diff options
| author | Pawan Gupta <pawan.kumar.gupta@linux.intel.com> | 2025-05-05 14:35:12 -0700 |
|---|---|---|
| committer | Greg Kroah-Hartman <gregkh@linuxfoundation.org> | 2025-05-18 08:21:25 +0200 |
| commit | db734ba733daf336e5815032244ff5ed96ec41c4 (patch) | |
| tree | b50958361c32008bd24c4e3b66fb3543fc873583 | |
| parent | cebc238b026de63479e7a43b9a7fc46896e1ce5a (diff) | |
| download | linux-db734ba733daf336e5815032244ff5ed96ec41c4.tar.gz linux-db734ba733daf336e5815032244ff5ed96ec41c4.tar.bz2 linux-db734ba733daf336e5815032244ff5ed96ec41c4.zip | |
x86/bhi: Do not set BHI_DIS_S in 32-bit mode
commit 073fdbe02c69c43fb7c0d547ec265c7747d4a646 upstream.
With the possibility of intra-mode BHI via cBPF, complete mitigation for
BHI is to use IBHF (history fence) instruction with BHI_DIS_S set. Since
this new instruction is only available in 64-bit mode, setting BHI_DIS_S in
32-bit mode is only a partial mitigation.
Do not set BHI_DIS_S in 32-bit mode so as to avoid reporting misleading
mitigated status. With this change IBHF won't be used in 32-bit mode, also
remove the CONFIG_X86_64 check from emit_spectre_bhb_barrier().
Suggested-by: Josh Poimboeuf <jpoimboe@kernel.org>
Signed-off-by: Pawan Gupta <pawan.kumar.gupta@linux.intel.com>
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Reviewed-by: Josh Poimboeuf <jpoimboe@kernel.org>
Reviewed-by: Alexandre Chartre <alexandre.chartre@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
| -rw-r--r-- | arch/x86/kernel/cpu/bugs.c | 5 | ||||
| -rw-r--r-- | arch/x86/net/bpf_jit_comp.c | 5 |
2 files changed, 6 insertions, 4 deletions
diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index 0be0edb07a2a..b22f72a177ba 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -1656,10 +1656,11 @@ static void __init bhi_select_mitigation(void) return; } - if (spec_ctrl_bhi_dis()) + if (!IS_ENABLED(CONFIG_X86_64)) return; - if (!IS_ENABLED(CONFIG_X86_64)) + /* Mitigate in hardware if supported */ + if (spec_ctrl_bhi_dis()) return; /* Mitigate KVM by default */ diff --git a/arch/x86/net/bpf_jit_comp.c b/arch/x86/net/bpf_jit_comp.c index e4ccc4136695..44b49a1fe579 100644 --- a/arch/x86/net/bpf_jit_comp.c +++ b/arch/x86/net/bpf_jit_comp.c @@ -971,8 +971,7 @@ static int emit_spectre_bhb_barrier(u8 **pprog, u8 *ip, /* Insert IBHF instruction */ if ((cpu_feature_enabled(X86_FEATURE_CLEAR_BHB_LOOP) && cpu_feature_enabled(X86_FEATURE_HYPERVISOR)) || - (cpu_feature_enabled(X86_FEATURE_CLEAR_BHB_HW) && - IS_ENABLED(CONFIG_X86_64))) { + cpu_feature_enabled(X86_FEATURE_CLEAR_BHB_HW)) { /* * Add an Indirect Branch History Fence (IBHF). IBHF acts as a * fence preventing branch history from before the fence from @@ -982,6 +981,8 @@ static int emit_spectre_bhb_barrier(u8 **pprog, u8 *ip, * hardware that doesn't need or support it. The REP and REX.W * prefixes are required by the microcode, and they also ensure * that the NOP is unlikely to be used in existing code. + * + * IBHF is not a valid instruction in 32-bit mode. */ EMIT5(0xF3, 0x48, 0x0F, 0x1E, 0xF8); /* ibhf */ } |