Patchwork questions regarding possible violation of AHCI spec in AHCI driver

login
register
mail settings
Submitter Jian Peng
Date Jan. 11, 2011, 6:55 p.m.
Message ID <E18F441196CA634DB8E1F1C56A50A874319FA7E37B@IRVEXCHCCR01.corp.ad.broadcom.com>
Download mbox | patch
Permalink /patch/78422/
State Not Applicable
Delegated to: David Miller
Headers show

Comments

Jian Peng - Jan. 11, 2011, 6:55 p.m.
Hi, Tejun,

Here is new patch based on your suggestion (against 2.6.37 release)


I will test host controller made by other vendors and report issues soon.

Thanks,
Jian
Harik - Jan. 12, 2011, 12:45 a.m.
> +           (tmp & 0x0f) != 0x03) {
> +               printk(KERN_DEBUG "START bit was not set in %s\n", __func__);
> +               return;
> +       }

Are there defines for the mask and result that you could use instead of raw hex?
--
To unsubscribe from this list: send the line "unsubscribe linux-ide" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Jian Peng - Jan. 12, 2011, 12:51 a.m.
Unfortunately, no

-----Original Message-----
From: Harik [mailto:harik.attar@gmail.com] 
Sent: Tuesday, January 11, 2011 4:46 PM
To: Jian Peng
Cc: Tejun Heo; Robert Hancock; linux-kernel@vger.kernel.org; jgarzik@pobox.com; ide
Subject: Re: questions regarding possible violation of AHCI spec in AHCI driver

> +           (tmp & 0x0f) != 0x03) {
> +               printk(KERN_DEBUG "START bit was not set in %s\n", __func__);
> +               return;
> +       }

Are there defines for the mask and result that you could use instead of raw hex?


--
To unsubscribe from this list: send the line "unsubscribe linux-ide" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Jeff Garzik - Jan. 19, 2011, 12:51 a.m.
On 01/11/2011 01:55 PM, Jian Peng wrote:
> Here is new patch based on your suggestion (against 2.6.37 release)
[...]
> I will test host controller made by other vendors and report issues soon.

Any updates?

	Jeff



--
To unsubscribe from this list: send the line "unsubscribe linux-ide" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Jian Peng - Jan. 19, 2011, 12:58 a.m.
Sorry that I switched to other tasks recently, and will report the testing results within this week.
Any suggestion on how to test the code path affected by this patch, AFAIK, bootup and hotplug will use this path, What else?

Thanks,
Jian

-----Original Message-----
From: Jeff Garzik [mailto:jgpobox@gmail.com] On Behalf Of Jeff Garzik
Sent: Tuesday, January 18, 2011 4:51 PM
To: Jian Peng
Cc: Tejun Heo; Robert Hancock; linux-kernel@vger.kernel.org; ide
Subject: Re: questions regarding possible violation of AHCI spec in AHCI driver

On 01/11/2011 01:55 PM, Jian Peng wrote:
> Here is new patch based on your suggestion (against 2.6.37 release)
[...]
> I will test host controller made by other vendors and report issues soon.

Any updates?

	Jeff





--
To unsubscribe from this list: send the line "unsubscribe linux-ide" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Jian Peng - Jan. 19, 2011, 11:35 p.m.
Hi, Jeff,

I bought a bunch of PCI cards and tested them, only one running AHCI using Marvell 88SE9125 controller chip. It worked well with the patched kernel at bootup, hotplug, and data transfer.
I put an order of a JMicron AHCI based PCI card and will test it.

Thanks,
Jian

-----Original Message-----
From: Jeff Garzik [mailto:jgpobox@gmail.com] On Behalf Of Jeff Garzik
Sent: Tuesday, January 18, 2011 4:51 PM
To: Jian Peng
Cc: Tejun Heo; Robert Hancock; linux-kernel@vger.kernel.org; ide
Subject: Re: questions regarding possible violation of AHCI spec in AHCI driver

On 01/11/2011 01:55 PM, Jian Peng wrote:
> Here is new patch based on your suggestion (against 2.6.37 release)
[...]
> I will test host controller made by other vendors and report issues soon.

Any updates?

	Jeff





--
To unsubscribe from this list: send the line "unsubscribe linux-ide" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Jian Peng - Jan. 19, 2011, 11:37 p.m.
Resent it since mailbox crashed last time

Hi, Jeff,

I bought a bunch of PCI cards and tested them, only one running AHCI using Marvell 88SE9125 controller chip. It worked well with the patched kernel at bootup, hotplug, and data transfer.
I put an order of a JMicron AHCI based PCI card and will test it.

Thanks,
Jian

-----Original Message-----
From: Jeff Garzik [mailto:jgpobox@gmail.com] On Behalf Of Jeff Garzik
Sent: Tuesday, January 18, 2011 4:51 PM
To: Jian Peng
Cc: Tejun Heo; Robert Hancock; linux-kernel@vger.kernel.org; ide
Subject: Re: questions regarding possible violation of AHCI spec in AHCI driver

On 01/11/2011 01:55 PM, Jian Peng wrote:
> Here is new patch based on your suggestion (against 2.6.37 release)
[...]
> I will test host controller made by other vendors and report issues soon.

Any updates?

	Jeff





--
To unsubscribe from this list: send the line "unsubscribe linux-ide" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Jian Peng - April 19, 2011, 9:48 p.m.
Hi, Jeff/Tejun,

I retested my patch using Marvell AHCI controller card and Intel built-in controller with 2.6.38 kernel, it worked well. I resubmitted the patch against Linus git tree in separate email through git send-email already. 

Please review it.

Thanks,
Jian

-----Original Message-----
From: Jeff Garzik [mailto:jgpobox@gmail.com] On Behalf Of Jeff Garzik
Sent: Tuesday, January 18, 2011 4:51 PM
To: Jian Peng
Cc: Tejun Heo; Robert Hancock; linux-kernel@vger.kernel.org; ide
Subject: Re: questions regarding possible violation of AHCI spec in AHCI driver

On 01/11/2011 01:55 PM, Jian Peng wrote:
> Here is new patch based on your suggestion (against 2.6.37 release)
[...]
> I will test host controller made by other vendors and report issues soon.

Any updates?

	Jeff





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

Patch

--- libahci.c.orig	2011-01-11 10:46:56.623991326 -0800
+++ libahci.c	2011-01-11 10:52:19.634036938 -0800
@@ -542,6 +542,15 @@ 
 {
 	void __iomem *port_mmio = ahci_port_base(ap);
 	u32 tmp;
+	u8 status;
+	
+	status = readl(port_mmio + PORT_TFDATA);
+	if (status & (ATA_BUSY | ATA_DRQ) ||
+	    ahci_scr_read(&ap->link, SCR_STATUS, &tmp) ||
+	    (tmp & 0x0f) != 0x03) {
+		printk(KERN_DEBUG "START bit was not set in %s\n", __func__);
+		return;
+	}	
 
 	/* start DMA */
 	tmp = readl(port_mmio + PORT_CMD);