linux-kernel - RE: [PATCH v2 3/3] platform/x86: dell-wmi: add new dmi keys to bios_to_linux

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Tue, 9 Jun 2020 19:49:18 +0000
From:   <Mario.Limonciello@...l.com>
To:     <pali@...nel.org>, <rdunlap@...radead.org>
CC:     <y.linux@...itcher.com>, <linux-kernel@...r.kernel.org>,
        <platform-driver-x86@...r.kernel.org>, <mjg59@...f.ucam.org>
Subject: RE: [PATCH v2 3/3] platform/x86: dell-wmi: add new dmi keys to
 bios_to_linux_keycode

> 
> Looking at the last two lines... and for me it looks like that 0x00FF
> and 0xFFFF are just "placeholders" or special values for unknown /
> custom / unsupported / reserved / special / ... codes.
> 
> It is really suspicious why first 38 values are defined, then there is
> gap, then one value 255 and then huge gap to 65535.
> 
> Mario, this looks like some mapping table between internal Dell BIOS key
> code and standard Linux key code. Are you able to get access to some
> documentation which contains explanation of those Dell key numbers?
> It could really help us to understand these gaps and what is correct
> interpretation of these numbers.
> 

The codes are actually 4 bytes in the table, but in practice nothing above the
first two bytes is used.

Those two called out are special though, here are their meanings:

0x00FF is user programmable function
0xFFFF is no function

For the purpose of memory consumption I think it's reasonable to ignore the
upper 2 bytes and special case these two.

> E.g. I remember that pressing Fn+Q or Fn+W on some Dell Latitude
> generates code 255, which could prove my thesis about "special codes"
> (which are probably not found in e.g. Windows or Linux mapping tables).
> 
> > >  };
> > >
> > >  /*
> > > @@ -503,10 +504,7 @@ static void handle_dmi_entry(const struct
> dmi_header *dm, void *opaque)
> > >  					&table->keymap[i];
> > >
> > >  		/* Uninitialized entries are 0 aka KEY_RESERVED. */
> > > -		u16 keycode = (bios_entry->keycode <
> > > -			       ARRAY_SIZE(bios_to_linux_keycode)) ?
> > > -			bios_to_linux_keycode[bios_entry->keycode] :
> > > -			KEY_RESERVED;
> > > +		u16 keycode = bios_to_linux_keycode[bios_entry->keycode];
> > >
> > >  		/*
> > >  		 * Log if we find an entry in the DMI table that we don't
> > >
> >
> > Something like:
> >
> > 		u16 keycode;
> >
> > 		keycode = bios_entry->keycode == 0xffff ? KEY_UNKNOWN :
> > 			(bios_entry->keycode <
> > 			       ARRAY_SIZE(bios_to_linux_keycode)) ?
> > 			bios_to_linux_keycode[bios_entry->keycode] :
> > 			KEY_RESERVED;
> >
> >
> >
> > Also please fix this:
> > (no To-header on input) <>
> 
> Hint: specifying git send-email with '--to' argument instead of '--cc'
> should help.
> 
> >
> > --
> > ~Randy
> >