[PATCH 08/22] ARM: omap1: move CF chipselect setup to board file

From: Arnd Bergmann <arnd@arndb.de>
To: Tony Lindgren <tony@atomide.com>,
	Aaro Koskinen <aaro.koskinen@iki.fi>,
	Dominik Brodowski <linux@dominikbrodowski.net>
Cc: linux-omap@vger.kernel.org, linux-arm-kernel@lists.infradead.org,
	Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	Linus Walleij <linus.walleij@linaro.org>,
	Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com>,
	Tomi Valkeinen <tomi.valkeinen@ti.com>,
	Arnd Bergmann <arnd@arndb.de>,
	linux-kernel@vger.kernel.org
Subject: [PATCH 08/22] ARM: omap1: move CF chipselect setup to board file
Date: Thu,  8 Aug 2019 23:22:17 +0200	[thread overview]
Message-ID: <20190808212234.2213262-9-arnd@arndb.de> (raw)
In-Reply-To: <20190808212234.2213262-1-arnd@arndb.de>

There is only one board that uses the omap_cf driver, so
moving the chipselect configuration there does not lead
to code duplication but avoids the use of mach/tc.h
in drivers.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 arch/arm/mach-omap1/board-osk.c | 38 ++++++++++++++++++++++++++++-----
 drivers/pcmcia/Kconfig          |  3 ++-
 drivers/pcmcia/omap_cf.c        | 38 ++++++---------------------------
 3 files changed, 42 insertions(+), 37 deletions(-)

diff --git a/arch/arm/mach-omap1/board-osk.c b/arch/arm/mach-omap1/board-osk.c
index 99ebe4503787..38d73da5d13d 100644
--- a/arch/arm/mach-omap1/board-osk.c
+++ b/arch/arm/mach-omap1/board-osk.c
@@ -149,14 +149,14 @@ static struct resource osk5912_cf_resources[] = {
 	[0] = {
 		.flags	= IORESOURCE_IRQ,
 	},
+	[1] = {
+		.flags = IORESOURCE_MEM,
+	},
 };
 
 static struct platform_device osk5912_cf_device = {
 	.name		= "omap_cf",
 	.id		= -1,
-	.dev = {
-		.platform_data	= (void *) 2 /* CS2 */,
-	},
 	.num_resources	= ARRAY_SIZE(osk5912_cf_resources),
 	.resource	= osk5912_cf_resources,
 };
@@ -267,13 +267,41 @@ static void __init osk_init_smc91x(void)
 	omap_writel(l, EMIFS_CCS(1));
 }
 
-static void __init osk_init_cf(void)
+static void __init osk_init_cf(int seg)
 {
+	struct resource *res = &osk5912_cf_resources[1];
+
 	omap_cfg_reg(M7_1610_GPIO62);
 	if ((gpio_request(62, "cf_irq")) < 0) {
 		printk("Error requesting gpio 62 for CF irq\n");
 		return;
 	}
+
+	switch (seg) {
+	/* NOTE: CS0 could be configured too ... */
+	case 1:
+		res->start = OMAP_CS1_PHYS;
+		break;
+	case 2:
+		res->start = OMAP_CS2_PHYS;
+		break;
+	case 3:
+		res->start = omap_cs3_phys();
+		break;
+	}
+
+	res->end = res->start + SZ_8K - 1;
+	osk5912_cf_device.dev.platform_data = (void *)(uintptr_t)seg;
+
+	/* NOTE:  better EMIFS setup might support more cards; but the
+	 * TRM only shows how to affect regular flash signals, not their
+	 * CF/PCMCIA variants...
+	 */
+	pr_debug("%s: cs%d, previous ccs %08x acs %08x\n", __func__,
+		seg, omap_readl(EMIFS_CCS(seg)), omap_readl(EMIFS_ACS(seg)));
+	omap_writel(0x0004a1b3, EMIFS_CCS(seg));	/* synch mode 4 etc */
+	omap_writel(0x00000000, EMIFS_ACS(seg));	/* OE hold/setup */
+
 	/* the CF I/O IRQ is really active-low */
 	irq_set_irq_type(gpio_to_irq(62), IRQ_TYPE_EDGE_FALLING);
 }
@@ -577,7 +605,7 @@ static void __init osk_init(void)
 	u32 l;
 
