The libfdt does not guarantee than fdt_getprop() returns a pointer
aligned to the property size.

Assuming the base of the fdt is aligned, a 32-bit property returns
a 32-bit aligned pointer. This is however not guaranteed for 64-bit
properties, where 64-bit loads might trigger unaligned access.

Fix this issue by using the ldst (host) API, which does a local copy
on the stack, thus guaranteeing a safe aligned access.

This fixes the following ASan warning:

  $ mips64el-softmmu/qemu-system-mips64el -M boston -kernel vmlinux.gz.itb -nographic
  hw/core/loader-fit.c:108:17: runtime error: load of misaligned address 0x7f95cd7e4264 for type 'fdt64_t', which requires 8 byte alignment
  0x7f95cd7e4264: note: pointer points here
    00 00 00 3e ff ff ff ff  80 7d 2a c0 00 00 00 01  68 61 73 68 40 30 00 00  00 00 00 03 00 00 00 14
                ^

Reported-by: AddressSanitizer
Suggested-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
---
 hw/core/loader-fit.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/hw/core/loader-fit.c b/hw/core/loader-fit.c
index 0c4a7207f4..628f854636 100644
--- a/hw/core/loader-fit.c
+++ b/hw/core/loader-fit.c
@@ -102,10 +102,10 @@ static int fit_image_addr(const void *itb, int img, const char *name,
 
     switch (len) {
     case 4:
-        *addr = fdt32_to_cpu(*(fdt32_t *)prop);
+        *addr = fdt32_to_cpu(ldl_he_p(prop));
         return 0;
     case 8:
-        *addr = fdt64_to_cpu(*(fdt64_t *)prop);
+        *addr = fdt64_to_cpu(ldq_he_p(prop));
         return 0;
     default:
         error_printf("invalid %s address length %d\n", name, len);
-- 
2.17.0

On 04/23/2018 06:25 AM, Philippe Mathieu-Daudé wrote:
> Assuming the base of the fdt is aligned, a 32-bit property returns
> a 32-bit aligned pointer...
...
>      case 4:
> -        *addr = fdt32_to_cpu(*(fdt32_t *)prop);
> +        *addr = fdt32_to_cpu(ldl_he_p(prop));
>          return 0;

So why are you changing the 32-bit case?


r~

On 04/23/2018 02:34 PM, Richard Henderson wrote:
> On 04/23/2018 06:25 AM, Philippe Mathieu-Daudé wrote:
>> Assuming the base of the fdt is aligned, a 32-bit property returns
>> a 32-bit aligned pointer...
> ...
>>      case 4:
>> -        *addr = fdt32_to_cpu(*(fdt32_t *)prop);
>> +        *addr = fdt32_to_cpu(ldl_he_p(prop));
>>          return 0;
> 
> So why are you changing the 32-bit case?

First to not assume base fdt is always 32-bit aligned,
second to keep calls similar and avoid someone use *(fdt64_t *) similar
to the 32-bit property. But this is avoidable with a simple comment
about why we have to use ldq_he_p().

On 23/04/2018 18:25, Philippe Mathieu-Daudé wrote:
> The libfdt does not guarantee than fdt_getprop() returns a pointer
> aligned to the property size.
> 
> Assuming the base of the fdt is aligned, a 32-bit property returns
> a 32-bit aligned pointer. This is however not guaranteed for 64-bit
> properties, where 64-bit loads might trigger unaligned access.
> 
> Fix this issue by using the ldst (host) API, which does a local copy
> on the stack, thus guaranteeing a safe aligned access.
> 
> This fixes the following ASan warning:
> 
>   $ mips64el-softmmu/qemu-system-mips64el -M boston -kernel vmlinux.gz.itb -nographic
>   hw/core/loader-fit.c:108:17: runtime error: load of misaligned address 0x7f95cd7e4264 for type 'fdt64_t', which requires 8 byte alignment
>   0x7f95cd7e4264: note: pointer points here
>     00 00 00 3e ff ff ff ff  80 7d 2a c0 00 00 00 01  68 61 73 68 40 30 00 00  00 00 00 03 00 00 00 14
>                 ^
> 
> Reported-by: AddressSanitizer
> Suggested-by: Peter Maydell <peter.maydell@linaro.org>
> Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
> ---
>  hw/core/loader-fit.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/hw/core/loader-fit.c b/hw/core/loader-fit.c
> index 0c4a7207f4..628f854636 100644
> --- a/hw/core/loader-fit.c
> +++ b/hw/core/loader-fit.c
> @@ -102,10 +102,10 @@ static int fit_image_addr(const void *itb, int img, const char *name,
>  
>      switch (len) {
>      case 4:
> -        *addr = fdt32_to_cpu(*(fdt32_t *)prop);
> +        *addr = fdt32_to_cpu(ldl_he_p(prop));
>          return 0;
>      case 8:
> -        *addr = fdt64_to_cpu(*(fdt64_t *)prop);
> +        *addr = fdt64_to_cpu(ldq_he_p(prop));
>          return 0;
>      default:
>          error_printf("invalid %s address length %d\n", name, len);
> 

Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>