Skip to contentSkip to conversation

clk: socfpga: Fix undefined behavior bug and add bounds-checking coverage

PATCH Applied Gustavo A. R. Silva 2 participants 5 messages Oct 24, 2023

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

v3 cover clk: socfpga: Fix undefined behavior bug and add bounds-checking coverage

This series aims to fix an undefined behavior bug in struct stratix10_clock_data and add bounds-checking coverage at run-time for flexible-array member hws in struct clk_hw_onecell_data when accessed throught struct stratix10_clock_data.

Changes in v3:
- None, really. Just Cc [email protected] this time.

Changes in v2:
- Mention -Wflex-array-member-not-at-end in the changelog text.
- Link: https://lore.kernel.org/linux-hardening/[email protected]/

v1:
- Link: https://lore.kernel.org/linux-hardening/[email protected]/

Gustavo A. R. Silva (2):
clk: socfpga: Fix undefined behavior bug in struct
stratix10_clock_data
clk: socfpga: agilex: Add bounds-checking coverage for struct
stratix10_clock_data

 drivers/clk/socfpga/clk-agilex.c    | 12 ++++++------
 drivers/clk/socfpga/clk-s10.c       |  6 +++---
 drivers/clk/socfpga/stratix10-clk.h |  4 +++-
 3 files changed, 12 insertions(+), 10 deletions(-)
Signature · 1 line
2.34.1
#1 lore raw

PATCH v3 1/2 clk: socfpga: Fix undefined behavior bug in struct stratix10_clock_data

struct clk_hw_onecell_data is a flexible structure, which means that it contains flexible-array member at the bottom, in this case array hws:

include/linux/clk-provider.h:
1380 struct clk_hw_onecell_data {
1381 unsigned int num;
1382 struct clk_hw *hws[] __counted_by(num);
1383 };

This could potentially lead to an overwrite of the objects following clk_data in struct stratix10_clock_data, in this case void __iomem *base; at run-time:

drivers/clk/socfpga/stratix10-clk.h:
9 struct stratix10_clock_data {
10 struct clk_hw_onecell_data clk_data;
11 void __iomem *base;
12 };

There are currently three different places where memory is allocated for struct stratix10_clock_data, including the flex-array hws in struct clk_hw_onecell_data:

drivers/clk/socfpga/clk-agilex.c:
469 clk_data = devm_kzalloc(dev, struct_size(clk_data, clk_data.hws,
470 num_clks), GFP_KERNEL);

drivers/clk/socfpga/clk-agilex.c:
509 clk_data = devm_kzalloc(dev, struct_size(clk_data, clk_data.hws,
510 num_clks), GFP_KERNEL);

drivers/clk/socfpga/clk-s10.c:
400 clk_data = devm_kzalloc(dev, struct_size(clk_data, clk_data.hws,
401 num_clks), GFP_KERNEL);

I'll use just one of them to describe the issue. See below.

Notice that a total of 440 bytes are allocated for flexible-array member hws at line 469:

include/dt-bindings/clock/agilex-clock.h:
70 #define AGILEX_NUM_CLKS 55

drivers/clk/socfpga/clk-agilex.c:
459 struct stratix10_clock_data *clk_data;
460 void __iomem *base;
...
466
467 num_clks = AGILEX_NUM_CLKS;
468
469 clk_data = devm_kzalloc(dev, struct_size(clk_data, clk_data.hws,
470 num_clks), GFP_KERNEL);

struct_size(clk_data, clk_data.hws, num_clks) above translates to
sizeof(struct stratix10_clock_data) + sizeof(struct clk_hw *) * 55 == 16 + 8 * 55 == 16 + 440

		    ^^^
		     |
	allocated bytes for flex-array `hws`

474 for (i = 0; i < num_clks; i++)
475 clk_data->clk_data.hws[i] = ERR_PTR(-ENOENT);
476
477 clk_data->base = base;

and then some data is written into both hws and base objects.

Fix this by placing the declaration of object clk_data at the end of struct stratix10_clock_data. Also, add a comment to make it clear that this object must always be last in the structure.

-Wflex-array-member-not-at-end is coming in GCC-14, and we are getting ready to enable it globally.

Fixes ba7e258425ac ("clk: socfpga: Convert to s10/agilex/n5x to use clk_hw")

---
Changes in v3:
- None, really. Just Cc [email protected] this time.

Changes in v2:
- Mention -Wflex-array-member-not-at-end in the changelog text.
- Link: https://lore.kernel.org/linux-hardening/bb2ae50abf622d4b6fda1a326ce830627356ab7c.1697493574.git.gustavoars@kernel.org/

v1:
- Link: https://lore.kernel.org/linux-hardening/5dd8483177dc8cd91d021170b6717f2e570bab03.1697059539.git.gustavoars@kernel.org/

 drivers/clk/socfpga/stratix10-clk.h | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/clk/socfpga/stratix10-clk.h +3 -1
index 75234e0783e1..83fe4eb3133c 100644
@@ -7,8 +7,10 @@
 #define	__STRATIX10_CLK_H
 
 struct stratix10_clock_data {
-	struct clk_hw_onecell_data	clk_data;
 	void __iomem		*base;
+
+	/* Must be last */
+	struct clk_hw_onecell_data	clk_data;
 };
 
 struct stratix10_pll_clock {
Signature · 1 line
2.34.1
maintainer Applied
#2 lore raw

Re: [PATCH v3 1/2][next] clk: socfpga: Fix undefined behavior bug in struct stratix10_clock_data

Quoting Gustavo A. R. Silva (2023-10-23 20:30:52)

Quoted 72 lines

struct clk_hw_onecell_data is a flexible structure, which means that it contains flexible-array member at the bottom, in this case array hws:

include/linux/clk-provider.h:
1380 struct clk_hw_onecell_data {
1381 unsigned int num;
1382 struct clk_hw *hws[] __counted_by(num);
1383 };

This could potentially lead to an overwrite of the objects following clk_data in struct stratix10_clock_data, in this case void __iomem *base; at run-time:

drivers/clk/socfpga/stratix10-clk.h:
9 struct stratix10_clock_data {
10 struct clk_hw_onecell_data clk_data;
11 void __iomem *base;
12 };

There are currently three different places where memory is allocated for struct stratix10_clock_data, including the flex-array hws in struct clk_hw_onecell_data:

drivers/clk/socfpga/clk-agilex.c:
469 clk_data = devm_kzalloc(dev, struct_size(clk_data, clk_data.hws,
470 num_clks), GFP_KERNEL);

drivers/clk/socfpga/clk-agilex.c:
509 clk_data = devm_kzalloc(dev, struct_size(clk_data, clk_data.hws,
510 num_clks), GFP_KERNEL);

drivers/clk/socfpga/clk-s10.c:
400 clk_data = devm_kzalloc(dev, struct_size(clk_data, clk_data.hws,
401 num_clks), GFP_KERNEL);

I'll use just one of them to describe the issue. See below.

Notice that a total of 440 bytes are allocated for flexible-array member hws at line 469:

include/dt-bindings/clock/agilex-clock.h:
70 #define AGILEX_NUM_CLKS 55

drivers/clk/socfpga/clk-agilex.c:
459 struct stratix10_clock_data *clk_data;
460 void __iomem *base;
...
466
467 num_clks = AGILEX_NUM_CLKS;
468
469 clk_data = devm_kzalloc(dev, struct_size(clk_data, clk_data.hws,
470 num_clks), GFP_KERNEL);

struct_size(clk_data, clk_data.hws, num_clks) above translates to sizeof(struct stratix10_clock_data) + sizeof(struct clk_hw *) * 55 == 16 + 8 * 55 == 16 + 440

                    ^^^
                     |
        allocated bytes for flex-array `hws`

474 for (i = 0; i < num_clks; i++)
475 clk_data->clk_data.hws[i] = ERR_PTR(-ENOENT);
476
477 clk_data->base = base;

and then some data is written into both hws and base objects.

Fix this by placing the declaration of object clk_data at the end of struct stratix10_clock_data. Also, add a comment to make it clear that this object must always be last in the structure.

Quoted 8 lines

-Wflex-array-member-not-at-end is coming in GCC-14, and we are getting ready to enable it globally.

Fixes ba7e258425ac ("clk: socfpga: Convert to s10/agilex/n5x to use clk_hw")

---

Applied to clk-next

#3 lore raw

PATCH v3 2/2 clk: socfpga: agilex: Add bounds-checking coverage for struct stratix10_clock_data

In reply to Gustavo A. R. Silva

In order to gain the bounds-checking coverage that __counted_by provides to flexible-array members at run-time via CONFIG_UBSAN_BOUNDS (for array indexing) and CONFIG_FORTIFY_SOURCE (for strcpy/memcpy-family functions), we must make sure that the counter member, in this case num, is updated before the first access to the flex-array member, in this case array hws.

commit f316cdff8d67 ("clk: Annotate struct clk_hw_onecell_data with __counted_by") introduced __counted_by for struct clk_hw_onecell_data together with changes to relocate some of assignments of counter num before hws is accessed:

include/linux/clk-provider.h:
1380 struct clk_hw_onecell_data {
1381 unsigned int num;
1382 struct clk_hw *hws[] __counted_by(num);
1383 };

However, this structure is used as a member in other structs, in this case in struct sstratix10_clock_data:

drivers/clk/socfpga/stratix10-clk.h:
9 struct stratix10_clock_data {
10 void __iomem *base;
11
12 /* Must be last */
13 struct clk_hw_onecell_data clk_data;
14 };

Hence, we need to move the assignments to clk_data->clk_data.num after allocations for struct stratix10_clock_data and before accessing the flexible array clk_data->clk_data.hws. And, as assignments for both clk_data->clk_data.num and clk_data->base are originally adjacent to each other, relocate both assignments together.

