linux-kernel - [PATCH AUTOSEL 3.18 70/98] dm9000: Fix irq trigger type setup on non-dt platforms

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: <20181025141853.214051-70-sashal@kernel.org>
Date:   Thu, 25 Oct 2018 10:18:25 -0400
From:   Sasha Levin <sashal@...nel.org>
To:     stable@...r.kernel.org, linux-kernel@...r.kernel.org
Cc:     Sylwester Nawrocki <s.nawrocki@...sung.com>,
        "David S . Miller" <davem@...emloft.net>,
        Sasha Levin <sashal@...nel.org>
Subject: [PATCH AUTOSEL 3.18 70/98] dm9000: Fix irq trigger type setup on non-dt platforms

From: Sylwester Nawrocki <s.nawrocki@...sung.com>

[ Upstream commit a96d3b7593a3eefab62dd930e5c99201c3678ee4 ]

Commit b5a099c67a1c36b "net: ethernet: davicom: fix devicetree irq
resource" causes an interrupt storm after the ethernet interface
is activated on S3C24XX platform (ARM non-dt), due to the interrupt
trigger type not being set properly.

It seems, after adding parsing of IRQ flags in commit 7085a7401ba54e92b
"drivers: platform: parse IRQ flags from resources", there is no path
for non-dt platforms where irq_set_type callback could be invoked when
we don't pass the trigger type flags to the request_irq() call.

In case of a board where the regression is seen the interrupt trigger
type flags are passed through a platform device's resource and it is
not currently handled properly without passing the irq trigger type
flags to the request_irq() call.  In case of OF an of_irq_get() call
within platform_get_irq() function seems to be ensuring required irq_chip
setup, but there is no equivalent code for non OF/ACPI platforms.

This patch mostly restores irq trigger type setting code which has been
removed in commit ("net: ethernet: davicom: fix devicetree irq resource").

Fixes: b5a099c67a1c36b913 ("net: ethernet: davicom: fix devicetree irq resource")

Signed-off-by: Sylwester Nawrocki <s.nawrocki@...sung.com>
Acked-by: Robert Jarzmik <robert.jarzmik@...e.fr>
Signed-off-by: David S. Miller <davem@...emloft.net>
Signed-off-by: Sasha Levin <sashal@...nel.org>
---
 drivers/net/ethernet/davicom/dm9000.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/davicom/dm9000.c b/drivers/net/ethernet/davicom/dm9000.c
index ef24b60b4d3f..d2e0a10a8511 100644
--- a/drivers/net/ethernet/davicom/dm9000.c
+++ b/drivers/net/ethernet/davicom/dm9000.c
@@ -1296,6 +1296,7 @@ static int
 dm9000_open(struct net_device *dev)
 {
 	struct board_info *db = netdev_priv(dev);
+	unsigned int irq_flags = irq_get_trigger_type(dev->irq);

 	if (netif_msg_ifup(db))
 		dev_dbg(db->dev, "enabling %s\n", dev->name);
@@ -1303,9 +1304,11 @@ dm9000_open(struct net_device *dev)
 	/* If there is no IRQ type specified, tell the user that this is a
 	 * problem
 	 */
-	if (irq_get_trigger_type(dev->irq) == IRQF_TRIGGER_NONE)
+	if (irq_flags == IRQF_TRIGGER_NONE)
 		dev_warn(db->dev, "WARNING: no IRQ resource flags set.\n");

+	irq_flags |= IRQF_SHARED;
+
 	/* GPIO0 on pre-activate PHY, Reg 1F is not set by reset */
 	iow(db, DM9000_GPR, 0);	/* REG_1F bit0 activate phyxcer */
 	mdelay(1); /* delay needs by DM9000B */
@@ -1313,8 +1316,7 @@ dm9000_open(struct net_device *dev)
 	/* Initialize DM9000 board */
 	dm9000_init_dm9000(dev);

-	if (request_irq(dev->irq, dm9000_interrupt, IRQF_SHARED,
-			dev->name, dev))
+	if (request_irq(dev->irq, dm9000_interrupt, irq_flags, dev->name, dev))
 		return -EAGAIN;
 	/* Now that we have an interrupt handler hooked up we can unmask
 	 * our interrupts
-- 
2.17.1