Hi Bin, On 6 August 2017 at 03:13, Bin Meng <bmeng...@gmail.com> wrote: > Hi Simon, > > On Sun, Aug 6, 2017 at 1:16 PM, Simon Glass <s...@chromium.org> wrote: >> Hi Bin, >> >> On 3 August 2017 at 18:17, Bin Meng <bmeng...@gmail.com> wrote: >>> Hi Simon, >>> >>> On Thu, Aug 3, 2017 at 11:24 PM, Simon Glass <s...@chromium.org> wrote: >>>> Hi Bin, >>>> >>>> On 1 August 2017 at 17:33, Bin Meng <bmeng...@gmail.com> wrote: >>>>> Convert SANDBOX_BITS_PER_LONG to Kconfig and assign it a correct >>>>> number depending on which host we are going to build and run. >>>>> >>>>> Signed-off-by: Bin Meng <bmeng...@gmail.com> >>>>> --- >>>>> >>>>> arch/sandbox/Kconfig | 5 +++++ >>>>> board/sandbox/README.sandbox | 7 +++---- >>>>> scripts/config_whitelist.txt | 1 - >>>>> 3 files changed, 8 insertions(+), 5 deletions(-) >>>> >>>> Reviewed-by: Simon Glass <s...@chromium.org> >>>> >>>> Is it OK to build 64-bit sandbox on a 32-bit machine? Does that actually >>>> work? >>>> >>>> If then I think we need a 3-way setting like: >>>> >>>> - 32-bit >>>> - 64-bit >>>> - native (i.e. whatever the host is) >>> >>> That means cross-compiling sandbox. So far this is not working. I will >>> take a look. >> >> Or perhaps we just require it to use the bit size of the host? Does >> compiling 64-bit U-Boot on a 32-bit machine actually work? >> > > I have not looked into that further but I suspect there is more work > than the bit size of the host, for example, linking 64-bit vs. 32-bit > libraries?
But does it compile 64-bit sandbox OK on a 32-bit machine? I would have thought it would complain. I'm not really asking for this feature, it's just that I don't understand how your patch works, and don't necessarily want to lose the ability to build a 32-bit sandbox. It is to some extent a helpful build/unit test for 32-bit boards which we still have many of. Regards, Simon _______________________________________________ U-Boot mailing list U-Boot@lists.denx.de https://lists.denx.de/listinfo/u-boot