[PATCH 0/6] Fix of_io_request_and_map error handling code

[PATCH 0/6] Fix of_io_request_and_map error handling code

[Maxime Ripard]

Hi,

This serie fixes a few drivers that use of_io_request_and_map without
properly checking for the error code.

That should fix all the callers present in the kernel so far.

Maxime

Maxime Ripard (6):
  clk: asm9260: Fix of_io_request_and_map error check
  clk: sunxi: Fix of_io_request_and_map error check
  clocksource: asm9260: Fix of_io_request_and_map error check
  clocksource: integrator: Fix of_io_request_and_map error check
  clocksource: sun5i: Fix of_io_request_and_map error check
  irqchip: vf610: Fix of_io_request_and_map error check

 drivers/clk/clk-asm9260.c                 |  2 +-
 drivers/clk/sunxi/clk-sun9i-core.c        | 10 +++++-----
 drivers/clk/sunxi/clk-sunxi.c             |  2 ++
 drivers/clocksource/asm9260_timer.c       |  2 +-
 drivers/clocksource/timer-integrator-ap.c |  2 +-
 drivers/clocksource/timer-sun5i.c         |  2 +-
 drivers/irqchip/irq-vf610-mscm-ir.c       |  5 ++---
 7 files changed, 13 insertions(+), 12 deletions(-)

-- 
2.3.6

[PATCH 1/6] clk: asm9260: Fix of_io_request_and_map error check

[Maxime Ripard]

of_io_request_and map returns an error pointer, but the current code assumes
that on error the returned pointer will be NULL.

Obviously, that makes the check completely useless. Change the test to actually
check for the proper error code.

Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Cc: Mike Turquette <mturquette@linaro.org>
Cc: Stephen Boyd <sboyd@codeaurora.org>
Cc: linux-clk@vger.kernel.org
---
 drivers/clk/clk-asm9260.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/clk/clk-asm9260.c b/drivers/clk/clk-asm9260.c
index 88f4ff6916fe..90897af8d9f7 100644
--- a/drivers/clk/clk-asm9260.c
+++ b/drivers/clk/clk-asm9260.c
@@ -274,7 +274,7 @@ static void __init asm9260_acc_init(struct device_node *np)
 	u32 accuracy = 0;
 
 	base = of_io_request_and_map(np, 0, np->name);
-	if (!base)
+	if (IS_ERR(base))
 		panic("%s: unable to map resource", np->name);
 
 	/* register pll */
-- 
2.3.6

Re: [PATCH 1/6] clk: asm9260: Fix of_io_request_and_map error check

[Maxime Ripard]

[-- Attachment #1: Type: text/plain, Size: 765 bytes --]

Mike, Stephen,

On Sat, May 02, 2015 at 05:03:21PM +0200, Maxime Ripard wrote:
> of_io_request_and map returns an error pointer, but the current code assumes
> that on error the returned pointer will be NULL.
> 
> Obviously, that makes the check completely useless. Change the test to actually
> check for the proper error code.
> 
> Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
> Cc: Mike Turquette <mturquette@linaro.org>
> Cc: Stephen Boyd <sboyd@codeaurora.org>
> Cc: linux-clk@vger.kernel.org

There doesn't seem to be a sub-maintainer for that driver, I guess it
should go through your tree?

Thanks,
Maxime

-- 
Maxime Ripard, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 819 bytes --]

Re: [PATCH 1/6] clk: asm9260: Fix of_io_request_and_map error check

[Stephen Boyd]

On 05/11, Maxime Ripard wrote:
> Mike, Stephen,
> 
> On Sat, May 02, 2015 at 05:03:21PM +0200, Maxime Ripard wrote:
> > of_io_request_and map returns an error pointer, but the current code assumes
> > that on error the returned pointer will be NULL.
> > 
> > Obviously, that makes the check completely useless. Change the test to actually
> > check for the proper error code.
> > 
> > Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
> > Cc: Mike Turquette <mturquette@linaro.org>
> > Cc: Stephen Boyd <sboyd@codeaurora.org>
> > Cc: linux-clk@vger.kernel.org
> 
> There doesn't seem to be a sub-maintainer for that driver, I guess it
> should go through your tree?
> 

Yep. Applied to clk-next.

