On Sun, May 08, 2011 at 10:24:39AM +0200, Stefan Weil wrote:
> cppcheck report:
> target-mips/helper.c:276: style:
> Variable 'access_type' is assigned a value that is never used
>
> 'access_type' is only used for system emulation, so the patch
> simply fixes the conditional compilation.
>
> There remains an issue with the return value 'ret' in user mode emulation.
> The assignment ret = TLBRET_NOMATCH is without effect because there is
> a ret = 1 later. Which one is correct?
ret = 1 is the correct version.
I have applied a slightly different patch also fixing the ret issue.
> Cc: Aurelien Jarno <aurel...@aurel32.net>
> Signed-off-by: Stefan Weil <w...@mail.berlios.de>
> ---
> target-mips/helper.c | 8 ++++----
> 1 files changed, 4 insertions(+), 4 deletions(-)
>
> diff --git a/target-mips/helper.c b/target-mips/helper.c
> index bdc1e53..f3cd521 100644
> --- a/target-mips/helper.c
> +++ b/target-mips/helper.c
> @@ -272,8 +272,8 @@ int cpu_mips_handle_mmu_fault (CPUState *env,
> target_ulong address, int rw,
> #if !defined(CONFIG_USER_ONLY)
> target_phys_addr_t physical;
> int prot;
> -#endif
> int access_type;
> +#endif
> int ret = 0;
>
> #if 0
> @@ -284,13 +284,13 @@ int cpu_mips_handle_mmu_fault (CPUState *env,
> target_ulong address, int rw,
>
> rw &= 1;
>
> +#if defined(CONFIG_USER_ONLY)
> + ret = TLBRET_NOMATCH;
> +#else
> /* data access */
> /* XXX: put correct access by using cpu_restore_state()
> correctly */
> access_type = ACCESS_INT;
> -#if defined(CONFIG_USER_ONLY)
> - ret = TLBRET_NOMATCH;
> -#else
> ret = get_physical_address(env, &physical, &prot,
> address, rw, access_type);
> qemu_log("%s address=" TARGET_FMT_lx " ret %d physical " TARGET_FMT_plx
> " prot %d\n",
> --
> 1.7.2.5
>
>
>
--
Aurelien Jarno GPG: 1024D/F1BCDB73
aurel...@aurel32.net http://www.aurel32.net
