[uClinux-dev] Kernel modules don't work on read-only filesystems ?

Marty Leisner leisner at rochester.rr.com
Wed Oct 10 19:05:45 EDT 2007


What does /proc/devices say?

Which version of the book are you using?  LDD v3  only covers 2.6...

Did you have it working on other systems...?

marty

Josue de Freitas <josue.freitas at gmail.com> writes  on Wed, 10 Oct 2007 17:59:56
 -0300
     > --===============0520366532==
     > Content-Type: multipart/alternative; 
     > 	boundary="----=_Part_13714_28569033.1192049996889"
     > 
     > ------=_Part_13714_28569033.1192049996889
     > Content-Type: text/plain; charset=ISO-8859-1
     > Content-Transfer-Encoding: 7bit
     > Content-Disposition: inline
     > 
     > On 10/10/07, Crane, Matthew <mcrane03 at harris.com> wrote:
     > >
     > >  Hi,
     > >
     > > Sounds like your module didn't actually get inserted.  Can you verify 
it
     > > is present with an lsmod?   If it failed to insert, then check your 
klog.
     > >
     > 
     > 
     > Yes it is up. It's very strange because it's up but seems not to be 
running,
     > a simple "hello world" module using printk  also don't work, it gets up 
but
     > apparently not run (damn my naivety to think that this will work :p).
     > 
     > The scull module is suppose to create an proc dir entry called 
scullmem, but
     > it's not created.
     > 
     > # lsmod
     > Module                  Size  Used by
     > scull                  10708   0 (unused)
     > #
     > 
     > An also:
     > 
     > # cat /proc/ksyms | grep scull
     > 30dd8000
     > __insmod_scull_O/lib/modules/2.4.34.5-uc0/kernel/drivers/char/scull/scul
l.o_M00000000_V-1
     > [scull]
     > 30dd8060 __insmod_scull_S.text_L9656    [scull]
     > 30dda618 __insmod_scull_S.rodata_L38    [scull]
     > 30dda640 __insmod_scull_S.data_L512     [scull]
     > #
     > 
     > 
     > 
     > You can build the module into the kernel statically to verify that it 
works
     > > with the device node.
     > >
     > 
     > 
     > It's a possibility. But to do it I just have to chance de obj-m to 
obj-y on
     > the Makefiles ? I think I must undefine de MODULE directive also, right 
?
     > Any other hint of how to make it statically ?
     > 
     > 
     > Matt
     > >
     > >
     > 
     > Thanks Matthew.
     > 
     > Regards, Josue.
     > 
     > ------=_Part_13714_28569033.1192049996889
     > Content-Type: text/html; charset=ISO-8859-1
     > Content-Transfer-Encoding: 7bit
     > Content-Disposition: inline
     > 
     > <br><br><div><span class="gmail_quote">On 10/10/07, <b 
class="gmail_sendername">Crane, Matthew</b> &lt;<a href="
     >mailto:mcrane03 at harris.com">mcrane03 at harris.com</a>&gt; wrote:</span>
<blockquote class="gmail_quote" style="borde
     >r-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; 
padding-left: 1ex;">
     > 
     > 
     > 
     > 
     > <div>
     > <div dir="ltr" align="left"><span><font color="#0000ff" face="Arial" 
size="2">Hi, </font></span></div>
     > <div dir="ltr" align="left"><span><font color="#0000ff" face="Arial" 
size="2"></font></span>&nbsp;</div>
     > <div dir="ltr" align="left"><span><font color="#0000ff" face="Arial" 
size="2">Sounds like your module didn&#39;t
     > actually get inserted.&nbsp; 
     > Can you verify it is present with an lsmod?&nbsp;&nbsp; If it failed to 
insert,
     > then check your klog.</font></span></div></div></blockquote><div><br>
<br>Yes it is up. It&#39;s very strange bec
     >ause it&#39;s up but seems not to be running, a simple &quot;hello 
world&quot; module using printk&nbsp; also don
     >&#39;t work, it gets up but apparently not run (damn my naivety to think 
that this will work :p).
     > <br><br>The scull module is suppose to create an proc dir entry called 
scullmem, but it&#39;s not created.<br><b
     >r># lsmod<br>Module&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbs
     >p;&nbsp; Size&nbsp; Used by<br>scull&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&
nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp
     >;&nbsp;&nbsp;&nbsp;&nbsp; 10708&nbsp;&nbsp; 0 (unused)<br>#<br><br>An 
also:
     > <br><br># cat /proc/ksyms | grep scull<br>30dd8000 
__insmod_scull_O/lib/modules/2.4.34.5-uc0/kernel/drivers/char
     >/scull/scull.o_M00000000_V-1&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; [scull]<br>
30dd8060 __insmod_scull_S.text_L9656&nbsp;&
     >nbsp;&nbsp; [scull]<br>30dda618 __insmod_scull_S.rodata_L38&nbsp;&nbsp;&n
bsp; [scull]
     > <br>30dda640 __insmod_scull_S.data_L512&nbsp;&nbsp;&nbsp;&nbsp; 
[scull]<br>#<br><br>&nbsp;</div><br><blockquote
     >class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); 
margin: 0pt 0pt 0pt 0.8ex; padding-left: 1e
     >x;"><div><div><span><font color="#0000ff" face="Arial" size="2">
     > You 
     > can build the module into the kernel statically to verify that it works 
with the
     > device node.&nbsp; </font></span></div></div></blockquote><div><br><br>
It&#39;s a possibility. But to do it I ju
     >st have to chance de obj-m to obj-y on the Makefiles ? I think I must 
undefine de MODULE directive also, right ?
     > Any other hint of how to make it statically ?
     > <br>&nbsp;<br></div><br><blockquote class="gmail_quote" 
style="border-left: 1px solid rgb(204, 204, 204); margin
     >: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"><div><div dir="ltr" 
align="left"><span><font color="#0000ff" face="Arial
     >" size="2">
     > Matt</font></span></div>
     > <div dir="ltr" align="left"><span><font color="#0000ff" face="Arial" 
size="2">&nbsp;</font></span></div></div></
     >blockquote></div><br>Thanks Matthew.<br><br>Regards, Josue.
     > 
     > ------=_Part_13714_28569033.1192049996889--
     > 
     > --===============0520366532==
     > Content-Type: text/plain; charset="us-ascii"
     > MIME-Version: 1.0
     > Content-Transfer-Encoding: 7bit
     > Content-Disposition: inline
     > 
     > _______________________________________________
     > uClinux-dev mailing list
     > uClinux-dev at uclinux.org
     > http://mailman.uclinux.org/mailman/listinfo/uclinux-dev
     > This message was resent by uclinux-dev at uclinux.org
     > To unsubscribe see:
     > http://mailman.uclinux.org/mailman/options/uclinux-dev
     > --===============0520366532==--




More information about the uClinux-dev mailing list