looking at that commit, it does look like what wmi_parse_guid +
wmi_swap_bytes did was to implement a LE uuid. So the conversion looks
correct, but the fact that it breaks something obviously means that
there's something missing. What's going on?

And dammit, how I hate uuid's and the idiotic byte ordering issue.
It's possibly the only idiotic standard that is even worse than xml.

            Linus

On Sun, May 29, 2016 at 11:35 AM, Kui Zhang <kuizhang@gmail.com> wrote:
> Hello,
>
> After commit 538d7eb86d58 ("drivers/platform/x86/wmi.c: use generic UUID library")
>
> some keyboard functions stopped working:
>
> clevo-wmi - git://git.code.sf.net/p/clevo-wmi/code
> [   83.877887] Clevo Get GUID not found
>
> /*
> #define CLWMI_GET_GUID "ABBC0F6D-8EA1-11d1-00A0-C90629100000"
>
> ...
>     if(!wmi_has_guid(CLWMI_GET_GUID)) {
>         pr_err("Clevo Get GUID not found\n");
>         return -ENODEV;
>     }
>
>
> */
>
> clevo-xsm-wmi - https://bitbucket.org/lynthium/clevo-xsm-wmi.git
> insmod: ERROR: could not insert module clevo-xsm-wmi.ko: No such device
>
> /*
>
> #define CLEVO_GET_GUID    "ABBC0F6D-8EA1-11D1-00A0-C90629100000"
>
> ...
>
>     if (!wmi_has_guid(CLEVO_GET_GUID)) {
>         CLEVO_XSM_INFO("No known WMI control method GUID found\n");
>         return -ENODEV;
>     }
>
> */
>
>
> thanks
> Kui.Z

[toc] | [next] | [standalone]


#1408834 — Re: WMI driver no longer load after switching to generic UUID library

FromAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Date2016-05-30 10:00 +0200
SubjectRe: WMI driver no longer load after switching to generic UUID library
Message-ID<rEqhb-1fU-9@gated-at.bofh.it>
In reply to#1408663
On Sun, 2016-05-29 at 13:22 -0700, Linus Torvalds wrote:
> Andy?
> 
> Looking at that commit, it does look like what wmi_parse_guid +
> wmi_swap_bytes did was to implement a LE uuid. So the conversion looks
> correct, but the fact that it breaks something obviously means that
> there's something missing. What's going on?

I proposed to add print to the loops to see what's going on there.

> 
> And dammit, how I hate uuid's and the idiotic byte ordering issue.
> It's possibly the only idiotic standard that is even worse than xml.

T'so seems has the same opinion http://www.spinics.net/lists/kernel/msg2
263333.html

> 
>             Linus
> 
> On Sun, May 29, 2016 at 11:35 AM, Kui Zhang <kuizhang@gmail.com>
> wrote:
> > Hello,
> > 
> > After commit 538d7eb86d58 ("drivers/platform/x86/wmi.c: use generic
> > UUID library")
> > 
> > some keyboard functions stopped working:
> > 
> > clevo-wmi - git://git.code.sf.net/p/clevo-wmi/code
> > [   83.877887] Clevo Get GUID not found
> > 
> > /*
> > #define CLWMI_GET_GUID "ABBC0F6D-8EA1-11d1-00A0-C90629100000"
> > 
> > ...
> >     if(!wmi_has_guid(CLWMI_GET_GUID)) {
> >         pr_err("Clevo Get GUID not found\n");
> >         return -ENODEV;
> >     }
> > 
> > 
> > */
> > 
> > clevo-xsm-wmi - https://bitbucket.org/lynthium/clevo-xsm-wmi.git
> > insmod: ERROR: could not insert module clevo-xsm-wmi.ko: No such
> > device
> > 
> > /*
> > 
> > #define CLEVO_GET_GUID    "ABBC0F6D-8EA1-11D1-00A0-C90629100000"
> > 
> > ...
> > 
> >     if (!wmi_has_guid(CLEVO_GET_GUID)) {
> >         CLEVO_XSM_INFO("No known WMI control method GUID found\n");
> >         return -ENODEV;
> >     }
> > 
> > */
> > 
> > 
> > thanks
> > Kui.Z

-- 
Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Intel Finland Oy

[toc] | [prev] | [next] | [standalone]


#1408985

FromBjørn Mork <bjorn@mork.no>
Date2016-05-30 13:00 +0200
Message-ID<rEt5o-33J-17@gated-at.bofh.it>
In reply to#1408834

[Multipart message — attachments visible in raw view] — view raw

How about the untested attached patch?


Bjørn

[toc] | [prev] | [next] | [standalone]


#1409175 — Re: WMI driver no longer load after switching to generic UUID library

FromAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Date2016-05-30 16:40 +0200
SubjectRe: WMI driver no longer load after switching to generic UUID library
Message-ID<rEwwh-5os-11@gated-at.bofh.it>
In reply to#1408985
On Mon, 2016-05-30 at 12:52 +0200, Bjørn Mork wrote:
> How about the untested attached patch?
> 

Facepalm.jpg!

Good catch!

The question is how it passed in our internal tree without being
notified (it even included a lot of ACPI code updated)!?

My explanation since I have no access to archive of internal mailing
list is that I submitted 'cleaned up' version which brought that and
wasn't tested. 

Anyway I just wrote a test module and is about to submit it with your
fix in a series.

Thanks!

> 
> Bjørn
> 

-- 
Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Intel Finland Oy

[toc] | [prev] | [next] | [standalone]


#1409797

FromKui Zhang <kuizhang@gmail.com>
Date2016-05-31 06:30 +0200
Message-ID<rEJtv-61m-3@gated-at.bofh.it>
In reply to#1409175
Thanks guys.

keyboard functions are working now.


kui.z

On Mon, May 30, 2016 at 7:40 AM, Andy Shevchenko
<andriy.shevchenko@linux.intel.com> wrote:
> On Mon, 2016-05-30 at 12:52 +0200, Bjørn Mork wrote:
>> How about the untested attached patch?
>>
>
> Facepalm.jpg!
>
> Good catch!
>
> The question is how it passed in our internal tree without being
> notified (it even included a lot of ACPI code updated)!?
>
> My explanation since I have no access to archive of internal mailing
> list is that I submitted 'cleaned up' version which brought that and
> wasn't tested.
>
> Anyway I just wrote a test module and is about to submit it with your
> fix in a series.
>
> Thanks!
>
>>
>> Bjørn
>>
>
> --
> Andy Shevchenko <andriy.shevchenko@linux.intel.com>
> Intel Finland Oy

[toc] | [prev] | [standalone]


Back to top | Article view | linux.kernel


csiph-web