* [PATCH] x86/rust: use `-Zindirect-branch-cs-prefix` for RETPOLINE builds
@ 2026-10-18 17:54 Miguel Ojeda
2026-10-18 17:56 ` Miguel Ojeda
0 siblings, 1 reply; 2+ messages in thread
From: Miguel Ojeda @ 2026-10-18 17:54 UTC (permalink / raw)
To: Thomas Gleixner, Borislav Petkov, Peter Zijlstra, Josh Poimboeuf,
Ingo Molnar, Dave Hansen, x86, Sami Tolvanen, Kees Cook,
Miguel Ojeda, Nathan Chancellor, Nicolas Schier
Cc: Pawan Gupta, H. Peter Anvin, linux-kernel, llvm, Boqun Feng,
Gary Guo, Björn Roy Baron, Benno Lossin, Andreas Hindborg,
Alice Ryhl, Trevor Gross, Danilo Krummrich, Daniel Almeida,
Tamir Duberstein, Alexandre Courbot, Onur Özkan,
rust-for-linux, linux-kbuild, stable, Kent Overstreet
The C compiler is asked to add a `cs` prefix to the `call` and `jmp`
instructions that go through the indirect thunks for the `r8`-`r15`
registers, which makes them 6 bytes long instead of 5. That gives
`patch_retpoline()` enough room to inline them, e.g. as `lfence; call
*%r11` for `spectre_v2=retpoline,lfence`.
However, Rust code does not get the prefix, e.g. for a
`CONFIG_MITIGATION_RETPOLINE=y` build:
$ objdump -dr rust/kernel.o
...
e2b: e8 00 00 00 00 call e30 <...>
e2c: R_X86_64_PLT32 __x86_indirect_thunk_r11-0x4
Moreover, since commit 4a1e02b15ac1 ("x86,retpoline: Optimize
patch_retpoline()"), `emit_indirect()` assumes there is enough room
for the inlined sequence -- e.g. with `spectre_v2=retpoline,lfence`,
the kernel warns and dies while patching a site inside `core`:
WARNING: arch/x86/kernel/alternative.c:736 at apply_retpolines+0x2ee/0x400, CPU#0: swapper/0/0
...
Call Trace:
<TASK>
? _RNvMs6_NtNtCshS2CUL9MEDn_4core3fmt8buildersNtB5_9DebugList21finish_non_exhaustive+0x5b/0xd0
...
alternative_instructions+0x1d/0x60
arch_cpu_finalize_init+0x5a/0xb0
start_kernel+0x1f4/0x280
Similarly, paranoid FineIBT rewrites each kCFI call site into a fixed
20-byte sequence, i.e. the 14 bytes of the kCFI check plus a 6-byte
call. For a 5-byte call, that overwrites the first byte of the next
instruction, corrupting Rust code:
SMP alternatives: Using paranoid FineIBT CFI
...
Oops: double fault: 0000 [#1]
...
RIP: 0010:_RNvNtCsdVypMZsLRAh_4core3fmt5write+0x1e9/0x210
Code: ... b8 68 81 16 9f 41 3b 43 f5 2e 4d 8d 5b f0 75 fd 41 ff e3 <cc> c0 eb 02 b0 01 ...
...
Kernel panic - not syncing: Fatal exception in interrupt
Non-paranoid FineIBT is not affected, since it only rewrites the kCFI
check and leaves the call itself to `apply_retpolines()`.
We added support for `-Zindirect-branch-cs-prefix` [1][2] in Rust 1.91.0
[3]. Thus pass it for `RETPOLINE` builds when possible, which makes all
sites indeed get the prefix:
e3c: 2e e8 00 00 00 00 cs call e42 <...>
e3e: R_X86_64_PLT32 __x86_indirect_thunk_r11-0x4
The kernel above boots fine with `spectre_v2=retpoline,lfence`, inlining
the Rust sites as well.
For older Rust compilers, set the LLVM module flag instead, which
has the same code-generation effect. This means that every Rust version
that the kernel supports gets the prefix, that the kernel boots fine with
`spectre_v2=retpoline,lfence`, and that paranoid FineIBT no longer
corrupts.
In addition, add it to our target modifier workaround for `rustdoc`.
