Skip to contentSkip to conversation

spl: implement CRC check on U-Boot uImage

PATCH Superseded Simon Goldschmidt 3 participants 12 messages Nov 28, 2018 – Feb 10, 2019

v2 of 3 Superseded Latest is v3 →

lore
Share & tools
lore
#0 lore raw

PATCH v2 1/1 spl: implement CRC check on U-Boot uImage

SPL currently does not check uImage CRCs when loading U-Boot.

This patch adds checking the uImage CRC when SPL loads U-Boot. It does this by reusing the existing config option SPL_CRC32_SUPPORT to allow leaving out the CRC check on boards where the additional code size or boot time is a problem (adding the CRC check currently adds ~1.4 kByte to flash).

The SPL_CRC32_SUPPORT config option now gets enabled by default if SPL support for legacy images is enabled to check the CRC on all boards that don't actively take countermeasures.

---

Changes in v2:
- added Kconfig option SPL_LEGACY_IMAGE_CRC_CHECK to enable/disable checking CRC on legacy images

 common/spl/Kconfig | 21 +++++++++++++++------
 common/spl/spl.c   | 30 +++++++++++++++++++++++++++++-
 include/spl.h      |  5 +++++
 3 files changed, 49 insertions(+), 7 deletions(-)
common/spl/Kconfig +15 -6
index d0564621d4..5c6c437ef0 100644
@@ -80,6 +80,15 @@ config SPL_LEGACY_IMAGE_SUPPORT
 	  is y. If this is not set, SPL will move on to other available
 	  boot media to find a suitable image.
 
+config SPL_LEGACY_IMAGE_CRC_CHECK
+	bool "Check CRC of Legacy images"
+	depends on SPL_LEGACY_IMAGE_SUPPORT
+	default y
+	select SPL_CRC32_SUPPORT
+	help
+	  Enable this to check the CRC of Legacy images. While this increases
+	  reliability, it affects both code size and boot duration.
+
 config SPL_SYS_MALLOC_SIMPLE
 	bool
 	prompt "Only use malloc_simple functions in the SPL"
@@ -207,13 +216,13 @@ config SYS_MMCSD_RAW_MODE_U_BOOT_PARTITION_TYPE
 
 config SPL_CRC32_SUPPORT
 	bool "Support CRC32"
-	depends on SPL_FIT
+	default y if SPL_LEGACY_IMAGE_SUPPORT
 	help
-	  Enable this to support CRC32 in FIT images within SPL. This is a
-	  32-bit checksum value that can be used to verify images. This is
-	  the least secure type of checksum, suitable for detected
-	  accidental image corruption. For secure applications you should
-	  consider SHA1 or SHA256.
+	  Enable this to support CRC32 in uImages or FIT images within SPL.
+	  This is a 32-bit checksum value that can be used to verify images.
+	  For FIT images, this is the least secure type of checksum, suitable
+	  for detected accidental image corruption. For secure applications you
+	  should consider SHA1 or SHA256.
 
 config SPL_MD5_SUPPORT
 	bool "Support MD5"
common/spl/spl.c +29 -1
index 12f9359c0a..f2ea87a0ec 100644
@@ -236,6 +236,14 @@ int spl_parse_image_header(struct spl_image_info *spl_image,
 #ifdef CONFIG_SPL_LEGACY_IMAGE_SUPPORT
 		u32 header_size = sizeof(struct image_header);
 
