* [PATCH] i.MX: hab: always lock SRK hash after fusing
@ 2026-03-06 10:38 Ulrich Ölmann
2026-03-06 10:50 ` Marco Felsch
2026-03-09 7:50 ` Sascha Hauer
0 siblings, 2 replies; 4+ messages in thread
From: Ulrich Ölmann @ 2026-03-06 10:38 UTC (permalink / raw)
To: Barebox List
The flag IMX_SRK_HASH_WRITE_LOCK has been present since the introduction of
barebox' hab command in [1], but only got its first user recently. Keeping SRK
hash locking optional is dangerous though: after programming the SRK hash,
leaving it writable allows later manipulations which can render a device
unbootable.
Make SRK hash programming always burn the corresponding lock fuses on all
supported i.MX variants (IIM/OCOTP), and remove IMX_SRK_HASH_WRITE_LOCK.
[1] 9dc622d5622c ("i.MX: hab: Add HAB fusebox related convenience functions / command")
Signed-off-by: Ulrich Ölmann <u.oelmann@pengutronix.de>
---
common/boards/hgs/common.c | 3 +--
drivers/hab/hab.c | 30 ++++++++++++------------------
include/hab.h | 4 ----
3 files changed, 13 insertions(+), 24 deletions(-)
diff --git a/common/boards/hgs/common.c b/common/boards/hgs/common.c
index 05edbd7e1df6..c4a31a27978f 100644
--- a/common/boards/hgs/common.c
+++ b/common/boards/hgs/common.c
@@ -479,8 +479,7 @@ struct hgs_fusemap {
*/
static int hgs_run_first_boot_setup(void)
{
- unsigned int flags = IMX_SRK_HASH_WRITE_PERMANENT |
- IMX_SRK_HASH_WRITE_LOCK;
+ unsigned int flags = IMX_SRK_HASH_WRITE_PERMANENT;
const struct hgs_fusemap common_fusemap[] = {
/* Security */
/* MX8M_OCOTP_SEC_CONFIG[1] is done during imx_hab_lockdown_device() */
diff --git a/drivers/hab/hab.c b/drivers/hab/hab.c
index de0a6c73d692..5a4be712d596 100644
--- a/drivers/hab/hab.c
+++ b/drivers/hab/hab.c
@@ -80,14 +80,12 @@ static int imx_hab_write_srk_hash_iim(const u8 *srk, unsigned flags)
return -EIO;
}
- if (flags & IMX_SRK_HASH_WRITE_LOCK) {
- ret = imx_iim_write_field(IMX25_IIM_SRK0_LOCK96, 1);
- if (ret < 0)
- return ret;
- ret = imx_iim_write_field(IMX25_IIM_SRK0_LOCK160, 1);
- if (ret < 0)
- return ret;
- }
+ ret = imx_iim_write_field(IMX25_IIM_SRK0_LOCK96, 1);
+ if (ret < 0)
+ return ret;
+ ret = imx_iim_write_field(IMX25_IIM_SRK0_LOCK160, 1);
+ if (ret < 0)
+ return ret;
return 0;
}
@@ -150,11 +148,9 @@ static int imx6_hab_write_srk_hash_ocotp(const u8 *newsrk, unsigned flags)
if (ret)
return ret;
- if (flags & IMX_SRK_HASH_WRITE_LOCK) {
- ret = imx_ocotp_write_field(OCOTP_SRK_LOCK, 1);
- if (ret < 0)
- return ret;
- }
+ ret = imx_ocotp_write_field(OCOTP_SRK_LOCK, 1);
+ if (ret < 0)
+ return ret;
return 0;
}
@@ -167,11 +163,9 @@ static int imx8m_hab_write_srk_hash_ocotp(const u8 *newsrk, unsigned flags)
if (ret)
return ret;
- if (flags & IMX_SRK_HASH_WRITE_LOCK) {
- ret = imx_ocotp_write_field(MX8M_OCOTP_SRK_LOCK, 1);
- if (ret < 0)
- return ret;
- }
+ ret = imx_ocotp_write_field(MX8M_OCOTP_SRK_LOCK, 1);
+ if (ret < 0)
+ return ret;
return 0;
}
diff --git a/include/hab.h b/include/hab.h
index 393cf0513e0c..212789dbfa59 100644
--- a/include/hab.h
+++ b/include/hab.h
@@ -38,10 +38,6 @@ static inline int habv4_get_state(void)
* are written.
*/
#define IMX_SRK_HASH_WRITE_PERMANENT BIT(1)
-/* When writing the super root key hash, also burn the write protection
- * fuses so that the key hash can not be modified.
