[PATCH 1/3] kbuild: make sha256sum command available generally

[PATCH 1/3] kbuild: make sha256sum command available generally

[Ahmad Fatoum]

We currently use the command only for generation a SHA256 sum of the
barebox proper binary. In preparation for using it in othe Makefiles
as well, move it to a central location and use the occasion to give
it a short comment.

Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
---
 images/Makefile      |  7 -------
 scripts/Makefile.lib | 11 +++++++++++
 2 files changed, 11 insertions(+), 7 deletions(-)

diff --git a/images/Makefile b/images/Makefile
index 7a8bb94fe0df..218a24ff1ddd 100644
--- a/images/Makefile
+++ b/images/Makefile
@@ -112,13 +112,6 @@ $(obj)/piggy.o: $(obj)/barebox.z FORCE
 
 $(obj)/sha_sum.o: $(obj)/barebox.sha.bin FORCE
 
-quiet_cmd_sha256bin ?= SHA-BIN $@
-      cmd_sha256bin = printf "$(shell sed 's/ .*$$//;s/../0x&\n/g;s/\n$$//' $(obj)/barebox.sum | \
-			while read -r byte; do printf '\%o' $$byte; done)" > $@
-
-quiet_cmd_sha256sum ?= SHA     $@
-      cmd_sha256sum ?= sha256sum $(obj)/barebox.z > $@
-
 $(obj)/barebox.sha.bin: $(obj)/barebox.sum FORCE
 	$(call if_changed,sha256bin)
 
diff --git a/scripts/Makefile.lib b/scripts/Makefile.lib
index 46042dab3c8f..16308497b845 100644
--- a/scripts/Makefile.lib
+++ b/scripts/Makefile.lib
@@ -272,6 +272,17 @@ cmd_ld = $(LD) $(KBUILD_LDFLAGS) $(EXTRA_LDFLAGS) $(LDFLAGS_$(@F)) \
 quiet_cmd_objcopy = OBJCOPY $@
 cmd_objcopy = $(OBJCOPY) $(OBJCOPYFLAGS) $(OBJCOPYFLAGS_$(@F)) $< $@
 
+# Hashing
+# ---------------------------------------------------------------------------
+# POSIX printf (e.g. dash's) doesn't support \xHH, but octal sequences are fine
+
+quiet_cmd_sha256bin ?= SHA-BIN $@
+      cmd_sha256bin = printf "$(shell sed 's/ .*$$//;s/../0x&\n/g;s/\n$$//' $< | \
+			while read -r byte; do printf '\%o' $$byte; done)" > $@
+
+quiet_cmd_sha256sum ?= SHA     $@
+      cmd_sha256sum ?= sha256sum $< > $@
+
 # Decompressor for barebox proper binary when using PBL
 # ---------------------------------------------------------------------------
 
-- 
2.30.2

[PATCH 2/3] pbl: export pbl_barebox_verify

[Ahmad Fatoum]

There's no downside to always build the digest verification code in PBL
and export pbl_barebox_verify to access it. This allows board code to
use the function for verifying other firmware blobs and
CONFIG_PBL_VERIFY_PIGGY=y will remain to enable the verification at
barebox proper extraction time. Code not using it will have the function
sections garbage collected by the linker, so no functional change.

Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
---
 crypto/Makefile | 3 +--
 include/pbl.h   | 3 +++
 pbl/decomp.c    | 6 +++---
 3 files changed, 7 insertions(+), 5 deletions(-)