---
Changes in v3:
- None, really. Just Cc [email protected] this time.

Changes in v2:
- None.
- Link: https://lore.kernel.org/linux-hardening/f531ae35ef434b62f48534c67ad3b1013bff0536.1697493574.git.gustavoars@kernel.org/

v1:
- Link: https://lore.kernel.org/linux-hardening/fd4cd8503316d536e1a84fa2ae5bdefdd4b24afe.1697059539.git.gustavoars@kernel.org/

 drivers/clk/socfpga/clk-agilex.c | 12 ++++++------
 drivers/clk/socfpga/clk-s10.c    |  6 +++---
 2 files changed, 9 insertions(+), 9 deletions(-)
drivers/clk/socfpga/clk-agilex.c +6 -6
index 6b65a74aefa6..8dd94f64756b 100644
@@ -471,12 +471,12 @@ static int agilex_clkmgr_init(struct platform_device *pdev)
 	if (!clk_data)
 		return -ENOMEM;
 
+	clk_data->clk_data.num = num_clks;
+	clk_data->base = base;
+
 	for (i = 0; i < num_clks; i++)
 		clk_data->clk_data.hws[i] = ERR_PTR(-ENOENT);
 
-	clk_data->base = base;
-	clk_data->clk_data.num = num_clks;
-
 	agilex_clk_register_pll(agilex_pll_clks, ARRAY_SIZE(agilex_pll_clks), clk_data);
 
 	agilex_clk_register_c_perip(agilex_main_perip_c_clks,
@@ -511,12 +511,12 @@ static int n5x_clkmgr_init(struct platform_device *pdev)
 	if (!clk_data)
 		return -ENOMEM;
 
-	for (i = 0; i < num_clks; i++)
-		clk_data->clk_data.hws[i] = ERR_PTR(-ENOENT);
-
 	clk_data->base = base;
 	clk_data->clk_data.num = num_clks;
 
+	for (i = 0; i < num_clks; i++)
+		clk_data->clk_data.hws[i] = ERR_PTR(-ENOENT);
+
 	n5x_clk_register_pll(agilex_pll_clks, ARRAY_SIZE(agilex_pll_clks), clk_data);
 
 	n5x_clk_register_c_perip(n5x_main_perip_c_clks,
drivers/clk/socfpga/clk-s10.c +3 -3
index 3752bd9c103c..b4bf4e2d38e1 100644
@@ -402,12 +402,12 @@ static int s10_clkmgr_init(struct platform_device *pdev)
 	if (!clk_data)
 		return -ENOMEM;
 
-	for (i = 0; i < num_clks; i++)
-		clk_data->clk_data.hws[i] = ERR_PTR(-ENOENT);
-
 	clk_data->base = base;
 	clk_data->clk_data.num = num_clks;
 
+	for (i = 0; i < num_clks; i++)
+		clk_data->clk_data.hws[i] = ERR_PTR(-ENOENT);
+
 	s10_clk_register_pll(s10_pll_clks, ARRAY_SIZE(s10_pll_clks), clk_data);
 
 	s10_clk_register_c_perip(s10_main_perip_c_clks,
Signature · 1 line
2.34.1
maintainer Applied
#4 lore raw

Re: [PATCH v3 2/2][next] clk: socfpga: agilex: Add bounds-checking coverage for struct stratix10_clock_data

Quoting Gustavo A. R. Silva (2023-10-23 20:31:42)

Quoted 34 lines

In order to gain the bounds-checking coverage that __counted_by provides to flexible-array members at run-time via CONFIG_UBSAN_BOUNDS (for array indexing) and CONFIG_FORTIFY_SOURCE (for strcpy/memcpy-family functions), we must make sure that the counter member, in this case num, is updated before the first access to the flex-array member, in this case array hws.

commit f316cdff8d67 ("clk: Annotate struct clk_hw_onecell_data with __counted_by") introduced __counted_by for struct clk_hw_onecell_data together with changes to relocate some of assignments of counter num before hws is accessed:

include/linux/clk-provider.h:
1380 struct clk_hw_onecell_data {
1381 unsigned int num;
1382 struct clk_hw *hws[] __counted_by(num);
1383 };

However, this structure is used as a member in other structs, in this case in struct sstratix10_clock_data:

drivers/clk/socfpga/stratix10-clk.h:
9 struct stratix10_clock_data {
10 void __iomem *base;
11
12 /* Must be last */
13 struct clk_hw_onecell_data clk_data;
14 };

Hence, we need to move the assignments to clk_data->clk_data.num after allocations for struct stratix10_clock_data and before accessing the flexible array clk_data->clk_data.hws. And, as assignments for both clk_data->clk_data.num and clk_data->base are originally adjacent to each other, relocate both assignments together.

Quoted 3 lines

---

Applied to clk-next