On Mon, Dec 8, 2014 at 9:03 PM, Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> wrote:
On Tue, 9 Dec 2014 02:40:09 +0800 kbuild test robot <fengguang.wu@xxxxxxxxx> wrote:
> tree: git://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git master
> head: cf12164be498180dc466ef97194ca7755ea39f3b
> commit: b4baa9e36be0651f7eb15077af5e0eff53b7691b [10653/11539] x86: hook up execveat system call
> reproduce:
> # apt-get install sparse
> git checkout b4baa9e36be0651f7eb15077af5e0eff53b7691b
> make ARCH=x86_64 allmodconfig
> make C=1 CF=-D__CHECK_ENDIAN__
>
>
> sparse warnings: (new ones prefixed by >>)
>
> arch/x86/ia32/audit.c:38:14: sparse: undefined identifier '__NR_execveat'
> >> arch/x86/ia32/audit.c:38:14: sparse: incompatible types for 'case' statement
> arch/x86/ia32/audit.c:38:14: sparse: Expected constant _expression_ in case statement
> arch/x86/ia32/audit.c: In function 'ia32_classify_syscall':
> arch/x86/ia32/audit.c:38:7: error: '__NR_execveat' undeclared (first use in this function)
> case __NR_execveat:
> ^
> arch/x86/ia32/audit.c:38:7: note: each undeclared identifier is reported only once for each function it appears in
> --
Confused. This makes no sense and I can't reproduce it.
Ditto.
Someone else did previously[1] have a build problem from a stale copy of
arch/x86/include/generated/asm/unistd_32.h in their tree, but I don't know
how that could happen.