target-ppc : Make hreg_store_msr returns exception
PATCH 1/1 target-ppc : Make hreg_store_msr returns exception
hreg_store_msr was both returning 0 or an exception number which could be POWERPC_EXCP_NONE. In case POWERPC_EXCP_NONE was returned, helper_store_msr was handling uncorrectly this return value and raise an exception. This patch correct this behavior by making hreg_store_msr always return an exception number, and helper_store_msr test for POWERPC_EXCP_NONE.
--- target-ppc/excp_helper.c | 2 +- target-ppc/helper_regs.h | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-)
target-ppc/excp_helper.c +1 -1
@@ -855,7 +855,7 @@ void helper_store_msr(CPUPPCState *env, target_ulong val)CPUState *cs; val =hreg_store_msr (env, val,0 ); -if (val !=0 ) { +if (val !=POWERPC_EXCP_NONE ) { cs =CPU (ppc_env_get_cpu (env)); cs->interrupt_request |=CPU_INTERRUPT_EXITTB ;helper_raise_exception (env, val);
target-ppc/helper_regs.h +1 -1
@@ -72,7 +72,7 @@ static inline int hreg_store_msr(CPUPPCState *env, target_ulong value,CPUState *cs =CPU (ppc_env_get_cpu (env));#endif - excp =0 ; + excp =POWERPC_EXCP_NONE ; value &= env->msr_mask;#if !defined(CONFIG_USER_ONLY )if (!alter_hv) {
Signature · 1 line
1.7.10.4
Quoted 7 lines · Pierre Mallard
hreg_store_msrwas both returning 0 or an exception number which could bePOWERPC_EXCP_NONE. In casePOWERPC_EXCP_NONEwas returned,helper_store_msrwas handling uncorrectly this return value and raise an exception. This patch correct this behavior by makinghreg_store_msralways return an exception number, andhelper_store_msrtest forPOWERPC_EXCP_NONE.Signed-off-by Pierre Mallard <[email protected]>
Are you sure that's a bug, not a feature? I think this idea behind POWERPC_EXCP_NONE in this code path is to pull the guest out of its TB so that no stale code could be executed after an MSR.IR/DR change.
Alex