On 03/24/09 20:52, Laszlo (Laca) Peter wrote:
> Brian,
>
> This is in a Solaris-specific file (sysdeps/solaris/netload.c)
> so there is no need to #ifdef __sun.
> I guess the question is if it works on s10.
>
s10? Why? The patch is for vermillion.
-Ghee (a bit puzzled)
> Could someone verify?
>
> Laca
>
> On Tue, 2009-03-24 at 15:46 -0500, Brian Cameron wrote:
>
>> Andras:
>>
>>
>>> it comes from kstat's link "module"
>>>
>>> try:
>>> kstat -m link
>>>
>>> it reports only the ethernet interface(s)' stats
>>>
>> If the need to change name to "link" is Solaris specific, shouldn't
>> this code change be "#ifdef __sun" or something so the change could
>> go upstream?
>>
>> Brian
>>
>>
>>
>>> On Tue, Mar 24, 2009 at 9:39 PM, Brian Cameron<Brian.Cameron at sun.com>
>>> wrote:
>>>
>>>> Andras:
>>>>
>>>>
>>>>>> +diff -Naur libgtop-2.26.0.old/sysdeps/solaris/netload.c
>>>>>> libgtop-2.26.0/sysdeps/solaris/netload.c
>>>>>> +--- libgtop-2.26.0.old/sysdeps/solaris/netload.c 2009-03-23
>>>>>> 20:02:34.031743396 +0200
>>>>>> ++++ libgtop-2.26.0/sysdeps/solaris/netload.c 2009-03-23
>>>>>> 20:00:36.697284713 +0200
>>>>>> +@@ -182,7 +182,7 @@
>>>>>> + /*
>>>>>> + * chop off the trailing interface
>>>>>> + */
>>>>>> +- module = strdup( name );
>>>>>> ++ module = strdup( "link" );
>>>>>> + ptr = module + strlen( module ) - 1;
>>>>>> + while( (ptr> module)&& isdigit( (int) *ptr ) ) {
>>>>>> + *ptr = '\0';
>>>>>> +
>>>>>>
>>>>>> Why is this one required?
>>>>>>
>>>>> take a look at the code:
>>>>> if( (ksp = kstat_lookup( kctl,
>>>>> module, 0, name )) == NULL ) {
>>>>>
>>>>> if you have let's say hme0:
>>>>> module = hme
>>>>> name = hme0
>>>>> which is correct, but let's say if you rename it...
>>>>> module = rename
>>>>> name = rename0
>>>>> which of course will fail...
>>>>> of course you can ask why it doesnt use the "output" from the
>>>>> netlist... dont ask me
>>>>>
>>>> I don't understand how this relates to changing name to "link" in
>>>> the netload.c file. Could you explain?
>>>>
>>>> Brian
>>>>
>>>>
>>>
>>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://mail.opensolaris.org/pipermail/jds-review/attachments/20090325/e7a57c66/attachment.html>
