Re: [PATCH 1/3] [ARC] Add arithmetic and logic instructions for nps

[Graham Markall <graham dot markall at embecosm dot com>]

>> +#define
>> MAKE_1BASED_INSERT_EXTRACT_FUNCS(NAME,SHIFT,UPPER,BITS)         \
>> +static unsigned                                                         \
>> +insert_nps_##NAME (unsigned insn ATTRIBUTE_UNUSED,                      \
>> +                   int value ATTRIBUTE_UNUSED,                          \
>> +                   const char **errmsg ATTRIBUTE_UNUSED)                \
>> +{                                                                       \
>> +  if (value < 0 || value > UPPER)                                       \
>> +    *errmsg = _("Value must be in the range 0 to " #UPPER);             \
>> +  if (value == UPPER)                                                   \
>> +    value = 0;                                                          \
>> +  return insn | (value << SHIFT);                                       \
>> +}                                                                       \
>> +                                                                        \
>> +static int                                                              \
>> +extract_nps_##NAME (unsigned insn ATTRIBUTE_UNUSED,                     \
>> +                    bfd_boolean * invalid ATTRIBUTE_UNUSED)             \
>> +{                                                                       \
>> +  int value = (insn >> SHIFT) & ((1 << BITS) - 1);                      \
>> +  if (value == 0)                                                       \
>> +    value = UPPER;                                                      \
>> +  return value;                                                         \
>> +}
> 
> I am not sure if you need those two functions as the default behavior for insert and extract behaves similarly.

Many thanks for looking at this so quickly.

I think that these functions have slightly different behaviour from the
default case - I have used these for operands which encode e.g. the
values 1-8 in 3 bits, where the values 1-7 are represented in the usual
way, and the value 8 is encoded as 0. If I understand it correctly, the
default behaviour will encode the values 0 to 7 literally. Does my
understanding seem correct?

This has however drawn my attention to the fact that there is an error
in the insert function's check - the condition should have been

  if (value < 1 || value > UPPER)

so I will provide a revised patch shortly.


Many thanks,
Graham.

Attachment: signature.asc
Description: OpenPGP digital signature