This is a note to let you know that I've just added the patch titled
x86/srso: Explain the untraining sequences a bit more
to the 6.1-stable tree which can be found at:
http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary
The filename of the patch is:
x86-srso-explain-the-untraining-sequences-a-bit-more.patch
and it can be found in the queue-6.1 subdirectory.
If you, or anyone else, feels it should not be added to the stable tree,
please let <stable@xxxxxxxxxxxxxxx> know about it.
>From 9dbd23e42ff0b10c9b02c9e649c76e5228241a8e Mon Sep 17 00:00:00 2001
From: "Borislav Petkov (AMD)" <bp@xxxxxxxxx>
Date: Mon, 14 Aug 2023 21:29:50 +0200
Subject: x86/srso: Explain the untraining sequences a bit more
From: Borislav Petkov (AMD) <bp@xxxxxxxxx>
commit 9dbd23e42ff0b10c9b02c9e649c76e5228241a8e upstream.
The goal is to eventually have a proper documentation about all this.
Signed-off-by: Borislav Petkov (AMD) <bp@xxxxxxxxx>
Link: https://lore.kernel.org/r/20230814164447.GFZNpZ/64H4lENIe94@fat_crate.local
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
---
arch/x86/lib/retpoline.S | 19 +++++++++++++++++++
1 file changed, 19 insertions(+)
--- a/arch/x86/lib/retpoline.S
+++ b/arch/x86/lib/retpoline.S
@@ -130,6 +130,25 @@ SYM_CODE_START(srso_alias_return_thunk)
SYM_CODE_END(srso_alias_return_thunk)
/*
+ * Some generic notes on the untraining sequences:
+ *
+ * They are interchangeable when it comes to flushing potentially wrong
+ * RET predictions from the BTB.
+ *
+ * The SRSO Zen1/2 (MOVABS) untraining sequence is longer than the
+ * Retbleed sequence because the return sequence done there
+ * (srso_safe_ret()) is longer and the return sequence must fully nest
+ * (end before) the untraining sequence. Therefore, the untraining
+ * sequence must fully overlap the return sequence.
+ *
+ * Regarding alignment - the instructions which need to be untrained,
+ * must all start at a cacheline boundary for Zen1/2 generations. That
+ * is, instruction sequences starting at srso_safe_ret() and
+ * the respective instruction sequences at retbleed_return_thunk()
+ * must start at a cacheline boundary.
+ */
+
+/*
* Safety details here pertain to the AMD Zen{1,2} microarchitecture:
* 1) The RET at retbleed_return_thunk must be on a 64 byte boundary, for
* alignment within the BTB.
Patches currently in stable-queue which might be from bp@xxxxxxxxx are
queue-6.1/x86-cpu-fix-up-srso_safe_ret-and-__x86_return_thunk.patch
queue-6.1/x86-cpu-cleanup-the-untrain-mess.patch
queue-6.1/objtool-x86-fixup-frame-pointer-vs-rethunk.patch
queue-6.1/x86-srso-correct-the-mitigation-status-when-smt-is-disabled.patch
queue-6.1/x86-retpoline-don-t-clobber-rflags-during-srso_safe_ret.patch
queue-6.1/x86-retpoline-kprobes-fix-position-of-thunk-sections-with-config_lto_clang.patch
queue-6.1/x86-cpu-amd-fix-the-div-0-initial-fix-attempt.patch
queue-6.1/x86-cpu-rename-srso_-.-_alias-to-srso_alias_-1.patch
queue-6.1/x86-srso-explain-the-untraining-sequences-a-bit-more.patch
queue-6.1/x86-alternative-make-custom-return-thunk-unconditional.patch
queue-6.1/x86-cpu-rename-original-retbleed-methods.patch
queue-6.1/x86-cpu-clean-up-srso-return-thunk-mess.patch
queue-6.1/x86-srso-disable-the-mitigation-on-unaffected-configurations.patch
queue-6.1/x86-cpu-fix-__x86_return_thunk-symbol-type.patch
