linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Peter Zijlstra <peterz@infradead.org>
To: x86@kernel.org, joao@overdrivepizza.com, hjl.tools@gmail.com,
	jpoimboe@redhat.com, andrew.cooper3@citrix.com
Cc: linux-kernel@vger.kernel.org, peterz@infradead.org,
	ndesaulniers@google.com, keescook@chromium.org,
	samitolvanen@google.com, mark.rutland@arm.com,
	alyssa.milburn@intel.com, mbenes@suse.cz, rostedt@goodmis.org,
	mhiramat@kernel.org, alexei.starovoitov@gmail.com
Subject: [PATCH v3 18/39] x86/ibt,ftrace: Make function-graph play nice
Date: Thu, 03 Mar 2022 12:23:39 +0100	[thread overview]
Message-ID: <20220303112826.044301664@infradead.org> (raw)
In-Reply-To: 20220303112321.422525803@infradead.org

Return trampoline must not use indirect branch to return; while this
preserves the RSB, it is fundamentally incompatible with IBT. Instead
use a retpoline like ROP gadget that defeats IBT while not unbalancing
the RSB.

And since ftrace_stub is no longer a plain RET, don't use it to copy
from. Since RET is a trivial instruction, poke it directly.

Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
---
 arch/x86/kernel/ftrace.c    |    9 ++-------
 arch/x86/kernel/ftrace_64.S |   16 ++++++++++++----
 2 files changed, 14 insertions(+), 11 deletions(-)

--- a/arch/x86/kernel/ftrace.c
+++ b/arch/x86/kernel/ftrace.c
@@ -316,12 +316,12 @@ create_trampoline(struct ftrace_ops *ops
 	unsigned long offset;
 	unsigned long npages;
 	unsigned long size;
-	unsigned long retq;
 	unsigned long *ptr;
 	void *trampoline;
 	void *ip;
 	/* 48 8b 15 <offset> is movq <offset>(%rip), %rdx */
 	unsigned const char op_ref[] = { 0x48, 0x8b, 0x15 };
+	unsigned const char retq[] = { RET_INSN_OPCODE, INT3_INSN_OPCODE };
 	union ftrace_op_code_union op_ptr;
 	int ret;
 
@@ -359,12 +359,7 @@ create_trampoline(struct ftrace_ops *ops
 		goto fail;
 
 	ip = trampoline + size;
-
-	/* The trampoline ends with ret(q) */
-	retq = (unsigned long)ftrace_stub;
-	ret = copy_from_kernel_nofault(ip, (void *)retq, RET_SIZE);
-	if (WARN_ON(ret < 0))
-		goto fail;
+	memcpy(ip, retq, RET_SIZE);
 
 	/* No need to test direct calls on created trampolines */
 	if (ops->flags & FTRACE_OPS_FL_SAVE_REGS) {
--- a/arch/x86/kernel/ftrace_64.S
+++ b/arch/x86/kernel/ftrace_64.S
@@ -176,10 +176,10 @@ SYM_FUNC_END(ftrace_caller);
 SYM_FUNC_START(ftrace_epilogue)
 /*
  * This is weak to keep gas from relaxing the jumps.
- * It is also used to copy the RET for trampolines.
  */
 SYM_INNER_LABEL_ALIGN(ftrace_stub, SYM_L_WEAK)
 	UNWIND_HINT_FUNC
+	ENDBR
 	RET
 SYM_FUNC_END(ftrace_epilogue)
 
@@ -284,6 +284,7 @@ SYM_FUNC_START(__fentry__)
 	jnz trace
 
 SYM_INNER_LABEL(ftrace_stub, SYM_L_GLOBAL)
+	ENDBR
 	RET
 
 trace:
@@ -307,7 +308,7 @@ EXPORT_SYMBOL(__fentry__)
 
 #ifdef CONFIG_FUNCTION_GRAPH_TRACER
 SYM_FUNC_START(return_to_handler)
-	subq  $24, %rsp
+	subq  $16, %rsp
 
