netdev - Re: [Linux-zigbee-devel] [PATCH 2/2] at86rf230: change irq handling to prevent lockups with edge type irq

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:	Fri, 5 Apr 2013 00:59:49 -0300
From:	Werner Almesberger <werner@...esberger.net>
To:	Sascha Herrmann <sascha@...nvbi.de>
Cc:	linux-zigbee-devel@...ts.sourceforge.net, netdev@...r.kernel.org
Subject: Re: [Linux-zigbee-devel] [PATCH 2/2] at86rf230: change irq handling
 to prevent lockups with edge type irq

Sascha Herrmann wrote:
> -	rc = request_irq(spi->irq, at86rf230_isr, IRQF_SHARED,
> +	rc = request_irq(spi->irq, at86rf230_isr,
> +			 IRQF_SHARED | IRQF_TRIGGER_RISING,

To clarify the purpose of your patch, I think it's only needed
for platforms that don't support level-triggered interrupts.
I.e., if you used IRQF_TRIGGER_HIGH (and happened to have a
platform that supports it, which you said on IRC you don't),
the existing code should work fine. Correct ?

I wonder if it wouldn't be better to make the code work with
both edge and level interrupts instead of having to choose.

E.g., this sort of loop in at86rf230_irqwork should work with
either:

	while (1) {
		irq = read_and_clear_interrupt();
		...
		if (!test_and_clear_bit(0, &lp->irq_disabled))
			break;
		enable_irq();
	}
	lp->irq_busy = 0; /* allow _xmit */

Disadvantage: you get extra IRQ_STATUS reads, which has a
slight penalty, given that all this is on SPI.

To achieve perfection, at86rf230_probe could try all four
possible trigger modes, pick one the platform supports, and
set TRX_CTRL_1.IRQ_POLARITY accordingly.

In any case, I gave your patch very light testing on ATBEN and
it performed as good as the original code.

- Werner
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html