[KJ] [PATCH] Centralize checking for adequate version of gcc.

[KJ] [PATCH] Centralize checking for adequate version of gcc.

[Robert P. J. Day]

  here's a patch i submitted to andrew to centralize checking for a
proper version of gcc when building the kernel, but andrew claims it
broke the build.  i'm trying to reproduce the problem, and i'd
appreciate it if anyone else wants to apply this patch and try a build
based on "allyesconfig" or "allmodconfig" on x86 to see if they run
into some kind of issue, cuz i'm just not seeing it here.  thanks.

diff --git a/include/linux/compiler.h b/include/linux/compiler.h
index 8287a72..50558f7 100644
--- a/include/linux/compiler.h
+++ b/include/linux/compiler.h
@@ -36,12 +36,20 @@ extern void __chk_io_ptr(const void __iomem *);

 #ifdef __KERNEL__

-#if __GNUC__ >= 4
+/*
+ *  Start with some sanity checking on the version of gcc.
+ */
+
+#if (__GNUC__ < 3) || (__GNUC__ = 3 && __GNUC_MINOR__ < 2)
+# error Sorry, your GCC is too old. It builds incorrect kernels.
+#elif __GNUC__ = 4 && __GNUC_MINOR__ = 1 && __GNUC_PATCHLEVEL__ = 0
+# warning gcc-4.1.0 is known to miscompile the kernel.  A different compiler version is recommended.
+#elif __GNUC__ = 4
 # include <linux/compiler-gcc4.h>
-#elif __GNUC__ = 3 && __GNUC_MINOR__ >= 2
+#elif __GNUC__ = 3
 # include <linux/compiler-gcc3.h>
 #else
-# error Sorry, your compiler is too old/not recognized.
+# error Sorry, your compiler is not recognized.
 #endif

 /* Intel compiler defines __GNUC__. So we will overwrite implementations
diff --git a/init/main.c b/init/main.c
index eb8bdba..db39a80 100644
--- a/init/main.c
+++ b/init/main.c
@@ -5,7 +5,6 @@
  *
  *  GK 2/5/95  -  Changed to support mounting root fs via NFS
  *  Added initrd & change_root: Werner Almesberger & Hans Lermen, Feb '96
- *  Moan early if gcc is old, avoiding bogus kernels - Paul Gortmaker, May '96
  *  Simplified starting of init:  Michael A. Griffith <grif@acm.org>
  */

@@ -66,23 +65,6 @@
 #include <asm/smp.h>
 #endif

-/*
- * This is one of the first .c files built. Error out early if we have compiler
- * trouble.
- *
- * Versions of gcc older than that listed below may actually compile and link
- * okay, but the end product can have subtle run time bugs.  To avoid associated
- * bogus bug reports, we flatly refuse to compile with a gcc that is known to be
- * too old from the very beginning.
- */
-#if (__GNUC__ < 3) || (__GNUC__ = 3 && __GNUC_MINOR__ < 2)
-#error Sorry, your GCC is too old. It builds incorrect kernels.
-#endif
-
-#if __GNUC__ = 4 && __GNUC_MINOR__ = 1 && __GNUC_PATCHLEVEL__ = 0
-#warning gcc-4.1.0 is known to miscompile the kernel.  A different compiler version is recommended.
-#endif
-
 static int kernel_init(void *);

 extern void init_IRQ(void);

-- 
====================================
Robert P. J. Day
Linux Consulting, Training and Annoying Kernel Pedantry
Waterloo, Ontario, CANADA

http://fsdev.net/wiki/index.php?title=Main_Page
====================================
_______________________________________________
Kernel-janitors mailing list
Kernel-janitors@lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/kernel-janitors

Re: [KJ] [PATCH] Centralize checking for adequate version of gcc.

[Vegard Nossum]

On 6/15/07, Robert P. J. Day <rpjday@mindspring.com> wrote:
>   here's a patch i submitted to andrew to centralize checking for a
> proper version of gcc when building the kernel, but andrew claims it
> broke the build.  i'm trying to reproduce the problem, and i'd
> appreciate it if anyone else wants to apply this patch and try a build
> based on "allyesconfig" or "allmodconfig" on x86 to see if they run
> into some kind of issue, cuz i'm just not seeing it here.  thanks.

