Re: [Xen-devel] [PATCH 03/10] tools/insn-fuzz: Don't use memcpy() for zero-length reads

[Andrew Cooper <andrew.cooper3@xxxxxxxxxx>]

On 27/03/17 12:32, Jan Beulich wrote:
>>>> On 27.03.17 at 13:05, <andrew.cooper3@xxxxxxxxxx> wrote:
>> On 27/03/17 12:02, George Dunlap wrote:
>>> On 27/03/17 10:56, Andrew Cooper wrote:
>>>> For control-flow changes, the emulator needs to perform a zero-length
>>>> instruction fetch at the target offset.  It also passes NULL for the
>>>> destination buffer, as there is no instruction stream to collect.
>>>>
>>>> This trips up UBSAN, even with a size of 0.  Exclude zero-length reads from
>>>> using memcpy(), rather than excluding NULL destination pointers, to still
>>>> catch unintentional uses of NULL.
>>> So memcpy() will actually try to write to dst even if bytes == 0?
>>>
>>> That seems a bit strange, but OK:
>>>
>>> Acked-by: George Dunlap <george.dunlap@xxxxxxxxxx>
>> This is the undefined behaviour sanitiser, which actually objects to
>> passing NULL to a function annotated with
>> __attribute__((notnull(...))).  The check is performed before making the
>> call, and doesn't account for nothing happening if size is 0.
> But then why is the function annotated "nonnull"? Iirc there's
> nothing in the standard disallowing NULL to be passed here so
> long as length is also zero, or even only calling this undefined
> behavior.

From n1570,

7.24.1 "String function conventions"

"Where an argument declared as size_t n specifies the length of the
array for a function, n can have the value zero on a call to that
function. Unless explicitly stated otherwise in the description of a
particular function in this subclause, pointer arguments on such a call
shall still have valid values, as described in 7.1.4. On such a call, a
function that locates a character finds no occurrence, a function that
compares two character sequences returns zero, and a function that
copies characters copies zero characters."

And from 7.1.4: "Use of library functions"

"If an argument to a function has an invalid value (such as a value
outside the domain of the function, or a pointer outside the address
space of the program, or a null pointer, or a pointer to non-modifiable
storage when the corresponding parameter is not const-qualified) or a
type (after promotion) not expected by a function with variable number
of arguments, the behavior is undefined."


Therefore, by my reading, the nonnull attribute is correct.

~Andrew

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
https://lists.xen.org/xen-devel