myhdl-list

[myhdl-list] Error in Intbv class?

[Forumulator B. <for...@gm...>]

Hello,

I may have found a potential bug in intbv:

>>> a=intbv(5)[4:0]
>>> a[6:0]=20
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/usr/local/lib/python3.4/dist-packages/myhdl/_intbv.py", line 180,
in __setitem__
    self._handleBounds()
  File "/usr/local/lib/python3.4/dist-packages/myhdl/_intbv.py", line 77,
in _handleBounds
    (self._val, self._max))
ValueError: intbv value 20 >= maximum 16
>>> a
intbv(20)

It gives a ValueError, but still sets the value of a to 20. Would this be
considered a bug since an exception is thrown?

Re: [myhdl-list] Error in Intbv class?

[Samuele D. <sm...@gm...>]

On Thu, Mar 3, 2016 at 1:14 PM, Forumulator Bing <for...@gm...>
wrote:

> Hello,
>
> I may have found a potential bug in intbv:
>
> >>> a=intbv(5)[4:0]
> >>> a[6:0]=20
> Traceback (most recent call last):
>   File "<stdin>", line 1, in <module>
>   File "/usr/local/lib/python3.4/dist-packages/myhdl/_intbv.py", line 180,
> in __setitem__
>     self._handleBounds()
>   File "/usr/local/lib/python3.4/dist-packages/myhdl/_intbv.py", line 77,
> in _handleBounds
>     (self._val, self._max))
> ValueError: intbv value 20 >= maximum 16
> >>> a
> intbv(20)
>
> It gives a ValueError, but still sets the value of a to 20. Would this be
> considered a bug since an exception is thrown?
>
>
>
> ------------------------------------------------------------------------------
> Site24x7 APM Insight: Get Deep Visibility into Application Performance
> APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
> Monitor end-to-end web transactions and take corrective actions now
> Troubleshoot faster and improve end-user experience. Signup Now!
> http://pubads.g.doubleclick.net/gampad/clk?id=272487151&iu=/4140
> _______________________________________________
> myhdl-list mailing list
> myh...@li...
> https://lists.sourceforge.net/lists/listinfo/myhdl-list
>
>

Re: [myhdl-list] Error in Intbv class?

[Samuele D. <sm...@gm...>]

Hello, Last message was empty, my mistake.

On Thu, Mar 3, 2016 at 2:10 PM, Samuele Disegna <sm...@gm...> wrote:

>
>
> On Thu, Mar 3, 2016 at 1:14 PM, Forumulator Bing <for...@gm...>
> wrote:
>
>> Hello,
>>
>> I may have found a potential bug in intbv:
>>
>> >>> a=intbv(5)[4:0]
>> >>> a[6:0]=20
>> Traceback (most recent call last):
>>   File "<stdin>", line 1, in <module>
>>   File "/usr/local/lib/python3.4/dist-packages/myhdl/_intbv.py", line
>> 180, in __setitem__
>>     self._handleBounds()
>>   File "/usr/local/lib/python3.4/dist-packages/myhdl/_intbv.py", line 77,
>> in _handleBounds
>>     (self._val, self._max))
>> ValueError: intbv value 20 >= maximum 16
>> >>> a
>> intbv(20)
>>
>> It gives a ValueError, but still sets the value of a to 20. Would this be
>> considered a bug since an exception is thrown?
>>
>>
> From what I see on myhdl code min/max bounds handling is done later than
assignment and it seems independent.
Therefore I would say it is not a bug. We could document this behaviour.
Is it possible to subclass the exception to a documented ValueBoundError
that define the behaviour?

You can find the related code at
https://github.com/jandecaluwe/myhdl/blob/master/myhdl/_intbv.py
def __setitem__(self, key, val):

Regards

Re: [myhdl-list] Error in Intbv class?

[Samuele D. <sm...@gm...>]

Hi again, while analyzing the _intbv code I noticed a related problem:

An Exception is Not risen when accessing a out of bound bit of a
constrained intbv.
This seems a real problem to me.

