Skip to contentSkip to conversation

firmware: turris-mox-rwtm: fix building without CONFIG_KEYS

PATCH In review Arnd Bergmann 2 participants 4 messages Apr 4, 2025 – Apr 6, 2025 Arm/cz.nic turris support

lore
Share & tools
lore
#0 lore raw

PATCH 1/2 firmware: turris-mox-rwtm: fix building without CONFIG_KEYS

From: Arnd Bergmann <[email protected]>

"struct key" is defined conditionally, so the code referencing it must be made conditional as well:

In file included from drivers/firmware/turris-mox-rwtm.c:29:
include/linux/turris-signing-key.h: In function 'turris_signing_key_get_dev':
include/linux/turris-signing-key.h:26:19: error: invalid use of undefined type 'const struct key'
26 | return key->payload.data[1];
| ^~

---
 include/linux/turris-signing-key.h | 2 ++
 1 file changed, 2 insertions(+)
include/linux/turris-signing-key.h +2 -0
index 032ca8cbf636..8a435b73c3a9 100644
@@ -11,6 +11,7 @@
 
 struct device;
 
+#ifdef CONFIG_KEYS
 struct turris_signing_key_subtype {
 	u16 key_size;
 	u8 data_size;
@@ -29,5 +30,6 @@ static inline struct device *turris_signing_key_get_dev(const struct key *key)
 int
 devm_turris_signing_key_create(struct device *dev, const struct turris_signing_key_subtype *subtype,
 			       const char *desc);
+#endif
 
 #endif /* __TURRIS_SIGNING_KEY_H */
Signature · 1 line
2.39.5
#1 lore raw

PATCH 2/2 platform: turris: use ffs() instead of __bf_shf()

From: Arnd Bergmann <[email protected]>

__bf_shf() on a 64-bit variable causes a link failure during compile-testing:

drivers-platform-cznic-turris-omnia-mcu-gpio.c:(.text):undefined-reference-to-__ffsdi2

Open-code this using ffs()-1, which has the same result but avoids the library call.

---
 drivers/platform/cznic/turris-omnia-mcu-gpio.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/platform/cznic/turris-omnia-mcu-gpio.c +1 -1
index 932383f7491a..c2df24ea8686 100644
@@ -1104,7 +1104,7 @@ int omnia_mcu_request_irq(struct omnia_mcu *mcu, u32 spec,
 	if (!spec)
 		return -EINVAL;
 
-	irq_idx = omnia_int_to_gpio_idx[__bf_shf(spec)];
+	irq_idx = omnia_int_to_gpio_idx[ffs(spec) - 1];
 	irq = gpiod_to_irq(gpio_device_get_desc(mcu->gc.gpiodev, irq_idx));
 	if (irq < 0)
 		return irq;
Signature · 1 line
2.39.5
maintainer
#2 lore raw

Re: [PATCH 2/2] platform: turris: use ffs() instead of __bf_shf()

Quoted 20 lines · Arnd Bergmann

From: Arnd Bergmann <[email protected]>

__bf_shf() on a 64-bit variable causes a link failure during compile-testing:

drivers-platform-cznic-turris-omnia-mcu-gpio.c:(.text):undefined-reference-to-__ffsdi2

Open-code this using ffs()-1, which has the same result but avoids the library call.

---
 drivers/platform/cznic/turris-omnia-mcu-gpio.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/platform/cznic/turris-omnia-mcu-gpio.c +0 -0
index 932383f7491a..c2df24ea8686 100644

drivers/platform/cznic/turris-omnia-mcu-gpio.c int omnia_mcu_request_irq(struct omnia_mcu *mcu, u32 spec, · line 1107

Quoted 5 lines
drivers/platform/cznic/turris-omnia-mcu-gpio.c +1 -1
@@ -1104,7 +1104,7 @@ int omnia_mcu_request_irq(struct omnia_mcu *mcu, u32 spec,
 	if (!spec)
 		return -EINVAL;
 
-	irq_idx = omnia_int_to_gpio_idx[__bf_shf(spec)];
+	irq_idx = omnia_int_to_gpio_idx[ffs(spec) - 1];

Weird that __bf_shf() procudes this error on non-constant input. But I guess a variant of ffs() makes more sense semantically, since the spec arugment isn't semantically a bitfield.

Shouldn't we consider using __ffs(), though?

With __ffs() we can drop the "- 1", i.e.:

	irq_idx = omnia_int_to_gpio_idx[__ffs(spec)];

Both __ffs() and ffs() are used within the drivers/ subdir. Of the 649 usages of ffs(), 482 are ffs(x) - 1, so they could be converted to __ffs(x).

But I don't particularly care.

Marek