+		if (CONFIG_IS_ENABLED(LEGACY_IMAGE_CRC_CHECK)) {
+			/* check uImage header CRC */
+			if (!image_check_hcrc(header)) {
+				puts("SPL: Image header CRC check failed!\n");
+				return -EINVAL;
+			}
+		}
+
 		if (spl_image->flags & SPL_COPY_PAYLOAD_ONLY) {
 			/*
 			 * On some system (e.g. powerpc), the load-address and
@@ -253,6 +261,13 @@ int spl_parse_image_header(struct spl_image_info *spl_image,
 			spl_image->size = image_get_data_size(header) +
 				header_size;
 		}
+#ifdef CONFIG_SPL_LEGACY_IMAGE_CRC_CHECK
+		/* store uImage data length and CRC to check later */
+		spl_image->dcrc_data = image_get_load(header);
+		spl_image->dcrc_length = image_get_data_size(header);
+		spl_image->dcrc = image_get_dcrc(header);
+#endif
+
 		spl_image->os = image_get_os(header);
 		spl_image->name = image_get_name(header);
 		debug("spl: payload image: %32s load addr: 0x%lx size: %d\n",
@@ -424,12 +439,25 @@ static struct spl_image_loader *spl_ll_find_loader(uint boot_device)
 static int spl_load_image(struct spl_image_info *spl_image,
 			  struct spl_image_loader *loader)
 {
+	int ret;
 	struct spl_boot_device bootdev;
 
 	bootdev.boot_device = loader->boot_device;
 	bootdev.boot_device_name = NULL;
 
-	return loader->load_image(spl_image, &bootdev);
+	ret = loader->load_image(spl_image, &bootdev);
+#ifdef CONFIG_SPL_LEGACY_IMAGE_CRC_CHECK
+	if (!ret && spl_image->dcrc_length) {
+		/* check data crc */
+		ulong dcrc = crc32_wd(0, (unsigned char *)spl_image->dcrc_data,
+				      spl_image->dcrc_length, CHUNKSZ_CRC32);
+		if (dcrc != spl_image->dcrc) {
+			puts("SPL: Image data CRC check failed!\n");
+			ret = -EINVAL;
+		}
+	}
+#endif
+	return ret;
 }
 
 /**
include/spl.h +5 -0
index 9a439f468b..1b1d9ebbfb 100644
@@ -33,6 +33,11 @@ struct spl_image_info {
 	u32 size;
 	u32 flags;
 	void *arg;
+#if CONFIG_IS_ENABLED(LEGACY_IMAGE_CRC_CHECK)
+	ulong dcrc_data;
+	ulong dcrc_length;
+	ulong dcrc;
+#endif
 };
 
 /*
Signature · 1 line
2.17.1
#1 lore raw

PATCH v2 1/1 spl: implement CRC check on U-Boot uImage

Quoted 16 lines · Simon Goldschmidt

SPL currently does not check uImage CRCs when loading U-Boot.

This patch adds checking the uImage CRC when SPL loads U-Boot. It does this by reusing the existing config option SPL_CRC32_SUPPORT to allow leaving out the CRC check on boards where the additional code size or boot time is a problem (adding the CRC check currently adds ~1.4 kByte to flash).

The SPL_CRC32_SUPPORT config option now gets enabled by default if SPL support for legacy images is enabled to check the CRC on all boards that don't actively take countermeasures.

---

Quoted 8 lines

Changes in v2:
- added Kconfig option SPL_LEGACY_IMAGE_CRC_CHECK to enable/disable checking CRC on legacy images

 common/spl/Kconfig | 21 +++++++++++++++------
 common/spl/spl.c   | 30 +++++++++++++++++++++++++++++-
 include/spl.h      |  5 +++++
 3 files changed, 49 insertions(+), 7 deletions(-)
#2 lore raw

PATCH v2 1/1 spl: implement CRC check on U-Boot uImage

Tom,

Am 11.12.2018 um 02:07 schrieb Simon Glass:

Quoted 22 lines
Quoted 20 lines · Simon Goldschmidt

SPL currently does not check uImage CRCs when loading U-Boot.

This patch adds checking the uImage CRC when SPL loads U-Boot. It does this by reusing the existing config option SPL_CRC32_SUPPORT to allow leaving out the CRC check on boards where the additional code size or boot time is a problem (adding the CRC check currently adds ~1.4 kByte to flash).

