linux-kernel - Re: [PATCH 12/13] mailbox: omap: Reverse FIFO busy check logic

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: Mon, 1 Apr 2024 18:47:05 -0500
From: Andrew Davis <afd@...com>
To: Hari Nagalla <hnagalla@...com>, Jassi Brar <jassisinghbrar@...il.com>,
        Nick Saulnier <nsaulnier@...com>,
        Bjorn Andersson <andersson@...nel.org>,
        Mathieu Poirier <mathieu.poirier@...aro.org>
CC: <linux-remoteproc@...r.kernel.org>, <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH 12/13] mailbox: omap: Reverse FIFO busy check logic

On 4/1/24 6:31 PM, Hari Nagalla wrote:
> On 3/25/24 12:20, Andrew Davis wrote:
>>   static int omap_mbox_chan_send_noirq(struct omap_mbox *mbox, u32 msg)
>>   {
>> -    int ret = -EBUSY;
>> +    if (mbox_fifo_full(mbox))
>> +        return -EBUSY;
>> -    if (!mbox_fifo_full(mbox)) {
>> -        omap_mbox_enable_irq(mbox, IRQ_RX);
>> -        mbox_fifo_write(mbox, msg);
>> -        ret = 0;
>> -        omap_mbox_disable_irq(mbox, IRQ_RX);
>> +    omap_mbox_enable_irq(mbox, IRQ_RX);
>> +    mbox_fifo_write(mbox, msg);
>> +    omap_mbox_disable_irq(mbox, IRQ_RX);
>> -        /* we must read and ack the interrupt directly from here */
>> -        mbox_fifo_read(mbox);
>> -        ack_mbox_irq(mbox, IRQ_RX);
>> -    }
>> +    /* we must read and ack the interrupt directly from here */
>> +    mbox_fifo_read(mbox);
>> +    ack_mbox_irq(mbox, IRQ_RX);
>> -    return ret;
>> +    return 0;
>>   }
> Is n't the interrupt supposed to be IRQ_TX above? i.e TX ready signal?

Hmm, could be, but this patch doesn't actually change anything, only moves code
around for readability. So if we were are ack'ing the wrong interrupt, then it
was wrong before. We should check that and fix it if needed in a follow up patch.

Andrew