[dpdk-dev] [PATCH] eal/x86: fix clang build with -O0

[dpdk-dev] [PATCH] eal/x86: fix clang build with -O0

[Thomas Monjalon]

From: Damjan Marion <damarion@cisco.com>

Clang seems to have a bug with asm inside inline function rte_xabort():

rte_rtm.h:56:15: error: invalid operand for inline asm constraint 'i'
        asm volatile(".byte 0xc6,0xf8,%P0" :: "i" (status) : "memory");
                     ^

It is seen only when building with EXTRA_CFLAGS=-O0.

The workaround is to replace the inline function by a macro.

Fixes: ba7468997ea6 ("spinlock: add HTM lock elision for x86")

Signed-off-by: Damjan Marion <damarion@cisco.com>
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
---
 lib/librte_eal/common/include/arch/x86/rte_rtm.h | 9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)

diff --git a/lib/librte_eal/common/include/arch/x86/rte_rtm.h b/lib/librte_eal/common/include/arch/x86/rte_rtm.h
index d935641..0649f79 100644
--- a/lib/librte_eal/common/include/arch/x86/rte_rtm.h
+++ b/lib/librte_eal/common/include/arch/x86/rte_rtm.h
@@ -50,11 +50,10 @@ void rte_xend(void)
 	 asm volatile(".byte 0x0f,0x01,0xd5" ::: "memory");
 }
 
-static __attribute__((__always_inline__)) inline
-void rte_xabort(const unsigned int status)
-{
-	asm volatile(".byte 0xc6,0xf8,%P0" :: "i" (status) : "memory");
-}
+/* not an inline function to workaround a clang bug with -O0 */
+#define rte_xabort(status) do { \
+	asm volatile(".byte 0xc6,0xf8,%P0" :: "i" (status) : "memory"); \
+} while (0)
 
 static __attribute__((__always_inline__)) inline
 int rte_xtest(void)
-- 
2.7.0

Re: [dpdk-dev] [PATCH] eal/x86: fix clang build with -O0

[Olivier MATZ]

Hi Thomas,

On 06/03/2016 10:15 AM, Thomas Monjalon wrote:
> From: Damjan Marion <damarion@cisco.com>
> 
> Clang seems to have a bug with asm inside inline function rte_xabort():
> 
> rte_rtm.h:56:15: error: invalid operand for inline asm constraint 'i'
>         asm volatile(".byte 0xc6,0xf8,%P0" :: "i" (status) : "memory");
>                      ^
> 
> It is seen only when building with EXTRA_CFLAGS=-O0.
> 
> The workaround is to replace the inline function by a macro.
> 
> Fixes: ba7468997ea6 ("spinlock: add HTM lock elision for x86")
> 
> Signed-off-by: Damjan Marion <damarion@cisco.com>
> Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
> ---
>  lib/librte_eal/common/include/arch/x86/rte_rtm.h | 9 ++++-----
>  1 file changed, 4 insertions(+), 5 deletions(-)
> 
> diff --git a/lib/librte_eal/common/include/arch/x86/rte_rtm.h b/lib/librte_eal/common/include/arch/x86/rte_rtm.h
> index d935641..0649f79 100644
> --- a/lib/librte_eal/common/include/arch/x86/rte_rtm.h
> +++ b/lib/librte_eal/common/include/arch/x86/rte_rtm.h
> @@ -50,11 +50,10 @@ void rte_xend(void)
>  	 asm volatile(".byte 0x0f,0x01,0xd5" ::: "memory");
>  }
>  
> -static __attribute__((__always_inline__)) inline
> -void rte_xabort(const unsigned int status)
> -{
> -	asm volatile(".byte 0xc6,0xf8,%P0" :: "i" (status) : "memory");
> -}
> +/* not an inline function to workaround a clang bug with -O0 */
> +#define rte_xabort(status) do { \
> +	asm volatile(".byte 0xc6,0xf8,%P0" :: "i" (status) : "memory"); \
> +} while (0)
>  
>  static __attribute__((__always_inline__)) inline
>  int rte_xtest(void)
> 

Generally speaking, it's not a good idea to replace a static inline by a
macro because it relaxes type checking. Looks ok in that case to
workaround the clang issue.

Acked-by: Olivier Matz <olivier.matz@6wind.com>

Re: [dpdk-dev] [PATCH] eal/x86: fix clang build with -O0

[Thomas Monjalon]

2016-06-03 10:16, Olivier MATZ:
> On 06/03/2016 10:15 AM, Thomas Monjalon wrote:
> > From: Damjan Marion <damarion@cisco.com>
> > 
> > Clang seems to have a bug with asm inside inline function rte_xabort():
> > 
> > rte_rtm.h:56:15: error: invalid operand for inline asm constraint 'i'
> >         asm volatile(".byte 0xc6,0xf8,%P0" :: "i" (status) : "memory");
> >                      ^
> > 
> > It is seen only when building with EXTRA_CFLAGS=-O0.
> > 
> > The workaround is to replace the inline function by a macro.
> > 
> > Fixes: ba7468997ea6 ("spinlock: add HTM lock elision for x86")
> > 
> > Signed-off-by: Damjan Marion <damarion@cisco.com>
> > Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
> 
> Generally speaking, it's not a good idea to replace a static inline by a
> macro because it relaxes type checking. Looks ok in that case to
> workaround the clang issue.
> 
> Acked-by: Olivier Matz <olivier.matz@6wind.com>

Thanks,
This patch has been appplied very quickly for the purpose of a live demo
at a VPP event: https://wiki.fd.io/view/Events