Fix check of stack size in debug

[Jentsch]

This behavior should be more intuitive. Otherwise Threads need to have KERNEL_CONF_STACKSIZE_PRINTF + 1 to run properly.

[thomaseichinger]

Seems nobody used DEVELHELP recently. The comment is somewhat confusing too, will open a PR for this.

ACK

[Jentsch]

We used it for our network scanner.

[Jentsch]

Are you going to accept or deny this PR?

[mehlis]

I think the size KERNEL_CONF_STACKSIZE_PRINTF is the memory printf will use, but for a stack you need to have a stacksize which can handle your code and printf.

I think using KERNEL_CONF_STACKSIZE_PRINTF for debugging isn't right, use KERNEL_CONF_STACKSIZE_MAIN in your code in case of debugging.

I will clarify this on Monday!

[Jentsch]

Ok, thanks

[OlegHahm]

@Jentsch, apparently the name of stack size constants is still confusing. I tried to document the idea in the Wiki but it seems that there is still room for improvement. Do you have any suggestion how to improve the documentation or the naming (see #732)?

[mehlis]

@OlegHahm thanks for giving the hint to the wiki!

to cite: "KERNEL_CONF_STACKSIZE_PRINTF defines additional stack space needed if the thread needs to call printf (which requires additional memory when using newlib. KERNEL_CONF_STACKSIZE_MAIN is the stack size for the main thread and probably a good size for your application."

@OlegHahm so I conclude that this PR is invalid, right?

[Jentsch]

May we rename the constant, but the comparison operator should stay >= in my opinion.

[Kijewski]

Yep, using >= is correct.

[Jentsch]

So we could merge this commit?

[OlegHahm]

Actually, I would say neither is correct.

Correct would be

sched_active_thread->stack_size >= KERNEL_CONF_STACKSIZE_PRINTF + MINIMUM_STACK_SIZE

[Jentsch]

So we all agree with >= ?
May we contiue our nameing discussion in #732?

[Kijewski]

ACK.

[OlegHahm]

See #1244 (comment)

[Kijewski]

Throughout all examples, tests and sys modules KERNEL_CONF_STACKSIZE_PRINTF is used like it was the whole needed stack size. Maybe it was not intended this way, but this is how it is used.

[OlegHahm]

That's simply wrong:

`--> git grep -l STACKSIZE_MAIN examples/ tests | wc -l 
9
`--> git grep -l STACKSIZE_PRINTF examples/ tests | wc -l
11

I would rather fix these examples.

[Kijewski]

Works for me, but first I'll get some sleep. :)

[mehlis]

@OlegHahm 's fix is merged: #1370

this PR is invalid so far, but thanks for spotting the issue!