It works for me (using gcc 4.1.1 for i386 on linus's tree). Removing
that explanatory comment is not very nice, though.

Vegard
_______________________________________________
Kernel-janitors mailing list
Kernel-janitors@lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/kernel-janitors

Re: [KJ] [PATCH] Centralize checking for adequate version of gcc.

[Robert P. J. Day]

On Fri, 15 Jun 2007, Vegard Nossum wrote:

> On 6/15/07, Robert P. J. Day <rpjday@mindspring.com> wrote:
> >   here's a patch i submitted to andrew to centralize checking for a
> > proper version of gcc when building the kernel, but andrew claims it
> > broke the build.  i'm trying to reproduce the problem, and i'd
> > appreciate it if anyone else wants to apply this patch and try a build
> > based on "allyesconfig" or "allmodconfig" on x86 to see if they run
> > into some kind of issue, cuz i'm just not seeing it here.  thanks.
>
> It works for me (using gcc 4.1.1 for i386 on linus's tree). Removing
> that explanatory comment is not very nice, though.

which explanatory comment?  the one from init/main.c?  but why would i
leave it in when the code it pertains to has been removed?

rday
-- 
====================================
Robert P. J. Day
Linux Consulting, Training and Annoying Kernel Pedantry
Waterloo, Ontario, CANADA

http://fsdev.net/wiki/index.php?title=Main_Page
====================================
_______________________________________________
Kernel-janitors mailing list
Kernel-janitors@lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/kernel-janitors

Re: [KJ] [PATCH] Centralize checking for adequate version of gcc.

[Vegard Nossum]

On 6/15/07, Robert P. J. Day <rpjday@mindspring.com> wrote:
> which explanatory comment?  the one from init/main.c?  but why would i
> leave it in when the code it pertains to has been removed?

Yes. Wouldn't it be good to keep it with the checks (so
include/linux/compiler.h)? It's a useful comment.

Vegard
_______________________________________________
Kernel-janitors mailing list
Kernel-janitors@lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/kernel-janitors

Re: [KJ] [PATCH] Centralize checking for adequate version of gcc.

[Jaco Kroon]

Robert P. J. Day wrote:
> On Fri, 15 Jun 2007, Vegard Nossum wrote:
> 
>> On 6/15/07, Robert P. J. Day <rpjday@mindspring.com> wrote:
>>>   here's a patch i submitted to andrew to centralize checking for a
>>> proper version of gcc when building the kernel, but andrew claims it
>>> broke the build.  i'm trying to reproduce the problem, and i'd
>>> appreciate it if anyone else wants to apply this patch and try a build
>>> based on "allyesconfig" or "allmodconfig" on x86 to see if they run
>>> into some kind of issue, cuz i'm just not seeing it here.  thanks.
>> It works for me (using gcc 4.1.1 for i386 on linus's tree). Removing
>> that explanatory comment is not very nice, though.
> 
> which explanatory comment?  the one from init/main.c?  but why would i
> leave it in when the code it pertains to has been removed?

Move the comment perhaps?
_______________________________________________
Kernel-janitors mailing list
Kernel-janitors@lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/kernel-janitors

Re: [KJ] [PATCH] Centralize checking for adequate version of gcc.

[Robert P. J. Day]

On Fri, 15 Jun 2007, Jaco Kroon wrote:

> Robert P. J. Day wrote:
> > On Fri, 15 Jun 2007, Vegard Nossum wrote:
> >
> > > On 6/15/07, Robert P. J. Day <rpjday@mindspring.com> wrote:
> > > >   here's a patch i submitted to andrew to centralize checking for a
> > > > proper version of gcc when building the kernel, but andrew claims it
> > > > broke the build.  i'm trying to reproduce the problem, and i'd
> > > > appreciate it if anyone else wants to apply this patch and try a build
> > > > based on "allyesconfig" or "allmodconfig" on x86 to see if they run
> > > > into some kind of issue, cuz i'm just not seeing it here.  thanks.
> > > It works for me (using gcc 4.1.1 for i386 on linus's tree). Removing
> > > that explanatory comment is not very nice, though.
> >
> > which explanatory comment?  the one from init/main.c?  but why would i
> > leave it in when the code it pertains to has been removed?
>
> Move the comment perhaps?

