linux-kernel - [PATCH 2/8] mfd: Prevent STMPE from abusing mfd_add_devices' irq

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]

Message-Id: <1352128236-11554-3-git-send-email-lee.jones@linaro.org>
Date:	Mon,  5 Nov 2012 16:10:30 +0100
From:	Lee Jones <lee.jones@...aro.org>
To:	linux-arm-kernel@...ts.infradead.org, linux-kernel@...r.kernel.org,
	sameo@...ux.intel.com
Cc:	arnd@...db.de, linus.walleij@...ricsson.com,
	Lee Jones <lee.jones@...aro.org>
Subject: [PATCH 2/8] mfd: Prevent STMPE from abusing mfd_add_devices' irq_base parameter

Originally IRQ incrementers were provided in some template resource
structures for keypad and touchscreen devices. These were passed as
IORESOURCE_IRQs to MFD core in the usual way. The true device IRQs
were instead added to the irq_base when mfd_add_devices was invoked.
This is clearly an abuse of the call, and does not scale when IRQ
Domains are brought into play. Before we can provide the STMPE with
its own IRQ Domain we must first fix this. This patche keeps most
of the driver's structure, keeping the template strategy. However,
instead of providing the IRQ as an increment to irq_base, we
dynamically populate the IORESOURCE_IRQ with the correct device IRQ.

Cc: Samuel Ortiz <sameo@...ux.intel.com>
Acked-by: Arnd Bergmann <arnd@...db.de>
Acked-by: Linus Walleij <linus.walleij@...aro.org>
Signed-off-by: Lee Jones <lee.jones@...aro.org>
---
 drivers/mfd/stmpe.c |   25 +++++++++++++------------
 1 file changed, 13 insertions(+), 12 deletions(-)

diff --git a/drivers/mfd/stmpe.c b/drivers/mfd/stmpe.c
index c94f521..ad13cb0 100644
--- a/drivers/mfd/stmpe.c
+++ b/drivers/mfd/stmpe.c
@@ -310,14 +310,10 @@ static struct mfd_cell stmpe_gpio_cell_noirq = {
 static struct resource stmpe_keypad_resources[] = {
 	{
 		.name	= "KEYPAD",
-		.start	= 0,
-		.end	= 0,
 		.flags	= IORESOURCE_IRQ,
 	},
 	{
 		.name	= "KEYPAD_OVER",
-		.start	= 1,
-		.end	= 1,
 		.flags	= IORESOURCE_IRQ,
 	},
 };
@@ -397,14 +393,10 @@ static struct stmpe_variant_info stmpe801_noirq = {
 static struct resource stmpe_ts_resources[] = {
 	{
 		.name	= "TOUCH_DET",
-		.start	= 0,
-		.end	= 0,
 		.flags	= IORESOURCE_IRQ,
 	},
 	{
 		.name	= "FIFO_TH",
-		.start	= 1,
-		.end	= 1,
 		.flags	= IORESOURCE_IRQ,
 	},
 };
@@ -959,10 +951,10 @@ static int __devinit stmpe_chip_init(struct stmpe *stmpe)
 }
 
 static int __devinit stmpe_add_device(struct stmpe *stmpe,
-				      struct mfd_cell *cell, int irq)
+				      struct mfd_cell *cell)
 {
 	return mfd_add_devices(stmpe->dev, stmpe->pdata->id, cell, 1,
-			       NULL, stmpe->irq_base + irq, NULL);
+			       NULL, stmpe->irq_base, NULL);
 }
 
 static int __devinit stmpe_devices_init(struct stmpe *stmpe)
@@ -970,7 +962,7 @@ static int __devinit stmpe_devices_init(struct stmpe *stmpe)
 	struct stmpe_variant_info *variant = stmpe->variant;
 	unsigned int platform_blocks = stmpe->pdata->blocks;
 	int ret = -EINVAL;
-	int i;
+	int i, j;
 
 	for (i = 0; i < variant->num_blocks; i++) {
 		struct stmpe_variant_block *block = &variant->blocks[i];
@@ -978,8 +970,17 @@ static int __devinit stmpe_devices_init(struct stmpe *stmpe)
 		if (!(platform_blocks & block->block))
 			continue;
 
+		for (j = 0; j < block->cell->num_resources; j++) {
+			struct resource *res =
+				(struct resource *) &block->cell->resources[j];
+
+			/* Dynamically fill in a variant's IRQ. */
+			if (res->flags & IORESOURCE_IRQ)
+				res->start = res->end = block->irq + j;
+		}
+
 		platform_blocks &= ~block->block;
-		ret = stmpe_add_device(stmpe, block->cell, block->irq);
+		ret = stmpe_add_device(stmpe, block->cell);
 		if (ret)
 			return ret;
 	}
-- 
1.7.9.5

--
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/