@@ -72,8 +72,23 @@
#define CHOOSE_VHE_SYM(sym) sym
#define CHOOSE_NVHE_SYM(sym) kvm_nvhe_sym(sym)
-#define this_cpu_ptr_nvhe(sym) this_cpu_ptr(&kvm_nvhe_sym(sym))
-#define per_cpu_ptr_nvhe(sym, cpu) per_cpu_ptr(&kvm_nvhe_sym(sym), cpu)
+/* Array of percpu base addresses. Length of the array is nr_cpu_ids. */
+extern unsigned long *kvm_arm_hyp_percpu_base;
+
+/*
+ * Compute pointer to a symbol defined in nVHE percpu region.
+ * Returns NULL if percpu memory has not been allocated yet.
+ */
+#define this_cpu_ptr_nvhe(sym) per_cpu_ptr_nvhe(sym, smp_processor_id())
+#define per_cpu_ptr_nvhe(sym, cpu) \
+ ({ \
+ unsigned long base, off; \
+ base = kvm_arm_hyp_percpu_base \
+ ? kvm_arm_hyp_percpu_base[cpu] : 0; \
+ off = (unsigned long)&kvm_nvhe_sym(sym) - \
+ (unsigned long)&kvm_nvhe_sym(__per_cpu_start); \
+ base ? (typeof(kvm_nvhe_sym(sym))*)(base + off) : NULL; \
+ })
#ifndef __KVM_NVHE_HYPERVISOR__
/*
@@ -18,5 +18,6 @@ extern char __exittext_begin[], __exittext_end[];
extern char __irqentry_text_start[], __irqentry_text_end[];
extern char __mmuoff_data_start[], __mmuoff_data_end[];
extern char __entry_tramp_text_start[], __entry_tramp_text_end[];
+extern char __kvm_nvhe___per_cpu_start[], __kvm_nvhe___per_cpu_end[];
#endif /* __ASM_SECTIONS_H */
@@ -16,6 +16,9 @@
#include "image.h"
+#define __CONCAT3(x, y, z) x ## y ## z
+#define CONCAT3(x, y, z) __CONCAT3(x, y, z)
+
OUTPUT_ARCH(aarch64)
ENTRY(_text)
@@ -187,6 +190,13 @@ SECTIONS
PERCPU_SECTION(L1_CACHE_BYTES)
+ /* KVM nVHE per-cpu section */
+ #undef PERCPU_SECTION_NAME
+ #undef PERCPU_SYMBOL_NAME
+ #define PERCPU_SECTION_NAME(suffix) CONCAT3(.hyp, PERCPU_SECTION_BASE_NAME, suffix)
+ #define PERCPU_SYMBOL_NAME(name) __kvm_nvhe_ ## name
+ PERCPU_SECTION(L1_CACHE_BYTES)
+
.rela.dyn : ALIGN(8) {
*(.rela .rela*)
}
@@ -47,6 +47,7 @@ __asm__(".arch_extension virt");
#endif
static DEFINE_PER_CPU(unsigned long, kvm_arm_hyp_stack_page);
+unsigned long *kvm_arm_hyp_percpu_base;
/* The VMID used in the VTTBR */
static atomic64_t kvm_vmid_gen = ATOMIC64_INIT(1);
@@ -1253,6 +1254,15 @@ long kvm_arch_vm_ioctl(struct file *filp,
}
}
+#define kvm_hyp_percpu_base(cpu) ((unsigned long)per_cpu_ptr_nvhe(__per_cpu_start, cpu))
+#define kvm_hyp_percpu_array_size (nr_cpu_ids * sizeof(*kvm_arm_hyp_percpu_base))
+#define kvm_hyp_percpu_array_order (get_order(kvm_hyp_percpu_array_size))
+#define kvm_hyp_percpu_begin CHOOSE_NVHE_SYM(__per_cpu_start)
+#define kvm_hyp_percpu_size ((unsigned long)CHOOSE_NVHE_SYM(__per_cpu_end) - \
+ (unsigned long)CHOOSE_NVHE_SYM(__per_cpu_start))
+#define kvm_hyp_percpu_order (kvm_hyp_percpu_size \
+ ? get_order(kvm_hyp_percpu_size) : 0)
+
static void cpu_init_hyp_mode(void)
{
phys_addr_t pgd_ptr;
@@ -1268,8 +1278,8 @@ static void cpu_init_hyp_mode(void)
* kernel's mapping to the linear mapping, and store it in tpidr_el2
* so that we can use adr_l to access per-cpu variables in EL2.
