Fei has reported that KASAN triggers during apply_alternatives() on
5-level paging machine:
BUG: KASAN: out-of-bounds in rcu_is_watching
Read of size 4 at addr ff110003ee6419a0 by task swapper/0/0
...
__asan_load4
rcu_is_watching
trace_hardirqs_on
text_poke_early
apply_alternatives
...
On machines with 5-level paging, cpu_feature_enabled(X86_FEATURE_LA57)
got patched. It includes KASAN code, where KASAN_SHADOW_START depends on
__VIRTUAL_MASK_SHIFT, which is defined with the cpu_feature_enabled().
KASAN gets confused when apply_alternatives() patches the
KASAN_SHADOW_START users. A test patch that makes KASAN_SHADOW_START
static, by replacing __VIRTUAL_MASK_SHIFT with 56, fixes the issue.
Disable KASAN while kernel patches alternatives.
Signed-off-by: Kirill A. Shutemov <[email protected]>
Reported-by: Fei Yang <[email protected]>
Fixes: 6657fca06e3f ("x86/mm: Allow to boot without LA57 if CONFIG_X86_5LEVEL=y")
Cc: [email protected]
---
v3:
- Summarize KASAN splat;
- Update comment in apply_alternatives();
v2:
- Move kasan_disable/_enable_current() to cover whole loop, not only
text_poke_early();
- Adjust commit message.
---
arch/x86/kernel/alternative.c | 13 +++++++++++++
1 file changed, 13 insertions(+)
diff --git a/arch/x86/kernel/alternative.c b/arch/x86/kernel/alternative.c
index 517ee01503be..73be3931e4f0 100644
--- a/arch/x86/kernel/alternative.c
+++ b/arch/x86/kernel/alternative.c
@@ -403,6 +403,17 @@ void __init_or_module noinline apply_alternatives(struct alt_instr *start,
u8 insn_buff[MAX_PATCH_LEN];
DPRINTK(ALT, "alt table %px, -> %px", start, end);
+
+ /*
+ * In the case CONFIG_X86_5LEVEL=y, KASAN_SHADOW_START is defined using
+ * cpu_feature_enabled(X86_FEATURE_LA57) and is therefore patched here.
+ * During the process, KASAN becomes confused seeing partial LA57
+ * conversion and triggers a false-positive out-of-bound report.
+ *
+ * Disable KASAN until the patching is complete.
+ */
+ kasan_disable_current();
+
/*
* The scan order should be from start to end. A later scanned
* alternative code can overwrite previously scanned alternative code.
@@ -452,6 +463,8 @@ void __init_or_module noinline apply_alternatives(struct alt_instr *start,
text_poke_early(instr, insn_buff, insn_buff_sz);
}
+
+ kasan_enable_current();
}
static inline bool is_jcc32(struct insn *insn)
--
2.41.0
The following commit has been merged into the x86/urgent branch of tip:
Commit-ID: d35652a5fc9944784f6f50a5c979518ff8dacf61
Gitweb: https://git.kernel.org/tip/d35652a5fc9944784f6f50a5c979518ff8dacf61
Author: Kirill A. Shutemov <[email protected]>
AuthorDate: Thu, 12 Oct 2023 13:04:24 +03:00
Committer: Ingo Molnar <[email protected]>
CommitterDate: Thu, 12 Oct 2023 20:27:16 +02:00
x86/alternatives: Disable KASAN in apply_alternatives()
Fei has reported that KASAN triggers during apply_alternatives() on
a 5-level paging machine:
BUG: KASAN: out-of-bounds in rcu_is_watching()
Read of size 4 at addr ff110003ee6419a0 by task swapper/0/0
...
__asan_load4()
rcu_is_watching()
trace_hardirqs_on()
text_poke_early()
apply_alternatives()
...
On machines with 5-level paging, cpu_feature_enabled(X86_FEATURE_LA57)
gets patched. It includes KASAN code, where KASAN_SHADOW_START depends on
__VIRTUAL_MASK_SHIFT, which is defined with cpu_feature_enabled().
KASAN gets confused when apply_alternatives() patches the
KASAN_SHADOW_START users. A test patch that makes KASAN_SHADOW_START
static, by replacing __VIRTUAL_MASK_SHIFT with 56, works around the issue.
Fix it for real by disabling KASAN while the kernel is patching alternatives.
[ mingo: updated the changelog ]
Fixes: 6657fca06e3f ("x86/mm: Allow to boot without LA57 if CONFIG_X86_5LEVEL=y")
Reported-by: Fei Yang <[email protected]>
Signed-off-by: Kirill A. Shutemov <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
Acked-by: Peter Zijlstra (Intel) <[email protected]>
Cc: Linus Torvalds <[email protected]>
Cc: [email protected]
Link: https://lore.kernel.org/r/[email protected]
---
arch/x86/kernel/alternative.c | 13 +++++++++++++
1 file changed, 13 insertions(+)
diff --git a/arch/x86/kernel/alternative.c b/arch/x86/kernel/alternative.c
index 517ee01..73be393 100644
--- a/arch/x86/kernel/alternative.c
+++ b/arch/x86/kernel/alternative.c
@@ -403,6 +403,17 @@ void __init_or_module noinline apply_alternatives(struct alt_instr *start,
u8 insn_buff[MAX_PATCH_LEN];
DPRINTK(ALT, "alt table %px, -> %px", start, end);
+
+ /*
+ * In the case CONFIG_X86_5LEVEL=y, KASAN_SHADOW_START is defined using
+ * cpu_feature_enabled(X86_FEATURE_LA57) and is therefore patched here.
+ * During the process, KASAN becomes confused seeing partial LA57
+ * conversion and triggers a false-positive out-of-bound report.
+ *
+ * Disable KASAN until the patching is complete.
+ */
+ kasan_disable_current();
+
/*
* The scan order should be from start to end. A later scanned
* alternative code can overwrite previously scanned alternative code.
@@ -452,6 +463,8 @@ void __init_or_module noinline apply_alternatives(struct alt_instr *start,
text_poke_early(instr, insn_buff, insn_buff_sz);
}
+
+ kasan_enable_current();
}
static inline bool is_jcc32(struct insn *insn)