diff --git a/crypto/Makefile b/crypto/Makefile
index 762d7e543be4..22035d4f69ee 100644
--- a/crypto/Makefile
+++ b/crypto/Makefile
@@ -10,8 +10,7 @@ obj-$(CONFIG_DIGEST_MD5_GENERIC)	+= md5.o
 obj-$(CONFIG_DIGEST_SHA1_GENERIC)	+= sha1.o
 obj-$(CONFIG_DIGEST_SHA224_GENERIC)	+= sha2.o
 obj-$(CONFIG_DIGEST_SHA256_GENERIC)	+= sha2.o
-pbl-$(CONFIG_PBL_VERIFY_PIGGY)		+= sha2.o
-pbl-$(CONFIG_PBL_VERIFY_PIGGY)		+= digest.o
+pbl-y					+= sha2.o digest.o
 obj-$(CONFIG_DIGEST_SHA384_GENERIC)	+= sha4.o
 obj-$(CONFIG_DIGEST_SHA512_GENERIC)	+= sha4.o
 obj-y	+= memneq.o
diff --git a/include/pbl.h b/include/pbl.h
index 0dc23c72dcf5..7dc4fa309257 100644
--- a/include/pbl.h
+++ b/include/pbl.h
@@ -30,4 +30,7 @@ const void *
 fdt_device_get_match_data(const void *fdt, const char *nodepath,
 			  const struct fdt_device_id ids[]);
 
+int pbl_barebox_verify(const void *compressed_start, unsigned int len,
+		       const void *hash, unsigned int hash_len);
+
 #endif /* __PBL_H__ */
diff --git a/pbl/decomp.c b/pbl/decomp.c
index 1e0ef81ada00..553895bac5e8 100644
--- a/pbl/decomp.c
+++ b/pbl/decomp.c
@@ -54,14 +54,14 @@ static void noinline errorfn(char *error)
 extern unsigned char sha_sum[];
 extern unsigned char sha_sum_end[];
 
-static int pbl_barebox_verify(void *compressed_start, unsigned int len, void *hash,
-			      unsigned int hash_len)
+int pbl_barebox_verify(const void *compressed_start, unsigned int len,
+		       const void *hash, unsigned int hash_len)
 {
 	struct sha256_state sha_state = { 0 };
 	struct digest d = { .ctx = &sha_state };
 	char computed_hash[SHA256_DIGEST_SIZE];
 	int i;
-	char *char_hash = hash;
+	const char *char_hash = hash;
 
 	if (hash_len != SHA256_DIGEST_SIZE)
 		return -1;
-- 
2.30.2

[PATCH 3/3] firmware: add external firmware PBL support

[Ahmad Fatoum]

Normally, barebox embds firmware into the binary referencing it, which
means that device tree blobs, RAM training code and e.g. TF-A for i.MX8M
end up in the prebootloader, while, e.g. Freescale FMan microcode ends
up in barebox proper. The only exception so far was barebox proper:
When only the PBL fits in on-chip SRAM, barebox proper is chainloaded
from the boot medium. To avoid TOCTOU attack, it's read fully into DRAM
after setup and then a SHA256 is calculated and compared against the
hash embedded in barebox PBL, which in a secure boot system would be
trusted by virtue of the PBL as a whole being verified beforehand by
the BootROM.

Reuse this mechanism to support arbitrary firmware, which is now termed
external firmware. Such firmware is placed beyond the piggydata (barebox
proper) and only offset and hash are included in the prebootloader
image. The new get_builtin_firmware_ext() is used to retrieve this
external firmware after integrity verification with SHA256.

This enables referencing firmware blobs from PBL that would bloat the
size of the PBL beyond what can fit into on-chip SRAM, e.g. very big
OP-TEE binaries. As users of get_builtin_firmware() didn't have to worry
about TOCTOU so far, we panic when a firmware verification fails to
ensure that we never load an OP-TEE that has been modified in-transit
in transit. We can't include the OP-TEE binary in barebox proper,
because we need to install it in EL3, but barebox proper on the i.MX8M
runs as BL33 in a lower exception level.

Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
---
 arch/arm/lib/pbl.lds.S | 11 +++++++++++
 firmware/Makefile      | 37 ++++++++++++++++++++++++++++++++++++-
 include/firmware.h     | 31 +++++++++++++++++++++++++++++++
 3 files changed, 78 insertions(+), 1 deletion(-)

diff --git a/arch/arm/lib/pbl.lds.S b/arch/arm/lib/pbl.lds.S
index e77b3220fcb0..87d9b297f0dc 100644
--- a/arch/arm/lib/pbl.lds.S
+++ b/arch/arm/lib/pbl.lds.S
@@ -62,6 +62,10 @@ SECTIONS
 	}
 	__shasum_end = .;
 
+	.pblext_shasum : {
+		KEEP(*(.pblext_shasum.*))
+	}
+
 	.rel_dyn_start : { *(.__rel_dyn_start) }
 #ifdef CONFIG_CPU_32
 	.rel.dyn : { *(.rel*) }
@@ -101,6 +105,13 @@ SECTIONS
 	}
 	__piggydata_end = .;
 
+	. = ALIGN(4);
+	__pblext_start = .;
+	.pblext : {
+		*(.pblext.*)
+	}
+	__pblext_end = .;
+
 	.image_end : { KEEP(*(.__image_end)) }
 
 	pbl_image_size =  . - BASE;
diff --git a/firmware/Makefile b/firmware/Makefile
index 87bd033f6e5c..eb8b59a3b1aa 100644
--- a/firmware/Makefile
+++ b/firmware/Makefile
@@ -51,16 +51,51 @@ filechk_fwbin = { \
 $(obj)/%.gen.S: FORCE
 	$(call filechk,fwbin)
 
+FWEXTNAME    = $(patsubst $(obj)/%.extgen.S,%,$@)
+FWEXTSTR     = $(subst /,_,$(subst .,_,$(subst -,_,$(FWEXTNAME))))
+
+filechk_fwextbin = { \
+	echo "/* Generated by $(src)/Makefile */"		;\
+	echo "    .section .pblext.$(FWEXTSTR)",\"a\"		;\
+	echo "    .p2align $(ASM_ALIGN)"			;\
+	echo ".global _fwext_$(FWEXTSTR)_start"			;\
+	echo "_fwext_$(FWEXTSTR)_start:"			;\
+	echo "    .incbin \"$(fwdir)/$(FWEXTNAME)\""		;\
+	echo ".global _fwext_$(FWEXTSTR)_end"			;\
+	echo "_fwext_$(FWEXTSTR)_end:"				;\
+	echo "    .section .pblext_shasum.$(FWEXTSTR), \"a\""	;\
+	echo "    .p2align $(ASM_ALIGN)"			;\
+	echo ".global _fwext_$(FWEXTSTR)_sha_start"		;\
+	echo "_fwext_$(FWEXTSTR)_sha_start:"			;\
+	echo "    .incbin \"$(objdir)/firmware/$(FWEXTNAME).sha.bin\""	;\
+	echo ".global _fwext_$(FWEXTSTR)_sha_end"		;\
+	echo "_fwext_$(FWEXTSTR)_sha_end:"			;\
+}
+
+$(obj)/%.extgen.S: $(obj)/%.sha.bin FORCE
+	$(call filechk,fwextbin)
+
+$(obj)/%.sha.bin: $(obj)/%.sum FORCE
+	$(call if_changed,sha256bin)
+
+$(obj)/%.sum: $(obj)/% FORCE
+	$(call if_changed,sha256sum)
+
+clean-files += *.sha.bin *.sum
+
 # The .o files depend on the binaries directly; the .S files don't.
 $(patsubst %,$(obj)/%.gen.o, $(obj-pbl-y)): $(obj)/%.gen.o: $(fwdir)/%
 
 # The same for pbl:
 $(patsubst %,$(obj)/%.gen.pbl.o, $(obj-pbl-y)): $(obj)/%.gen.pbl.o: $(fwdir)/%
+$(patsubst %,$(obj)/%.extgen.pbl.o, $(pbl-y)): $(obj)/%.extgen.pbl.o: $(fwdir)/%
 
