Bug report: kernel paniced when system hibernates

[Conor Dooley]

On Thu, May 25, 2023 at 07:29:46PM +0530, Anup Patel wrote:
> On Thu, May 25, 2023 at 7:26 PM Conor Dooley <conor.dooley at microchip.com> wrote:
> >
> > On Thu, May 25, 2023 at 07:13:11PM +0530, Anup Patel wrote:
> > > On Thu, May 25, 2023 at 7:08 PM Conor Dooley <conor.dooley at microchip.com> wrote:
> > > >
> > > > On Thu, May 25, 2023 at 06:51:28PM +0530, Anup Patel wrote:
> > > >
> > > > > > We should only rely on this node name for known bad versions of opensbi
> > > > > > IMO. Going forward, if something needs to be reserved for firmware, the
> > > > > > firmware should make sure that it is reserved by using the property for
> > > > > > that purpose :)
> > > >
> > > > > There is no issue with OpenSBI since it does the right thing by marking
> > > > > memory as reserved in the DT. This real issue is with the kernel handling
> > > > > of reserved memory for hibernate.
> > > >
> > > > I don't think we are talking about the same thing here. I meant the
> > > > no-map property which OpenSBI does not set.
> > >
> > > Yes, we are talking about the same thing. It's not just OpenSBI not
> > > setting no-map property in reserved memory node because other
> > > SBI implementations would be doing the same thing (i.e. not setting
> > > no-map property)
> >
> > Other SBI implementations doing the same thing doesn't make it any more
> > correct though, right?
> 
> Like multiple folks suggested, we need DT binding for distinguishing
> firmware reserved memory from other reserved memory.

And I have agreed with multiple times!

> Until that
> happens we should either mark hibernate support as experimental
> or revert it.

That works for me. How about the below?

-- >8 --
From 1d4381290a1600eff9b29b8ace6be73955d9726c Mon Sep 17 00:00:00 2001
From: Conor Dooley <conor.dooley at microchip.com>
Date: Thu, 25 May 2023 15:09:08 +0100
Subject: [PATCH] RISC-V: mark hibernation as broken

Hibernation support depends on firmware marking its reserved
regions as not mappable by Linux. As things stand, the de-facto SBI
implementation (OpenSBI) does not do this, and other implementations may
not do so either, resulting in kernel panics during hibernation ([1],
[2]).

Disable support for hibernation until such time that an SBI
implementation independent way to communicate what regions are reserved
has been agreed upon.

Reported-by: Song Shuai <suagrfillet at gmail.com>
Link: https://lore.kernel.org/all/CAAYs2=gQvkhTeioMmqRDVGjdtNF_vhB+vm_1dHJxPNi75YDQ_Q@mail.gmail.com/ [1]
Reported-by: JeeHeng Sia <jeeheng.sia at starfivetech.com>
Link: https://groups.google.com/a/groups.riscv.org/g/sw-dev/c/ITXwaKfA6z8
Signed-off-by: Conor Dooley <conor.dooley at microchip.com>
---
 arch/riscv/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/riscv/Kconfig b/arch/riscv/Kconfig
index 13f058490608..b2495192f35a 100644
--- a/arch/riscv/Kconfig
+++ b/arch/riscv/Kconfig
@@ -801,7 +801,7 @@ menu "Power management options"
 source "kernel/power/Kconfig"
 
 config ARCH_HIBERNATION_POSSIBLE
-	def_bool y
+	def_bool n
 
 config ARCH_HIBERNATION_HEADER
 	def_bool HIBERNATION
-- 
2.39.2

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 228 bytes
Desc: not available
URL: <http://lists.infradead.org/pipermail/linux-riscv/attachments/20230525/2e94063b/attachment.sig>