Hi everyone, I use slackware to program a lot.
So in interested in LTO. But now I need a stable machine to finish the projects to my degree. So has some one tried LTO?

Can report some values to see the difference?

Here is a information of hoe to use LTO in gcc.

It's been a month w/ no replies, perhaps some clarifications would help you get some help.
How does this impact the problem you are asking about here?

I'm sure there are LQ members who have, but I'm not 1 of them, & I don't know if any have read this. Worse, I fear that any who did, don't understand what you are asking.

This line is probably the problem -- I don't understand what you mean. I suspect that "Aveiro" means that your native language is Portuguese, & therefore clarity in English isn't easy. Why don't you rephrase. we'll be patient.

Thank you for the link. It's very considerate to save your potential helpers from duplicating the search.


BTW, which Aveiro: Aveiro, Pará, Brazil o Aveiro, Portugal?

I'm from Aveiro Portugal. Sorry for my English.

I'm finishing my master's degree in Informatics. I'm working in a agent to detect and classify objects in pictures.
So for now my laptop isn't available to run Slackware-current.

I'm asking if someone has Slackware-current, and program a little. If can run some program with LTO and without LTO. In some test that i saw in other website LTO produce code 6% faster in terms of time.

I'm interested because all my code is modular, so LTO can have some impact in performance.

Don't worry too much about your English, you're doing better than more than a few native speakers.

I don't run Slack myself, but here's hoping some Slackware-current user will see this & run (& report) the tests you need.

If you don't get a reply soon, try asking the mods to move this to the Slackware forum. (Use the "Report" button on your own post.)

Good luck.

So, I had the chance of installing Slackware64 -current, and run some tests with LTO.
This is my result.

I make a small implementation of a queue with a linked-list.
And I fill the queue and empty the queue. Run three times with LTO and without, make and average of times and compute the difference.
It's 3,44% faster, and with only two modules to optimize it's significant.

If some one want's the code to see just ask.

Yes, it is interesting.

Can you give an email so I send the code?
I tried put here as a file but only lets 3 files and don't support .cpp or .h