Add RAM -> physical addr mapping in MCE simulation
PATCH 1/1 Add RAM -> physical addr mapping in MCE simulation
In QEMU-KVM, physical address != RAM address. While MCE simulation needs physical address instead of RAM address. So kvm_physical_memory_addr_from_ram() is implemented to do the conversion, and it is invoked before being filled in the IA32_MCi_ADDR MSR.
--- kvm-all.c | 18 ++++++++++++++++++ kvm.h | 3 +++ qemu-kvm.c | 13 ++++++++++--- 3 files changed, 31 insertions(+), 3 deletions(-)
qemu-kvm.c +10 -3
@@ -1137,12 +1137,15 @@ static void sigbus_handler(int n, structif (first_cpu->mcg_cap && siginfo->ssi_addr && siginfo->ssi_code ==BUS_MCEERR_AO ) {uint64_t status; +void *vaddr; +ram_addr_t ram_addr;unsigned long paddr;CPUState *cenv;/* Hope we are lucky for AO MCE */ -if (do_qemu_ram_addr_from_host ((void *)(intptr_t )siginfo->ssi_addr, - &paddr)) { + vaddr = (void *)(intptr_t )siginfo->ssi_addr; +if (do_qemu_ram_addr_from_host (vaddr, &ram_addr) || + !kvm_physical_memory_addr_from_ram (kvm_state, ram_addr, &paddr)) {fprintf (stderr,"Hardware memory error for memory used by " "QEMU itself instead of guest system!: %llx\n" , (unsigned long long )siginfo->ssi_addr);
@@ -1316,6 +1319,8 @@ static void kvm_on_sigbus(CPUState *env,struct kvm_x86_mce mce = { .bank =9 , }; +void *vaddr; +ram_addr_t ram_addr;unsigned long paddr;int r;
@@ -1347,7 +1352,9 @@ static void kvm_on_sigbus(CPUState *env, mce.misc = (MCM_ADDR_PHYS <<6 ) |0xc ; mce.mcg_status =MCG_STATUS_MCIP |MCG_STATUS_RIPV ; } -if (do_qemu_ram_addr_from_host ((void *)siginfo->si_addr, &paddr)) { + vaddr = (void *)siginfo->si_addr; +if (do_qemu_ram_addr_from_host (vaddr, &ram_addr) || + !kvm_physical_memory_addr_from_ram (kvm_state, ram_addr, &paddr)) {fprintf (stderr,"Hardware memory error for memory used by " "QEMU itself instaed of guest system!\n" );/* Hope we are lucky for AO MCE */
kvm-all.c +18 -0
@@ -141,6 +141,24 @@ static KVMSlot *kvm_lookup_overlapping_sreturn found; } +int kvm_physical_memory_addr_from_ram (KVMState *s,ram_addr_t ram_addr, +target_phys_addr_t *phys_addr) +{ +int i; + +for (i =0 ; i <ARRAY_SIZE (s->slots); i++) { +KVMSlot *mem = &s->slots[i]; + +if (ram_addr >= mem->phys_offset && + ram_addr < mem->phys_offset + mem->memory_size) { + *phys_addr = mem->start_addr + (ram_addr - mem->phys_offset); +return 1 ; + } + } + +return 0 ; +} +static int kvm_set_user_memory_region (KVMState *s,KVMSlot *slot) {struct kvm_userspace_memory_region mem;
kvm.h +3 -0
@@ -195,4 +195,7 @@ int kvm_set_irqfd(int gsi, int fd, bool#endif int kvm_set_ioeventfd_pio_word (int fd,uint16_t adr,uint16_t val,bool assign); + +int kvm_physical_memory_addr_from_ram (KVMState *s,ram_addr_t ram_addr, +target_phys_addr_t *phys_addr);#endif
Marcelo Tosatti
AppliedQuoted 8 lines · Huang Ying
In QEMU-KVM, physical address != RAM address. While MCE simulation needs physical address instead of RAM address. So
kvm_physical_memory_addr_from_ram()is implemented to do the conversion, and it is invoked before being filled in theIA32_MCi_ADDRMSR.Reported-by Dean Nelson <[email protected]>Signed-off-by Huang Ying <[email protected]>
Applied, thanks.
Can you please submit the code necessary to run the test suite upstream? Is there anything else needed other than p2v monitor command?
TIA
Hi, Marcelo,
Quoted 7 lines · Marcelo Tosatti
Quoted 6 lines · Huang Ying
In QEMU-KVM, physical address != RAM address. While MCE simulation needs physical address instead of RAM address. So
kvm_physical_memory_addr_from_ram()is implemented to do the conversion, and it is invoked before being filled in theIA32_MCi_ADDRMSR.
Quoted 7 lines
Quoted 2 lines
Reported-by Dean Nelson <[email protected]>Signed-off-by Huang Ying <[email protected]>Applied, thanks.
Can you please submit the code necessary to run the test suite upstream? Is there anything else needed other than p2v monitor command?
The only missing part is p2v monitor command. I will re-post it.
Do you want to merge the test suite into kvm autotest tool? If so, Shaohui can help to do that. But it seems that he is busy on some other project now.
Best Regards,
Huang Ying