Upstream a long-standing OpenWrt patch for RALINK MT7621 SoC. Selecting
WEAK_REORDERING_BEYOND_LLSC fixes random kernel hangs. This bug and fix
was reported by MediaTek WCN division [0].
[0]: https://lists.infradead.org/pipermail/lede-commits/2017-August/004537.html
Signed-off-by: Ilya Lipnitskiy <[email protected]>
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
---
arch/mips/ralink/Kconfig | 1 +
1 file changed, 1 insertion(+)
diff --git a/arch/mips/ralink/Kconfig b/arch/mips/ralink/Kconfig
index c20c44788b62..31671bbf26ff 100644
--- a/arch/mips/ralink/Kconfig
+++ b/arch/mips/ralink/Kconfig
@@ -62,6 +62,7 @@ choice
select CLKSRC_MIPS_GIC
select HAVE_PCI if PCI_MT7621
select SOC_BUS
+ select WEAK_REORDERING_BEYOND_LLSC
endchoice
choice
--
2.30.1
On Sun, Mar 07, 2021 at 11:08:06AM -0800, Ilya Lipnitskiy wrote:
> Upstream a long-standing OpenWrt patch for RALINK MT7621 SoC. Selecting
> WEAK_REORDERING_BEYOND_LLSC fixes random kernel hangs. This bug and fix
> was reported by MediaTek WCN division [0].
>
> [0]: https://lists.infradead.org/pipermail/lede-commits/2017-August/004537.html
I don't see the Mediatek WCN reporting there. Looking at the 1004K user
manual I couldn't find a notice about such behaviour. So this looks
more like a fix for papering over the real bug by just introducing
a few more syncs. Could you please point me to where this is
really documented ?
Thomas.
--
Crap can work. Given enough thrust pigs will fly, but it's not necessarily a
good idea. [ RFC1925, 2.3 ]
Hi Thomas,
On Fri, Mar 12, 2021 at 2:34 AM Thomas Bogendoerfer
<[email protected]> wrote:
>
> On Sun, Mar 07, 2021 at 11:08:06AM -0800, Ilya Lipnitskiy wrote:
> > Upstream a long-standing OpenWrt patch for RALINK MT7621 SoC. Selecting
> > WEAK_REORDERING_BEYOND_LLSC fixes random kernel hangs. This bug and fix
> > was reported by MediaTek WCN division [0].
> >
> > [0]: https://lists.infradead.org/pipermail/lede-commits/2017-August/004537.html
>
> I don't see the Mediatek WCN reporting there. Looking at the 1004K user
> manual I couldn't find a notice about such behaviour. So this looks
> more like a fix for papering over the real bug by just introducing
> a few more syncs. Could you please point me to where this is
> really documented ?
You may actually be right. After doing some testing we may not need
this change after all. Let me drop this patch for now. FYI, the
following commit may have fixed LL SC issues this change was covering
up: https://lore.kernel.org/lkml/[email protected]/
>
> Thomas.
>
> --
> Crap can work. Given enough thrust pigs will fly, but it's not necessarily a
> good idea. [ RFC1925, 2.3 ]
Ilya