The SPL_CRC32_SUPPORT config option now gets enabled by default if SPL support for legacy images is enabled to check the CRC on all boards that don't actively take countermeasures.

---

Changes in v2:
- added Kconfig option SPL_LEGACY_IMAGE_CRC_CHECK to enable/disable checking CRC on legacy images

Quoted 6 lines
Quoted 4 lines
  common/spl/Kconfig | 21 +++++++++++++++------
  common/spl/spl.c   | 30 +++++++++++++++++++++++++++++-
  include/spl.h      |  5 +++++
  3 files changed, 49 insertions(+), 7 deletions(-)

This is assigned to you in patchwork, any reason you did not add it for the 2019.01 release? Is there anything missing or was this just lost?

Thanks,
Simon

#3 lore raw

PATCH v2 1/1 spl: implement CRC check on U-Boot uImage

Quoted 30 lines · Simon Goldschmidt

Tom,

Am 11.12.2018 um 02:07 schrieb Simon Glass:

Quoted 27 lines
Quoted 24 lines · Simon Goldschmidt

SPL currently does not check uImage CRCs when loading U-Boot.

This patch adds checking the uImage CRC when SPL loads U-Boot. It does this by reusing the existing config option SPL_CRC32_SUPPORT to allow leaving out the CRC check on boards where the additional code size or boot time is a problem (adding the CRC check currently adds ~1.4 kByte to flash).

The SPL_CRC32_SUPPORT config option now gets enabled by default if SPL support for legacy images is enabled to check the CRC on all boards that don't actively take countermeasures.

---

Changes in v2:
- added Kconfig option SPL_LEGACY_IMAGE_CRC_CHECK to enable/disable checking CRC on legacy images

 common/spl/Kconfig | 21 +++++++++++++++------
 common/spl/spl.c   | 30 +++++++++++++++++++++++++++++-
 include/spl.h      |  5 +++++
 3 files changed, 49 insertions(+), 7 deletions(-)
Quoted 4 lines
Quoted 1 line

This is assigned to you in patchwork, any reason you did not add it for the 2019.01 release? Is there anything missing or was this just lost?

I think I might have filed this off mentally as "Did we still want to do this?", so I'll pick it up for the next release, thanks!

Signature · 8 lines
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: not available
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20190114/bcdf3c72/attachment.sig>
#4 lore raw

PATCH v2 1/1 spl: implement CRC check on U-Boot uImage

Am 14.01.2019 um 17:27 schrieb Tom Rini:

Quoted 31 lines · Simon Goldschmidt
Quoted 30 lines

Tom,

Am 11.12.2018 um 02:07 schrieb Simon Glass:

Quoted 27 lines
Quoted 24 lines · Simon Goldschmidt

SPL currently does not check uImage CRCs when loading U-Boot.

This patch adds checking the uImage CRC when SPL loads U-Boot. It does this by reusing the existing config option SPL_CRC32_SUPPORT to allow leaving out the CRC check on boards where the additional code size or boot time is a problem (adding the CRC check currently adds ~1.4 kByte to flash).

The SPL_CRC32_SUPPORT config option now gets enabled by default if SPL support for legacy images is enabled to check the CRC on all boards that don't actively take countermeasures.

---

Changes in v2:
- added Kconfig option SPL_LEGACY_IMAGE_CRC_CHECK to enable/disable checking CRC on legacy images

  common/spl/Kconfig | 21 +++++++++++++++------
  common/spl/spl.c   | 30 +++++++++++++++++++++++++++++-
  include/spl.h      |  5 +++++
  3 files changed, 49 insertions(+), 7 deletions(-)
Quoted 7 lines
Quoted 4 lines
Quoted 1 line

This is assigned to you in patchwork, any reason you did not add it for the 2019.01 release? Is there anything missing or was this just lost?

I think I might have filed this off mentally as "Did we still want to do this?", so I'll pick it up for the next release, thanks!

