Menu
▾
▴
tpctl-users
|
From: fago <nu...@gm...> - 2004-09-15 13:42:58
|
hi I am using debian sarge/sid with kernel 2.6.8 on my thinkpad t23. I have successfully compiled the thinkpad modules 5.6 (manually because the debian source package is still 5.5). Loading the modules is working, but udev (0.030-1) only creates this device: ls -l /dev/thinkpad crw-rw---- 1 root root 10, 170 2004-09-15 14:21 /dev/thinkpad but according to the readme, i miss the devices for smapi,superio,rtcmosram and thinkpadpm Are the modules intended to work with udev? MfG, fago |
|
From: Christian K. <ck...@mp...> - 2004-09-15 13:53:51
|
On Wed, Sep 15, 2004 at 03:42:48PM +0200, fago wrote: > I am using debian sarge/sid with kernel 2.6.8 on my thinkpad t23. > I have successfully compiled the thinkpad modules 5.6 (manually because > the debian source package is still 5.5). > AFAIK 5.5 and 5.6 differ only in small makefile and doc changes. Hence it may work with the debian package. -- CU/2, Christian "M(i)ndflay(e)r" Klein http://www.garfield.com/comics/comics_archives_strip.html?2003-ga030816 EMail: ck...@mp... Homepage: http://www.klex.handshake.de/ |
|
From: fago <nu...@gm...> - 2004-09-15 14:56:46
|
> AFAIK 5.5 and 5.6 differ only in small makefile and doc changes. Hence it
> may work with the debian package.
ok, it seems like there are not too much changes. but the debian package
wo'nt help me, so with it i could only compile the modules with the help
of make-kpkg.
i'm using the other debian packages (tpctl, thinkpad-base), i just
compiled the modules without make-kpkg.
perhaps the problem is not udev. according to the README the thinkpad
device has to be created, what udev does. so its not clear to me, if
there should be more devices under /dev/thinkpad?
however smapi is not working:
tpctl --standby
tpctl: SMAPI BIOS error 0x91 ("system is invalid") -- exiting.
tpctl -ib
tpctl: SMAPI BIOS error 0x86 ("function is not supported") -- exiting.
also the fn-keys for standby and hibernate don't work.
Regards, fago
|
|
From: Charles L. <cl...@gh...> - 2004-09-15 15:22:24
|
fago said:
> however smapi is not working:
> tpctl --standby
> tpctl: SMAPI BIOS error 0x91 ("system is invalid") -- exiting.
> tpctl -ib
> tpctl: SMAPI BIOS error 0x86 ("function is not supported") -- exiting.
If the error messages are to be believed, then tpctl has received an erro=
r
message from the SMAPI BIOS itself, and therefore tpctl is probably going
through the main /dev/thinkpad device.
You can always run tpctl under strace and see if it is trying to open
other device nodes.
> also the fn-keys for standby and hibernate don't work.
I don't think they usually do. AFAIR they just set some bits in nvram, an=
d
I have yet to see a program that polls for them. (This is different from
actual keyboard multimedia keys, which send events that programs can
efficiently watch for.)
If you are using ACPI, they may be mapped to ACPI events.
--=20
Charles Lepple
cl...@gh...
|
|
From: Christian K. <ck...@mp...> - 2004-09-15 15:36:45
|
On Wed, Sep 15, 2004 at 11:22:21AM -0400, Charles Lepple wrote: > > also the fn-keys for standby and hibernate don't work. > > I don't think they usually do. AFAIR they just set some bits in nvram, and > I have yet to see a program that polls for them. (This is different from You can see one at http://www.nongnu.org/tpb/ -- CU/2, Christian "M(i)ndflay(e)r" Klein http://ars.userfriendly.org/cartoons/?id=20030819&mode=classic EMail: ck...@mp... Homepage: http://www.klex.handshake.de/ |
|
From: fago <nu...@gm...> - 2004-09-15 16:35:11
|
> You can always run tpctl under strace and see if it is trying to open
> other device nodes.
