Re: [kvm-unit-tests] x86: pkeys: add test for PKEYS

[Paolo Bonzini <pbonzini@redhat.com>]

On 16/11/2015 08:53, Huaitong Han wrote:
> Signed-off-by: Huaitong Han <huaitong.han@intel.com>
> 
> diff --git a/config/config-x86-common.mak b/config/config-x86-common.mak
> index c2f9908..2ef98cc 100644
> --- a/config/config-x86-common.mak
> +++ b/config/config-x86-common.mak
> @@ -36,7 +36,8 @@ tests-common = $(TEST_DIR)/vmexit.flat $(TEST_DIR)/tsc.flat \
>                 $(TEST_DIR)/kvmclock_test.flat  $(TEST_DIR)/eventinj.flat \
>                 $(TEST_DIR)/s3.flat $(TEST_DIR)/pmu.flat \
>                 $(TEST_DIR)/tsc_adjust.flat $(TEST_DIR)/asyncpf.flat \
> -               $(TEST_DIR)/init.flat $(TEST_DIR)/smap.flat
> +               $(TEST_DIR)/init.flat $(TEST_DIR)/smap.flat \
> +               $(TEST_DIR)/pku.flat

This needs to be in config-x86_64.mak, because the 32-bit version will
have EFER.LMA=0.  Otherwise looks good (and I've tested it with a
quick-and-dirty implementation of pkeys in QEMU).

Paolo

>  ifdef API
>  tests-common += api/api-sample
> @@ -104,6 +105,8 @@ $(TEST_DIR)/pcid.elf: $(cstart.o) $(TEST_DIR)/pcid.o
>  
>  $(TEST_DIR)/smap.elf: $(cstart.o) $(TEST_DIR)/smap.o
>  
> +$(TEST_DIR)/pku.elf: $(cstart.o) $(TEST_DIR)/pku.o
> +
>  $(TEST_DIR)/vmx.elf: $(cstart.o) $(TEST_DIR)/vmx.o $(TEST_DIR)/vmx_tests.o
>  
>  $(TEST_DIR)/debug.elf: $(cstart.o) $(TEST_DIR)/debug.o
> diff --git a/lib/x86/processor.h b/lib/x86/processor.h
> index 7973879..f7aa5ec 100644
> --- a/lib/x86/processor.h
> +++ b/lib/x86/processor.h
> @@ -26,6 +26,7 @@
>  #define X86_CR4_PAE    0x00000020
>  #define X86_CR4_PCIDE  0x00020000
>  #define X86_CR4_SMAP   0x00200000
> +#define X86_CR4_PKE    0x00400000
>  
>  #define X86_IA32_EFER          0xc0000080
>  #define X86_EFER_LMA           (1UL << 8)
> diff --git a/x86/pku.c b/x86/pku.c
> new file mode 100644
> index 0000000..0e00b99
> --- /dev/null
> +++ b/x86/pku.c
> @@ -0,0 +1,161 @@
> +#include "libcflat.h"
> +#include "x86/desc.h"
> +#include "x86/processor.h"
> +#include "x86/vm.h"
> +#include "x86/msr.h"
> +
> +#define X86_FEATURE_PKU  3
> +#define CR0_WP_MASK      (1UL << 16)
> +#define PTE_PKEY_BIT     59
> +#define USER_BASE        (1 << 24)
> +#define USER_VAR(v)      (*((__typeof__(&(v))) (((unsigned long)&v) + USER_BASE)))
> +
> +volatile int pf_count = 0;
> +volatile unsigned save;
> +volatile unsigned test;
> +
> +void set_cr0_wp(int wp)
> +{
> +    unsigned long cr0 = read_cr0();
> +
> +    cr0 &= ~CR0_WP_MASK;
> +    if (wp)
> +        cr0 |= CR0_WP_MASK;
> +    write_cr0(cr0);
> +}
> +
> +static inline u32 read_pkru(void)
> +{
> +    unsigned int eax, edx;
> +    unsigned int ecx = 0;
> +    unsigned int pkru;
> +
> +    asm volatile(".byte 0x0f,0x01,0xee\n\t"
> +                 : "=a" (eax), "=d" (edx)
> +                 : "c" (ecx));
> +    pkru = eax;
> +    return pkru;
> +}
> +
> +static void write_pkru(u32 pkru)
> +{
> +    unsigned int eax = pkru;
> +    unsigned int ecx = 0;
> +    unsigned int edx = 0;
> +
> +    asm volatile(".byte 0x0f,0x01,0xef\n\t"
> +        : : "a" (eax), "c" (ecx), "d" (edx));
> +}
> +
> +void do_pf_tss(unsigned long error_code)
> +{
> +    pf_count++;
> +    save = test;
> +    write_pkru(0);
> +}
> +
> +extern void pf_tss(void);
> +
> +asm ("pf_tss: \n\t"
> +#ifdef __x86_64__
> +    // no task on x86_64, save/restore caller-save regs
> +    "push %rax; push %rcx; push %rdx; push %rsi; push %rdi\n"
> +    "push %r8; push %r9; push %r10; push %r11\n"
> +#endif
> +    "call do_pf_tss \n\t"
> +#ifdef __x86_64__
> +    "pop %r11; pop %r10; pop %r9; pop %r8\n"
> +    "pop %rdi; pop %rsi; pop %rdx; pop %rcx; pop %rax\n"
> +#endif
> +    "add $"S", %"R "sp\n\t" // discard error code
> +    "iret"W" \n\t"
> +    "jmp pf_tss\n\t"
> +    );
> +
> +static void init_test()
> +{
> +    pf_count = 0;
> +
> +    invlpg(&test);
> +    invlpg(&USER_VAR(test));
> +    write_pkru(0);
> +    set_cr0_wp(0);
> +}
> +
> +int main(int ac, char **av)
> +{
> +    unsigned long i;
> +    unsigned int pkey = 0x2;
> +    unsigned int pkru_ad = 0x10;
> +    unsigned int pkru_wd = 0x20;
> +
> +    if (!(cpuid_indexed(7, 0).c & (1 << X86_FEATURE_PKU))) {
> +        printf("PKU not enabled, exiting\n");
> +        exit(1);
> +    }
> +
> +    setup_vm();
> +    setup_alt_stack();
> +    set_intr_alt_stack(14, pf_tss);
> +    wrmsr(MSR_EFER, rdmsr(MSR_EFER) | EFER_LMA);
> +
> +    for (i = 0; i < USER_BASE; i += PAGE_SIZE) {
> +        *get_pte(phys_to_virt(read_cr3()), phys_to_virt(i)) &= ~PTE_USER;
> +        *get_pte(phys_to_virt(read_cr3()), phys_to_virt(i)) |= ((unsigned long)pkey << PTE_PKEY_BIT);
> +        invlpg((void *)i);
> +    }
> +
> +    for (i = USER_BASE; i < 2 * USER_BASE; i += PAGE_SIZE) {
> +        *get_pte(phys_to_virt(read_cr3()), phys_to_virt(i)) &= ~USER_BASE;
> +        *get_pte(phys_to_virt(read_cr3()), phys_to_virt(i)) |= ((unsigned long)pkey << PTE_PKEY_BIT);
> +        invlpg((void *)i);
> +    }
> +
> +    write_cr4(read_cr4() | X86_CR4_PKE);
> +    write_cr3(read_cr3());
> +
> +    init_test();
> +    set_cr0_wp(1);
> +    write_pkru(pkru_ad);
> +    test = 21;
> +    report("write to supervisor page when pkru is ad and wp == 1", pf_count == 0 && test == 21);
> +
> +    init_test();
> +    set_cr0_wp(0);
> +    write_pkru(pkru_ad);
> +    test = 22;
> +    report("write to supervisor page when pkru is ad and wp == 0", pf_count == 0 && test == 22);
> +
> +    init_test();
> +    set_cr0_wp(1);
> +    write_pkru(pkru_wd);
> +    test = 23;
> +    report("write to supervisor page when pkru is wd and wp == 1", pf_count == 0 && test == 23);
> +
> +    init_test();
> +    set_cr0_wp(0);
> +    write_pkru(pkru_wd);
> +    test = 24;
> +    report("write to supervisor page when pkru is wd and wp == 0", pf_count == 0 && test == 24);
> +
> +    init_test();
> +    write_pkru(pkru_wd);
> +    set_cr0_wp(0);
> +    USER_VAR(test) = 25;
> +    report("write to user page when pkru is wd and wp == 0", pf_count == 0 && test == 25);
> +
> +    init_test();
> +    write_pkru(pkru_wd);
> +    set_cr0_wp(1);
> +    USER_VAR(test) = 26;
> +    report("write to user page when pkru is wd and wp == 1", pf_count == 1 && test == 26 && save == 25);
> +
> +    init_test();
> +    write_pkru(pkru_ad);
> +    (void)USER_VAR(test);
> +    report("read from user page when pkru is ad", pf_count == 1 && save == 26);
> +
> +    // TODO: implicit kernel access from ring 3 (e.g. int)
> +
> +    return report_summary();
> +}
> diff --git a/x86/unittests.cfg b/x86/unittests.cfg
> index 337cc19..14e36a4 100644
> --- a/x86/unittests.cfg
> +++ b/x86/unittests.cfg
> @@ -72,6 +72,11 @@ groups = vmexit
>  file = access.flat
>  arch = x86_64
>  
> +[pku]
> +file = pku.flat
> +arch = x86_64
> +extra_params = -cpu host
> +
>  #[asyncpf]
>  #file = asyncpf.flat
>  
>