Ping?

Regards,
Simon

#5 lore raw

PATCH v2 1/1 spl: implement CRC check on U-Boot uImage

In reply to Simon Goldschmidt
Quoted 8 lines · Simon Goldschmidt

SPL currently does not check uImage CRCs when loading U-Boot.

This patch adds checking the uImage CRC when SPL loads U-Boot. It does this by reusing the existing config option SPL_CRC32_SUPPORT to allow leaving out the CRC check on boards where the additional code size or boot time is a problem (adding the CRC check currently adds ~1.4 kByte to flash).

Quoted 6 lines

The SPL_CRC32_SUPPORT config option now gets enabled by default if SPL support for legacy images is enabled to check the CRC on all boards that don't actively take countermeasures.

Really sorry for the delay on this, especially as I've found one or two problems. The first problem is that with this vyasa-rk3288 and a few others fail to build due to a number of errors such as:
../common/spl/spl.c:269:12: error: 'struct spl_image_info' has no member named 'dcrc_data'

Second, I believe this is causing a number of platforms with very tight SPL constraints, namely all of 64bit sunxi, to fail to link as they overflow SRAM now. This can be fixed at least by making the new behavior opt-in, but I would fix the vyasa-rk3288 problem first. Thanks in advance!

Signature · 8 lines
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: not available
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20190208/c158814c/attachment.sig>
#6 lore raw

PATCH v2 1/1 spl: implement CRC check on U-Boot uImage

Quoted 15 lines · Tom Rini
Quoted 12 lines · Simon Goldschmidt

SPL currently does not check uImage CRCs when loading U-Boot.

This patch adds checking the uImage CRC when SPL loads U-Boot. It does this by reusing the existing config option SPL_CRC32_SUPPORT to allow leaving out the CRC check on boards where the additional code size or boot time is a problem (adding the CRC check currently adds ~1.4 kByte to flash).

The SPL_CRC32_SUPPORT config option now gets enabled by default if SPL support for legacy images is enabled to check the CRC on all boards that don't actively take countermeasures.

Quoted 8 lines
Quoted 2 lines

Really sorry for the delay on this, especially as I've found one or two problems. The first problem is that with this vyasa-rk3288 and a few others fail to build due to a number of errors such as:
../common/spl/spl.c:269:12: error: 'struct spl_image_info' has no member named 'dcrc_data'

Hmm, let me check what's wrong there.

Quoted 6 lines

Second, I believe this is causing a number of platforms with very tight SPL constraints, namely all of 64bit sunxi, to fail to link as they overflow SRAM now. This can be fixed at least by making the new behavior opt-in, but I would fix the vyasa-rk3288 problem first. Thanks in advance!

Well, I thought it would be better to have it default to 'y' since I think it is what is expected. Shouldn't we explicitly work on those platforms? I must say I was pretty shocked to see that SPL did not detect an invalid CRC...

Can you point me to a config that fails?

Regards,
Simon

#7 lore raw

PATCH v2 1/1 spl: implement CRC check on U-Boot uImage

Quoted 34 lines · Simon Goldschmidt
Quoted 15 lines · Tom Rini
Quoted 12 lines · Simon Goldschmidt

SPL currently does not check uImage CRCs when loading U-Boot.

This patch adds checking the uImage CRC when SPL loads U-Boot. It does this by reusing the existing config option SPL_CRC32_SUPPORT to allow leaving out the CRC check on boards where the additional code size or boot time is a problem (adding the CRC check currently adds ~1.4 kByte to flash).

The SPL_CRC32_SUPPORT config option now gets enabled by default if SPL support for legacy images is enabled to check the CRC on all boards that don't actively take countermeasures.

Quoted 8 lines
Quoted 2 lines

Really sorry for the delay on this, especially as I've found one or two problems. The first problem is that with this vyasa-rk3288 and a few others fail to build due to a number of errors such as:
../common/spl/spl.c:269:12: error: 'struct spl_image_info' has no member named 'dcrc_data'

Hmm, let me check what's wrong there.