-obj-pbl-y			 += $(patsubst %,%.gen.o, $(fw-external-y))
+pbl-y := $(addsuffix .extgen.o, $(fw-external-y))
 
 targets := $(patsubst $(obj)/%,%, \
                                 $(shell find $(obj) -name \*.gen.S 2>/dev/null))
+targets += $(patsubst $(obj)/%,%, \
+                                $(shell find $(obj) -name \*.extgen.S 2>/dev/null))
 
 # just to build a built-in.o. Otherwise compilation fails when no
 # firmware is built.
diff --git a/include/firmware.h b/include/firmware.h
index 2583342230ae..b96726bd4416 100644
--- a/include/firmware.h
+++ b/include/firmware.h
@@ -6,8 +6,11 @@
 #ifndef FIRMWARE_H
 #define FIRMWARE_H
 
+#include <pbl.h>
 #include <types.h>
 #include <driver.h>
+#include <debug_ll.h>
+#include <linux/kernel.h>
 
 struct firmware_handler {
 	char *id; /* unique identifier for this firmware device */
@@ -57,6 +60,16 @@ static inline void firmware_set_searchpath(const char *path)
 
 void firmwaremgr_list_handlers(void);
 
+static inline void firmware_ext_verify(const void *data_start, size_t data_size,
+				       const void *hash_start, size_t hash_size)
+{
+	if (pbl_barebox_verify(data_start, data_size,
+			       hash_start, hash_size) != 0) {
+		putc_ll('!');
+		panic("hash mismatch, refusing to decompress");
+	}
+}
+
 #define get_builtin_firmware(name, start, size) \
 	{							\
 		extern char _fw_##name##_start[];		\
@@ -65,4 +78,22 @@ void firmwaremgr_list_handlers(void);
 		*size = _fw_##name##_end - _fw_##name##_start;	\
 	}
 
+#define get_builtin_firmware_ext(name, base, start, size)		\
+	{								\
+		extern char _fwext_##name##_start[];			\
+		extern char _fwext_##name##_end[];			\
+		extern char _fwext_##name##_sha_start[];		\
+		extern char _fwext_##name##_sha_end[];			\
+		long offset = (long)bl33 - (long)_text;			\
+		*start = (typeof(*start)) _fwext_##name##_start;	\
+		*size = _fwext_##name##_end - _fwext_##name##_start;	\
+		*start += offset;					\
+		firmware_ext_verify(					\
+			*start, *size,					\
+			_fwext_##name##_sha_start,			\
+			_fwext_##name##_sha_end -			\
+			_fwext_##name##_sha_start			\
+		);							\
+	}
+
 #endif /* FIRMWARE_H */
-- 
2.30.2

[PATCH 3/4] pbl: replace __piggydata_end with __image_end

[Ahmad Fatoum]

__piggydata_end and __image_end used to be synonyms before the addition
of external firmware. Now that external firmware is located after
__piggydata_end, code using it needs to be revisited.

There's no reason to have code reference __piggydata_end. Either they
want all the rest of the image, so they should use __image_end instead
or they want just the piggy data, in which case they can read the data
size embedded into the piggydata itself.

Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
---
 arch/arm/cpu/setupc_64.S       | 2 +-
 arch/arm/lib/pbl.lds.S         | 1 -
 arch/arm/mach-imx/romapi.c     | 2 +-
 arch/mips/lib/pbl.lds.S        | 3 ++-
 arch/riscv/lib/pbl.lds.S       | 2 --
 include/asm-generic/sections.h | 1 -
 6 files changed, 4 insertions(+), 7 deletions(-)

diff --git a/arch/arm/cpu/setupc_64.S b/arch/arm/cpu/setupc_64.S
index b5f4a643fa42..d64281c148fc 100644
--- a/arch/arm/cpu/setupc_64.S
+++ b/arch/arm/cpu/setupc_64.S
@@ -29,7 +29,7 @@ ENDPROC(setup_c)
 					/* x0: target address */
 #ifdef __PBL__
 ENTRY(relocate_to_adr_full)
-	ldr	x2, =__piggydata_end
+	ldr	x2, =__image_end
 	b	1f
 #endif
 
