Re: arch/x86/include/asm/cmpxchg.h:245:2: error: 'asm' operand has impossible constraints

[Juergen Gross <jgross@suse.com>]

On 30/11/2018 07:44, kbuild test robot wrote:
> Hi Juergen,
> 
> FYI, the error/warning still remains.

This problem is one of gcc. It is in no way the (direct) result of my
patch. Please see my gcc bug report:

https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=908154


Juergen

> 
> tree:   https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
> head:   94f371cb73944b410a269d570d6946c042f2ddd0
> commit: 6da63eb241a05b0e676d68975e793c0521387141 x86/paravirt: Move the pv_irq_ops under the PARAVIRT_XXL umbrella
> date:   3 months ago
> config: i386-randconfig-sb0-11301144 (attached as .config)
> compiler: gcc-4.9 (Debian 4.9.4-2) 4.9.4
> reproduce:
>         git checkout 6da63eb241a05b0e676d68975e793c0521387141
>         # save the attached .config to linux build tree
>         make ARCH=i386 
> 
> All errors (new ones prefixed by >>):
> 
>    In file included from arch/x86/include/asm/atomic.h:8:0,
>                     from arch/x86/include/asm/msr.h:67,
>                     from arch/x86/include/asm/processor.h:21,
>                     from arch/x86/include/asm/cpufeature.h:5,
>                     from arch/x86/include/asm/thread_info.h:53,
>                     from include/linux/thread_info.h:38,
>                     from arch/x86/include/asm/preempt.h:7,
>                     from include/linux/preempt.h:81,
>                     from include/linux/spinlock.h:51,
>                     from include/linux/mmzone.h:8,
>                     from include/linux/gfp.h:6,
>                     from include/linux/mm.h:10,
>                     from mm/slub.c:13:
>    mm/slub.c: In function '__slab_free':
>>> arch/x86/include/asm/cmpxchg.h:245:2: error: 'asm' operand has impossible constraints
>      asm volatile(pfx "cmpxchg%c4b %2; sete %0"   \
>      ^
>    arch/x86/include/asm/cmpxchg.h:254:2: note: in expansion of macro '__cmpxchg_double'
>      __cmpxchg_double(LOCK_PREFIX, p1, p2, o1, o2, n1, n2)
>      ^
>    include/asm-generic/atomic-instrumented.h:457:2: note: in expansion of macro 'arch_cmpxchg_double'
>      arch_cmpxchg_double(__ai_p1, (p2), (o1), (o2), (n1), (n2)); \
>      ^
>    mm/slub.c:403:7: note: in expansion of macro 'cmpxchg_double'
>       if (cmpxchg_double(&page->freelist, &page->counters,
>           ^
> 
> vim +/asm +245 arch/x86/include/asm/cmpxchg.h
> 
> 3d94ae0c Jeremy Fitzhardinge 2011-09-28  235  
> cdcd6298 Jan Beulich         2012-01-02  236  #define __cmpxchg_double(pfx, p1, p2, o1, o2, n1, n2)			\
> cdcd6298 Jan Beulich         2012-01-02  237  ({									\
> cdcd6298 Jan Beulich         2012-01-02  238  	bool __ret;							\
> cdcd6298 Jan Beulich         2012-01-02  239  	__typeof__(*(p1)) __old1 = (o1), __new1 = (n1);			\
> cdcd6298 Jan Beulich         2012-01-02  240  	__typeof__(*(p2)) __old2 = (o2), __new2 = (n2);			\
> cdcd6298 Jan Beulich         2012-01-02  241  	BUILD_BUG_ON(sizeof(*(p1)) != sizeof(long));			\
> cdcd6298 Jan Beulich         2012-01-02  242  	BUILD_BUG_ON(sizeof(*(p2)) != sizeof(long));			\
> cdcd6298 Jan Beulich         2012-01-02  243  	VM_BUG_ON((unsigned long)(p1) % (2 * sizeof(long)));		\
> cdcd6298 Jan Beulich         2012-01-02  244  	VM_BUG_ON((unsigned long)((p1) + 1) != (unsigned long)(p2));	\
> cdcd6298 Jan Beulich         2012-01-02 @245  	asm volatile(pfx "cmpxchg%c4b %2; sete %0"			\
> cdcd6298 Jan Beulich         2012-01-02  246  		     : "=a" (__ret), "+d" (__old2),			\
> cdcd6298 Jan Beulich         2012-01-02  247  		       "+m" (*(p1)), "+m" (*(p2))			\
> cdcd6298 Jan Beulich         2012-01-02  248  		     : "i" (2 * sizeof(long)), "a" (__old1),		\
> cdcd6298 Jan Beulich         2012-01-02  249  		       "b" (__new1), "c" (__new2));			\
> cdcd6298 Jan Beulich         2012-01-02  250  	__ret;								\
> cdcd6298 Jan Beulich         2012-01-02  251  })
> cdcd6298 Jan Beulich         2012-01-02  252  
> 
> :::::: The code at line 245 was first introduced by commit
> :::::: cdcd629869fabcd38ebd24a03b0a05ec1cbcafb0 x86: Fix and improve cmpxchg_double{,_local}()
> 
> :::::: TO: Jan Beulich <JBeulich@suse.com>
> :::::: CC: Ingo Molnar <mingo@elte.hu>
> 
> ---
> 0-DAY kernel test infrastructure                Open Source Technology Center
> https://lists.01.org/pipermail/kbuild-all                   Intel Corporation
>