linux-kernel - Re: [PATCH v8] input: keyboard: Add keys driver for the LPC32xx SoC

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, 10 Jul 2012 11:36:25 +0200
From:	Roland Stigge <stigge@...com.de>
To:	Dmitry Torokhov <dmitry.torokhov@...il.com>
CC:	axel.lin@...il.com, riyer@...dia.com, michael.hennerich@...log.com,
	grant.likely@...retlab.ca, linux-input@...r.kernel.org,
	linux-kernel@...r.kernel.org, linux-arm-kernel@...ts.infradead.org,
	kevin.wells@....com, srinivas.bakki@....com,
	devicetree-discuss@...ts.ozlabs.org, rob.herring@...xeda.com,
	aletes.xgr@...il.com
Subject: Re: [PATCH v8] input: keyboard: Add keys driver for the LPC32xx SoC

Hi Dmitry!

On 07/10/2012 08:36 AM, Dmitry Torokhov wrote:
>> +static void lpc32xx_mod_states(struct lpc32xx_kscan_drv *kscandat, int col)
>> +{
>> +	u8 key;
>> +	int row;
>> +	unsigned changed, scancode, keycode;
>> +
>> +	key = readl(LPC32XX_KS_DATA(kscandat->kscan_base, col));
>> +	changed = key ^ kscandat->lastkeystates[col];
>> +	if (changed) {
>> +		for (row = 0; row < kscandat->matrix_sz; row++) {
>> +			if (changed & (1 << row)) {
> 
> I think it could be optimized a bit of you do not scan entire "changed"
> but shift it until it reaches 0 instead.
> 
>> +	of_property_read_u32(np, "nxp,debounce-delay-ms", &kscandat->deb_clks);
>> +	of_property_read_u32(np, "nxp,scan-delay-ms", &kscandat->scan_delay);
>> +	if (!kscandat->deb_clks || !kscandat->scan_delay) {
>> +		dev_err(dev, "debounce or scan delay not specified\n");
>> +		return -ENXIO;
> 
> EINVAL suits better.
> 
>> +
>> +static int __devexit lpc32xx_kscan_remove(struct platform_device *pdev)
>> +{
>> +	struct lpc32xx_kscan_drv *kscandat = platform_get_drvdata(pdev);
>> +
>> +	kfree(kscandat->keymap);
> 
> This seems dangerous in case we manage IRQ fire here.
> 
>> +	free_irq(platform_get_irq(pdev, 0), pdev);
>> +	clk_put(kscandat->clk);
>> +	iounmap(kscandat->kscan_base);
>> +	release_mem_region(kscandat->io_p_start, kscandat->io_p_size);
>> +	input_unregister_device(kscandat->input);
>> +	kfree(kscandat);
>> +
>> +	return 0;
>> +}
>> +
> 
> Does the following patch (on top of your) still work with your device?

Thanks for your suggestions and for the incremental patch!

The direct implementation of your above suggestions looks good.
Unfortunately, you reordered things in probe() which leads to breakage,
e.g. NULL pointer dereference on lpc32xx_parse_dt() because you decided
to fill input->dev.parent only _afterwards_.

However, I'm fine with the 2 above suggested changes. I can post an
updated patch addressing this later today.

Are there any other changes you would require?

Thanks in advance,

Roland
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/