Cc: stable@vger.kernel.org # 6.18.y+
Link: https://github.com/rust-lang/rust/issues/116852 [1]
Link: https://github.com/rust-lang/compiler-team/issues/899 [2]
Link: https://github.com/rust-lang/rust/pull/140740 [3]
Reported-by: Kent Overstreet <kent.overstreet@linux.dev>
Assisted-by: LLM
Signed-off-by: Miguel Ojeda <ojeda@kernel.org>
---
arch/x86/Makefile | 1 +
rust/Makefile | 2 +-
2 files changed, 2 insertions(+), 1 deletion(-)
diff --git a/arch/x86/Makefile b/arch/x86/Makefile
index 8af6b80cffdd..01c848855f99 100644
--- a/arch/x86/Makefile
+++ b/arch/x86/Makefile
@@ -21,6 +21,7 @@ RETPOLINE_CFLAGS := -mretpoline-external-thunk
RETPOLINE_VDSO_CFLAGS := -mretpoline
endif
RETPOLINE_CFLAGS += $(call cc-option,-mindirect-branch-cs-prefix)
+RETPOLINE_RUSTFLAGS += $(if $(call rustc-min-version,109100),-Zindirect-branch-cs-prefix,-Zllvm_module_flag=indirect_branch_cs_prefix:u32:1:override)
ifdef CONFIG_MITIGATION_RETHUNK
RETHUNK_CFLAGS := -mfunction-return=thunk-extern
diff --git a/rust/Makefile b/rust/Makefile
index da1a7409d984..35868a47e18b 100644
--- a/rust/Makefile
+++ b/rust/Makefile
@@ -167,7 +167,7 @@ pin_init-flags := \
rustdoc_modifiers_workaround := $(if $(call rustc-min-version,108800),-Cunsafe-allow-abi-mismatch=fixed-x18)
# Similarly, for doctests (https://github.com/rust-lang/rust/issues/146465).
-doctests_modifiers_workaround := $(rustdoc_modifiers_workaround)$(if $(call rustc-min-version,109100),$(comma)sanitizer)
+doctests_modifiers_workaround := $(rustdoc_modifiers_workaround)$(if $(call rustc-min-version,109100),$(comma)sanitizer$(comma)indirect-branch-cs-prefix)
quiet_cmd_rustdoc = RUSTDOC $(if $(rustdoc_host),H, ) $<
cmd_rustdoc = \
base-commit: 24a19b953e2a3d0ea5a8c4951a0ab8f32720c907
--
2.56.0
^ permalink raw reply related [flat|nested] 2+ messages in thread* Re: [PATCH] x86/rust: use `-Zindirect-branch-cs-prefix` for RETPOLINE builds
2026-10-18 17:54 [PATCH] x86/rust: use `-Zindirect-branch-cs-prefix` for RETPOLINE builds Miguel Ojeda
@ 2026-10-18 17:56 ` Miguel Ojeda
0 siblings, 0 replies; 2+ messages in thread
From: Miguel Ojeda @ 2026-10-18 17:56 UTC (permalink / raw)
To: Miguel Ojeda
Cc: Thomas Gleixner, Borislav Petkov, Peter Zijlstra, Josh Poimboeuf,
Ingo Molnar, Dave Hansen, x86, Sami Tolvanen, Kees Cook,
Nathan Chancellor, Nicolas Schier, Pawan Gupta, H. Peter Anvin,
linux-kernel, llvm, Boqun Feng, Gary Guo, Björn Roy Baron,
Benno Lossin, Andreas Hindborg, Alice Ryhl, Trevor Gross,
Danilo Krummrich, Daniel Almeida, Tamir Duberstein,
Alexandre Courbot, Onur Özkan, rust-for-linux, linux-kbuild,
stable, Kent Overstreet
On Sun, Oct 18, 2026 at 7:55 PM Miguel Ojeda <ojeda@kernel.org> wrote:
>
> The C compiler is asked to add a `cs` prefix to the `call` and `jmp`
> instructions that go through the indirect thunks for the `r8`-`r15`
> registers, which makes them 6 bytes long instead of 5. That gives
> `patch_retpoline()` enough room to inline them, e.g. as `lfence; call
> *%r11` for `spectre_v2=retpoline,lfence`.
