On Wed, Aug 10, 2011 at 09:59:26PM -0400, Stephen Rothwell wrote:
> Hi Linus,
>
> With Linus' tree, today's linux-next build (x86_64 allmodconfig) failed
> like this:
>
> drivers/edac/i7core_edac.c:259:19: error: field 'edac_mce' has incomplete type
> drivers/edac/i7core_edac.c:262:24: error: 'MCE_LOG_LEN' undeclared here (not in a function)
> drivers/edac/i7core_edac.c: In function 'i7core_mce_output_error':
> drivers/edac/i7core_edac.c:1671:25: error: dereferencing pointer to incomplete type
> drivers/edac/i7core_edac.c:1672:20: error: dereferencing pointer to incomplete type
> drivers/edac/i7core_edac.c:1673:23: error: dereferencing pointer to incomplete type
> drivers/edac/i7core_edac.c:1674:15: error: dereferencing pointer to incomplete type
> drivers/edac/i7core_edac.c:1675:18: error: dereferencing pointer to incomplete type
> drivers/edac/i7core_edac.c:1676:18: error: dereferencing pointer to incomplete type
> drivers/edac/i7core_edac.c:1680:7: error: dereferencing pointer to incomplete type
> drivers/edac/i7core_edac.c:1742:22: error: dereferencing pointer to incomplete type
> drivers/edac/i7core_edac.c:1742:31: error: dereferencing pointer to incomplete type
> drivers/edac/i7core_edac.c:1743:39: error: dereferencing pointer to incomplete type
> drivers/edac/i7core_edac.c:1744:15: error: dereferencing pointer to incomplete type
> drivers/edac/i7core_edac.c:1751:7: error: dereferencing pointer to incomplete type
> drivers/edac/i7core_edac.c: In function 'i7core_check_error':
> drivers/edac/i7core_edac.c:1787:50: error: dereferencing pointer to incomplete type
> drivers/edac/i7core_edac.c:1791:3: error: invalid use of undefined type 'struct mce'
> drivers/edac/i7core_edac.c:1793:49: error: dereferencing pointer to incomplete type
> drivers/edac/i7core_edac.c: In function 'i7core_mce_check_error':
> drivers/edac/i7core_edac.c:1838:11: error: dereferencing pointer to incomplete type
> drivers/edac/i7core_edac.c:1842:9: error: dereferencing pointer to incomplete type
> drivers/edac/i7core_edac.c:1847:6: error: dereferencing pointer to incomplete type
> drivers/edac/i7core_edac.c:1859:52: error: dereferencing pointer to incomplete type
> drivers/edac/i7core_edac.c:1864:9: error: dereferencing pointer to incomplete type
> drivers/edac/i7core_edac.c: In function 'i7core_unregister_mci':
> drivers/edac/i7core_edac.c:1910:2: error: implicit declaration of function 'edac_mce_unregister'
> drivers/edac/i7core_edac.c: In function 'i7core_register_mci':
> drivers/edac/i7core_edac.c:2008:2: error: implicit declaration of function 'edac_mce_register'
>
> Caused by commit af9d220bac41 ("EDAC: Correct Kconfig dependencies").
> Since CONFIG_EDAC_MCE is no longer selected most of the stuff in
> linux/edac_mce.h is not available. I have reverted that commit for today.
§$%§$&&§! (please excuse my french).
And I build-tested this patch with randconfigs a whole night and didn't
see any issues. But I think I see what happens, the above is the result
of a chain of unfortunate events :-)
* Randy mistakenly reported the original randconfig issue against
3.1-rc1. However, in reality it was against linux-next. It was caused by
the fact that i7core_edac now needs to depend on X86_MCE_INTEL (which
pulls in X86_MCE for the notifier chain).
* I sent the proper fix which was for Mauro to apply _ontop_ of
git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/edac.git linux_next
where the rest of the i7core_edac changes are. This patch depends on them.
* I wasn't clear enough with the patch though, and I see Linus has
applied it already on his tree, which is clearly too early - it has to
go in in the next merge window together with the rest of the stuff in
Mauro's branch.
So, I guess we're reverting for now, Linus?
Thanks and sorry for the trouble.
--
Regards/Gruss,
Boris.
Advanced Micro Devices GmbH
Einsteinring 24, 85609 Dornach
GM: Alberto Bozzo
Reg: Dornach, Landkreis Muenchen
HRB Nr. 43632 WEEE Registernr: 129 19551
--
To unsubscribe from this list: send the line "unsubscribe linux-next" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