-- 
Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum,
a Linux Foundation Collaborative Project

[PATCH 2/6] clk: sunxi: Fix of_io_request_and_map error check

[Maxime Ripard]

of_io_request_and map returns an error pointer, but the current code assumes
that on error the returned pointer will be NULL.

Obviously, that makes the check completely useless. Change the test to actually
check for the proper error code.

Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Cc: Mike Turquette <mturquette@linaro.org>
Cc: Stephen Boyd <sboyd@codeaurora.org>
Cc: linux-clk@vger.kernel.org
---
 drivers/clk/sunxi/clk-sun9i-core.c | 10 +++++-----
 drivers/clk/sunxi/clk-sunxi.c      |  2 ++
 2 files changed, 7 insertions(+), 5 deletions(-)

diff --git a/drivers/clk/sunxi/clk-sun9i-core.c b/drivers/clk/sunxi/clk-sun9i-core.c
index d8da77d72861..887f4ea161bb 100644
--- a/drivers/clk/sunxi/clk-sun9i-core.c
+++ b/drivers/clk/sunxi/clk-sun9i-core.c
@@ -93,7 +93,7 @@ static void __init sun9i_a80_pll4_setup(struct device_node *node)
 	void __iomem *reg;
 
 	reg = of_io_request_and_map(node, 0, of_node_full_name(node));
-	if (!reg) {
+	if (IS_ERR(reg)) {
 		pr_err("Could not get registers for a80-pll4-clk: %s\n",
 		       node->name);
 		return;
@@ -154,7 +154,7 @@ static void __init sun9i_a80_gt_setup(struct device_node *node)
 	struct clk *gt;
 
 	reg = of_io_request_and_map(node, 0, of_node_full_name(node));
-	if (!reg) {
+	if (IS_ERR(reg)) {
 		pr_err("Could not get registers for a80-gt-clk: %s\n",
 		       node->name);
 		return;
@@ -218,7 +218,7 @@ static void __init sun9i_a80_ahb_setup(struct device_node *node)
 	void __iomem *reg;
 
 	reg = of_io_request_and_map(node, 0, of_node_full_name(node));
-	if (!reg) {
+	if (IS_ERR(reg)) {
 		pr_err("Could not get registers for a80-ahb-clk: %s\n",
 		       node->name);
 		return;
@@ -244,7 +244,7 @@ static void __init sun9i_a80_apb0_setup(struct device_node *node)
 	void __iomem *reg;
 
 	reg = of_io_request_and_map(node, 0, of_node_full_name(node));
-	if (!reg) {
+	if (IS_ERR(reg)) {
 		pr_err("Could not get registers for a80-apb0-clk: %s\n",
 		       node->name);
 		return;
@@ -310,7 +310,7 @@ static void __init sun9i_a80_apb1_setup(struct device_node *node)
 	void __iomem *reg;
 
 	reg = of_io_request_and_map(node, 0, of_node_full_name(node));
-	if (!reg) {
+	if (IS_ERR(reg)) {
 		pr_err("Could not get registers for a80-apb1-clk: %s\n",
 		       node->name);
 		return;
diff --git a/drivers/clk/sunxi/clk-sunxi.c b/drivers/clk/sunxi/clk-sunxi.c
index 7e1e2bd189b6..9a82f17d2d73 100644
--- a/drivers/clk/sunxi/clk-sunxi.c
+++ b/drivers/clk/sunxi/clk-sunxi.c
@@ -198,6 +198,8 @@ static void __init sun6i_ahb1_clk_setup(struct device_node *node)
 	int i = 0;
 
 	reg = of_io_request_and_map(node, 0, of_node_full_name(node));
+	if (IS_ERR(reg))
+		return;
 
 	/* we have a mux, we will have >1 parents */
 	while (i < SUN6I_AHB1_MAX_PARENTS &&
-- 
2.3.6

Re: [PATCH 2/6] clk: sunxi: Fix of_io_request_and_map error check

[Maxime Ripard]

[-- Attachment #1: Type: text/plain, Size: 658 bytes --]

On Sat, May 02, 2015 at 05:03:22PM +0200, Maxime Ripard wrote:
> of_io_request_and map returns an error pointer, but the current code assumes
> that on error the returned pointer will be NULL.
> 
> Obviously, that makes the check completely useless. Change the test to actually
> check for the proper error code.
> 
> Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
> Cc: Mike Turquette <mturquette@linaro.org>
> Cc: Stephen Boyd <sboyd@codeaurora.org>
> Cc: linux-clk@vger.kernel.org

Applied on my tree.

