Re: [PATCH v5 2/2] csu: Implement and use _dl_early_allocate during static startup

[Adhemerval Zanella <adhemerval.zanella@linaro.org>]

On 12/05/2022 15:37, Florian Weimer wrote:
> * Adhemerval Zanella:
> 
>>> +void *
>>> +_dl_early_allocate (size_t size)
>>> +{
>>> +  void *result;
>>> +
>>> +  if (__curbrk != NULL)
>>> +    /* If the break has been initialized, brk must have run before,
>>> +       so just call it once more.  */
>>> +    {
>>> +      result = __sbrk (size);
>>> +      if (result == (void *) -1)
>>> +        result = NULL;
>>> +    }
>>> +  else
>>> +    {
>>> +      /* If brk has not been invoked, there is no need to update
>>> +         __curbrk.  The first call to brk will take care of that.  */
>>> +      void *previous = __brk_call (0);
>>> +      result = __brk_call (previous + size);
>>> +      if (result == previous)
>>> +        result = NULL;
>>> +      else
>>> +        result = previous;
>>> +    }
>>
>> Why do you need to avoid update __curbrk here? Otherwise it seems that a
>> __sbrk() should be suffice here.
> 
> A subsequent call to _dl_early_allocate would then take the first (sbrk)
> branch, which may or may not be correct, depending on whether the TCB
> has been initialized at that point or not.

OK, LGTM then.  I think I might try later to consolidate the mmap
syscall, although not sure if it would be worth.

Reviewed-by: Adhemerval Zanella  <adhemerval.zanella@linaro.org>