Subject: Re: indirect calls [Re: issues with com and non-PCI platforms....
To: None <pavel.cahyna@st.mff.cuni.cz,>
From: Garrett D'Amore <garrett_damore@tadpole.com>
List: tech-kern
Date: 03/14/2006 13:46:52
Pavel Cahyna wrote:
> On Tue, Mar 14, 2006 at 12:57:01PM -0800, Garrett D'Amore wrote:
>
>> Pavel Cahyna wrote:
>>
>>> On Wed, Mar 08, 2006 at 04:30:35PM -0800, Matt Thomas wrote:
>>>
>>>
>>>> On Mar 8, 2006, at 3:59 PM, Garrett D'Amore wrote:
>>>>
>>>>
>>>>> How does an indirect function call compare against say the cost of
>>>>> another array index lookup and an extra shift/mask operation>
>>>>>
>>>>>
>>>> for vax, about an order of magnitude.
>>>>
>>>>
>>> What can be so slow? An indirect function call, or function calls in
>>> general? And which architectures especially suffer from expensive indirect
>>> function calls?
>>>
>>>
>> My understanding, from what others have said, is that the problem is
>> that function calls *in general* are expensive on a VAX.
>>
>> It is mind boggling to me, that we have to design our software to
>> eliminate function calls where-ever possible, in order to get good
>> performance.
>>
>
> I do not see where this particular problem is - does VAXen use the com
> driver at all? From a quick look at files.vax, intro(4) and GENERIC, it
> does not seem so.
>
No, but apparently the same complaint was made about low end 386
hardware. And in any case, the objections more or less came about as
part of a larger discussion about bus_space stuff, not just com.
That conversation was on the developers list, and hence didn't include
the larger tech-kern@ audience.
-- Garrett
--
Garrett D'Amore, Principal Software Engineer
Tadpole Computer / Computing Technologies Division,
General Dynamics C4 Systems
http://www.tadpolecomputer.com/
Phone: 951 325-2134 Fax: 951 325-2191