for what reason, exactly?  the purpose of that preprocessor content is
to warn the user about a bad version of gcc.  it does that when the
user tries to build.  why comment it?  i'm guessing most users will
never look at that header file, anyway.

and, more to the point, that content is fairly self-explanatory.  how
would adding redundant comments improve the situation?  it would be
like having a line of code such as

   i = i + 1;

and commenting that with:

  /* add 1 to i */

i dropped the comment since it didn't seem to add any information
beyond the C preprocessor content.

rday
-- 
====================================
Robert P. J. Day
Linux Consulting, Training and Annoying Kernel Pedantry
Waterloo, Ontario, CANADA

http://fsdev.net/wiki/index.php?title=Main_Page
====================================
_______________________________________________
Kernel-janitors mailing list
Kernel-janitors@lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/kernel-janitors

Re: [KJ] [PATCH] Centralize checking for adequate version of gcc.

[Jaco Kroon]

Robert P. J. Day wrote:
> On Fri, 15 Jun 2007, Jaco Kroon wrote:
> 
>> Robert P. J. Day wrote:
>>> On Fri, 15 Jun 2007, Vegard Nossum wrote:
>>>
>>>> On 6/15/07, Robert P. J. Day <rpjday@mindspring.com> wrote:
>>>>>   here's a patch i submitted to andrew to centralize checking for a
>>>>> proper version of gcc when building the kernel, but andrew claims it
>>>>> broke the build.  i'm trying to reproduce the problem, and i'd
>>>>> appreciate it if anyone else wants to apply this patch and try a build
>>>>> based on "allyesconfig" or "allmodconfig" on x86 to see if they run
>>>>> into some kind of issue, cuz i'm just not seeing it here.  thanks.
>>>> It works for me (using gcc 4.1.1 for i386 on linus's tree). Removing
>>>> that explanatory comment is not very nice, though.
>>> which explanatory comment?  the one from init/main.c?  but why would i
>>> leave it in when the code it pertains to has been removed?
>> Move the comment perhaps?
> 
> for what reason, exactly?  the purpose of that preprocessor content is
> to warn the user about a bad version of gcc.  it does that when the
> user tries to build.  why comment it?  i'm guessing most users will
> never look at that header file, anyway.
> 
> and, more to the point, that content is fairly self-explanatory.  how
> would adding redundant comments improve the situation?  it would be
> like having a line of code such as
> 
>    i = i + 1;
> 
> and commenting that with:
> 
>   /* add 1 to i */

The comment explained _why_ those checks were done, ie, it generates bad 
builds.  The pre-processor code simply states _which_ compilers are 
acceptable.

Jaco
_______________________________________________
Kernel-janitors mailing list
Kernel-janitors@lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/kernel-janitors

Re: [KJ] [PATCH] Centralize checking for adequate version of gcc.

[burns.ethan]

On Fri, Jun 15, 2007 at 06:47:01PM +0200, Jaco Kroon wrote:

[...]

>  The comment explained _why_ those checks were done, ie, it generates bad
>  builds.  The pre-processor code simply states _which_ compilers are
>  acceptable.

Is it not enough that the error and warning messages them selves state *why*
the checks are being done?

#if (__GNUC__ < 3) || (__GNUC__ = 3 && __GNUC_MINOR__ < 2)
# error Sorry, your GCC is too old. It builds incorrect kernels.
#elif __GNUC__ = 4 && __GNUC_MINOR__ = 1 && __GNUC_PATCHLEVEL__ = 0
# warning gcc-4.1.0 is known to miscompile the kernel.  A different compiler
version is recommended.
#elif __GNUC__ = 4
...

I think that it is pretty explicit *why* the error and warning message are
being given in this code snip.


Ethan
_______________________________________________
Kernel-janitors mailing list
Kernel-janitors@lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/kernel-janitors

Re: [KJ] [PATCH] Centralize checking for adequate version of gcc.