 	/* Save the return values */
 	movq %rax, (%rsp)
@@ -319,7 +320,14 @@ SYM_FUNC_START(return_to_handler)
 	movq %rax, %rdi
 	movq 8(%rsp), %rdx
 	movq (%rsp), %rax
-	addq $24, %rsp
-	JMP_NOSPEC rdi
+
+	addq $16, %rsp
+	ANNOTATE_INTRA_FUNCTION_CALL
+	call .Ldo_rop
+	int3
+.Ldo_rop:
+	mov %rdi, (%rsp)
+	UNWIND_HINT_FUNC
+	RET
 SYM_FUNC_END(return_to_handler)
 #endif



  parent reply	other threads:[~2022-03-03 11:31 UTC|newest]

Thread overview: 64+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-03-03 11:23 [PATCH v3 00/39] x86: Kernel IBT Peter Zijlstra
2022-03-03 11:23 ` [PATCH v3 01/39] static_call: Avoid building empty .static_call_sites Peter Zijlstra
2022-03-03 11:23 ` [PATCH v3 02/39] x86/module: Fix the paravirt vs alternative order Peter Zijlstra
2022-03-08 13:58   ` [tip: x86/urgent] " tip-bot2 for Peter Zijlstra
2022-03-03 11:23 ` [PATCH v3 03/39] objtool: Add --dry-run Peter Zijlstra
2022-03-03 11:23 ` [PATCH v3 04/39] x86/ibt: Base IBT bits Peter Zijlstra
2022-03-07 11:17   ` Peter Zijlstra
2022-03-07 12:25     ` Joao Moreira
2022-03-07 18:19       ` Nick Desaulniers
2022-03-03 11:23 ` [PATCH v3 05/39] x86/ibt: Add ANNOTATE_NOENDBR Peter Zijlstra
2022-03-04 18:59   ` Josh Poimboeuf
2022-03-04 20:22     ` Peter Zijlstra
2022-03-03 11:23 ` [PATCH v3 06/39] x86/text-patching: Make text_gen_insn() play nice with ANNOTATE_NOENDBR Peter Zijlstra
2022-03-03 11:23 ` [PATCH v3 07/39] x86/ibt,paravirt: Use text_gen_insn() for paravirt_patch() Peter Zijlstra
2022-03-03 11:23 ` [PATCH v3 08/39] x86/entry: Cleanup PARAVIRT Peter Zijlstra
2022-03-03 11:23 ` [PATCH v3 09/39] x86/entry,xen: Early rewrite of restore_regs_and_return_to_kernel() Peter Zijlstra
2022-03-03 11:23 ` [PATCH v3 10/39] x86/ibt,xen: Sprinkle the ENDBR Peter Zijlstra
2022-03-03 11:23 ` [PATCH v3 11/39] x86/ibt,entry: Sprinkle ENDBR dust Peter Zijlstra
2022-03-03 11:23 ` [PATCH v3 12/39] x86/linkage: Add ENDBR to SYM_FUNC_START*() Peter Zijlstra
2022-03-03 11:23 ` [PATCH v3 13/39] x86/ibt,paravirt: Sprinkle ENDBR Peter Zijlstra
2022-03-03 11:23 ` [PATCH v3 14/39] x86/ibt,crypto: Add ENDBR for the jump-table entries Peter Zijlstra
2022-03-03 11:23 ` [PATCH v3 15/39] x86/ibt,kvm: Add ENDBR to fastops Peter Zijlstra
2022-03-03 11:23 ` [PATCH v3 16/39] x86/ibt,ftrace: Search for __fentry__ location Peter Zijlstra
2022-03-03 11:23 ` [PATCH v3 17/39] x86/livepatch: Validate " Peter Zijlstra
2022-03-03 11:23 ` Peter Zijlstra [this message]
2022-03-04 17:51   ` [PATCH v3 18/39] x86/ibt,ftrace: Make function-graph play nice Josh Poimboeuf
2022-03-04 19:48     ` Peter Zijlstra
2022-03-04 21:03       ` Josh Poimboeuf
2022-03-04 21:44         ` Peter Zijlstra
2022-03-04 22:05           ` Josh Poimboeuf
2022-03-04 23:08       ` David Laight
2022-03-03 11:23 ` [PATCH v3 19/39] x86/ibt,kprobes: Cure sym+0 equals fentry woes Peter Zijlstra
2022-03-07  9:44   ` Masami Hiramatsu