Quoted 6 lines

Second, I believe this is causing a number of platforms with very tight SPL constraints, namely all of 64bit sunxi, to fail to link as they overflow SRAM now. This can be fixed at least by making the new behavior opt-in, but I would fix the vyasa-rk3288 problem first. Thanks in advance!

Quoted 6 lines

Well, I thought it would be better to have it default to 'y' since I think it is what is expected. Shouldn't we explicitly work on those platforms? I must say I was pretty shocked to see that SPL did not detect an invalid CRC...

Can you point me to a config that fails?

In general, pine64_plus_defconfig is one that might (as I re-run my build without this patch, I still see some failures but they've scrolled off screen, but 64bit sunxi _is_ very sensitive to SPL growth).

Signature · 8 lines
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: not available
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20190208/094c4b89/attachment.sig>
#8 lore raw

PATCH v2 1/1 spl: implement CRC check on U-Boot uImage

Quoted 21 lines · Tom Rini
Quoted 19 lines · Simon Goldschmidt
Quoted 18 lines · Tom Rini
Quoted 14 lines · Simon Goldschmidt

SPL currently does not check uImage CRCs when loading U-Boot.

This patch adds checking the uImage CRC when SPL loads U-Boot. It does this by reusing the existing config option SPL_CRC32_SUPPORT to allow leaving out the CRC check on boards where the additional code size or boot time is a problem (adding the CRC check currently adds ~1.4 kByte to flash).

The SPL_CRC32_SUPPORT config option now gets enabled by default if SPL support for legacy images is enabled to check the CRC on all boards that don't actively take countermeasures.

Quoted 7 lines
Quoted 7 lines
Quoted 5 lines

Really sorry for the delay on this, especially as I've found one or two problems. The first problem is that with this vyasa-rk3288 and a few others fail to build due to a number of errors such as:
../common/spl/spl.c:269:12: error: 'struct spl_image_info' has no member named 'dcrc_data'

Hmm, let me check what's wrong there.

OK, so the vyasa-rk3288 uses TPL, that's what I got wrong.

Quoted 12 lines
Quoted 12 lines
Quoted 6 lines

Second, I believe this is causing a number of platforms with very tight SPL constraints, namely all of 64bit sunxi, to fail to link as they overflow SRAM now. This can be fixed at least by making the new behavior opt-in, but I would fix the vyasa-rk3288 problem first. Thanks in advance!

Well, I thought it would be better to have it default to 'y' since I think it is what is expected. Shouldn't we explicitly work on those platforms? I must say I was pretty shocked to see that SPL did not detect an invalid CRC...

Quoted 5 lines
Quoted 1 line

Can you point me to a config that fails?

In general, pine64_plus_defconfig is one that might (as I re-run my build without this patch, I still see some failures but they've scrolled off screen, but 64bit sunxi _is_ very sensitive to SPL growth).

And I also did compile that one but I don't see any error message. How am I supposed to detect the SRAM overflowing?

I assume it's .text or .rodata which makes it grow too big? If so, we could still check the CRC by using a smaller algorithm that does not rely on a precalculated table...

Regards,
Simon

#9 lore raw

PATCH v2 1/1 spl: implement CRC check on U-Boot uImage

Quoted 50 lines · Simon Goldschmidt
Quoted 21 lines · Tom Rini
Quoted 19 lines · Simon Goldschmidt
Quoted 18 lines · Tom Rini
Quoted 14 lines · Simon Goldschmidt

SPL currently does not check uImage CRCs when loading U-Boot.

This patch adds checking the uImage CRC when SPL loads U-Boot. It does this by reusing the existing config option SPL_CRC32_SUPPORT to allow leaving out the CRC check on boards where the additional code size or boot time is a problem (adding the CRC check currently adds ~1.4 kByte to flash).

