From c08bb6a3f9c30e3ad63a6246a59b841c1fe06140 Mon Sep 17 00:00:00 2001 From: bunnie Date: Wed, 13 Jul 2022 04:43:20 +0800 Subject: [PATCH] add notes about the second ECC error reported --- services/spinor/src/main.rs | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/services/spinor/src/main.rs b/services/spinor/src/main.rs index dc5658087..d26d98fda 100644 --- a/services/spinor/src/main.rs +++ b/services/spinor/src/main.rs @@ -835,6 +835,13 @@ fn main() -> ! { This is the raw log: ERR :spinor: ECC error reported: 0xfffffffc 0xb3b30000 0x3305080 0x7305030 (services\spinor\src\main.rs:830) Archived here: https://ci.betrusted.io/view/Enabled/job/ctap2-tests/64/console + - Second ECC failure noted July 13, 2022 on the high-cycle dev unit. The failure actually may + be linked to an aborted write during backup generation; it was reported at an address that + up until now was never used. This error was different from the previous one in that after + tripping the ROM would only return 0xFF, and it would not clear. The error address + was 0x01D7_F0A0 - just inside the backup block. The backup code has been fixed to not + use two disjoint patch operations to merge its data, and to instead merge the write data + before patching. Error was cleared by erasing the block, and has not since been observed again. */ if !ecc_errors.contains(&(hw_rep as u32, status as u32, lower_addr as u32, upper_addr as u32)) { if ecc_errors.len() < MAX_ERRLOG_LEN {