I tested it by using the example code at
http://docs.myhdl.org/en/stable/manual/hwtypes.html#bit-indexing and
modifing it for using a constrained intbv:

from myhdl import Signal, delay, Simulation, always_comb, instance, intbv,
bin

def bin2gray(B, G, width):
    """ Gray encoder.

    B -- input intbv signal, binary encoded
    G -- output intbv signal, gray encoded
    width -- bit width
    """

    @always_comb
    def logic():
        for i in range(width):
            G.next[i] = B[i+1] ^ B[i]

    return logic

def testBench(width):

    B = Signal(intbv(0)[3:])
    G = Signal(intbv(0)[3:])

    dut = bin2gray(B, G, width)

    @instance
    def stimulus():
        for i in range(2**width):
            B.next = intbv(i)[3:]
            yield delay(10)
            print("B: " + bin(B, width) + "| G: " + bin(G, width))

    return dut, stimulus

sim = Simulation(testBench(width=3))
sim.run()


On Thu, Mar 3, 2016 at 2:24 PM, Samuele Disegna <sm...@gm...> wrote:

> Hello, Last message was empty, my mistake.
>
> On Thu, Mar 3, 2016 at 2:10 PM, Samuele Disegna <sm...@gm...> wrote:
>
>>
>>
>> On Thu, Mar 3, 2016 at 1:14 PM, Forumulator Bing <for...@gm...>
>> wrote:
>>
>>> Hello,
>>>
>>> I may have found a potential bug in intbv:
>>>
>>> >>> a=intbv(5)[4:0]
>>> >>> a[6:0]=20
>>> Traceback (most recent call last):
>>>   File "<stdin>", line 1, in <module>
>>>   File "/usr/local/lib/python3.4/dist-packages/myhdl/_intbv.py", line
>>> 180, in __setitem__
>>>     self._handleBounds()
>>>   File "/usr/local/lib/python3.4/dist-packages/myhdl/_intbv.py", line
>>> 77, in _handleBounds
>>>     (self._val, self._max))
>>> ValueError: intbv value 20 >= maximum 16
>>> >>> a
>>> intbv(20)
>>>
>>> It gives a ValueError, but still sets the value of a to 20. Would this
>>> be considered a bug since an exception is thrown?
>>>
>>>
>> From what I see on myhdl code min/max bounds handling is done later than
> assignment and it seems independent.
> Therefore I would say it is not a bug. We could document this behaviour.
> Is it possible to subclass the exception to a documented ValueBoundError
> that define the behaviour?
>
> You can find the related code at
> https://github.com/jandecaluwe/myhdl/blob/master/myhdl/_intbv.py
> def __setitem__(self, key, val):
>
> Regards
>

Re: [myhdl-list] Error in Intbv class?

[Forumulator B. <for...@gm...>]

True, because the value wrapped by the class is still a standard
unconstrained int. Thus, accessing something like:

>>a=intbv(5)[4:]
>>a[6]
false

Where as it should raise an out of bounds exception.Can someone comment on
this so that I could possibly try to fix this.

While we're on the topic, i also think that the __setitem__ method should
be seperately implemented for modbv too, so that wrap around behaviour can
be extended to slices of an object. For example.

>>a=modbv(5)[4:]
>>a[4:1]=20

should give a value of 12 for a instead of giving an error. Are these
issues trivial?

On Thu, Mar 3, 2016 at 8:35 PM, Samuele Disegna <sm...@gm...> wrote:

