Re: [Qemu-devel] [PATCH 6/6] softmmu-semi: fix lock_user* functions not to deref NULL upon OOM

[Jim Meyering]

Peter Maydell wrote:

> On 16 May 2012 14:08, Jim Meyering <address@hidden> wrote:
>> From: Jim Meyering <address@hidden>
>>
>> Use g_malloc/g_free in place of malloc/free.
>>
>> Signed-off-by: Jim Meyering <address@hidden>
>> ---
>>  softmmu-semi.h | 6 +++---
>>  1 file changed, 3 insertions(+), 3 deletions(-)
>>
>> diff --git a/softmmu-semi.h b/softmmu-semi.h
>> index 648cb95..996e0f7 100644
>> --- a/softmmu-semi.h
>> +++ b/softmmu-semi.h
>> @@ -39,7 +39,7 @@ static void *softmmu_lock_user(CPUArchState *env,
>> uint32_t addr, uint32_t len,
>>  {
>>     uint8_t *p;
>>     /* TODO: Make this something that isn't fixed size.  */
>> -    p = malloc(len);
>> +    p = g_malloc(len);
>>     if (copy)
>>         cpu_memory_rw_debug(env, addr, p, len, 0);
>>     return p;
>
> Nak. This function is called with a length passed from the guest, so
> killing qemu if the length is too large is a bad idea. The callers
> should handle it returning NULL on failure. (Most of them do already,
> if any do not that's a bug.) The bug in this function is passing
> NULL to cpu_memory_rw_debug().

That makes sense.
I've adjusted as you suggest and posted a V2.