[Robert P. J. Day]

On Fri, 15 Jun 2007, Vegard Nossum wrote:

> On 6/15/07, Robert P. J. Day <rpjday@mindspring.com> wrote:
> >   here's a patch i submitted to andrew to centralize checking for a
> > proper version of gcc when building the kernel, but andrew claims it
> > broke the build.  i'm trying to reproduce the problem, and i'd
> > appreciate it if anyone else wants to apply this patch and try a build
> > based on "allyesconfig" or "allmodconfig" on x86 to see if they run
> > into some kind of issue, cuz i'm just not seeing it here.  thanks.
>
> It works for me (using gcc 4.1.1 for i386 on linus's tree).

well, that's just weird.  andrew insisted that the build failed for
him -- *i* can't reproduce it, and it seems to work for you, and that
patch is so innocuous that i just don't understand how it could have
screwed things up.

how odd.

rday
-- 
====================================
Robert P. J. Day
Linux Consulting, Training and Annoying Kernel Pedantry
Waterloo, Ontario, CANADA

http://fsdev.net/wiki/index.php?title=Main_Page
====================================
_______________________________________________
Kernel-janitors mailing list
Kernel-janitors@lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/kernel-janitors

Re: [KJ] [PATCH] Centralize checking for adequate version of gcc.

[Adrian Bunk]

On Fri, Jun 15, 2007 at 06:52:17AM -0400, Robert P. J. Day wrote:
>...
> --- a/include/linux/compiler.h
> +++ b/include/linux/compiler.h
>...
> +#elif __GNUC__ = 4 && __GNUC_MINOR__ = 1 && __GNUC_PATCHLEVEL__ = 0
> +# warning gcc-4.1.0 is known to miscompile the kernel.  A different compiler version is recommended.
> +#elif __GNUC__ = 4
>  # include <linux/compiler-gcc4.h>
>...

This breaks compilation with gcc 4.1.0

cu
Adrian

-- 

       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed

_______________________________________________
Kernel-janitors mailing list
Kernel-janitors@lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/kernel-janitors

Re: [KJ] [PATCH] Centralize checking for adequate version of gcc.

[Matthew Wilcox]

On Wed, Jun 20, 2007 at 11:38:07PM +0200, Adrian Bunk wrote:
> On Fri, Jun 15, 2007 at 06:52:17AM -0400, Robert P. J. Day wrote:
> >...
> > --- a/include/linux/compiler.h
> > +++ b/include/linux/compiler.h
> >...
> > +#elif __GNUC__ = 4 && __GNUC_MINOR__ = 1 && __GNUC_PATCHLEVEL__ = 0
> > +# warning gcc-4.1.0 is known to miscompile the kernel.  A different compiler version is recommended.
> > +#elif __GNUC__ = 4
> >  # include <linux/compiler-gcc4.h>
> >...
> 
> This breaks compilation with gcc 4.1.0

Also perhaps a more serious problem is that instead of emitting the
warning once, it's now emitted for every single file.
_______________________________________________
Kernel-janitors mailing list
Kernel-janitors@lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/kernel-janitors

Re: [KJ] [PATCH] Centralize checking for adequate version of gcc.

[Robert P. J. Day]

On Wed, 20 Jun 2007, Matthew Wilcox wrote:

> On Wed, Jun 20, 2007 at 11:38:07PM +0200, Adrian Bunk wrote:
> > On Fri, Jun 15, 2007 at 06:52:17AM -0400, Robert P. J. Day wrote:
> > >...
> > > --- a/include/linux/compiler.h
> > > +++ b/include/linux/compiler.h
> > >...
> > > +#elif __GNUC__ = 4 && __GNUC_MINOR__ = 1 && __GNUC_PATCHLEVEL__ = 0
> > > +# warning gcc-4.1.0 is known to miscompile the kernel.  A different compiler version is recommended.
> > > +#elif __GNUC__ = 4
> > >  # include <linux/compiler-gcc4.h>
> > >...
> >
> > This breaks compilation with gcc 4.1.0
>
> Also perhaps a more serious problem is that instead of emitting the
> warning once, it's now emitted for every single file.

ok, i see the problem.  an #error is once, but a #warning is forever.

rday
-- 
====================================
Robert P. J. Day
Linux Consulting, Training and Annoying Kernel Pedantry
Waterloo, Ontario, CANADA

http://fsdev.net/wiki/index.php?title=Main_Page
====================================
_______________________________________________
Kernel-janitors mailing list
Kernel-janitors@lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/kernel-janitors

Re: [KJ] [PATCH] Centralize checking for adequate version of gcc.

