From: Peter Zijlstra <peterz@infradead.org>
To: x86@kernel.org
Cc: linux-kernel@vger.kernel.org, peterz@infradead.org,
David.Kaplan@amd.com, Andrew.Cooper3@citrix.com,
jpoimboe@kernel.org, gregkh@linuxfoundation.org
Subject: [RFC][PATCH 17/17] x86/cpu: Use fancy alternatives to get rid of entry_untrain_ret()
Date: Wed, 09 Aug 2023 09:12:35 +0200 [thread overview]
Message-ID: <20230809072201.596268530@infradead.org> (raw)
In-Reply-To: 20230809071218.000335006@infradead.org
Use the new nested alternatives to create what is effectively
ALTERNATIVE_5 and merge the dummy entry_untrain_ret stub into
UNTRAIN_RET properly.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
---
arch/x86/include/asm/nospec-branch.h | 33 ++++++++++++++++++---------------
arch/x86/lib/retpoline.S | 7 -------
2 files changed, 18 insertions(+), 22 deletions(-)
--- a/arch/x86/include/asm/nospec-branch.h
+++ b/arch/x86/include/asm/nospec-branch.h
@@ -271,11 +271,15 @@
.endm
#ifdef CONFIG_CPU_UNRET_ENTRY
-#define CALL_UNTRAIN_RET "call entry_untrain_ret"
+#define ALT_UNRET(old) \
+ __ALTERNATIVE(__ALTERNATIVE(__ALTERNATIVE(old, call zen_untrain_ret, X86_FEATURE_UNRET), \
+ call srso_untrain_ret, X86_FEATURE_SRSO), \
+ call srso_alias_untrain_ret, X86_FEATURE_SRSO_ALIAS)
#else
-#define CALL_UNTRAIN_RET ""
+#define ALT_UNRET(old) old
#endif
+
/*
* Mitigate RETBleed for AMD/Hygon Zen uarch. Requires KERNEL CR3 because the
* return thunk isn't mapped into the userspace tables (then again, AMD
@@ -290,30 +294,30 @@
.macro UNTRAIN_RET
#if defined(CONFIG_CPU_UNRET_ENTRY) || defined(CONFIG_CALL_DEPTH_TRACKING)
VALIDATE_UNRET_END
- ALTERNATIVE_3 "", \
- CALL_UNTRAIN_RET, X86_FEATURE_UNRET, \
- "call entry_ibpb", X86_FEATURE_ENTRY_IBPB, \
- __stringify(RESET_CALL_DEPTH), X86_FEATURE_CALL_DEPTH
+
+ __ALTERNATIVE(__ALTERNATIVE(ALT_UNRET(;),
+ call entry_ibpb, X86_FEATURE_ENTRY_IBPB),
+ RESET_CALL_DEPTH, X86_FEATURE_CALL_DEPTH)
#endif
.endm
.macro UNTRAIN_RET_VM
#if defined(CONFIG_CPU_UNRET_ENTRY) || defined(CONFIG_CALL_DEPTH_TRACKING)
VALIDATE_UNRET_END
- ALTERNATIVE_3 "", \
- CALL_UNTRAIN_RET, X86_FEATURE_UNRET, \
- "call entry_ibpb", X86_FEATURE_IBPB_ON_VMEXIT, \
- __stringify(RESET_CALL_DEPTH), X86_FEATURE_CALL_DEPTH
+
+ __ALTERNATIVE(__ALTERNATIVE(ALT_UNRET(;),
+ call entry_ibpb, X86_FEATURE_IBPB_ON_VMEXIT),
+ RESET_CALL_DEPTH, X86_FEATURE_CALL_DEPTH)
#endif
.endm
.macro UNTRAIN_RET_FROM_CALL
#if defined(CONFIG_CPU_UNRET_ENTRY) || defined(CONFIG_CALL_DEPTH_TRACKING)
VALIDATE_UNRET_END
- ALTERNATIVE_3 "", \
- CALL_UNTRAIN_RET, X86_FEATURE_UNRET, \
- "call entry_ibpb", X86_FEATURE_ENTRY_IBPB, \