2022-03-03 11:23 ` [PATCH v3 20/39] x86/ibt,bpf: Add ENDBR instructions to prologue and trampoline Peter Zijlstra
2022-03-03 11:23 ` [PATCH v3 21/39] x86/ibt,ftrace: Add ENDBR to samples/ftrace Peter Zijlstra
2022-03-03 11:23 ` [PATCH v3 22/39] x86/ibt: Add IBT feature, MSR and #CP handling Peter Zijlstra
2022-03-04 17:57   ` Josh Poimboeuf
2022-03-04 20:38     ` Peter Zijlstra
2022-03-04 18:07   ` Josh Poimboeuf
2022-03-04 20:39     ` Peter Zijlstra
2022-03-03 11:23 ` [PATCH v3 23/39] x86/alternative: Simplify int3_selftest_ip Peter Zijlstra
2022-03-04 18:02   ` Josh Poimboeuf
2022-03-04 18:09     ` Josh Poimboeuf
2022-03-03 11:23 ` [PATCH v3 24/39] x86/ibt: Disable IBT around firmware Peter Zijlstra
2022-03-03 11:23 ` [PATCH v3 25/39] x86/bugs: Disable Retpoline when IBT Peter Zijlstra
2022-03-04 17:39   ` Josh Poimboeuf
2022-03-03 11:23 ` [PATCH v3 26/39] x86/ibt: Annotate text references Peter Zijlstra
2022-03-03 11:23 ` [PATCH v3 27/39] x86/ibt,ftrace: Annotate ftrace code patching Peter Zijlstra
2022-03-03 11:23 ` [PATCH v3 28/39] x86/ibt,sev: Annotations Peter Zijlstra
2022-03-03 11:23 ` [PATCH v3 29/39] x86/ibt: Dont generate ENDBR in .discard.text Peter Zijlstra
2022-03-03 11:23 ` [PATCH v3 30/39] x86/ibt: Ensure module init/exit points have references Peter Zijlstra
2022-03-03 11:23 ` [PATCH v3 31/39] objtool: Rename --duplicate to --lto Peter Zijlstra
2022-03-03 11:23 ` [PATCH v3 32/39] Kbuild: Allow whole module objtool runs Peter Zijlstra
2022-03-03 11:23 ` [PATCH v3 33/39] objtool: Read the NOENDBR annotation Peter Zijlstra
2022-03-03 11:23 ` [PATCH v3 34/39] objtool: Have WARN_FUNC fall back to sym+off Peter Zijlstra
2022-03-03 11:23 ` [PATCH v3 35/39] objtool: Add IBT/ENDBR decoding Peter Zijlstra
2022-03-03 11:23 ` [PATCH v3 36/39] objtool: Validate IBT assumptions Peter Zijlstra
2022-03-03 11:23 ` [PATCH v3 37/39] objtool: Optionally WARN about unused ANNOTATE_NOENDBR Peter Zijlstra
2022-03-04 18:27   ` Josh Poimboeuf
2022-03-03 11:23 ` [PATCH v3 38/39] objtool: Find unused ENDBR instructions Peter Zijlstra
2022-03-03 11:24 ` [PATCH v3 39/39] x86/alternative: Use .ibt_endbr_seal to seal indirect calls Peter Zijlstra
2022-03-04 19:09 ` [PATCH v3 00/39] x86: Kernel IBT Josh Poimboeuf
2022-03-04 20:22   ` Peter Zijlstra
2022-03-04 21:39   ` 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=20220303112826.044301664@infradead.org \
    --to=peterz@infradead.org \
    --cc=alexei.starovoitov@gmail.com \
    --cc=alyssa.milburn@intel.com \
    --cc=andrew.cooper3@citrix.com \
    --cc=hjl.tools@gmail.com \
    --cc=joao@overdrivepizza.com \
    --cc=jpoimboe@redhat.com \
    --cc=keescook@chromium.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mark.rutland@arm.com \
    --cc=mbenes@suse.cz \
    --cc=mhiramat@kernel.org \
    --cc=ndesaulniers@google.com \
    --cc=rostedt@goodmis.org \
    --cc=samitolvanen@google.com \
    --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;
as well as URLs for NNTP newsgroup(s).