Re: [Suspend-devel] Help

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Thu, 24 Sep 2009 00:28:03 +0200
"Rafael J. Wysocki" <rjw@xxxxxxx> wrote:

> On Wednesday 23 September 2009, Rodolfo (kix) wrote:
> > Hi Rafael,
> 
> Hi,
> 
> > I am thinking about the maintaining ...

Good luck! I know that you'll need it ;)

Let me throw my 2 cents into the bowl.

> > 1. The users sends the hardware info to be added to the database
> > (now by mail). Then:
> > 
> > 1. Probably we need a little script (or web script? or post directly
> > form the s2ram -i?):
> >   1.a The script checks if the machine is previously added to the
> > database (in a text file). Probably this script can read from the
> > mailing list format
> > 
> > xxxx:% sudo /root/s2ram -i
> > This machine can be identified by:
> >     sys_vendor   = "TOSHIBA"
> >     sys_product  = "Satellite U305"
> >     sys_version  = "PSU34U-00L003"
> >     bios_version = "V1.70   "
> > See http://en.opensuse.org/S2ram for details.
> > 
> >   1.b If is not added, the script add the machine to a text file.
> 
> However you arrange that is up to you.  My experience is that the
> users generally send whitelist information by email.
> 
> > 2. The text file can be processed using a little script to convert
> > it to the "whitelist.[c,h]" files and converted to web page
> > (s2whitelist and s2web :-))
> > 3. The script can do a git update if needed.
> 
> Well, I'm surely not going to allow the script to update the master
> git tree at kernel.org directly.

There's also the issue that some sanity checking should be done before
adding an entry. Pretty often, users send some hilarious option
combinations, on machines that should nowadays just work without any
options (e.g. intel-based).

The next thing is, that I, personally, would reject any machines that
should work out of the box nowadays with in-kernel drivers (means:
intel, nvidia- and ATI binary-module driven). You can only make things
worse on such machines with userspace workarounds, and they should be
recognized automatically by recent (means: newer than 2 years)
pm-utils, so that no hack is needed. IMHO the whitelist should not be
cluttered with them.

> > In my opinion, the file should have this format:
> > 
> > <author info for the comment's
> > line>·<sys_vendor>·<sys_product>·<sys_version>·<bios_version>·<flags>
> > 
> > I am using the "·" separator because probably is not used in any
> > field.
> 
> What's the code of this character?

Something strange ;-)

I'd use the pipe character: '|'. I have not yet encountered it in
s2ram-relevant DMI data. It still is readable and not subject to
encoding issues AFAICT.

I'd put the author info / comment at the end, since then you can
sort/grep for sys_vendor easily, which is probably most useful.

And I still think that nowadays s2ram is the wrong tool for the
whitelist, but well... :-)

Good luck,

	Stefan
-- 
Stefan Seyfried

"Ceterum censeo whitelist.c esse delendam"
_______________________________________________
linux-pm mailing list
linux-pm@xxxxxxxxxxxxxxxxxxxxxxxxxx
https://lists.linux-foundation.org/mailman/listinfo/linux-pm


[Index of Archives]     [Linux ACPI]     [Netdev]     [Ethernet Bridging]     [Linux Wireless]     [CPU Freq]     [Kernel Newbies]     [Fedora Kernel]     [Security]     [Linux for Hams]     [Netfilter]     [Bugtraq]     [Yosemite News]     [MIPS Linux]     [ARM Linux]     [Linux RAID]     [Linux Admin]     [Samba]

  Powered by Linux