The test checks that when hypervisor prefaults swapped in page it does
not generates spurious #PFs in case page was unmapped while it was
swapped out.
diff --git a/config-x86-common.mak b/config-x86-common.mak
index c8fbda7..6976f78 100644
--- a/config-x86-common.mak
+++ b/config-x86-common.mak
@@ -34,7 +34,7 @@ tests-common = $(TEST_DIR)/vmexit.flat $(TEST_DIR)/tsc.flat \
$(TEST_DIR)/realmode.flat $(TEST_DIR)/msr.flat \
$(TEST_DIR)/hypercall.flat $(TEST_DIR)/sieve.flat \
$(TEST_DIR)/kvmclock_test.flat $(TEST_DIR)/eventinj.flat \
- $(TEST_DIR)/s3.flat $(TEST_DIR)/pmu.flat
+ $(TEST_DIR)/s3.flat $(TEST_DIR)/pmu.flat $(TEST_DIR)/asyncpf.flat
ifdef API
tests-common += api/api-sample
@@ -90,6 +90,8 @@ $(TEST_DIR)/s3.elf: $(cstart.o) $(TEST_DIR)/s3.o
$(TEST_DIR)/pmu.elf: $(cstart.o) $(TEST_DIR)/pmu.o
+$(TEST_DIR)/asyncpf.elf: $(cstart.o) $(TEST_DIR)/asyncpf.o
+
arch_clean:
$(RM) $(TEST_DIR)/*.o $(TEST_DIR)/*.flat $(TEST_DIR)/*.elf \
$(TEST_DIR)/.*.d $(TEST_DIR)/lib/.*.d $(TEST_DIR)/lib/*.o
diff --git a/lib/x86/processor.h b/lib/x86/processor.h
index c7e1afb..e46d8d0 100644
--- a/lib/x86/processor.h
+++ b/lib/x86/processor.h
@@ -302,4 +302,9 @@ static inline void invlpg(void *va)
{
asm volatile("invlpg (%0)" ::"r" (va) : "memory");
}
+
+static inline void safe_halt(void)
+{
+ asm volatile("sti; hlt");
+}
#endif
diff --git a/lib/x86/vm.c b/lib/x86/vm.c
index 550ec9b..71b70fd 100644
--- a/lib/x86/vm.c
+++ b/lib/x86/vm.c
@@ -219,6 +219,11 @@ void *vmalloc(unsigned long size)
return mem;
}
+uint64_t virt_to_phys_cr3(void *mem)
+{
+ return (get_pte(phys_to_virt(read_cr3()), mem) & PTE_ADDR) + ((ulong)mem & (PAGE_SIZE - 1));
+}
+
void vfree(void *mem)
{
unsigned long size = ((unsigned long *)mem)[-1];
diff --git a/lib/x86/vm.h b/lib/x86/vm.h
index 71ab4a8..3473f8d 100644
--- a/lib/x86/vm.h
+++ b/lib/x86/vm.h
@@ -22,6 +22,7 @@ void vfree(void *mem);
void *vmap(unsigned long long phys, unsigned long size);
void *alloc_vpage(void);
void *alloc_vpages(ulong nr);
+uint64_t virt_to_phys_cr3(void *mem);
void install_pte(unsigned long *cr3,
int pte_level,
diff --git a/x86/asyncpf.c b/x86/asyncpf.c
new file mode 100644
index 0000000..62c0455
--- /dev/null
+++ b/x86/asyncpf.c
@@ -0,0 +1,113 @@
+/*
+ * Async PF test. For the test to actually do anywathing it ineeds to be started
+ * in memory cgroup with 512M of memory and with more then 1G memory provided
+ * to the guest.
+ *
+ * To create cgroup do as root:
+ * mkdir /dev/cgroup
+ * mount -t cgroup none -omemory /dev/cgroup
+ * chmod a+rxw /dev/cgroup/
+ *
+ * From a shell you will start qemu from:
+ * mkdir /dev/cgroup/1
+ * echo $$ > /dev/cgroup/1/tasks
+ * echo 512M > /dev/cgroup/1/memory.limit_in_bytes
+ *
+ */
+#include "x86/msr.h"
+#include "x86/processor.h"
+#include "x86/apic-defs.h"
+#include "x86/apic.h"
+#include "x86/desc.h"
+#include "x86/isr.h"
+#include "x86/vm.h"
+
+#include "libcflat.h"
+#include <stdint.h>
+
+#define KVM_PV_REASON_PAGE_NOT_PRESENT 1
+#define KVM_PV_REASON_PAGE_READY 2
+
+#define MSR_KVM_ASYNC_PF_EN 0x4b564d02
+
+#define KVM_ASYNC_PF_ENABLED (1 << 0)
+#define KVM_ASYNC_PF_SEND_ALWAYS (1 << 1)
+
+volatile uint32_t apf_reason __attribute__((aligned(64)));
+char *buf;
+volatile uint64_t i;
+volatile uint64_t phys;
+bool fail;
+
+static inline uint32_t get_apf_reason(void)
+{
+ uint32_t r = apf_reason;
+ apf_reason = 0;
+ return r;
+}
+
+static void pf_isr(struct ex_regs *r)
+{
+ void* virt = (void*)((ulong)(buf+i) & ~(PAGE_SIZE-1));
+ uint32_t reason = get_apf_reason();
+
+ switch (reason) {
+ case 0:
+ printf("unexpected #PF at %p\n", read_cr2());
+ fail = true;
+ break;
+ case KVM_PV_REASON_PAGE_NOT_PRESENT:
+ phys = virt_to_phys_cr3(virt);
+ install_pte(phys_to_virt(read_cr3()), 1, virt, phys, 0);
+ write_cr3(read_cr3());
+ printf("Got not present #PF token %x virt addr %p phys addr %p\n", read_cr2(), virt, phys);
+ while(phys) {
+ safe_halt(); /* enables irq */
+ irq_disable();
+ }
+ break;
+ case KVM_PV_REASON_PAGE_READY:
+ printf("Got present #PF token %x\n", read_cr2());
+ if ((uint32_t)read_cr2() == ~0)
+ break;
+ install_pte(phys_to_virt(read_cr3()), 1, virt, phys | PTE_PRESENT | PTE_WRITE, 0);
+ write_cr3(read_cr3());
+ phys = 0;
+ break;
+ default:
+ printf("unexpected async pf reason %d\n", reason);
+ fail = true;
+ break;
+ }
+}
+
+#define MEM 1ull*1024*1024*1024
+
+int main(int ac, char **av)
+{
+ int loop = 2;
+
+ setup_vm();
+ setup_idt();
+ setup_gdt();
+ printf("install handler\n");
+ handle_exception(14, pf_isr);
+ apf_reason = 0;
+ printf("enable async pf\n");
+ wrmsr(MSR_KVM_ASYNC_PF_EN, virt_to_phys((void*)&apf_reason) |
+ KVM_ASYNC_PF_SEND_ALWAYS | KVM_ASYNC_PF_ENABLED);
+ printf("alloc memory\n");
+ buf = vmalloc(MEM);
+ irq_enable();
+ while(loop--) {
+ printf("start loop\n");
+ /* access a lot of memory to make host swap it out */
+ for (i=0; i < MEM; i+=4096)
+ buf[i] = 1;
+ printf("end loop\n");
+ }
+ irq_disable();
+
+ printf("%s\n", fail ? "FAIL" : "PASS");
+ return fail;
+}
--
Gleb.
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
