C::B adn Mingw: compiler options

[nenin]

I found on latest nighties kind of distortion of the compiler options tree (see attach.)
I wander is it issue of the nighties or my global config is somehow corrupted?

[oBFusCATed]

Can you be more exact? What is the problem?

[gd_on]

I can't answer for the original author, but you can see that options -O1, -O2 ... are displayed two times : one without any comment, one with a comment. This is not new. I have seen that for a long time (and live with that without problems, but ...)

[oBFusCATed]

These are passed to the linker I think, when doing LTO.

Generally someone should add a capability to show comboboxes here and probably specify a way to specify options which are compiler and linker or even if they depend on other options.

[nenin]

Can you be more exact? What is the problem?

A lot of "march" options  disappeared,  all  after "nocona" for Intel, for example.

[nenin]

These are passed to the linker I think, when doing LTO.

Generally someone should add a capability to show comboboxes here and probably specify a way to specify options which are compiler and linker or even if they depend on other options.

Concerning compler-linker options I found a strange glitch, but cannot describe it clearly now. Problem is that -Wl,warp=symbol does not emitted properly by C::B. However looks like it not works properly under Win by any way.

[oBFusCATed]

Problem is that -Wl,warp=symbol does not emitted properly by C::B.

What does this mean? Where have you added it? Example project?

[nenin]

Looks like "--wrap=symbol" not working under windows/mingw gcc  by any way. I tried few examples from command line, and all were erratic.

[oBFusCATed]

Is this related to C::B in any way?

[nenin]

I cannot check it under Linux.
Under Win implementation of the option in ld does not work. It looked like that during compilation from CB gcc considered  this option as erratic or ignored it (depending of some altrerations), and from command line this option was accepted, but compilation ended with errors.
I suggest to drop this topic.

[omlk]

https://stackoverflow.com/questions/13961774/gnu-gcc-ld-wrapping-a-call-to-symbol-with-caller-and-callee-defined-in-the-sam