qemu-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [PATCH v2] hw/char/pl011: Avoid division-by-zero in pl011_get_baudra


From: Zheyu Ma
Subject: Re: [PATCH v2] hw/char/pl011: Avoid division-by-zero in pl011_get_baudrate()
Date: Tue, 2 Jul 2024 17:56:18 +0200

Hi Philippe,

On Tue, Jul 2, 2024 at 5:44 PM Philippe Mathieu-Daudé <philmd@linaro.org> wrote:
On 2/7/24 17:40, Philippe Mathieu-Daudé wrote:
> Hi Zheyu,
>
> On 2/7/24 17:27, Zheyu Ma wrote:
>> In pl011_get_baudrate(), when we calculate the baudrate we can
>> accidentally divide by zero. This happens because although (as the
>> specification requires) we treat UARTIBRD = 0 as invalid, we aren't
>> correctly limiting UARTIBRD and UARTFBRD values to the 16-bit and 6-bit
>> ranges the hardware allows, and so some non-zero values of UARTIBRD can
>> result in a zero divisor.
>>
>> Enforce the correct register field widths on guest writes and on inbound
>> migration to avoid the division by zero.
>>
>> ASAN log:
>> ==2973125==ERROR: AddressSanitizer: FPE on unknown address 0x55f72629b348
>> (pc 0x55f72629b348 bp 0x7fffa24d0e00 sp 0x7fffa24d0d60 T0)
>>      #0 0x55f72629b348 in pl011_get_baudrate hw/char/pl011.c:255:17
>>      #1 0x55f726298d94 in pl011_trace_baudrate_change
>> hw/char/pl011.c:260:33
>>      #2 0x55f726296fc8 in pl011_write hw/char/pl011.c:378:9
>>
>> Reproducer:
>> cat << EOF | qemu-system-aarch64 -display \
>> none -machine accel=qtest, -m 512M -machine realview-pb-a8 -qtest stdio
>> writeq 0x1000b024 0xf8000000
>> EOF
>>
>> Signed-off-by: Zheyu Ma <zheyuma97@gmail.com>
>> ---
>> Changes in v2:
>> - Enforce the correct register field widths on writes to both UARTIBRD
>>    and UARTFBRD registers.
>> - Mask UARTIBRD to 16 bits and UARTFBRD to 6 bits in the pl011_post_load
>>    function to prevent division by zero during inbound migration.
>> ---
>>   hw/char/pl011.c | 7 +++++--
>>   1 file changed, 5 insertions(+), 2 deletions(-)
>>
>> diff --git a/hw/char/pl011.c b/hw/char/pl011.c
>> index 8753b84a84..ba5f7cfbda 100644
>> --- a/hw/char/pl011.c
>> +++ b/hw/char/pl011.c
>> @@ -374,11 +374,11 @@ static void pl011_write(void *opaque, hwaddr
>> offset,
>>           s->ilpr = value;
>>           break;
>>       case 9: /* UARTIBRD */
>> -        s->ibrd = value;
>> +        s->ibrd = value & 0xffff;
>
> LGTM but could you add a definition instead of these magic values?
>
> Similarly to the ones in the top if this file:
>
>    /* Flag Register, UARTFR */
>    #define PL011_FLAG_RI   0x100
>    #define PL011_FLAG_TXFE 0x80
>    ...
>
>    /* Integer Baud Rate Divider, UARTIBRD */
>    #define IBRD_MASK 0x3f
>
>    /* Fractional Baud Rate Divider, UARTFBRD */
>    #define FBRD_MASK 0xffff
>
> Then use these definitions :)

Since I have the patch, let's share it to save you some time:

-- >8 --
diff --git a/hw/char/pl011.c b/hw/char/pl011.c
index 8753b84a84..7f9a85708e 100644
--- a/hw/char/pl011.c
+++ b/hw/char/pl011.c
@@ -87,6 +87,12 @@ DeviceState *pl011_create(hwaddr addr, qemu_irq irq,
Chardev *chr)
  #define CR_DTR      (1 << 10)
  #define CR_LBE      (1 << 7)

+/* Integer Baud Rate Divider, UARTIBRD */
+#define IBRD_MASK 0x3f
+
+/* Fractional Baud Rate Divider, UARTFBRD */
+#define FBRD_MASK 0xffff
+
  static const unsigned char pl011_id_arm[8] =
    { 0x11, 0x10, 0x14, 0x00, 0x0d, 0xf0, 0x05, 0xb1 };
  static const unsigned char pl011_id_luminary[8] =
@@ -374,11 +380,11 @@ static void pl011_write(void *opaque, hwaddr offset,
          s->ilpr = value;
          break;
      case 9: /* UARTIBRD */
-        s->ibrd = value;
+        s->ibrd = value & IBRD_MASK;
          pl011_trace_baudrate_change(s);
          break;
      case 10: /* UARTFBRD */
-        s->fbrd = value;
+        s->fbrd = value & FBRD_MASK;
          pl011_trace_baudrate_change(s);
          break;
      case 11: /* UARTLCR_H */
@@ -531,6 +537,9 @@ static int pl011_post_load(void *opaque, int version_id)
          s->read_pos = 0;
      }

+    s->ibrd &= IBRD_MASK;
+    s->fbrd &= FBRD_MASK;
+
      return 0;
  }

Thanks for your helpful patch :)

regards,
Zheyu

reply via email to

[Prev in Thread] Current Thread [Next in Thread]