[Jaco Kroon]

Robert P. J. Day wrote:
> On Wed, 20 Jun 2007, Matthew Wilcox wrote:
> 
>> On Wed, Jun 20, 2007 at 11:38:07PM +0200, Adrian Bunk wrote:
>>> On Fri, Jun 15, 2007 at 06:52:17AM -0400, Robert P. J. Day wrote:
>>>> ...
>>>> --- a/include/linux/compiler.h
>>>> +++ b/include/linux/compiler.h
>>>> ...
>>>> +#elif __GNUC__ = 4 && __GNUC_MINOR__ = 1 && __GNUC_PATCHLEVEL__ = 0
>>>> +# warning gcc-4.1.0 is known to miscompile the kernel.  A different compiler version is recommended.
>>>> +#elif __GNUC__ = 4
>>>>  # include <linux/compiler-gcc4.h>
>>>> ...
>>> This breaks compilation with gcc 4.1.0
>> Also perhaps a more serious problem is that instead of emitting the
>> warning once, it's now emitted for every single file.
> 
> ok, i see the problem.  an #error is once, but a #warning is forever.

Perhaps that'll get people to take the warning seriously? :p

Jaco
_______________________________________________
Kernel-janitors mailing list
Kernel-janitors@lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/kernel-janitors

Re: [KJ] [PATCH] Centralize checking for adequate version of gcc.

[Robert P. J. Day]

On Thu, 21 Jun 2007, Jaco Kroon wrote:

> Robert P. J. Day wrote:
> > On Wed, 20 Jun 2007, Matthew Wilcox wrote:
> >
> > > On Wed, Jun 20, 2007 at 11:38:07PM +0200, Adrian Bunk wrote:
> > > > On Fri, Jun 15, 2007 at 06:52:17AM -0400, Robert P. J. Day wrote:
> > > > > ...
> > > > > --- a/include/linux/compiler.h
> > > > > +++ b/include/linux/compiler.h
> > > > > ...
> > > > > +#elif __GNUC__ = 4 && __GNUC_MINOR__ = 1 && __GNUC_PATCHLEVEL__ =
> > > > > 0
> > > > > +# warning gcc-4.1.0 is known to miscompile the kernel.  A different
> > > > > compiler version is recommended.
> > > > > +#elif __GNUC__ = 4
> > > > >  # include <linux/compiler-gcc4.h>
> > > > > ...
> > > > This breaks compilation with gcc 4.1.0
> > > Also perhaps a more serious problem is that instead of emitting the
> > > warning once, it's now emitted for every single file.
> >
> > ok, i see the problem.  an #error is once, but a #warning is forever.
>
> Perhaps that'll get people to take the warning seriously? :p

actually, it made me wonder why gcc < 3.2.0 is a *fatal* choice, but
gcc = 4.1.0 is just a *warning*.  if 4.1.0 is "known to miscompile
the kernel," why not just make it unacceptable?  why the double
standard here?

rday
-- 
====================================
Robert P. J. Day
Linux Consulting, Training and Annoying Kernel Pedantry
Waterloo, Ontario, CANADA

http://fsdev.net/wiki/index.php?title=Main_Page
====================================
_______________________________________________
Kernel-janitors mailing list
Kernel-janitors@lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/kernel-janitors

Re: [KJ] [PATCH] Centralize checking for adequate version of gcc.

[Alexey Dobriyan]

On 6/21/07, Robert P. J. Day <rpjday@mindspring.com> wrote:
> actually, it made me wonder why gcc < 3.2.0 is a *fatal* choice, but
> gcc = 4.1.0 is just a *warning*.  if 4.1.0 is "known to miscompile
> the kernel," why not just make it unacceptable?  why the double
> standard here?

