linux-kernel - Re: [PATCH RESEND v2] interconnect: drop unused icc

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:   Tue, 20 Jun 2023 14:57:17 -0700
From:   Mike Tipton <quic_mdtipton@...cinc.com>
To:     Johan Hovold <johan+linaro@...nel.org>,
        Georgi Djakov <djakov@...nel.org>
CC:     Bjorn Andersson <andersson@...nel.org>, <linux-pm@...r.kernel.org>,
        <linux-kernel@...r.kernel.org>,
        Konrad Dybcio <konrad.dybcio@...aro.org>,
        Odelu Kukatla <quic_okukatla@...cinc.com>
Subject: Re: [PATCH RESEND v2] interconnect: drop unused icc_get() interface

On 5/23/2023 2:52 AM, Johan Hovold wrote:
> The icc_get() interface can be used to lookup an interconnect path based
> on global node ids. There has never been any users of this interface and
> all lookups are currently done from the devicetree.
> 
> Remove the unused icc_get() interface.
> 
> Reviewed-by: Konrad Dybcio <konrad.dybcio@...aro.org>
> Signed-off-by: Johan Hovold <johan+linaro@...nel.org>
> ---
> 
> Hi Georgi,
> 
> I just noticed that this patch never made it into 6.4 along with the
> rest of the series:
> 
> 	https://lore.kernel.org/lkml/20230306075651.2449-23-johan+linaro@kernel.org/
> 
> This interface is still unused in mainline and should be removed so
> resending the patch again.
> 
> Johan
> 
> 
>   drivers/interconnect/core.c  | 52 ++----------------------------------
>   include/linux/interconnect.h |  8 ------
>   2 files changed, 2 insertions(+), 58 deletions(-)
> 

We have downstream debug/test modules that removing icc_get() will 
break. I'd like to get equivalent debug support in mainline, but until 
then I'd prefer we not remove this.

Our debug module adds debugfs files for requesting a path and voting BW 
to it. The path is defined by supplying the raw src_id / dst_id and 
calling icc_get() for them. This allows us to issue any request to any 
path from debugfs without needing to recompile or reflash devicetree. We 
use this extensively.

That said, I don't like the current implementation of icc_get(). It 
should take strings for src/dst rather than integer IDs. This would be 
similar to other interfaces like clk_get() and regulator_get(). And 
would allow users to specify their path in debugfs using logical names 
rather than raw integers.

I suspect having a mainline approach for voting paths from debugfs would 
be useful to others as well. There are similar debugfs control 
mechanisms in other frameworks already, e.g. clock.

Instead of removing icc_get() immediately, can we wait for a future 
patch series that adds debugfs as a consumer?