Re: [Qemu-devel] [PULL 00/30] target-arm queue

[Peter Maydell]

On 27 February 2017 at 18:04, Peter Maydell <address@hidden> wrote:
> ARM queu; includes all the NVIC rewrite patches.
> The QOMify-armv7m patchset hasn't got enough review just
> yet but I may be able to sneak it in before freeze
> tomorrow if it gets review. Didn't want to hold this lot
> up waiting, anyway.
>
> thanks
> -- PMM
>
>
> The following changes since commit 8f2d7c341184a95d05476ea3c45dbae2b9ddbe51:
>
>   Merge remote-tracking branch 
> 'remotes/berrange/tags/pull-qcrypto-2017-02-27-1' into staging (2017-02-27 
> 15:33:21 +0000)
>
> are available in the git repository at:
>
>   git://git.linaro.org/people/pmaydell/qemu-arm.git 
> tags/pull-target-arm-20170227
>
> for you to fetch changes up to 94d5bcf5a7f3799660b62098a5183f161aad0601:
>
>   hw/arm/exynos: Fix proper mapping of CPUs by providing real cluster ID 
> (2017-02-27 17:23:16 +0000)
>
> ----------------------------------------------------------------
> target-arm queue:
>  * raspi2: implement RNG module, GPIO and new SD card controller
>    (sufficient to boot new raspbian kernels)
>  * sdhci: bugfixes for block transfers
>  * virt: fix cpu object reference leak
>  * Add missing fp_access_check() to aarch64 crypto instructions
>  * cputlb: Don't assume do_unassigned_access() never returns
>  * virt: Add a user option to disallow ITS instantiation
>  * i.MX timers: fix reset handling
>  * ARMv7M NVIC: rewrite to fix broken priority handling and masking
>  * exynos: Fix proper mapping of CPUs by providing real cluster ID
>  * exynos: Fix Linux kernel division by zero for PLLs
>

Unfortunately I see crashes or assert failures on the raspi2
model on OSX hosts. Not sure why OSX only, probably just that
the malloc/free has different patterns of reusing freed memory.
In any case we clearly didn't get the reparent-the-sdcard code
right, so I'm going to have to drop those patches for the moment :-(

thanks
-- PMM