>
> However, Rust code does not get the prefix, e.g. for a
> `CONFIG_MITIGATION_RETPOLINE=y` build:
>
> $ objdump -dr rust/kernel.o
> ...
> e2b: e8 00 00 00 00 call e30 <...>
> e2c: R_X86_64_PLT32 __x86_indirect_thunk_r11-0x4
>
> Moreover, since commit 4a1e02b15ac1 ("x86,retpoline: Optimize
> patch_retpoline()"), `emit_indirect()` assumes there is enough room
> for the inlined sequence -- e.g. with `spectre_v2=retpoline,lfence`,
> the kernel warns and dies while patching a site inside `core`:
>
> WARNING: arch/x86/kernel/alternative.c:736 at apply_retpolines+0x2ee/0x400, CPU#0: swapper/0/0
> ...
> Call Trace:
> <TASK>
> ? _RNvMs6_NtNtCshS2CUL9MEDn_4core3fmt8buildersNtB5_9DebugList21finish_non_exhaustive+0x5b/0xd0
> ...
> alternative_instructions+0x1d/0x60
> arch_cpu_finalize_init+0x5a/0xb0
> start_kernel+0x1f4/0x280
>
> Similarly, paranoid FineIBT rewrites each kCFI call site into a fixed
> 20-byte sequence, i.e. the 14 bytes of the kCFI check plus a 6-byte
> call. For a 5-byte call, that overwrites the first byte of the next
> instruction, corrupting Rust code:
>
> SMP alternatives: Using paranoid FineIBT CFI
> ...
> Oops: double fault: 0000 [#1]
> ...
> RIP: 0010:_RNvNtCsdVypMZsLRAh_4core3fmt5write+0x1e9/0x210
> Code: ... b8 68 81 16 9f 41 3b 43 f5 2e 4d 8d 5b f0 75 fd 41 ff e3 <cc> c0 eb 02 b0 01 ...
> ...
> Kernel panic - not syncing: Fatal exception in interrupt
>
> Non-paranoid FineIBT is not affected, since it only rewrites the kCFI
> check and leaves the call itself to `apply_retpolines()`.
>
> We added support for `-Zindirect-branch-cs-prefix` [1][2] in Rust 1.91.0
> [3]. Thus pass it for `RETPOLINE` builds when possible, which makes all
> sites indeed get the prefix:
>
> e3c: 2e e8 00 00 00 00 cs call e42 <...>
> e3e: R_X86_64_PLT32 __x86_indirect_thunk_r11-0x4
>
> The kernel above boots fine with `spectre_v2=retpoline,lfence`, inlining
> the Rust sites as well.
>
> For older Rust compilers, set the LLVM module flag instead, which
> has the same code-generation effect. This means that every Rust version
> that the kernel supports gets the prefix, that the kernel boots fine with
> `spectre_v2=retpoline,lfence`, and that paranoid FineIBT no longer
> corrupts.
>
> In addition, add it to our target modifier workaround for `rustdoc`.
>
> Cc: stable@vger.kernel.org # 6.18.y+
> Link: https://github.com/rust-lang/rust/issues/116852 [1]
> Link: https://github.com/rust-lang/compiler-team/issues/899 [2]
> Link: https://github.com/rust-lang/rust/pull/140740 [3]
> Reported-by: Kent Overstreet <kent.overstreet@linux.dev>
> Assisted-by: LLM
> Signed-off-by: Miguel Ojeda <ojeda@kernel.org>
Not sure if `Fixes:` tags may be wanted here, since this was simply a
matter of not having added the support after we got it into the
upstream compiler, but two candidates are:
Fixes: 97e59672a9d2 ("x86/ibt: Add paranoid FineIBT mode")
Fixes: 4a1e02b15ac1 ("x86,retpoline: Optimize patch_retpoline()")
I hope this helps.
Cheers,
Miguel
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2026-10-18 17:57 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-10-18 17:54 [PATCH] x86/rust: use `-Zindirect-branch-cs-prefix` for RETPOLINE builds Miguel Ojeda
2026-10-18 17:56 ` Miguel Ojeda
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox