[PATCH] firmware_loader: use SHA-256 library API instead of crypto_shash API

[PATCH] firmware_loader: use SHA-256 library API instead of crypto_shash API

[Eric Biggers]

From: Eric Biggers <ebiggers@google.com>

This user of SHA-256 does not support any other algorithm, so the
crypto_shash abstraction provides no value.  Just use the SHA-256
library API instead, which is much simpler and easier to use.

Also take advantage of printk's built-in hex conversion using %*phN.

Signed-off-by: Eric Biggers <ebiggers@google.com>
---

This patch is targeting the firmware_loader tree for 6.16.

 drivers/base/firmware_loader/Kconfig |  4 +---
 drivers/base/firmware_loader/main.c  | 34 ++++------------------------
 2 files changed, 5 insertions(+), 33 deletions(-)

diff --git a/drivers/base/firmware_loader/Kconfig b/drivers/base/firmware_loader/Kconfig
index a037016742651..752b9a9bea038 100644
--- a/drivers/base/firmware_loader/Kconfig
+++ b/drivers/base/firmware_loader/Kconfig
@@ -1,12 +1,11 @@
 # SPDX-License-Identifier: GPL-2.0
 menu "Firmware loader"
 
 config FW_LOADER
 	tristate "Firmware loading facility" if EXPERT
-	select CRYPTO_HASH if FW_LOADER_DEBUG
-	select CRYPTO_SHA256 if FW_LOADER_DEBUG
+	select CRYPTO_LIB_SHA256 if FW_LOADER_DEBUG
 	default y
 	help
 	  This enables the firmware loading facility in the kernel. The kernel
 	  will first look for built-in firmware, if it has any. Next, it will
 	  look for the requested firmware in a series of filesystem paths:
@@ -26,11 +25,10 @@ config FW_LOADER
 	  You also want to be sure to enable this built-in if you are going to
 	  enable built-in firmware (CONFIG_EXTRA_FIRMWARE).
 
 config FW_LOADER_DEBUG
 	bool "Log filenames and checksums for loaded firmware"
-	depends on CRYPTO = FW_LOADER || CRYPTO=y
 	depends on DYNAMIC_DEBUG
 	depends on FW_LOADER
 	default FW_LOADER
 	help
 	  Select this option to use dynamic debug to log firmware filenames and
diff --git a/drivers/base/firmware_loader/main.c b/drivers/base/firmware_loader/main.c
index cb0912ea3e627..44486b2c71728 100644
--- a/drivers/base/firmware_loader/main.c
+++ b/drivers/base/firmware_loader/main.c
@@ -804,45 +804,19 @@ static void fw_abort_batch_reqs(struct firmware *fw)
 		fw_state_aborted(fw_priv);
 	mutex_unlock(&fw_lock);
 }
 
 #if defined(CONFIG_FW_LOADER_DEBUG)
-#include <crypto/hash.h>
 #include <crypto/sha2.h>
 
 static void fw_log_firmware_info(const struct firmware *fw, const char *name, struct device *device)
 {
-	struct shash_desc *shash;
-	struct crypto_shash *alg;
-	u8 *sha256buf;
-	char *outbuf;
+	u8 digest[SHA256_DIGEST_SIZE];
 
-	alg = crypto_alloc_shash("sha256", 0, 0);
-	if (IS_ERR(alg))
-		return;
-
-	sha256buf = kmalloc(SHA256_DIGEST_SIZE, GFP_KERNEL);
-	outbuf = kmalloc(SHA256_BLOCK_SIZE + 1, GFP_KERNEL);
-	shash = kmalloc(sizeof(*shash) + crypto_shash_descsize(alg), GFP_KERNEL);
-	if (!sha256buf || !outbuf || !shash)
-		goto out_free;
-
-	shash->tfm = alg;
-
-	if (crypto_shash_digest(shash, fw->data, fw->size, sha256buf) < 0)
-		goto out_free;
-
-	for (int i = 0; i < SHA256_DIGEST_SIZE; i++)
-		sprintf(&outbuf[i * 2], "%02x", sha256buf[i]);
-	outbuf[SHA256_BLOCK_SIZE] = 0;
-	dev_dbg(device, "Loaded FW: %s, sha256: %s\n", name, outbuf);
-
-out_free:
-	kfree(shash);
-	kfree(outbuf);
-	kfree(sha256buf);
-	crypto_free_shash(alg);
+	sha256(fw->data, fw->size, digest);
+	dev_dbg(device, "Loaded FW: %s, sha256: %*phN\n",
+		name, SHA256_DIGEST_SIZE, digest);
 }
 #else
 static void fw_log_firmware_info(const struct firmware *fw, const char *name,
 				 struct device *device)
 {}

base-commit: 33035b665157558254b3c21c3f049fd728e72368
-- 
2.49.0

Re: [PATCH] firmware_loader: use SHA-256 library API instead of crypto_shash API

[Greg Kroah-Hartman]

On Mon, Apr 28, 2025 at 12:09:09PM -0700, Eric Biggers wrote:
> From: Eric Biggers <ebiggers@google.com>
> 
> This user of SHA-256 does not support any other algorithm, so the
> crypto_shash abstraction provides no value.  Just use the SHA-256
> library API instead, which is much simpler and easier to use.
> 
> Also take advantage of printk's built-in hex conversion using %*phN.
> 
> Signed-off-by: Eric Biggers <ebiggers@google.com>
> ---
> 
> This patch is targeting the firmware_loader tree for 6.16.


Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Re: [PATCH] firmware_loader: use SHA-256 library API instead of crypto_shash API

[Eric Biggers]

On Tue, Apr 29, 2025 at 07:54:42AM +0200, Greg Kroah-Hartman wrote:
> On Mon, Apr 28, 2025 at 12:09:09PM -0700, Eric Biggers wrote:
> > From: Eric Biggers <ebiggers@google.com>
> > 
> > This user of SHA-256 does not support any other algorithm, so the
> > crypto_shash abstraction provides no value.  Just use the SHA-256
> > library API instead, which is much simpler and easier to use.
> > 
> > Also take advantage of printk's built-in hex conversion using %*phN.
> > 
> > Signed-off-by: Eric Biggers <ebiggers@google.com>
> > ---
> > 
> > This patch is targeting the firmware_loader tree for 6.16.
> 
> 
> Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Greg, it looks like you normally take patches to this file.  You can go ahead
and take it if you want.  Thanks!

- Eric

Re: [PATCH] firmware_loader: use SHA-256 library API instead of crypto_shash API

[Danilo Krummrich]

On Mon, Apr 28, 2025 at 12:09:09PM -0700, Eric Biggers wrote:
> From: Eric Biggers <ebiggers@google.com>
> 
> This user of SHA-256 does not support any other algorithm, so the
> crypto_shash abstraction provides no value.  Just use the SHA-256
> library API instead, which is much simpler and easier to use.
> 
> Also take advantage of printk's built-in hex conversion using %*phN.
> 
> Signed-off-by: Eric Biggers <ebiggers@google.com>

Applied to driver-core-testing, thanks!