linux-kernel - Re: [PATCH 1/2] bindings: ipmi: Add property for skipping SBMR boot progress response

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date: Thu, 13 Jun 2024 13:55:01 -0500
From: Corey Minyard <corey@...yard.net>
To: Rob Herring <robh@...nel.org>
Cc: Potin Lai <potin.lai.pt@...il.com>, Corey Minyard <minyard@....org>,
	Krzysztof Kozlowski <krzysztof.kozlowski+dt@...aro.org>,
	Conor Dooley <conor+dt@...nel.org>,
	Quan Nguyen <quan@...amperecomputing.com>,
	openipmi-developer@...ts.sourceforge.net,
	devicetree@...r.kernel.org, linux-kernel@...r.kernel.org,
	Patrick Williams <patrick@...cx.xyz>,
	Cosmo Chou <cosmo.chou@...ntatw.com>,
	Potin Lai <potin.lai@...ntatw.com>
Subject: Re: [PATCH 1/2] bindings: ipmi: Add property for skipping SBMR boot
 progress response

On Thu, Jun 13, 2024 at 11:59:46AM -0600, Rob Herring wrote:
> On Wed, Jun 12, 2024 at 12:32:54PM +0800, Potin Lai wrote:
> > In ARM Server Base Manageability Requirements (SBMR) document, Callers can
> > choose to not read back Response Data after sending the command "Send Boot
> > Progress Code".
> 
> Got a link to that document?
> 
> > Define "arm-sbmr,skip-bootprogress-response" property for skipping the
> > response of "Send Boot Progress Code" from userspace.
> 
> I don't understand why this would be conditional? How can you define in 
> the BMC what the host behavior is? Doesn't the host side decide 
> that? So don't you always have to support no response?

Yeah, this doesn't make any sense for two reasons:

What if the host wanted to read back the response?  You make no
provision for that, as I believe Rob said above.

The BMC should be able to start a new transaction without the previous
response being read.  So this should be pointless.  If that's not
happening, it's a bug and should be fixed.  Otherwise an untimely reset
could hang the SSIF interface.

-corey

> 
> > 
> > Signed-off-by: Potin Lai <potin.lai.pt@...il.com>
> > ---
> >  Documentation/devicetree/bindings/ipmi/ssif-bmc.yaml | 5 +++++
> >  1 file changed, 5 insertions(+)
> > 
> > diff --git a/Documentation/devicetree/bindings/ipmi/ssif-bmc.yaml b/Documentation/devicetree/bindings/ipmi/ssif-bmc.yaml
> > index 02b662d780bbb..b21e958efc184 100644
> > --- a/Documentation/devicetree/bindings/ipmi/ssif-bmc.yaml
> > +++ b/Documentation/devicetree/bindings/ipmi/ssif-bmc.yaml
> > @@ -19,6 +19,11 @@ properties:
> >    reg:
> >      maxItems: 1
> >  
> > +  arm-sbmr,skip-bootprogress-response:
> 
> Form is vendor,property-name where vendor is defined in 
> vendor-prefixes.yaml. 'arm-sbmr' is not a vendor.
> 
> > +    type: boolean
> > +    description:
> > +      Skipping ARM SBMR “Send Boot Progress Code” response.
> > +
> >  required:
> >    - compatible
> >    - reg
> > -- 
> > 2.31.1
> >