so i tried strace
stat64("/dev/thinkpad/thinkpad", 0xbffffb3c) = -1 ENOTDIR (Not a
directory)
stat64("/dev/thinkpad", {st_mode=S_IFCHR|0660,
st_rdev=makedev(10, 170), ...}) = 0
open("/dev/thinkpad", O_RDONLY) = 3
tpctl is finding the correct device.
the next thing concerning a device:
fstat64(1, {st_mode=S_IFCHR|0600, st_rdev=makedev(136, 1), ...}) = 0
mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1,
0) = 0x40018000
write(1, "tpctl: SMAPI BIOS error 0x86 (\"f"..., 71) = 71
so is it looking for a device 136,1 or does it not need a device?
fago
|
|
From: Charles L. <cl...@gh...> - 2004-09-15 17:10:46
|
fago said:
>
>> You can always run tpctl under strace and see if it is trying to open
>> other device nodes.
>
> so i tried strace
>
> stat64("/dev/thinkpad/thinkpad", 0xbffffb3c) =3D -1 ENOTDIR (No=
t a
> directory)
> stat64("/dev/thinkpad", {st_mode=3DS_IFCHR|0660,
> st_rdev=3Dmakedev(10, 170), ...}) =3D 0
> open("/dev/thinkpad", O_RDONLY) =3D 3
>
> tpctl is finding the correct device.
> the next thing concerning a device:
>
> fstat64(1, {st_mode=3DS_IFCHR|0600, st_rdev=3Dmakedev(136, 1), ...}) =3D=
0
> mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1,
> 0) =3D 0x40018000
> write(1, "tpctl: SMAPI BIOS error 0x86 (\"f"..., 71) =3D 71
>
>
> so is it looking for a device 136,1 or does it not need a device?
No, (136,1) is the result of the fstat call (fstat fills in the structure=
,
but strace prints it like it was passed in).
That's probably not relevant, though, because file descriptor 1 is usuall=
y
standard output. (And the mmap2 is just part of the way glibc implements
malloc.)
--=20
Charles Lepple
cl...@gh...
|
|
From: fago <nu...@gm...> - 2004-09-15 22:12:35
|
> No, (136,1) is the result of the fstat call (fstat fills in the structure, > but strace prints it like it was passed in). > > That's probably not relevant, though, because file descriptor 1 is usually > standard output. (And the mmap2 is just part of the way glibc implements > malloc.) ok, so far so good ;) thanks for your help so should smapi work with a thinkpad t23? |
|
From: Charles L. <cl...@gh...> - 2004-09-15 22:34:41
|
fago said: > >> No, (136,1) is the result of the fstat call (fstat fills in the >> structure, >> but strace prints it like it was passed in). >> >> That's probably not relevant, though, because file descriptor 1 is >> usually >> standard output. (And the mmap2 is just part of the way glibc implemen= ts >> malloc.) > > ok, so far so good ;) > thanks for your help > > so should smapi work with a thinkpad t23? that I don't know. I was just pointing out that tpctl was talking to the devices properly. --=20 Charles Lepple cl...@gh... |
|
From: Cheuksan E. W. <wan...@gm...> - 2004-09-16 00:16:01
|
On Thu, 16 Sep 2004 00:12:25 +0200, fago <nu...@gm...> wrote: > > > No, (136,1) is the result of the fstat call (fstat fills in the structure, > > but strace prints it like it was passed in). > > > > That's probably not relevant, though, because file descriptor 1 is usually > > standard output. (And the mmap2 is just part of the way glibc implements > > malloc.) > > ok, so far so good ;) > thanks for your help > > so should smapi work with a thinkpad t23? > > Yes, it should, but not with ACPI. You should try disabling ACPI. Cheuksan Wang |
×
Want the latest updates on software, tech news, and AI?
Get latest updates about software, tech news, and AI from SourceForge directly in your inbox once a month.
Thanks for helping keep SourceForge clean.
X