The SPL_CRC32_SUPPORT config option now gets enabled by default if SPL support for legacy images is enabled to check the CRC on all boards that don't actively take countermeasures.

Quoted 7 lines
Quoted 7 lines
Quoted 5 lines

Really sorry for the delay on this, especially as I've found one or two problems. The first problem is that with this vyasa-rk3288 and a few others fail to build due to a number of errors such as:
../common/spl/spl.c:269:12: error: 'struct spl_image_info' has no member named 'dcrc_data'

Hmm, let me check what's wrong there.

OK, so the vyasa-rk3288 uses TPL, that's what I got wrong.

Quoted 12 lines
Quoted 12 lines
Quoted 6 lines

Second, I believe this is causing a number of platforms with very tight SPL constraints, namely all of 64bit sunxi, to fail to link as they overflow SRAM now. This can be fixed at least by making the new behavior opt-in, but I would fix the vyasa-rk3288 problem first. Thanks in advance!

Well, I thought it would be better to have it default to 'y' since I think it is what is expected. Shouldn't we explicitly work on those platforms? I must say I was pretty shocked to see that SPL did not detect an invalid CRC...

Quoted 5 lines
Quoted 1 line

Can you point me to a config that fails?

In general, pine64_plus_defconfig is one that might (as I re-run my build without this patch, I still see some failures but they've scrolled off screen, but 64bit sunxi _is_ very sensitive to SPL growth).

Quoted 6 lines

And I also did compile that one but I don't see any error message. How am I supposed to detect the SRAM overflowing?

I assume it's .text or .rodata which makes it grow too big? If so, we could still check the CRC by using a smaller algorithm that does not rely on a precalculated table...

That could have been the other patch then. Does all of sunxi build for you? If so, we're good then.

Signature · 8 lines
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: not available
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20190209/1284fe30/attachment.sig>
#10 lore raw

PATCH v2 1/1 spl: implement CRC check on U-Boot uImage

Quoted 55 lines · Tom Rini
Quoted 53 lines · Simon Goldschmidt
Quoted 21 lines · Tom Rini
Quoted 19 lines · Simon Goldschmidt
Quoted 18 lines · Tom Rini
Quoted 14 lines · Simon Goldschmidt

SPL currently does not check uImage CRCs when loading U-Boot.

This patch adds checking the uImage CRC when SPL loads U-Boot. It does this by reusing the existing config option SPL_CRC32_SUPPORT to allow leaving out the CRC check on boards where the additional code size or boot time is a problem (adding the CRC check currently adds ~1.4 kByte to flash).

The SPL_CRC32_SUPPORT config option now gets enabled by default if SPL support for legacy images is enabled to check the CRC on all boards that don't actively take countermeasures.

Quoted 7 lines
Quoted 7 lines
Quoted 5 lines

Really sorry for the delay on this, especially as I've found one or two problems. The first problem is that with this vyasa-rk3288 and a few others fail to build due to a number of errors such as:
../common/spl/spl.c:269:12: error: 'struct spl_image_info' has no member named 'dcrc_data'

Hmm, let me check what's wrong there.

OK, so the vyasa-rk3288 uses TPL, that's what I got wrong.

Quoted 12 lines
Quoted 12 lines
Quoted 6 lines

Second, I believe this is causing a number of platforms with very tight SPL constraints, namely all of 64bit sunxi, to fail to link as they overflow SRAM now. This can be fixed at least by making the new behavior opt-in, but I would fix the vyasa-rk3288 problem first. Thanks in advance!

Well, I thought it would be better to have it default to 'y' since I think it is what is expected. Shouldn't we explicitly work on those platforms? I must say I was pretty shocked to see that SPL did not detect an invalid CRC...

Quoted 5 lines
Quoted 1 line

Can you point me to a config that fails?