> Hi again, while analyzing the _intbv code I noticed a related problem:
>
> An Exception is Not risen when accessing a out of bound bit of a
> constrained intbv.
> This seems a real problem to me.
>
> I tested it by using the example code at
> http://docs.myhdl.org/en/stable/manual/hwtypes.html#bit-indexing and
> modifing it for using a constrained intbv:
>
> from myhdl import Signal, delay, Simulation, always_comb, instance, intbv,
> bin
>
> def bin2gray(B, G, width):
>     """ Gray encoder.
>
>     B -- input intbv signal, binary encoded
>     G -- output intbv signal, gray encoded
>     width -- bit width
>     """
>
>     @always_comb
>     def logic():
>         for i in range(width):
>             G.next[i] = B[i+1] ^ B[i]
>
>     return logic
>
> def testBench(width):
>
>     B = Signal(intbv(0)[3:])
>     G = Signal(intbv(0)[3:])
>
>     dut = bin2gray(B, G, width)
>
>     @instance
>     def stimulus():
>         for i in range(2**width):
>             B.next = intbv(i)[3:]
>             yield delay(10)
>             print("B: " + bin(B, width) + "| G: " + bin(G, width))
>
>     return dut, stimulus
>
> sim = Simulation(testBench(width=3))
> sim.run()
>
>
> On Thu, Mar 3, 2016 at 2:24 PM, Samuele Disegna <sm...@gm...> wrote:
>
>> Hello, Last message was empty, my mistake.
>>
>> On Thu, Mar 3, 2016 at 2:10 PM, Samuele Disegna <sm...@gm...> wrote:
>>
>>>
>>>
>>> On Thu, Mar 3, 2016 at 1:14 PM, Forumulator Bing <for...@gm...>
>>> wrote:
>>>
>>>> Hello,
>>>>
>>>> I may have found a potential bug in intbv:
>>>>
>>>> >>> a=intbv(5)[4:0]
>>>> >>> a[6:0]=20
>>>> Traceback (most recent call last):
>>>>   File "<stdin>", line 1, in <module>
>>>>   File "/usr/local/lib/python3.4/dist-packages/myhdl/_intbv.py", line
>>>> 180, in __setitem__
>>>>     self._handleBounds()
>>>>   File "/usr/local/lib/python3.4/dist-packages/myhdl/_intbv.py", line
>>>> 77, in _handleBounds
>>>>     (self._val, self._max))
>>>> ValueError: intbv value 20 >= maximum 16
>>>> >>> a
>>>> intbv(20)
>>>>
>>>> It gives a ValueError, but still sets the value of a to 20. Would this
>>>> be considered a bug since an exception is thrown?
>>>>
>>>>
>>> From what I see on myhdl code min/max bounds handling is done later than
>> assignment and it seems independent.
>> Therefore I would say it is not a bug. We could document this behaviour.
>> Is it possible to subclass the exception to a documented ValueBoundError
>> that define the behaviour?
>>
>> You can find the related code at
>> https://github.com/jandecaluwe/myhdl/blob/master/myhdl/_intbv.py
>> def __setitem__(self, key, val):
>>
>> Regards
>>
>
>
>
> ------------------------------------------------------------------------------
> Site24x7 APM Insight: Get Deep Visibility into Application Performance
> APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
> Monitor end-to-end web transactions and take corrective actions now
> Troubleshoot faster and improve end-user experience. Signup Now!
> http://pubads.g.doubleclick.net/gampad/clk?id=272487151&iu=/4140
> _______________________________________________
> myhdl-list mailing list
> myh...@li...
> https://lists.sourceforge.net/lists/listinfo/myhdl-list
>
>

Re: [myhdl-list] Error in Intbv class?

[Christopher F. <chr...@gm...>]

On 3/3/2016 1:19 PM, Forumulator Bing wrote:
> True, because the value wrapped by the class is still a standard
> unconstrained int. Thus, accessing something like:
>
>>> a=intbv(5)[4:] a[6]
> false
>
> Where as it should raise an out of bounds exception.Can someone
> comment on this so that I could possibly try to fix this.

I don't know if this was an original design decision
to allow the out of index range access to occur?  It does
return the correct value for the position if you imagine
it was sign-extended.

     >>> a = intbv(0)[8:]
     >>> b = intbv(-1, min=-1000, max=1000)
     >>> print("{:1d} and {:1d}".format(a[1000000], b[1000000]))
     0 and 1

But with that said, it seems like we are missing an
opportunity to flag errors, especially since one off
errors are so common.

You might need to wait for Jan D. to comment or someone
else that might know the history (or someone with strong
feelings one way or the other :)

Regards,
Chris