Warning in '__loc_network_tree_count()'

Warning in '__loc_network_tree_count()'

[Gisle Vanem]

[-- Attachment #1: Type: text/plain, Size: 607 bytes --]

Hello list.

Compiling with 'clang-cl' on Windows-10, I saw this
important warning:
network.c(796,10): warning: variable 'counter' set but not used [-Wunused-but-set-variable]
    size_t* counter = (size_t*)data;
            ^

So shouldn't the function be:

static int __loc_network_tree_count(struct loc_network* network, void* data) {
	size_t* counter = (size_t*)data;

	// Increase the counter for each network
	(*counter)++;
	return 0;
}

But I fail to see this ret-val is used for anything.

BTW, disasm now is correct:
  inc qword ptr [rdx]
  xor eax,eax
  ret


-- 
--gv

Re: Warning in '__loc_network_tree_count()'

[Michael Tremer]

[-- Attachment #1: Type: text/plain, Size: 1264 bytes --]

Hello Gisle,

Thanks for your email.

> On 4 Mar 2023, at 10:44, Gisle Vanem <gvanem(a)online.no> wrote:
> 
> Hello list.
> 
> Compiling with 'clang-cl' on Windows-10, I saw this
> important warning:
> network.c(796,10): warning: variable 'counter' set but not used [-Wunused-but-set-variable]
>   size_t* counter = (size_t*)data;
>           ^
> 
> So shouldn't the function be:
> 
> static int __loc_network_tree_count(struct loc_network* network, void* data) {
> size_t* counter = (size_t*)data;
> 
> // Increase the counter for each network
> (*counter)++;
> return 0;

Yes, you are incorrect. I am incrementing the pointer when I wanted to increment the value. Probably this is an artefact from rewriting the code a few times.

But indeed this has not been a problem since the function is not being called at all in the code, and it is not being exported.

So assuming that it is not very useful, I will drop loc_network_tree_count_networks and the callback function.

Thank you for helping to improve libloc. How does it run on Windows?

Best,
-Michael

> }
> 
> But I fail to see this ret-val is used for anything.
> 
> BTW, disasm now is correct:
> inc qword ptr [rdx]
> xor eax,eax
> ret
> 
> 
> -- 
> --gv

Re: Warning in '__loc_network_tree_count()'

[Gisle Vanem]

[-- Attachment #1: Type: text/plain, Size: 765 bytes --]

Michael Tremer wrote:

>> So shouldn't the function be:
>>
>> static int __loc_network_tree_count(struct loc_network* network, void* data) {
>> size_t* counter = (size_t*)data;
>>
>> // Increase the counter for each network
>> (*counter)++;
>> return 0;
> 
> Yes, you are incorrect. I am incrementing the pointer when I wanted to increment the value. 

You mean "you are correct"?

> Thank you for helping to improve libloc. How does it run on Windows?
I doesn't run at all w/o much patching by myself. I've added
support for it and 'location.db' files in my project Wsock-trace:
   https://github.com/gvanem/wsock-trace

The patched up 'libloc' version:
   https://github.com/gvanem/wsock-trace/tree/master/src/Geo-IP/IPFire/src

-- 
--gv

Re: Warning in '__loc_network_tree_count()'

[Michael Tremer]

[-- Attachment #1: Type: text/plain, Size: 963 bytes --]

Hello,

> On 4 Mar 2023, at 11:00, Gisle Vanem <gvanem(a)online.no> wrote:
> 
> Michael Tremer wrote:
> 
>>> So shouldn't the function be:
>>> 
>>> static int __loc_network_tree_count(struct loc_network* network, void* data) {
>>> size_t* counter = (size_t*)data;
>>> 
>>> // Increase the counter for each network
>>> (*counter)++;
>>> return 0;
>> Yes, you are incorrect. I am incrementing the pointer when I wanted to increment the value. 
> 
> You mean "you are correct"?
> 
>> Thank you for helping to improve libloc. How does it run on Windows?
> I doesn't run at all w/o much patching by myself. I've added
> support for it and 'location.db' files in my project Wsock-trace:
>  https://github.com/gvanem/wsock-trace
> 
> The patched up 'libloc' version:
>  https://github.com/gvanem/wsock-trace/tree/master/src/Geo-IP/IPFire/src

Do you intend to upstream those changes at some point (if that is possible)?

> 
> -- 
> --gv