diff --git a/arch/arm/lib/pbl.lds.S b/arch/arm/lib/pbl.lds.S
index 87d9b297f0dc..4c1057bbce35 100644
--- a/arch/arm/lib/pbl.lds.S
+++ b/arch/arm/lib/pbl.lds.S
@@ -103,7 +103,6 @@ SECTIONS
 	.piggydata : {
 		*(.piggydata)
 	}
-	__piggydata_end = .;
 
 	. = ALIGN(4);
 	__pblext_start = .;
diff --git a/arch/arm/mach-imx/romapi.c b/arch/arm/mach-imx/romapi.c
index f7d421d73757..5d00d71154d7 100644
--- a/arch/arm/mach-imx/romapi.c
+++ b/arch/arm/mach-imx/romapi.c
@@ -28,7 +28,7 @@ static int imx8m_bootrom_load_image(struct rom_api *rom_api)
 {
 	return imx8m_bootrom_load(rom_api,
 				  (void *)MX8M_ATF_BL33_BASE_ADDR + barebox_pbl_size,
-				  __piggydata_end - __piggydata_start);
+				  __image_end - __piggydata_start);
 }
 
 int imx8mp_bootrom_load_image(void)
diff --git a/arch/mips/lib/pbl.lds.S b/arch/mips/lib/pbl.lds.S
index 75069b0c50d6..413f24b9ab05 100644
--- a/arch/mips/lib/pbl.lds.S
+++ b/arch/mips/lib/pbl.lds.S
@@ -48,7 +48,8 @@ SECTIONS
 	.piggydata : {
 		*(.piggydata)
 	}
-	__piggydata_end = . - BASE;
+
+	.image_end : { KEEP(*(.__image_end)) }
 
 	pbl_image_size =  .;
 
diff --git a/arch/riscv/lib/pbl.lds.S b/arch/riscv/lib/pbl.lds.S
index e238b2bfd34e..ccf64fc6d3aa 100644
--- a/arch/riscv/lib/pbl.lds.S
+++ b/arch/riscv/lib/pbl.lds.S
@@ -74,8 +74,6 @@ SECTIONS
 	.piggydata : {
 		*(.piggydata)
 	}
-	__piggydata_end = .;
-
 	.image_end : { KEEP(*(.__image_end)) }
 
 	pbl_image_size =  .;
diff --git a/include/asm-generic/sections.h b/include/asm-generic/sections.h
index e54123de0ea7..0b2ed5615bd6 100644
--- a/include/asm-generic/sections.h
+++ b/include/asm-generic/sections.h
@@ -11,7 +11,6 @@ extern char _end[];
 extern char __image_start[];
 extern char __image_end[];
 extern char __piggydata_start[];
-extern char __piggydata_end[];
 extern void *_barebox_image_size;
 extern void *_barebox_bare_init_size;
 extern void *_barebox_pbl_size;
-- 
2.30.2

Re: [PATCH 3/3] firmware: add external firmware PBL support

[Sascha Hauer]

On Mon, Aug 15, 2022 at 03:09:55PM +0200, Ahmad Fatoum wrote:
> Normally, barebox embds firmware into the binary referencing it, which
> means that device tree blobs, RAM training code and e.g. TF-A for i.MX8M
> end up in the prebootloader, while, e.g. Freescale FMan microcode ends
> up in barebox proper. The only exception so far was barebox proper:
> When only the PBL fits in on-chip SRAM, barebox proper is chainloaded
> from the boot medium. To avoid TOCTOU attack, it's read fully into DRAM
> after setup and then a SHA256 is calculated and compared against the
> hash embedded in barebox PBL, which in a secure boot system would be
> trusted by virtue of the PBL as a whole being verified beforehand by
> the BootROM.
> 
> Reuse this mechanism to support arbitrary firmware, which is now termed
> external firmware. Such firmware is placed beyond the piggydata (barebox
> proper) and only offset and hash are included in the prebootloader
> image. The new get_builtin_firmware_ext() is used to retrieve this
> external firmware after integrity verification with SHA256.