- __stringify(RESET_CALL_DEPTH_FROM_CALL), X86_FEATURE_CALL_DEPTH
+
+ __ALTERNATIVE(__ALTERNATIVE(ALT_UNRET(;),
+ call entry_ibpb, X86_FEATURE_ENTRY_IBPB),
+ RESET_CALL_DEPTH_FROM_CALL, X86_FEATURE_CALL_DEPTH)
#endif
.endm
@@ -348,7 +352,6 @@ extern void zen_untrain_ret(void);
extern void srso_untrain_ret(void);
extern void srso_alias_untrain_ret(void);
-extern void entry_untrain_ret(void);
extern void entry_ibpb(void);
extern void (*x86_return_thunk)(void);
--- a/arch/x86/lib/retpoline.S
+++ b/arch/x86/lib/retpoline.S
@@ -263,13 +263,6 @@ SYM_CODE_END(srso_safe_ret)
SYM_FUNC_END(srso_untrain_ret)
__EXPORT_THUNK(srso_untrain_ret)
-SYM_FUNC_START(entry_untrain_ret)
- ALTERNATIVE_2 "jmp zen_untrain_ret", \
- "jmp srso_untrain_ret", X86_FEATURE_SRSO, \
- "jmp srso_alias_untrain_ret", X86_FEATURE_SRSO_ALIAS
-SYM_FUNC_END(entry_untrain_ret)
-__EXPORT_THUNK(entry_untrain_ret)
-
/*
* Both these do an unbalanced CALL to mess up the RSB, terminate with UD2
* to indicate noreturn.
next prev parent reply other threads:[~2023-08-09 7:27 UTC|newest]
Thread overview: 80+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-08-09 7:12 [RFC][PATCH 00/17] Fix up the recent SRSO patches Peter Zijlstra
2023-08-09 7:12 ` [RFC][PATCH 01/17] x86/alternative: Unconditional custom return thunk Peter Zijlstra
2023-08-09 9:31 ` Nikolay Borisov
2023-08-10 11:37 ` Borislav Petkov
2023-08-09 7:12 ` [RFC][PATCH 02/17] x86/cpu: Clean up SRSO return thunk mess Peter Zijlstra
2023-08-09 15:45 ` Nikolay Borisov
2023-08-10 11:51 ` Borislav Petkov
2023-08-10 12:37 ` Peter Zijlstra
2023-08-10 12:56 ` Borislav Petkov
2023-08-10 13:22 ` Peter Zijlstra
2023-08-11 7:01 ` Peter Zijlstra
2023-08-11 17:00 ` Nick Desaulniers
2023-08-12 11:20 ` Peter Zijlstra
2023-08-09 7:12 ` [RFC][PATCH 03/17] x86/cpu: Make srso_untrain_ret consistent Peter Zijlstra
2023-08-10 12:00 ` Borislav Petkov
2023-08-09 7:12 ` [RFC][PATCH 04/17] objtool/x86: Fix SRSO mess Peter Zijlstra
2023-08-10 12:06 ` Borislav Petkov
2023-08-10 12:48 ` Peter Zijlstra
2023-08-10 12:50 ` Peter Zijlstra
2023-08-10 15:02 ` Borislav Petkov
2023-08-10 15:22 ` Peter Zijlstra
2023-08-09 7:12 ` [RFC][PATCH 05/17] x86/cpu: Cleanup the untrain mess Peter Zijlstra
2023-08-09 12:51 ` Josh Poimboeuf
2023-08-09 13:12 ` Peter Zijlstra
2023-08-09 13:26 ` Peter Zijlstra
2023-08-12 18:30 ` Borislav Petkov
2023-08-09 7:12 ` [RFC][PATCH 06/17] x86/cpu: Add SRSO untrain to retbleed= Peter Zijlstra
2023-08-09 13:42 ` Josh Poimboeuf
2023-08-09 14:06 ` Peter Zijlstra
2023-08-09 14:28 ` Josh Poimboeuf
2023-08-09 15:08 ` Peter Zijlstra
2023-08-09 15:43 ` Josh Poimboeuf
2023-08-09 14:31 ` Andrew.Cooper3