Runtime checking of gcc for known miscompilation bugs is welcome. :-)
_______________________________________________
Kernel-janitors mailing list
Kernel-janitors@lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/kernel-janitors

Re: [KJ] [PATCH] Centralize checking for adequate version of gcc.

[Robert P. J. Day]

On Thu, 21 Jun 2007, Alexey Dobriyan wrote:

> On 6/21/07, Robert P. J. Day <rpjday@mindspring.com> wrote:
> > actually, it made me wonder why gcc < 3.2.0 is a *fatal* choice, but
> > gcc = 4.1.0 is just a *warning*.  if 4.1.0 is "known to miscompile
> > the kernel," why not just make it unacceptable?  why the double
> > standard here?
>
> Runtime checking of gcc for known miscompilation bugs is welcome. :-)

but that wasn't the question -- the question was whether trying to
compile the kernel with gcc-4.1.0 should be an error rather than just
a warning.  personally, i don't know what the problem is but the
current source tree does warn that gcc-4.1.0 will miscompile the
kernel.

if that's really true, why allow people to keep using it?

rday
-- 
====================================
Robert P. J. Day
Linux Consulting, Training and Annoying Kernel Pedantry
Waterloo, Ontario, CANADA

http://fsdev.net/wiki/index.php?title=Main_Page
====================================
_______________________________________________
Kernel-janitors mailing list
Kernel-janitors@lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/kernel-janitors

Re: [KJ] [PATCH] Centralize checking for adequate version of gcc.

[Alexey Dobriyan]

On Thu, Jun 21, 2007 at 12:04:08PM -0400, Robert P. J. Day wrote:
> On Thu, 21 Jun 2007, Alexey Dobriyan wrote:
> > On 6/21/07, Robert P. J. Day <rpjday@mindspring.com> wrote:
> > > actually, it made me wonder why gcc < 3.2.0 is a *fatal* choice, but
> > > gcc = 4.1.0 is just a *warning*.  if 4.1.0 is "known to miscompile
> > > the kernel," why not just make it unacceptable?  why the double
> > > standard here?
> >
> > Runtime checking of gcc for known miscompilation bugs is welcome. :-)
>
> but that wasn't the question -- the question was whether trying to
> compile the kernel with gcc-4.1.0 should be an error rather than just
> a warning.  personally, i don't know what the problem is but the
> current source tree does warn that gcc-4.1.0 will miscompile the
> kernel.
>
> if that's really true, why allow people to keep using it?

Distros in theory can backport fix for gcc and leave version intact.
Checking if compiler really miscompiles would be way more smart than
outright rejecting compiler.

_______________________________________________
Kernel-janitors mailing list
Kernel-janitors@lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/kernel-janitors

Re: [KJ] [PATCH] Centralize checking for adequate version of gcc.

[Thomas De Schampheleire]

2007/6/21, Alexey Dobriyan <adobriyan@gmail.com>:
> On Thu, Jun 21, 2007 at 12:04:08PM -0400, Robert P. J. Day wrote:
> > On Thu, 21 Jun 2007, Alexey Dobriyan wrote:
> > > On 6/21/07, Robert P. J. Day <rpjday@mindspring.com> wrote:
> > > > actually, it made me wonder why gcc < 3.2.0 is a *fatal* choice, but
> > > > gcc = 4.1.0 is just a *warning*.  if 4.1.0 is "known to miscompile
> > > > the kernel," why not just make it unacceptable?  why the double
> > > > standard here?
> > >
> > > Runtime checking of gcc for known miscompilation bugs is welcome. :-)
> >
> > but that wasn't the question -- the question was whether trying to
> > compile the kernel with gcc-4.1.0 should be an error rather than just
> > a warning.  personally, i don't know what the problem is but the
> > current source tree does warn that gcc-4.1.0 will miscompile the
> > kernel.
> >
> > if that's really true, why allow people to keep using it?
>
> Distros in theory can backport fix for gcc and leave version intact.

Yes, but this would also be the case for gcc < 3.2.0.
_______________________________________________
Kernel-janitors mailing list
Kernel-janitors@lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/kernel-janitors