*/
- tpidr_el2 = ((unsigned long)this_cpu_ptr(&kvm_host_data) -
- (unsigned long)kvm_ksym_ref(&kvm_host_data));
+ tpidr_el2 = (unsigned long)this_cpu_ptr_nvhe(__per_cpu_start) -
+ (unsigned long)kvm_ksym_ref(kvm_hyp_percpu_begin);
pgd_ptr = kvm_mmu_get_httbr();
hyp_stack_ptr = __this_cpu_read(kvm_arm_hyp_stack_page) + PAGE_SIZE;
@@ -1501,8 +1511,11 @@ static void teardown_hyp_mode(void)
int cpu;
free_hyp_pgds();
- for_each_possible_cpu(cpu)
+ for_each_possible_cpu(cpu) {
free_hyp_page(per_cpu(kvm_arm_hyp_stack_page, cpu));
+ free_hyp_pages(kvm_hyp_percpu_base(cpu), kvm_hyp_percpu_order);
+ }
+ free_hyp_pages((unsigned long)kvm_arm_hyp_percpu_base, kvm_hyp_percpu_array_order);
}
/**
@@ -1535,6 +1548,28 @@ static int init_hyp_mode(void)
per_cpu(kvm_arm_hyp_stack_page, cpu) = stack_page;
}
+ /*
+ * Allocate and initialize pages for Hypervisor-mode percpu regions.
+ */
+ kvm_arm_hyp_percpu_base = (unsigned long*)alloc_hyp_pages(
+ GFP_KERNEL | __GFP_ZERO, kvm_hyp_percpu_array_order);
+ if (!kvm_arm_hyp_percpu_base) {
+ err = -ENOMEM;
+ goto out_err;
+ }
+ for_each_possible_cpu(cpu) {
+ unsigned long percpu_base;
+
+ percpu_base = alloc_hyp_pages(GFP_KERNEL, kvm_hyp_percpu_order);
+ if (!percpu_base) {
+ err = -ENOMEM;
+ goto out_err;
+ }
+
+ memcpy((void*)percpu_base, kvm_hyp_percpu_begin, kvm_hyp_percpu_size);
+ kvm_arm_hyp_percpu_base[cpu] = percpu_base;
+ }
+
/*
* Map the Hyp-code called directly from the host
*/
@@ -1579,6 +1614,20 @@ static int init_hyp_mode(void)
}
}
+ /*
+ * Map Hyp percpu pages
+ */
+ for_each_possible_cpu(cpu) {
+ char *percpu_begin = (char *)kvm_arm_hyp_percpu_base[cpu];
+ char *percpu_end = percpu_begin + PAGE_ALIGN(kvm_hyp_percpu_size);
+ err = create_hyp_mappings(percpu_begin, percpu_end, PAGE_HYP);
+
+ if (err) {
+ kvm_err("Cannot map hyp percpu region\n");
+ goto out_err;
+ }
+ }
+
for_each_possible_cpu(cpu) {
kvm_host_data_t *cpu_data;
@@ -23,6 +23,8 @@ $(obj)/%.hyp.o: $(obj)/%.hyp.tmp.o FORCE
quiet_cmd_hypcopy = HYPCOPY $@
cmd_hypcopy = $(OBJCOPY) --prefix-symbols=__kvm_nvhe_ \
--rename-section=.text=.hyp.text \
+ --rename-section=.data..percpu=.hyp.data..percpu \
+ --rename-section=.data..percpu..read_mostly=.hyp.data..percpu..read_mostly \
$< $@
# Remove ftrace and Shadow Call Stack CFLAGS.
@@ -33,7 +33,7 @@ void kvm_set_pmu_events(u32 set, struct perf_event_attr *attr)
{
struct kvm_host_data *ctx = this_cpu_ptr_hyp(kvm_host_data);
- if (!kvm_pmu_switch_needed(attr))
+ if (!ctx || !kvm_pmu_switch_needed(attr))
return;
if (!attr->exclude_host)
@@ -49,6 +49,9 @@ void kvm_clr_pmu_events(u32 clr)
{
struct kvm_host_data *ctx = this_cpu_ptr_hyp(kvm_host_data);
+ if (!ctx)
+ return;
+
ctx->pmu_events.events_host &= ~clr;
ctx->pmu_events.events_guest &= ~clr;
}
Add hyp percpu section to linker script and rename the corresponding ELF sections of hyp/nvhe object files. This moves all nVHE-specific percpu variables to the new hyp percpu section. Allocate sufficient amount of memory for all percpu hyp regions at global KVM init time, and create corresponding hyp mappings. The base addresses of hyp percpu regions are kept in a dynamically allocated array in the kernel. Add NULL checks in PMU event-reset code as it may run before KVM memory is initialized. Signed-off-by: David Brazdil <dbrazdil@google.com> --- arch/arm64/include/asm/kvm_asm.h | 19 +++++++++-- arch/arm64/include/asm/sections.h | 1 + arch/arm64/kernel/vmlinux.lds.S | 10 ++++++ arch/arm64/kvm/arm.c | 55 +++++++++++++++++++++++++++++-- arch/arm64/kvm/hyp/nvhe/Makefile | 2 ++ arch/arm64/kvm/pmu.c | 5 ++- 6 files changed, 86 insertions(+), 6 deletions(-)