linux-kernel - Re: [PATCH v7 2/2] add new platform driver for PCI RC

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

Date:	Tue, 02 Feb 2016 21:25:25 +0100
From:	Arnd Bergmann <arnd@...db.de>
To:	Joao Pinto <Joao.Pinto@...opsys.com>
Cc:	Vineet.Gupta1@...opsys.com, helgaas@...nel.org,
	linux-pci@...r.kernel.org, linux-kernel@...r.kernel.org,
	linux-snps-arc@...ts.infradead.org, CARLOS.PALMINHA@...opsys.com,
	Alexey.Brodkin@...opsys.com, robh+dt@...nel.org,
	pawel.moll@....com, mark.rutland@....com,
	ijc+devicetree@...lion.org.uk, galak@...eaurora.org
Subject: Re: [PATCH v7 2/2] add new platform driver for PCI RC

On Monday 01 February 2016 18:07:45 Joao Pinto wrote:
> This patch adds a new driver that will be the reference platform driver
> for all PCI RC IP Protoyping Kits based on ARC SDP.
> This patch is composed by:
> 
> -MAINTAINERS file was updated to include the new driver
> -Documentation/devicetree/bindings/pci was updated to include the new
> driver documentation
> -New driver called pcie-synopsys
> 
> Signed-off-by: Joao Pinto <jpinto@...opsys.com>

I must have completely missed all the earlier submissions and just happened
to see this one now that it was merged.

I have a couple of comments, maybe you can send a follow up patch to address
them:

> +----------------------------------
> +
> +This is the reference platform driver to be used in the Synopsys PCI Root
> +Complex IP Prototyping Kit.
> +
> +Required properties:
> +- compatible: set to "snps,pcie-synopsys";

Please also include the exact version of the designware PCIe part that
is used in here, such as "snps,dw-pcie-1.234". I'm sure you can find
out the version.

"snps,pcie-synopsys" by itself is about the worst possible compatible
string because it says nothing about the specific hardware. Half the
machines we support all have a designware PCIe host that was made
by synopsys. Please use the name of the SoC for the most specific
string.

> +- reg: base address and length of the pcie controller registers.
> +- #address-cells: set to <3>
> +- #size-cells: set to <2>
> +- device_type: set to "pci"
> +- ranges: ranges for the PCI memory and I/O regions.
> +- interrupts: one interrupt source for MSI interrupts, followed by interrupt
> +	source for hardware related interrupts.
> +- #interrupt-cells: set to <1>
> +- num-lanes: set to <1>;
> +
> +Example configuration:
> +
> +	pcie: pcie@...ffff000 {
> +		compatible = "snps,pcie-synopsys";
> +		reg = <0xdffff000 0x1000>;
> +		#address-cells = <3>;
> +		#size-cells = <2>;
> +		device_type = "pci";
> +		ranges = <0x00000800 0 0xd0000000 0xd0000000 0 0x00002000>,

This line looks misplaced here and does not match the documentation. It's
probably a leftover from an earlier version that had the config space
in the ranges, so you need to remove that.

> +/* This handler was created for future work */
> +static irqreturn_t synopsys_pcie_irq_handler(int irq, void *arg)
> +{
> +	return IRQ_NONE;
> +}

It's not harmful, but we generally don't add code "for future work",
better remove it for now.

> +static void synopsys_pcie_establish_link(struct pcie_port *pp)
> +{
> +	int retries = 0;
> +
> +	/* check if the link is up or not */
> +	for (retries = 0; retries < 10; retries++) {
> +		if (dw_pcie_link_up(pp)) {
> +			dev_info(pp->dev, "Link up\n");
> +			return;
> +		}
> +		mdelay(100);
> +	}
> +}

That mdelay() needs to be an msleep(). You should never waste
a whole second worth of CPU time in a driver.

	Arnd