Does it make sense to use this mechanism for barebox proper as well?

>  #define get_builtin_firmware(name, start, size) \
>  	{							\
>  		extern char _fw_##name##_start[];		\
> @@ -65,4 +78,22 @@ void firmwaremgr_list_handlers(void);
>  		*size = _fw_##name##_end - _fw_##name##_start;	\
>  	}
>  
> +#define get_builtin_firmware_ext(name, base, start, size)		\

base is unused.

It would be nice if users could use the same macro for both internal and
external firmware, but I have no idea how and if this could be
implemented.

Sascha

-- 
Pengutronix e.K.                           |                             |
Steuerwalder Str. 21                       | http://www.pengutronix.de/  |
31137 Hildesheim, Germany                  | Phone: +49-5121-206917-0    |
Amtsgericht Hildesheim, HRA 2686           | Fax:   +49-5121-206917-5555 |

Re: [PATCH 3/3] firmware: add external firmware PBL support

[Ahmad Fatoum]

On 16.08.22 10:42, Sascha Hauer wrote:
> On Mon, Aug 15, 2022 at 03:09:55PM +0200, Ahmad Fatoum wrote:
>> Normally, barebox embds firmware into the binary referencing it, which
>> means that device tree blobs, RAM training code and e.g. TF-A for i.MX8M
>> end up in the prebootloader, while, e.g. Freescale FMan microcode ends
>> up in barebox proper. The only exception so far was barebox proper:
>> When only the PBL fits in on-chip SRAM, barebox proper is chainloaded
>> from the boot medium. To avoid TOCTOU attack, it's read fully into DRAM
>> after setup and then a SHA256 is calculated and compared against the
>> hash embedded in barebox PBL, which in a secure boot system would be
>> trusted by virtue of the PBL as a whole being verified beforehand by
>> the BootROM.
>>
>> Reuse this mechanism to support arbitrary firmware, which is now termed
>> external firmware. Such firmware is placed beyond the piggydata (barebox
>> proper) and only offset and hash are included in the prebootloader
>> image. The new get_builtin_firmware_ext() is used to retrieve this
>> external firmware after integrity verification with SHA256.
> 
> Does it make sense to use this mechanism for barebox proper as well?

I'd rather we leave this as future exercise..

> 
>>  #define get_builtin_firmware(name, start, size) \
>>  	{							\
>>  		extern char _fw_##name##_start[];		\
>> @@ -65,4 +78,22 @@ void firmwaremgr_list_handlers(void);
>>  		*size = _fw_##name##_end - _fw_##name##_start;	\
>>  	}
>>  
>> +#define get_builtin_firmware_ext(name, base, start, size)		\
> 
> base is unused.

Ouch. It's called bl33 in the macro. Will fix.

> It would be nice if users could use the same macro for both internal and
> external firmware, but I have no idea how and if this could be
> implemented.

We can unconditionally emit _fw_*_sha_{start,end} symbols.
If they happen to be equal, we skip the SHA verification.
The problem however is that we need to know the offset
between the currently running binary and the chainloaded one
to arrive at the address of external firmware. We can use 0 for
that (i.e. base = _text) to have internal firmware a special case
of external firmware, but for external firmware, we must specify
explicit an offset. So we can't avoid here.

Cheers,
Ahmad

> 
> Sascha
> 


-- 
Pengutronix e.K.                           |                             |
Steuerwalder Str. 21                       | http://www.pengutronix.de/  |
31137 Hildesheim, Germany                  | Phone: +49-5121-206917-0    |
Amtsgericht Hildesheim, HRA 2686           | Fax:   +49-5121-206917-5555 |