linux-kernel - Re: [PATCH v1] bus: mhi: ep: Clear channel interrupts that are unmasked

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 for Android: free password hash cracker in your pocket

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

Date:   Mon, 14 Aug 2023 13:44:35 +0530
From:   Manivannan Sadhasivam <manivannan.sadhasivam@...aro.org>
To:     Subramanian Ananthanarayanan <quic_skananth@...cinc.com>
Cc:     helgaas@...nel.org, linux-pci@...r.kernel.org,
        linux-arm-msm@...r.kernel.org, linux-kernel@...r.kernel.org,
        quic_vbadigan@...cinc.com, quic_nitegupt@...cinc.com,
        quic_ramkri@...cinc.com, quic_parass@...cinc.com,
        krzysztof.kozlowski@...aro.org,
        Manivannan Sadhasivam <mani@...nel.org>,
        Jeffrey Hugo <quic_jhugo@...cinc.com>,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        Dan Carpenter <error27@...il.com>,
        "open list:MHI BUS" <mhi@...ts.linux.dev>
Subject: Re: [PATCH v1] bus: mhi: ep: Clear channel interrupts that are
 unmasked

On Fri, Aug 11, 2023 at 02:57:08AM -0700, Subramanian Ananthanarayanan wrote:
> Change is to avoid clearing of interrupts that could have been masked.
> 
> Signed-off-by: Subramanian Ananthanarayanan <quic_skananth@...cinc.com>

Is this scenario possible in real life? We unmask all the interrupts during
power_up and later mask the interrupts specific to channels when they are
started. So I do not see a possibility where the CHDB status would be set for an
unmasked channel.

Looking at the code now, I think we can get rid of the mask check, even the
local copy of mask if there is a guarantee that doorbell interrupts won't be
enabled behind the back of the MHI stack.

IIRC, I added the mask check as the HW channel interrupts were enabled behind
the back. But that was fixed with all unmask during power_up and mask only
started channels strategy.

PS: The term "mask" is horribly used in an inverted manner in MHI.

- Mani

> ---
>  drivers/bus/mhi/ep/main.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/drivers/bus/mhi/ep/main.c b/drivers/bus/mhi/ep/main.c
> index 6008818..a818781 100644
> --- a/drivers/bus/mhi/ep/main.c
> +++ b/drivers/bus/mhi/ep/main.c
> @@ -867,7 +867,7 @@ static void mhi_ep_check_channel_interrupt(struct mhi_ep_cntrl *mhi_cntrl)
>  		if (ch_int) {
>  			mhi_ep_queue_channel_db(mhi_cntrl, ch_int, ch_idx);
>  			mhi_ep_mmio_write(mhi_cntrl, MHI_CHDB_INT_CLEAR_n(i),
> -							mhi_cntrl->chdb[i].status);
> +							ch_int);
>  		}
>  	}
>  }
> -- 
> 2.7.4
> 

-- 
மணிவண்ணன் சதாசிவம்