2023-08-09 14:39 ` Josh Poimboeuf
2023-08-10 15:44 ` Borislav Petkov
2023-08-10 16:10 ` Josh Poimboeuf
2023-08-11 10:27 ` Borislav Petkov
2023-08-12 11:32 ` Peter Zijlstra
2023-08-12 12:12 ` Borislav Petkov
2023-08-14 15:45 ` David Laight
2023-08-12 11:24 ` Peter Zijlstra
2023-08-12 12:10 ` Borislav Petkov
2023-08-14 10:56 ` Peter Zijlstra
2023-08-09 7:12 ` [RFC][PATCH 07/17] x86/cpu/kvm: Provide UNTRAIN_RET_VM Peter Zijlstra
2023-08-09 13:50 ` Josh Poimboeuf
2023-08-09 14:06 ` Peter Zijlstra
2023-08-09 14:30 ` Josh Poimboeuf
2023-08-09 15:10 ` Peter Zijlstra
2023-08-13 10:36 ` Borislav Petkov
2023-08-14 10:35 ` Peter Zijlstra
2023-08-09 7:12 ` [RFC][PATCH 08/17] x86/cpu: Add IBPB on VMEXIT to retbleed= Peter Zijlstra
2023-08-09 7:12 ` [RFC][PATCH 09/17] x86: Remove CONFIG_CPU_SRSO Peter Zijlstra
2023-08-09 13:57 ` Josh Poimboeuf
2023-08-09 7:12 ` [RFC][PATCH 10/17] x86: Remove CPU_IBPB_ENTRY Peter Zijlstra
2023-08-09 7:12 ` [RFC][PATCH 11/17] x86/cpu: Remove all SRSO interface nonsense Peter Zijlstra
2023-08-09 13:10 ` Andrew.Cooper3
2023-08-09 13:36 ` Peter Zijlstra
2023-08-09 14:05 ` Josh Poimboeuf
2023-08-09 14:43 ` Peter Zijlstra
2023-08-09 14:51 ` Josh Poimboeuf
2023-08-09 15:34 ` Josh Poimboeuf
2023-08-09 7:12 ` [RFC][PATCH 12/17] x86/cpu: Rename original retbleed return thunk Peter Zijlstra
2023-08-09 14:20 ` Josh Poimboeuf
2023-08-09 14:22 ` Peter Zijlstra
2023-08-10 11:06 ` Andrew.Cooper3
2023-08-10 13:02 ` Peter Zijlstra
2023-08-13 15:23 ` Andrew.Cooper3
2023-08-14 10:34 ` Peter Zijlstra
2023-08-14 11:31 ` Andrew.Cooper3
2023-08-14 12:06 ` Peter Zijlstra
2023-08-09 7:12 ` [RFC][PATCH 13/17] objtool/x86: Add arch_is_offset_insn() Peter Zijlstra
2023-08-09 9:56 ` Nikolay Borisov
2023-08-09 14:34 ` Josh Poimboeuf
2023-08-09 7:12 ` [RFC][PATCH 14/17] objtool: Add comments to the arch_is_$foo() magic symbols Peter Zijlstra
2023-08-09 7:12 ` [RFC][PATCH 15/17] x86/cpu: Rename srso_(.*)_alias to srso_alias_\1 Peter Zijlstra
2023-08-09 7:12 ` [RFC][PATCH 16/17] x86/alternatives: Simplify ALTERNATIVE_n() Peter Zijlstra
2023-08-09 7:12 ` Peter Zijlstra [this message]
2023-08-09 9:04 ` [RFC][PATCH 00/17] Fix up the recent SRSO patches Nikolay Borisov
2023-08-09 10:04 ` Andrew.Cooper3
2023-08-09 11:58 ` Peter Zijlstra
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20230809072201.596268530@infradead.org \
--to=peterz@infradead.org \
--cc=Andrew.Cooper3@citrix.com \
--cc=David.Kaplan@amd.com \
--cc=gregkh@linuxfoundation.org \
--cc=jpoimboe@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=x86@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox