On Fri, Feb 25, 2022 at 01:52:52PM +0000, Peter Maydell wrote:
> On Fri, 25 Feb 2022 at 09:19, Stafford Horne <shorne@gmail.com> wrote:
> >
> > The following changes since commit 4aa2e497a98bafe962e72997f67a369e4b52d9c1:
> >
> > Merge remote-tracking branch 'remotes/berrange-gitlab/tags/misc-next-pull-request' into staging (2022-02-23 09:25:05 +0000)
> >
> > are available in the Git repository at:
> >
> > git://github.com/stffrdhrn/qemu.git tags/or1k-pull-request
> >
> > for you to fetch changes up to 94c71f14e9ca15ede4172e0826d690b15069a7f8:
> >
> > hw/openrisc/openrisc_sim: Add support for initrd loading (2022-02-25 15:42:23 +0900)
> >
> > ----------------------------------------------------------------
> > OpenRISC patches
> >
> > - Add automatic DTS generation to openrisc_sim
> >
> > ----------------------------------------------------------------
> > Stafford Horne (6):
> > hw/openrisc/openrisc_sim: Create machine state for or1ksim
> > hw/openrisc/openrisc_sim: Parameterize initialization
> > hw/openrisc/openrisc_sim: Use IRQ splitter when connecting UART
> > hw/openrisc/openrisc_sim: Increase max_cpus to 4
> > hw/openrisc/openrisc_sim: Add automatic device tree generation
> > hw/openrisc/openrisc_sim: Add support for initrd loading
>
> Hi; this fails to build on various CI configs, eg:
> https://gitlab.com/qemu-project/qemu/-/jobs/2137393314
> https://gitlab.com/qemu-project/qemu/-/jobs/2137393335
>
> ../hw/openrisc/openrisc_sim.c:40:10: fatal error: libfdt.h: No such
> file or directory
> 40 | #include <libfdt.h>
> | ^~~~~~~~~~
>
>
> This happens because meson doesn't put the include path for libfdt
> on the include path for every .c file -- you have to do something
> special in the meson.build file for the files that include it.
> Paolo can tell you what that is, I expect.
OK, I missed the CI results as it was all working for me. I will fix and test
with the same as CI configs.
-Stafford
> Paolo: are we going to be able to stop doing this at some point
> and get meson to just DTRT and put includes on the path for
> every C file ?
>
> thanks
> -- PMM