Re: [PATCH v2 1/1] BaseTools/tools_def: Add GCC49xASL Toolchain support

[Chris Co <Christopher.Co@microsoft.com>]

> -----Original Message-----
> From: Ard Biesheuvel <ard.biesheuvel@linaro.org>
> Sent: Tuesday, June 26, 2018 11:52 AM
> To: Chris Co <Christopher.Co@microsoft.com>
> Cc: Gao, Liming <liming.gao@intel.com>; Leif Lindholm
> <leif.lindholm@linaro.org>; edk2-devel@lists.01.org; Charles Garcia-Tobin
> <Charles.Garcia-Tobin@arm.com>
> Subject: Re: [edk2] [PATCH v2 1/1] BaseTools/tools_def: Add GCC49xASL
> Toolchain support
> 
> On 26 June 2018 at 20:42, Chris Co <Christopher.Co@microsoft.com> wrote:
> >
> >
> >> -----Original Message-----
> >> From: Ard Biesheuvel <ard.biesheuvel@linaro.org>
> >> Sent: Tuesday, June 26, 2018 1:22 AM
> >> To: Gao, Liming <liming.gao@intel.com>
> >> Cc: Chris Co <Christopher.Co@microsoft.com>; edk2-devel@lists.01.org;
> >> Leif Lindholm <leif.lindholm@linaro.org>; Charles Garcia-Tobin
> >> <Charles.Garcia- Tobin@arm.com>
> >> Subject: Re: [edk2] [PATCH v2 1/1] BaseTools/tools_def: Add GCC49xASL
> >> Toolchain support
> >>
> >> On 26 June 2018 at 06:40, Gao, Liming <liming.gao@intel.com> wrote:
> >> > Chris:
> >> >   Here is the example usage. You can add below [BuildOptions] in
> >> > your
> >> platform DSC. Then, it will use override ASL path and flag.
> >> >
> >> > [BuildOptions]
> >> >   *_*_*_ASL_PATH                     == C:\Asl\asl.exe
> >> >   *_*_*_ASL_FLAGS                    == /MsftInternal
> >> >   *_*_*_ASL_OUTFLAGS                 == /Fo=
> >> >
> >>
> >> I strongly prefer this over adding a new MS specific toolchain definition.
> >>
> >
> > Thanks Liming!  I added these overrides and they works great.  Since we
> can do platform dsc overrides, I also used it to override the objcopy_path
> and symrename_path which was the other change we needed for building
> on Windows.
> >   GCC:*_*_*_OBJCOPY_PATH = noop
> >   GCC:*_*_*_SYMRENAME_PATH = noop
> >
> > Ard, I agree this is the better method.  We can drop this patch in favor of
> the platform dsc overrides.
> >
> > Leif, I tested GCC7.3 arm-eabi and ran into build errors.  Found out
> > this was an issue encountered before:
> >
> https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fbugs.
> >
> linaro.org%2Fshow_bug.cgi%3Fid%3D2909&amp;data=02%7C01%7CChristop
> her.C
> >
> o%40microsoft.com%7C0960f252b3dd4f2e79ed08d5db95e411%7C72f988bf86
> f141a
> >
> f91ab2d7cd011db47%7C1%7C0%7C636656359185404698&amp;sdata=ZEt9sY
> %2F94us
> > UaTDkvHQrzkkIiaXTPPfWqMlPh2NK%2BgM%3D&amp;reserved=0
> > I made a couple modifications to Trim.py to fix the errors and was able to
> successfully build and test.  Will submit the fix for review and update our
> toolchain to use GCC7.3 with GCC5 profile.
> >
> 
> That is good news.
> 
> So I guess we're all set to start incorporating this platform.
> 
> Any clue whether there is also a device tree available for this platform? I
> would prefer it if we could make the firmware OS agnostic, rather than have
> Windows-only and Linux-only platforms in the edk2-platforms tree.

We did not include a device tree since Windows does not comprehend device trees, however all of our ARM platforms have a u-boot config which generates a device tree and boots linux.