From a9bb416c822d20c1b94fb7b2bfab282d25e11125 Mon Sep 17 00:00:00 2001 From: Ian McIntyre Date: Wed, 11 Feb 2026 16:29:27 -0500 Subject: Preload RAM for ECC ciphers Testing shows that we must blow fuses in order to activate either ECC engine. Once the fuse is blown, NXP's boot ROM will enable the ECC engine. So by the time the reset handler is running, ECC may already be enabled. The 32-bit copy loops won't work for initializing ITCM and OCRAM. Since these bus widths are 64-bit and the ECC is computed over 64 bits, these 32 bit accesses result in a read-modify-write (RMW) on the underlying interface. If these RAM regions aren't already initialized, we'll fail an ECC check; the read in RMW compares bogus RAM contents against a bogus ECC. This seems to dramatically slow down initialization, or stall it (I've seen the CPU spinning in the boot ROM). If we listen to NXP and use 64-bit writes to preload ECC-protected RAM regions, the 32-bit copy loops will work as expected. Note that we can still issue 64-bit accesses towards DTCM; they're scattered across the two 32-bit interfaces. There's an NXP whitepaper [1] describing that. Once TCM ECC is enabled, we need to tell the TCM interfaces to RMW smaller accesses. Otherwise, we won't generate correct ciphers. We also might need to retry operations if there was a correction by the RAM controller. We can perform these with predicated instructions. Tested on an 1010EVK, where I used the debugger to show the control registers weren't touched; and tested on the 1170EVK, with the same approach showing the expected set bits. [1]: https://www.nxp.com/docs/en/white-paper/CORTEXM7WP.pdf [2]: https://www.nxp.com/docs/en/application-note/AN13204.pdf --- src/target.rs | 52 ++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 52 insertions(+) (limited to 'src/target.rs') diff --git a/src/target.rs b/src/target.rs index 96ae9f7..4515934 100644 --- a/src/target.rs +++ b/src/target.rs @@ -55,6 +55,22 @@ global_asm! {r#" 999: .endm +.macro zero_section64 beg, end + ldr r0, =\beg + ldr r1, =\end + movw r2, #0 + + 888: + cmp r0, r1 + bge 999f + + strd r2, r2, [r0], #8 + b 888b + + 999: + +.endm + __pre_init: ldr r0, =__imxrt_rt_v0.2 @ Need to know which chip family we're initializing. ldr r1, =0x1180 @@ -85,6 +101,42 @@ __pre_init: ldr r1, [r0, #64] @ r1 = *(IMXRT_IOMUXC_GPR + 16) orr r1, r1, #1<<2 @ r1 |= 1 << 2 str r1, [r0, #64] @ *(IMXRT_IOMUXC_GPR + 16) = r1 + + # If there's an ITCM ECC region, make sure + # + # - the CPU issues RMW operations for smaller accesses. + # - the CPU retries operations when signaled. + # + # If there's no ECC region, the host sets the ends + # to zero, and we skip the instructions. + movw r3, #0 + ldr r0, =__eitcm + cmp r0, r3 + itttt ne + ldrne r0, =0xE000EF90 @ ITCMCR + ldrne r1, [r0] @ r1 = *ITCMCR + orrne r1, r1, #3<<1 @ r1 |= 3 << 1, setting RMW and RETEN + strne r1, [r0] @ *ITCMCR = r1 + + # Same as above, but for DTCM. + ldr r0, =__edtcm + cmp r0, r3 + itttt ne + ldrne r0, =0xE000EF94 @ DTCMCR + ldrne r1, [r0] @ r1 = *DTCMCR + orrne r1, r1, #3<<1 @ r1 |= 3 << 1, setting RMW and RETEN + strne r1, [r0] @ *DTCMCR = r1 + + dsb + isb + + # Zero memory regions for any ECC support. + # If ECC isn't enabled, then the host generates + # values that short circuit the loops. + zero_section64 __sitcm , __eitcm + zero_section64 __sdtcm , __edtcm + zero_section64 __socram , __eocram + b 1000f 1180: -- cgit v1.2.3