 	osk_init_smc91x();
-	osk_init_cf();
+	osk_init_cf(2); /* CS2 */
 
 	/* Workaround for wrong CS3 (NOR flash) timing
 	 * There are some U-Boot versions out there which configure
diff --git a/drivers/pcmcia/Kconfig b/drivers/pcmcia/Kconfig
index e004d8da03dc..ca6e2ac2a92f 100644
--- a/drivers/pcmcia/Kconfig
+++ b/drivers/pcmcia/Kconfig
@@ -250,7 +250,8 @@ config PCMCIA_VRC4173
 
 config OMAP_CF
 	tristate "OMAP CompactFlash Controller"
-	depends on PCMCIA && ARCH_OMAP16XX
+	depends on PCMCIA
+	depends on ARCH_OMAP16XX || (ARM && COMPILE_TEST)
 	help
 	  Say Y here to support the CompactFlash controller on OMAP.
 	  Note that this doesn't support "True IDE" mode.
diff --git a/drivers/pcmcia/omap_cf.c b/drivers/pcmcia/omap_cf.c
index 0a04eb04f3a2..98df6473034d 100644
--- a/drivers/pcmcia/omap_cf.c
+++ b/drivers/pcmcia/omap_cf.c
@@ -16,13 +16,12 @@
 
 #include <pcmcia/ss.h>
 
-#include <mach/hardware.h>
 #include <asm/io.h>
 #include <linux/sizes.h>
 
-#include <mach/mux.h>
-#include <mach/tc.h>
-
+#include <linux/soc/ti/omap1-io.h>
+#include <linux/soc/ti/omap1-soc.h>
+#include <linux/soc/ti/omap1-mux.h>
 
 /* NOTE:  don't expect this to support many I/O cards.  The 16xx chips have
  * hard-wired timings to support Compact Flash memory cards; they won't work
@@ -205,6 +204,7 @@ static int __init omap_cf_probe(struct platform_device *pdev)
 	struct omap_cf_socket	*cf;
 	int			irq;
 	int			status;
+	struct resource		*res;
 
 	seg = (int) pdev->dev.platform_data;
 	if (seg == 0 || seg > 3)
@@ -215,6 +215,8 @@ static int __init omap_cf_probe(struct platform_device *pdev)
 	if (irq < 0)
 		return -EINVAL;
 
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+
 	cf = kzalloc(sizeof *cf, GFP_KERNEL);
 	if (!cf)
 		return -ENOMEM;
@@ -230,24 +232,7 @@ static int __init omap_cf_probe(struct platform_device *pdev)
 		goto fail0;
 	cf->irq = irq;
 	cf->socket.pci_irq = irq;
-
-	switch (seg) {
-	/* NOTE: CS0 could be configured too ... */
-	case 1:
-		cf->phys_cf = OMAP_CS1_PHYS;
-		break;
-	case 2:
-		cf->phys_cf = OMAP_CS2_PHYS;
-		break;
-	case 3:
-		cf->phys_cf = omap_cs3_phys();
-		break;
-	default:
-		goto  fail1;
-	}
-	cf->iomem.start = cf->phys_cf;
-	cf->iomem.end = cf->iomem.end + SZ_8K - 1;
-	cf->iomem.flags = IORESOURCE_MEM;
+	cf->phys_cf = res->start;
 
 	/* pcmcia layer only remaps "real" memory */
 	cf->socket.io_offset = (unsigned long)
@@ -269,15 +254,6 @@ static int __init omap_cf_probe(struct platform_device *pdev)
 
 	pr_info("%s: cs%d on irq %d\n", driver_name, seg, irq);
 
-	/* NOTE:  better EMIFS setup might support more cards; but the
-	 * TRM only shows how to affect regular flash signals, not their
-	 * CF/PCMCIA variants...
-	 */
-	pr_debug("%s: cs%d, previous ccs %08x acs %08x\n", driver_name,
-		seg, omap_readl(EMIFS_CCS(seg)), omap_readl(EMIFS_ACS(seg)));
-	omap_writel(0x0004a1b3, EMIFS_CCS(seg));	/* synch mode 4 etc */
-	omap_writel(0x00000000, EMIFS_ACS(seg));	/* OE hold/setup */
-
 	/* CF uses armxor_ck, which is "always" available */
 
 	pr_debug("%s: sts %04x cfg %04x control %04x %s\n", driver_name,
-- 
2.20.0

