When I try to build a ppc64 target on a ppc64 host (gcc 8.1.1), I have:

.../target/ppc/int_helper.c: In function 'helper_vinsertb':
.../target/ppc/int_helper.c:1954:32: error: array subscript 18446744073709551608 is above array bounds of 'uint8_t[16]' {aka 'unsigned char[16]'} [-Werror=array-bounds]
         memmove(&r->u8[index], &b->u8[8 - sizeof(r->element)],              \
                                ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
.../target/ppc/int_helper.c:1965:1: note: in expansion of macro 'VINSERT'

If we compare with the macro for ppc64le, we can see
sizeof(r->element[0]) should be used instead of sizeof(r->element).

And VINSERT uses only u8, u16, u32 and u64, so the maximum value
of sizeof(r->element[0]) is 8

Suggested-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
---
 target/ppc/int_helper.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/target/ppc/int_helper.c b/target/ppc/int_helper.c
index 03d37da79f..d52338ed71 100644
--- a/target/ppc/int_helper.c
+++ b/target/ppc/int_helper.c
@@ -1951,7 +1951,7 @@ VSPLT(w, u32)
 #define VINSERT(suffix, element)                                            \
     void helper_vinsert##suffix(ppc_avr_t *r, ppc_avr_t *b, uint32_t index) \
     {                                                                       \
-        memmove(&r->u8[index], &b->u8[8 - sizeof(r->element)],              \
+        memmove(&r->u8[index], &b->u8[8 - sizeof(r->element[0])],           \
                sizeof(r->element[0]));                                      \
     }
 #else
-- 
2.17.1

On 07/04/2018 11:38 AM, Laurent Vivier wrote:
> When I try to build a ppc64 target on a ppc64 host (gcc 8.1.1), I have:
> 
> .../target/ppc/int_helper.c: In function 'helper_vinsertb':
> .../target/ppc/int_helper.c:1954:32: error: array subscript 18446744073709551608 is above array bounds of 'uint8_t[16]' {aka 'unsigned char[16]'} [-Werror=array-bounds]
>          memmove(&r->u8[index], &b->u8[8 - sizeof(r->element)],              \
>                                 ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> .../target/ppc/int_helper.c:1965:1: note: in expansion of macro 'VINSERT'
> 
> If we compare with the macro for ppc64le, we can see
> sizeof(r->element[0]) should be used instead of sizeof(r->element).
> 
> And VINSERT uses only u8, u16, u32 and u64, so the maximum value
> of sizeof(r->element[0]) is 8
> 
> Suggested-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
> Signed-off-by: Laurent Vivier <laurent@vivier.eu>
> ---
>  target/ppc/int_helper.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)

Reviewed-by: Richard Henderson <richard.henderson@linaro.org>


r~

On Wed, Jul 04, 2018 at 08:38:55PM +0200, Laurent Vivier wrote:
1;5202;0c> When I try to build a ppc64 target on a ppc64 host (gcc 8.1.1), I have:
> 
> .../target/ppc/int_helper.c: In function 'helper_vinsertb':
> .../target/ppc/int_helper.c:1954:32: error: array subscript 18446744073709551608 is above array bounds of 'uint8_t[16]' {aka 'unsigned char[16]'} [-Werror=array-bounds]
>          memmove(&r->u8[index], &b->u8[8 - sizeof(r->element)],              \
>                                 ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> .../target/ppc/int_helper.c:1965:1: note: in expansion of macro 'VINSERT'
> 
> If we compare with the macro for ppc64le, we can see
> sizeof(r->element[0]) should be used instead of sizeof(r->element).
> 
> And VINSERT uses only u8, u16, u32 and u64, so the maximum value
> of sizeof(r->element[0]) is 8
> 
> Suggested-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
> Signed-off-by: Laurent Vivier <laurent@vivier.eu>

Applied to ppc-for-3.0, thanks.

> ---
>  target/ppc/int_helper.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/target/ppc/int_helper.c b/target/ppc/int_helper.c
> index 03d37da79f..d52338ed71 100644
> --- a/target/ppc/int_helper.c
> +++ b/target/ppc/int_helper.c
> @@ -1951,7 +1951,7 @@ VSPLT(w, u32)
>  #define VINSERT(suffix, element)                                            \
>      void helper_vinsert##suffix(ppc_avr_t *r, ppc_avr_t *b, uint32_t index) \
>      {                                                                       \
> -        memmove(&r->u8[index], &b->u8[8 - sizeof(r->element)],              \
> +        memmove(&r->u8[index], &b->u8[8 - sizeof(r->element[0])],           \
>                 sizeof(r->element[0]));                                      \
>      }
>  #else

-- 
David Gibson			| I'll have my music baroque, and my code
david AT gibson.dropbear.id.au	| minimalist, thank you.  NOT _the_ _other_
				| _way_ _around_!
http://www.ozlabs.org/~dgibson