Maxime

-- 
Maxime Ripard, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 819 bytes --]

[PATCH 3/6] clocksource: asm9260: Fix of_io_request_and_map error check

[Maxime Ripard]

of_io_request_and map returns an error pointer, but the current code assumes
that on error the returned pointer will be NULL.

Obviously, that makes the check completely useless. Change the test to actually
check for the proper error code.

Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Cc: Daniel Lezcano <daniel.lezcano@linaro.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
---
 drivers/clocksource/asm9260_timer.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/clocksource/asm9260_timer.c b/drivers/clocksource/asm9260_timer.c
index 2c9c993727c8..4c2ba59897e8 100644
--- a/drivers/clocksource/asm9260_timer.c
+++ b/drivers/clocksource/asm9260_timer.c
@@ -178,7 +178,7 @@ static void __init asm9260_timer_init(struct device_node *np)
 	unsigned long rate;
 
 	priv.base = of_io_request_and_map(np, 0, np->name);
-	if (!priv.base)
+	if (IS_ERR(priv.base))
 		panic("%s: unable to map resource", np->name);
 
 	clk = of_clk_get(np, 0);
-- 
2.3.6

[tip:timers/core] clocksource: asm9260: Fix of_io_request_and_map error check

[tip-bot for Maxime Ripard]

Commit-ID:  ff569fcd7123d9a9176d59174a27bab9a4fad328
Gitweb:     http://git.kernel.org/tip/ff569fcd7123d9a9176d59174a27bab9a4fad328
Author:     Maxime Ripard <maxime.ripard@free-electrons.com>
AuthorDate: Sat, 2 May 2015 17:03:23 +0200
Committer:  Thomas Gleixner <tglx@linutronix.de>
CommitDate: Tue, 5 May 2015 10:35:58 +0200

clocksource: asm9260: Fix of_io_request_and_map error check

of_io_request_and map returns an error pointer, but the current code assumes
that on error the returned pointer will be NULL.

Obviously, that makes the check completely useless. Change the test to actually
check for the proper error code.

Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Cc: Daniel Lezcano <daniel.lezcano@linaro.org>
Link: http://lkml.kernel.org/r/1430579006-32702-4-git-send-email-maxime.ripard@free-electrons.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 drivers/clocksource/asm9260_timer.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/clocksource/asm9260_timer.c b/drivers/clocksource/asm9260_timer.c
index 2c9c993..4c2ba59 100644
--- a/drivers/clocksource/asm9260_timer.c
+++ b/drivers/clocksource/asm9260_timer.c
@@ -178,7 +178,7 @@ static void __init asm9260_timer_init(struct device_node *np)
 	unsigned long rate;
 
 	priv.base = of_io_request_and_map(np, 0, np->name);
-	if (!priv.base)
+	if (IS_ERR(priv.base))
 		panic("%s: unable to map resource", np->name);
 
 	clk = of_clk_get(np, 0);

[PATCH 4/6] clocksource: integrator: Fix of_io_request_and_map error check

[Maxime Ripard]

of_io_request_and map returns an error pointer, but the current code assumes
that on error the returned pointer will be NULL.

Obviously, that makes the check completely useless. Change the test to actually
check for the proper error code.

Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Cc: Daniel Lezcano <daniel.lezcano@linaro.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
---
 drivers/clocksource/timer-integrator-ap.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/clocksource/timer-integrator-ap.c b/drivers/clocksource/timer-integrator-ap.c
index b9efd30513d5..c97d1980c0f8 100644
--- a/drivers/clocksource/timer-integrator-ap.c
+++ b/drivers/clocksource/timer-integrator-ap.c
@@ -166,7 +166,7 @@ static void __init integrator_ap_timer_init_of(struct device_node *node)
 	struct device_node *sec_node;
 
 	base = of_io_request_and_map(node, 0, "integrator-timer");
-	if (!base)
+	if (IS_ERR(base))
 		return;
 
 	clk = of_clk_get(node, 0);
-- 
2.3.6

[tip:timers/core] clocksource: integrator: Fix of_io_request_and_map error check

[tip-bot for Maxime Ripard]

Commit-ID:  bd580e7ed4add8ce9b2c1dd1911f8fb9d444128b
Gitweb:     http://git.kernel.org/tip/bd580e7ed4add8ce9b2c1dd1911f8fb9d444128b
Author:     Maxime Ripard <maxime.ripard@free-electrons.com>
AuthorDate: Sat, 2 May 2015 17:03:24 +0200
Committer:  Thomas Gleixner <tglx@linutronix.de>
CommitDate: Tue, 5 May 2015 10:35:58 +0200

clocksource: integrator: Fix of_io_request_and_map error check

of_io_request_and map returns an error pointer, but the current code assumes
that on error the returned pointer will be NULL.

Obviously, that makes the check completely useless. Change the test to actually
check for the proper error code.

Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Cc: Daniel Lezcano <daniel.lezcano@linaro.org>
Link: http://lkml.kernel.org/r/1430579006-32702-5-git-send-email-maxime.ripard@free-electrons.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 drivers/clocksource/timer-integrator-ap.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/clocksource/timer-integrator-ap.c b/drivers/clocksource/timer-integrator-ap.c
index b9efd30..c97d198 100644
--- a/drivers/clocksource/timer-integrator-ap.c
+++ b/drivers/clocksource/timer-integrator-ap.c
@@ -166,7 +166,7 @@ static void __init integrator_ap_timer_init_of(struct device_node *node)
 	struct device_node *sec_node;
 
 	base = of_io_request_and_map(node, 0, "integrator-timer");
-	if (!base)
+	if (IS_ERR(base))
 		return;
 
 	clk = of_clk_get(node, 0);

[PATCH 5/6] clocksource: sun5i: Fix of_io_request_and_map error check

[Maxime Ripard]

of_io_request_and map returns an error pointer, but the current code assumes
that on error the returned pointer will be NULL.

Obviously, that makes the check completely useless. Change the test to actually
check for the proper error code.

Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Cc: Daniel Lezcano <daniel.lezcano@linaro.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
---
 drivers/clocksource/timer-sun5i.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/clocksource/timer-sun5i.c b/drivers/clocksource/timer-sun5i.c
index 28aa4b7bb602..0ffb4ea7c925 100644
--- a/drivers/clocksource/timer-sun5i.c
+++ b/drivers/clocksource/timer-sun5i.c
@@ -324,7 +324,7 @@ static void __init sun5i_timer_init(struct device_node *node)
 	int irq;
 
 	timer_base = of_io_request_and_map(node, 0, of_node_full_name(node));
-	if (!timer_base)
+	if (IS_ERR(timer_base))
 		panic("Can't map registers");
 
 	irq = irq_of_parse_and_map(node, 0);
-- 
2.3.6

[tip:timers/core] clocksource: sun5i: Fix of_io_request_and_map error check

[tip-bot for Maxime Ripard]

Commit-ID:  9fa8cc0a85c774b75218ba5f385792be387c190b
Gitweb:     http://git.kernel.org/tip/9fa8cc0a85c774b75218ba5f385792be387c190b
Author:     Maxime Ripard <maxime.ripard@free-electrons.com>
AuthorDate: Sat, 2 May 2015 17:03:25 +0200
Committer:  Thomas Gleixner <tglx@linutronix.de>
CommitDate: Tue, 5 May 2015 10:35:58 +0200

clocksource: sun5i: Fix of_io_request_and_map error check

of_io_request_and map returns an error pointer, but the current code assumes
that on error the returned pointer will be NULL.

Obviously, that makes the check completely useless. Change the test to actually
check for the proper error code.

Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Cc: Daniel Lezcano <daniel.lezcano@linaro.org>
Link: http://lkml.kernel.org/r/1430579006-32702-6-git-send-email-maxime.ripard@free-electrons.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 drivers/clocksource/timer-sun5i.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/clocksource/timer-sun5i.c b/drivers/clocksource/timer-sun5i.c
index 28aa4b7..0ffb4ea 100644
--- a/drivers/clocksource/timer-sun5i.c
+++ b/drivers/clocksource/timer-sun5i.c
@@ -324,7 +324,7 @@ static void __init sun5i_timer_init(struct device_node *node)
 	int irq;
 
 	timer_base = of_io_request_and_map(node, 0, of_node_full_name(node));
-	if (!timer_base)
+	if (IS_ERR(timer_base))
 		panic("Can't map registers");
 
 	irq = irq_of_parse_and_map(node, 0);

[PATCH 6/6] irqchip: vf610: Fix of_io_request_and_map error check

[Maxime Ripard]

of_io_request_and map returns an error pointer, but the current code assumes
that on error the returned pointer will be NULL.

Obviously, that makes the check completely useless. Change the test to actually
check for the proper error code.

Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Cc: Jason Cooper <jason@lakedaemon.net>
Cc: Thomas Gleixner <tglx@linutronix.de>
---
 drivers/irqchip/irq-vf610-mscm-ir.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/irqchip/irq-vf610-mscm-ir.c b/drivers/irqchip/irq-vf610-mscm-ir.c
index 9521057d4744..d0f940a3516d 100644
--- a/drivers/irqchip/irq-vf610-mscm-ir.c
+++ b/drivers/irqchip/irq-vf610-mscm-ir.c
@@ -174,10 +174,9 @@ static int __init vf610_mscm_ir_of_init(struct device_node *node,
 		return -ENOMEM;
 
 	mscm_ir_data->mscm_ir_base = of_io_request_and_map(node, 0, "mscm-ir");
-
-	if (!mscm_ir_data->mscm_ir_base) {
+	if (IS_ERR(mscm_ir_data->mscm_ir_base)) {
 		pr_err("vf610_mscm_ir: unable to map mscm register\n");
-		ret = -ENOMEM;
+		ret = PTR_ERR(mscm_ir_data->mscm_ir_base);
 		goto out_free;
 	}
 
-- 
2.3.6

[tip:irq/core] irqchip: vf610: Fix of_io_request_and_map error check

[tip-bot for Maxime Ripard]

Commit-ID:  dbf07cf0c846c3a0d3f97976931cb4ae78f730e2
Gitweb:     http://git.kernel.org/tip/dbf07cf0c846c3a0d3f97976931cb4ae78f730e2
Author:     Maxime Ripard <maxime.ripard@free-electrons.com>
AuthorDate: Sat, 2 May 2015 17:03:26 +0200
Committer:  Thomas Gleixner <tglx@linutronix.de>
CommitDate: Tue, 5 May 2015 10:45:59 +0200

irqchip: vf610: Fix of_io_request_and_map error check

of_io_request_and map returns an error pointer, but the current code assumes
that on error the returned pointer will be NULL.

Obviously, that makes the check completely useless. Change the test to actually
check for the proper error code.

Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Cc: Jason Cooper <jason@lakedaemon.net>
Link: http://lkml.kernel.org/r/1430579006-32702-7-git-send-email-maxime.ripard@free-electrons.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 drivers/irqchip/irq-vf610-mscm-ir.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/irqchip/irq-vf610-mscm-ir.c b/drivers/irqchip/irq-vf610-mscm-ir.c
index 9521057..d0f940a 100644
--- a/drivers/irqchip/irq-vf610-mscm-ir.c
+++ b/drivers/irqchip/irq-vf610-mscm-ir.c
@@ -174,10 +174,9 @@ static int __init vf610_mscm_ir_of_init(struct device_node *node,
 		return -ENOMEM;
 
 	mscm_ir_data->mscm_ir_base = of_io_request_and_map(node, 0, "mscm-ir");
-
-	if (!mscm_ir_data->mscm_ir_base) {
+	if (IS_ERR(mscm_ir_data->mscm_ir_base)) {
 		pr_err("vf610_mscm_ir: unable to map mscm register\n");
-		ret = -ENOMEM;
+		ret = PTR_ERR(mscm_ir_data->mscm_ir_base);
 		goto out_free;
 	}