Hi. I see these warnings while compiling modules in 2.4.0-test10. This
is with RH 7.0's kgcc (why-oh-why did they base their system on
2.96!!). It doesn't seem to break anything--I'm just curious as to what
the warnings signify.
{standard input}: Assembler messages:
{standard input}:8: Warning: Ignoring changed section attributes for
.modinfo
--
Matthew Vanecek
perl -e 'print
$i=pack(c5,(41*2),sqrt(7056),(unpack(c,H)-2),oct(115),10);'
********************************************************************************
For 93 million miles, there is nothing between the sun and my shadow
except me.
I'm always getting in the way of something...
On Mon, 20 Nov 2000 22:11:04 -0600,
Matthew Vanecek <[email protected]> wrote:
>Hi. I see these warnings while compiling modules in 2.4.0-test10. This
>is with RH 7.0's kgcc (why-oh-why did they base their system on
>2.96!!). It doesn't seem to break anything--I'm just curious as to what
>the warnings signify.
>
>{standard input}: Assembler messages:
>{standard input}:8: Warning: Ignoring changed section attributes for
>.modinfo
Firstly you should not be compiling the kernel with gcc 2.96, you
should be using kgcc. Search recent l-k archives for kgcc.
Secondly that is a warning that section .modinfo was initially created
without the allocation bit but later usage of the section implies that
the allocation bit should be set. Creating .modinfo as non-allocated
was an old kludge to stop the module descriptive data being loaded into
memory as part of the module. modutils 2.3.19 onwards force .modinfo
to be non-allocated (and hence not loaded into kernel memory), even if
the allocate bit is set. Feel free to try this patch, it is not urgent
so I have not sent it to Linus "code freeze" Torvalds yet.
Against 2.4.0-test11-pre6, it might fit 2.4.0-test11.
Index: 0-test11-pre6.1/include/linux/module.h
--- 0-test11-pre6.1/include/linux/module.h Sun, 12 Nov 2000 14:59:01 +1100 kaos (linux-2.4/W/33_module.h 1.1.2.1.2.1.2.1.2.1.1.3 644)
+++ 0-test11-pre6.1(w)/include/linux/module.h Tue, 21 Nov 2000 15:46:58 +1100 kaos (linux-2.4/W/33_module.h 1.1.2.1.2.1.2.1.2.1.1.3 644)
@@ -247,12 +247,6 @@ static const struct gtype##_id * __modul
__attribute__ ((unused)) = name
#define MODULE_DEVICE_TABLE(type,name) \
MODULE_GENERIC_TABLE(type##_device,name)
-/* not put to .modinfo section to avoid section type conflicts */
-
-/* The attributes of a section are set the first time the section is
- seen; we want .modinfo to not be allocated. */
-
-__asm__(".section .modinfo\n\t.previous");
/* Define the module variable, and usage macros. */
extern struct module __this_module;
Keith Owens wrote:
>
> On Mon, 20 Nov 2000 22:11:04 -0600,
> Matthew Vanecek <[email protected]> wrote:
> >Hi. I see these warnings while compiling modules in 2.4.0-test10. This
> >is with RH 7.0's kgcc (why-oh-why did they base their system on
> >2.96!!). It doesn't seem to break anything--I'm just curious as to what
> >the warnings signify.
> >
> >{standard input}: Assembler messages:
> >{standard input}:8: Warning: Ignoring changed section attributes for
> >.modinfo
>
> Firstly you should not be compiling the kernel with gcc 2.96, you
> should be using kgcc. Search recent l-k archives for kgcc.
>
I said above the compiler I'm using, which is kgcc. =P AFAICT, a 2.96
kernel wouldn't even boot properly (well, I listened to others, so I've
only had the benefit of *other* people's experience!).
> Secondly that is a warning that section .modinfo was initially created
> without the allocation bit but later usage of the section implies that
> the allocation bit should be set. Creating .modinfo as non-allocated
> was an old kludge to stop the module descriptive data being loaded into
> memory as part of the module. modutils 2.3.19 onwards force .modinfo
> to be non-allocated (and hence not loaded into kernel memory), even if
> the allocate bit is set. Feel free to try this patch, it is not urgent
> so I have not sent it to Linus "code freeze" Torvalds yet.
>
I'll give your patch a try on the morrow, or day after. Gotta get my
new Visor working w/USB cradle first! I just wanted to find out if
those messages indicated a problem, or if they are harmless.
Personally, I hate seeing even warnings--I'd rather the code compiled
without any at all (although with some compilers, that's just not
possible!!).
Thanks for the reply.
> Against 2.4.0-test11-pre6, it might fit 2.4.0-test11.
>
> Index: 0-test11-pre6.1/include/linux/module.h
> --- 0-test11-pre6.1/include/linux/module.h Sun, 12 Nov 2000 14:59:01 +1100 kaos (linux-2.4/W/33_module.h 1.1.2.1.2.1.2.1.2.1.1.3 644)
> +++ 0-test11-pre6.1(w)/include/linux/module.h Tue, 21 Nov 2000 15:46:58 +1100 kaos (linux-2.4/W/33_module.h 1.1.2.1.2.1.2.1.2.1.1.3 644)
> @@ -247,12 +247,6 @@ static const struct gtype##_id * __modul
> __attribute__ ((unused)) = name
> #define MODULE_DEVICE_TABLE(type,name) \
> MODULE_GENERIC_TABLE(type##_device,name)
> -/* not put to .modinfo section to avoid section type conflicts */
> -
> -/* The attributes of a section are set the first time the section is
> - seen; we want .modinfo to not be allocated. */
> -
> -__asm__(".section .modinfo\n\t.previous");
>
> /* Define the module variable, and usage macros. */
> extern struct module __this_module;
--
Matthew Vanecek
perl -e 'print
$i=pack(c5,(41*2),sqrt(7056),(unpack(c,H)-2),oct(115),10);'
********************************************************************************
For 93 million miles, there is nothing between the sun and my shadow
except me.
I'm always getting in the way of something...