- */
-#define IMX_SRK_HASH_WRITE_LOCK BIT(2)
bool imx_hab_srk_hash_valid(const void *buf);
int imx_hab_write_srk_hash(const void *buf, unsigned flags);
--
2.47.3
^ permalink raw reply [flat|nested] 4+ messages in thread* Re: [PATCH] i.MX: hab: always lock SRK hash after fusing
2026-03-06 10:38 [PATCH] i.MX: hab: always lock SRK hash after fusing Ulrich Ölmann
@ 2026-03-06 10:50 ` Marco Felsch
2026-03-09 7:50 ` Sascha Hauer
1 sibling, 0 replies; 4+ messages in thread
From: Marco Felsch @ 2026-03-06 10:50 UTC (permalink / raw)
To: Ulrich Ölmann; +Cc: Barebox List
On 26-03-06, Ulrich Ölmann wrote:
> The flag IMX_SRK_HASH_WRITE_LOCK has been present since the introduction of
> barebox' hab command in [1], but only got its first user recently. Keeping SRK
> hash locking optional is dangerous though: after programming the SRK hash,
> leaving it writable allows later manipulations which can render a device
> unbootable.
>
> Make SRK hash programming always burn the corresponding lock fuses on all
> supported i.MX variants (IIM/OCOTP), and remove IMX_SRK_HASH_WRITE_LOCK.
>
> [1] 9dc622d5622c ("i.MX: hab: Add HAB fusebox related convenience functions / command")
>
> Signed-off-by: Ulrich Ölmann <u.oelmann@pengutronix.de>
Reviewed-by: Marco Felsch <m.felsch@pengutronix.de>
^ permalink raw reply [flat|nested] 4+ messages in thread* Re: [PATCH] i.MX: hab: always lock SRK hash after fusing
2026-03-06 10:38 [PATCH] i.MX: hab: always lock SRK hash after fusing Ulrich Ölmann
2026-03-06 10:50 ` Marco Felsch
@ 2026-03-09 7:50 ` Sascha Hauer
2026-03-09 8:07 ` Sascha Hauer
1 sibling, 1 reply; 4+ messages in thread
From: Sascha Hauer @ 2026-03-09 7:50 UTC (permalink / raw)
To: Barebox List, Ulrich Ölmann
On Fri, 06 Mar 2026 11:38:12 +0100, Ulrich Ölmann wrote:
> The flag IMX_SRK_HASH_WRITE_LOCK has been present since the introduction of
> barebox' hab command in [1], but only got its first user recently. Keeping SRK
> hash locking optional is dangerous though: after programming the SRK hash,
> leaving it writable allows later manipulations which can render a device
> unbootable.
>
> Make SRK hash programming always burn the corresponding lock fuses on all
> supported i.MX variants (IIM/OCOTP), and remove IMX_SRK_HASH_WRITE_LOCK.
>
> [...]
Applied, thanks!
[1/1] i.MX: hab: always lock SRK hash after fusing
https://git.pengutronix.de/cgit/barebox/commit/?id=1f4d4a2b75b7 (link may not be stable)
Best regards,
--
Sascha Hauer <s.hauer@pengutronix.de>
^ permalink raw reply [flat|nested] 4+ messages in thread* Re: [PATCH] i.MX: hab: always lock SRK hash after fusing
2026-03-09 7:50 ` Sascha Hauer
@ 2026-03-09 8:07 ` Sascha Hauer
0 siblings, 0 replies; 4+ messages in thread
From: Sascha Hauer @ 2026-03-09 8:07 UTC (permalink / raw)
To: Barebox List, Ulrich Ölmann
On Mon, Mar 09, 2026 at 08:50:38AM +0100, Sascha Hauer wrote:
>
> On Fri, 06 Mar 2026 11:38:12 +0100, Ulrich Ölmann wrote:
> > The flag IMX_SRK_HASH_WRITE_LOCK has been present since the introduction of
> > barebox' hab command in [1], but only got its first user recently. Keeping SRK
> > hash locking optional is dangerous though: after programming the SRK hash,
> > leaving it writable allows later manipulations which can render a device
> > unbootable.
> >
> > Make SRK hash programming always burn the corresponding lock fuses on all
> > supported i.MX variants (IIM/OCOTP), and remove IMX_SRK_HASH_WRITE_LOCK.
> >
> > [...]
>
> Applied, thanks!
>
> [1/1] i.MX: hab: always lock SRK hash after fusing
> https://git.pengutronix.de/cgit/barebox/commit/?id=1f4d4a2b75b7 (link may not be stable)
Also added a note to the migration guide.
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 |
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2026-03-09 8:08 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2026-03-06 10:38 [PATCH] i.MX: hab: always lock SRK hash after fusing Ulrich Ölmann
2026-03-06 10:50 ` Marco Felsch
2026-03-09 7:50 ` Sascha Hauer
2026-03-09 8:07 ` Sascha Hauer
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox