[PATCH] ARM: mvebu: don't set the PL310 in I/O coherency mode when I/O coherency is disabled
Gregory CLEMENT
gregory.clement at free-electrons.com
Wed Jan 28 04:29:56 PST 2015
Hi Thomas, Andrew,
On 28/01/2015 12:55, Thomas Petazzoni wrote:
> Since commit f2c3c67f00 (merge commit that adds commit "ARM: mvebu:
> completely disable hardware I/O coherency"), we disable I/O coherency
> on Armada EBU platforms.
>
> However, we continue to initialize the coherency fabric, because this
> coherency fabric is needed on Armada XP for inter-CPU
> coherency. Unfortunately, due to this, we also continued to execute
> the coherency fabric initialization code for Armada 375/38x, which
> switched the PL310 into I/O coherent mode. This has the effect of
> disabling the outer cache sync operation: this is needed when I/O
> coherency is enabled to work around a PCIe/L2 deadlock. But obviously,
> when I/O coherency is disabled, having the outer cache sync operation
> is crucial.
>
> Therefore, this commit fixes the armada_375_380_coherency_init() so
> that the PL310 is switched to I/O coherent mode only if I/O coherency
> is enabled.
>
> Without this fix, all devices using DMA are broken on Armada 375/38x.
Without this fix on my Armada 375 DB board, the kernel hang during the XOR
engine self-test. This fix is really something we should have in 3.19-rc.
Acked-by: Gregory CLEMENT <gregory.clement at free-electrons.com>
and also
Tested-by: Gregory CLEMENT <gregory.clement at free-electrons.com>
Thanks,
Gregory
>
> Signed-off-by: Thomas Petazzoni <thomas.petazzoni at free-electrons.com>
> Cc: <stable at vger.kernel.org>
> ---
> This fix is needed in 3.19-rc, and in all stable kernel releases in
> which f2c3c67f00 was added.
>
> In mainline, it is not strictly needed, since we have since then
> re-enabled I/O coherency using automatic barriers.
>
> Commit applies on top of 3.19-rc6, and has been tested on Armada XP,
> Armada 375 and Armada 38x, by checking that the XOR engine self-tests
> are working properly, which is a good indication that DMA and I/O
> coherency is minimally working. Without the fix, the XOR engine
> self-tests were not even working on Armada 38x.
>
> arch/arm/mach-mvebu/coherency.c | 7 +++++++
> 1 file changed, 7 insertions(+)
>
> diff --git a/arch/arm/mach-mvebu/coherency.c b/arch/arm/mach-mvebu/coherency.c
> index caa21e9..ccef880 100644
> --- a/arch/arm/mach-mvebu/coherency.c
> +++ b/arch/arm/mach-mvebu/coherency.c
> @@ -190,6 +190,13 @@ static void __init armada_375_380_coherency_init(struct device_node *np)
> arch_ioremap_caller = armada_pcie_wa_ioremap_caller;
>
> /*
> + * We should switch the PL310 to I/O coherency mode only if
> + * I/O coherency is actually enabled.
> + */
> + if (!coherency_available())
> + return;
> +
> + /*
> * Add the PL310 property "arm,io-coherent". This makes sure the
> * outer sync operation is not used, which allows to
> * workaround the system erratum that causes deadlocks when
>
--
Gregory Clement, Free Electrons
Kernel, drivers, real-time and embedded Linux
development, consulting, training and support.
http://free-electrons.com
More information about the linux-arm-kernel
mailing list