Dear Jean-Christophe PLAGNIOL-VILLARD,
In message 20090708211902.GH12394@game.jcrosoft.org you wrote:
With this patch and the following one we move away from all these problems and we will be able to have full control to have a functions embedded into u-boot.
Please fix the comment.
In git history, there will be no "following one" patch, and here you did not post one either. If you don't have it ready yet, then please use a description of what it will do.
the precedent patch still apply cleany so I do not resend it
You don't get it.
Your commit message uses the phrase "the following [patch]", but nobody knows what this "following one" is. I do not know it even now, while we are discussing it. Please use some form of reference that can be understood.
And "we will be able to have full control to have a functions embedded into u-boot" makes no sense. Either it is "a function" (singular), or it is "functions" (plural, without "a"). But what do you want to tell us? "have functions embedded into U-Boot"? We do this all the time by linking them in - we already have full control over it (except that it's not working as expected sometimes), and your patch does not really change anything of this.
gcc for arm is really boring about abi support and float support so on arm we do need to control it and be sure about what we use for basic function and do not use the libgcc provide by the tool chains as we can not trust them so for linux as example we implement it correctly and are able to manage against what armv we want to compile or be comptatible ditto when we will be able to support Thumb2 as needed for cortex-m3
I do understand the situation and what you are trying to do.
What I do not understand is the sentence you wrote in the commit message.
You wrote: "gcc does not provide necessarely cored functions". Sorry, but I have no idea what you are talking about here. What is a "cored function"?
correct function
Ah.
Best regards,
Wolfgang Denk