Return-path: Received: from oproxy5-pub.bluehost.com ([67.222.38.55]:46272 "HELO oproxy5-pub.bluehost.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with SMTP id S1759211Ab1F1SUH (ORCPT ); Tue, 28 Jun 2011 14:20:07 -0400 Message-ID: <4E0A1B4F.8080700@boundarydevices.com> (sfid-20110628_203844_531788_1234E3A8) Date: Tue, 28 Jun 2011 11:19:59 -0700 From: Troy Kisky MIME-Version: 1.0 To: "Levi, Shahar" CC: linux-wireless@vger.kernel.org Subject: Re: WL1271: WLAN_IRQ not working References: <4E0942CA.9010402@boundarydevices.com> In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1 Sender: linux-wireless-owner@vger.kernel.org List-ID: On 6/27/2011 11:19 PM, Levi, Shahar wrote: > On Tue, Jun 28, 2011 at 5:56 AM, Troy Kisky > wrote: >> Hi > Hi Troy, > >> >> I trying to get an LSR WL1271 sdio module to work. Everything seems to >> work fine until the WLAN_IRQ pin should go high, but it never does. Checking >> with a scope shows constant low. proc/interrupts also says 0. I'm using >> 3.0.0-rc4 with patches for non-mainline board. If I make the interrupt >> gp a high output, I can raise the level of the WLAN_IRQ pin slightly, >> but it is still a logic 0. >> >> >> tiwi_r2_datasheet.pdf says interrupts are "active-low or falling-edge" >> >> So, why does the driver request an active high interrupt? I'm guessing >> datasheet bug. >> >> Anyway, here are the messages I'm getting with most debugging turned on, >> if someone is willing to take a look. >> >> Thanks >> Troy > Could you possibly check the irq pin setting fit to your platform? It > is read via wl12xx_get_platform_data(). > The driver irq is HIGH polarity level. In case your platform not > support level you can switch to edge via > WL12XX_PLATFORM_QUIRK_EDGE_IRQ. > > All the best, > Shahar > Thanks for taking a look. I'm pretty sure it is a hardware problem and not software. proc/interrupts is showing the correct irq as claimed. Looking at a scope, I'm just getting no activity on the WLAN_IRQ pin. The datasheet says that the slow clock is 32768 HZ +/- 150 ppm. Mine currently measures 32650. This needs fixed, but is unlikely to be my problem isn't it? Perhaps I have a power up sequence problem ? wbr Troy