In general, pine64_plus_defconfig is one that might (as I re-run my build without this patch, I still see some failures but they've scrolled off screen, but 64bit sunxi _is_ very sensitive to SPL growth).

And I also did compile that one but I don't see any error message. How am I supposed to detect the SRAM overflowing?

Quoted 6 lines
Quoted 3 lines

I assume it's .text or .rodata which makes it grow too big? If so, we could still check the CRC by using a smaller algorithm that does not rely on a precalculated table...

That could have been the other patch then. Does all of sunxi build for you? If so, we're good then.

Well, I really can't tell: especially with other patches in the queue that might get puhed before mine. text and rodata do grow by ~1.6 KiB summed up, so that might of course trigger some platforms.

On the other hand, I think this is a critical feature for reliable boot unless using FIT. If you don't think so (and in fact, noone seems to have noticed this for some years), then I'm happy to re-post this without default 'y'.

Regards,
Simon

#11 lore raw

PATCH v2 1/1 spl: implement CRC check on U-Boot uImage

Quoted 63 lines · Simon Goldschmidt
Quoted 55 lines · Tom Rini
Quoted 53 lines · Simon Goldschmidt
Quoted 21 lines · Tom Rini
Quoted 19 lines · Simon Goldschmidt
Quoted 18 lines · Tom Rini
Quoted 14 lines · Simon Goldschmidt

SPL currently does not check uImage CRCs when loading U-Boot.

This patch adds checking the uImage CRC when SPL loads U-Boot. It does this by reusing the existing config option SPL_CRC32_SUPPORT to allow leaving out the CRC check on boards where the additional code size or boot time is a problem (adding the CRC check currently adds ~1.4 kByte to flash).

The SPL_CRC32_SUPPORT config option now gets enabled by default if SPL support for legacy images is enabled to check the CRC on all boards that don't actively take countermeasures.

Quoted 7 lines
Quoted 7 lines
Quoted 5 lines

Really sorry for the delay on this, especially as I've found one or two problems. The first problem is that with this vyasa-rk3288 and a few others fail to build due to a number of errors such as:
../common/spl/spl.c:269:12: error: 'struct spl_image_info' has no member named 'dcrc_data'

Hmm, let me check what's wrong there.

OK, so the vyasa-rk3288 uses TPL, that's what I got wrong.

Quoted 12 lines
Quoted 12 lines
Quoted 6 lines

Second, I believe this is causing a number of platforms with very tight SPL constraints, namely all of 64bit sunxi, to fail to link as they overflow SRAM now. This can be fixed at least by making the new behavior opt-in, but I would fix the vyasa-rk3288 problem first. Thanks in advance!

Well, I thought it would be better to have it default to 'y' since I think it is what is expected. Shouldn't we explicitly work on those platforms? I must say I was pretty shocked to see that SPL did not detect an invalid CRC...

Quoted 5 lines
Quoted 1 line

Can you point me to a config that fails?

In general, pine64_plus_defconfig is one that might (as I re-run my build without this patch, I still see some failures but they've scrolled off screen, but 64bit sunxi _is_ very sensitive to SPL growth).

And I also did compile that one but I don't see any error message. How am I supposed to detect the SRAM overflowing?

Quoted 6 lines
Quoted 3 lines

I assume it's .text or .rodata which makes it grow too big? If so, we could still check the CRC by using a smaller algorithm that does not rely on a precalculated table...

That could have been the other patch then. Does all of sunxi build for you? If so, we're good then.

Quoted 7 lines

Well, I really can't tell: especially with other patches in the queue that might get puhed before mine. text and rodata do grow by ~1.6 KiB summed up, so that might of course trigger some platforms.

On the other hand, I think this is a critical feature for reliable boot unless using FIT. If you don't think so (and in fact, noone seems to have noticed this for some years), then I'm happy to re-post this without default 'y'.

Yeah, I think we should leave default y off. Thanks!

Signature · 8 lines
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: not available
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20190210/123e4fdb/attachment.sig>