This is the mail archive of the crossgcc@sourceware.org mailing list for the crossgcc project.

See the CrossGCC FAQ for lots more information.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [Crosstool-NG] Blackfin support, RFC


Thomas, All,

On Wednesday 07 April 2010 10:33:49 Thomas Petazzoni wrote:
> I've recently started adding Blackfin support to Crosstool-NG.
[--SNIP--]

I am having a look at your patches. See my comments below...

> The attached patches implement the needed modifications:

I've re-ordered the patch series as thus:

>  * 1839, Fix the uClibc clean-up-O_CLOEXEC-handling patch
>  * 1840, Re-enable the FD_PIC ELF configuration option

Applied as 6cebc65d4a55 and 5d872df743fe, respectively.

Those two are needed for bfin to work, so they should come first.
1839 is related to bfin, but is a typo in an existing patch, so better
fix it sooner than later.

>  * 1838, Add basic support for the Blackfin architecture
>  * 1841, Add uClibc support for Blackfin

Please fold those two patches together. They are a single entity named
"Add balckfin support". Two new files, so no new regression expected.
And the way the uClibc script is changed is far from dubious, either.

Still:
- 1838: blackfin.in says that user can set ARCH, CPU, TUNE and FPU, but you
  force the gcc ./configure options (--with-arch=.... et al.) to empty in
  blackfin.sh. Why is that?

>  * 1842, Add Blackfin-related uClibc patch

Ideally, this one should also be folded with the previous two ones, but I
could not find any reference to that in the uClibc tree (the current daemon
impl. is completely different now, anyway).

Seems it should have no side effect to any MMU arch, and bfin is now the
only noMMU arch that CT-NG builds uClibc for. So we can't check for regs.

Please, fold with the two previous ones as well.

>  * 1843, Add Blackfin sample

No comment, I'll trust you on the sample. :-)
Keep the commit separate.

> However, in addition to these patches, I had to make a very ugly
> modification :
> diff -r e4d7a4cd72cc scripts/build/kernel/linux.sh
> --- a/scripts/build/kernel/linux.sh	Wed Apr 07 09:26:12 2010 +0200
> +++ b/scripts/build/kernel/linux.sh	Wed Apr 07 10:30:07 2010 +0200
> @@ -6,7 +6,7 @@
>      if [ "${CT_ARCH_USE_MMU}" = "y" ]; then
>          CT_TARGET_KERNEL="linux"
>      else
> -        CT_TARGET_KERNEL="uclinux"
> +        CT_TARGET_KERNEL="linux"
>      fi
>  }

Do you want to submit the corresponding patch, please? Keep the if/else/fi,
and just comment out the uclinux, adding an explanation why you did that.
Then we can revisit the code when it's time.

And it should come before the 3 folded patches, as it's needed for bfin.

Thank you!

Regards,
Yann E. MORIN.

-- 
.-----------------.--------------------.------------------.--------------------.
|  Yann E. MORIN  | Real-Time Embedded | /"\ ASCII RIBBON | Erics' conspiracy: |
| +33 662 376 056 | Software  Designer | \ / CAMPAIGN     |  ___               |
| +33 223 225 172 `------------.-------:  X  AGAINST      |  \e/  There is no  |
| http://ymorin.is-a-geek.org/ | _/*\_ | / \ HTML MAIL    |   v   conspiracy.  |
'------------------------------^-------^------------------^--------------------'



--
For unsubscribe information see http://sourceware.org/lists.html#faq


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]