cpu/speculation: Add 'mitigations=' cmdline option
Subsystems (9)
v2 cover cpu/speculation: Add 'mitigations=' cmdline option
v2:
- docs improvements: [Randy, Michael]
- Rename to "mitigations=" [Michael]
- Add cpu_mitigations_off() function wrapper [Michael]
- x86: Simplify logic [Boris]
- powerpc: Fix no_rfi_flush checking bug (use '&&' instead of '||')
- arm64: Rebase onto Jeremy Linton's v7 patches [Will]
- arm64: "kpti command line option" [Steve P]
- arm64: Add nospectre_v2 support
-----------
Keeping track of the number of mitigations for all the CPU speculation bugs has become overwhelming for many users. It's getting more and more complicated to decide which mitigations are needed for a given architecture. Complicating matters is the fact that each arch tends to have its own custom way to mitigate the same vulnerability.
Most users fall into a few basic categories:
a) they want all mitigations off;
b) they want all reasonable mitigations on, with SMT enabled even if it's vulnerable; or
c) they want all reasonable mitigations on, with SMT disabled if vulnerable.
Define a set of curated, arch-independent options, each of which is an aggregation of existing options:
- mitigations=off: Disable all mitigations.
- mitigations=auto: [default] Enable all the default mitigations, but leave SMT enabled, even if it's vulnerable.
- mitigations=auto,nosmt: Enable all the default mitigations, disabling SMT if needed by a mitigation.
Josh Poimboeuf (5):
cpu/speculation: Add 'mitigations=' cmdline option x86/speculation: Support 'mitigations=' cmdline option powerpc/speculation: Support 'mitigations=' cmdline option s390/speculation: Support 'mitigations=' cmdline option arm64/speculation: Support 'mitigations=' cmdline option
.../admin-guide/kernel-parameters.txt | 34 +++++++++++++++++++ arch/arm64/kernel/cpu_errata.c | 6 +++- arch/arm64/kernel/cpufeature.c | 8 ++++- arch/powerpc/kernel/security.c | 6 ++-- arch/powerpc/kernel/setup_64.c | 2 +- arch/s390/kernel/nospec-branch.c | 3 +- arch/x86/kernel/cpu/bugs.c | 11 ++++-- arch/x86/mm/pti.c | 4 ++- include/linux/cpu.h | 24 +++++++++++++ kernel/cpu.c | 15 ++++++++ 10 files changed, 103 insertions(+), 10 deletions(-)
Signature · 1 line
2.17.2
PATCH v2 1/5 cpu/speculation: Add 'mitigations=' cmdline option
Keeping track of the number of mitigations for all the CPU speculation bugs has become overwhelming for many users. It's getting more and more complicated to decide which mitigations are needed for a given architecture. Complicating matters is the fact that each arch tends to have its own custom way to mitigate the same vulnerability.
Most users fall into a few basic categories:
a) they want all mitigations off;
b) they want all reasonable mitigations on, with SMT enabled even if it's vulnerable; or
c) they want all reasonable mitigations on, with SMT disabled if vulnerable.
Define a set of curated, arch-independent options, each of which is an aggregation of existing options:
- mitigations=off: Disable all mitigations.
- mitigations=auto: [default] Enable all the default mitigations, but leave SMT enabled, even if it's vulnerable.
- mitigations=auto,nosmt: Enable all the default mitigations, disabling SMT if needed by a mitigation.
Currently, these options are placeholders which don't actually do anything. They will be fleshed out in upcoming patches.
--- .../admin-guide/kernel-parameters.txt | 24 +++++++++++++++++++ include/linux/cpu.h | 24 +++++++++++++++++++ kernel/cpu.c | 15 ++++++++++++ 3 files changed, 63 insertions(+)
Documentation/admin-guide/kernel-parameters.txt +24 -0
@@ -2544,6 +2544,30 @@ in the "bleeding edge" mini2440 support kernel at http://repo.or.cz/w/linux-2.6/mini2440.git + mitigations= + Control optional mitigations for CPU vulnerabilities. + This is a set of curated, arch-independent options, each + of which is an aggregation of existing arch-specific + options. + + off + Disable all optional CPU mitigations. This + improves system performance, but it may also + expose users to several CPU vulnerabilities. + + auto (default) + Mitigate all CPU vulnerabilities, but leave SMT + enabled, even if it's vulnerable. This is for + users who don't want to be surprised by SMT + getting disabled across kernel upgrades, or who + have other ways of avoiding SMT-based attacks. + This is the default behavior. + + auto,nosmt + Mitigate all CPU vulnerabilities, disabling SMT + if needed. This is for users who always want to + be fully mitigated, even if it means losing SMT. + mminit_loglevel= [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this parameter allows control of the logging verbosity for
include/linux/cpu.h +24 -0
@@ -188,4 +188,28 @@ static inline void cpu_smt_disable(bool force) { }static inline void cpu_smt_check_topology (void ) { }#endif +/* +* These are used for a global "mitigations=" cmdline option for toggling +* optional CPU mitigations. +*/ +enum cpu_mitigations { +CPU_MITIGATIONS_OFF , +CPU_MITIGATIONS_AUTO , +CPU_MITIGATIONS_AUTO_NOSMT , +}; + +extern enum cpu_mitigations cpu_mitigations; + +/* mitigations=off */ +static inline bool cpu_mitigations_off (void ) +{ +return cpu_mitigations ==CPU_MITIGATIONS_OFF ; +} + +/* mitigations=auto,nosmt */ +static inline bool cpu_mitigations_auto_nosmt (void ) +{ +return cpu_mitigations ==CPU_MITIGATIONS_AUTO_NOSMT ; +} +#endif /* _LINUX_CPU_H_ */
kernel/cpu.c +15 -0
@@ -2320,3 +2320,18 @@ void __init boot_cpu_hotplug_init(void)#endif this_cpu_write (cpuhp_state.state,CPUHP_ONLINE ); } + +enum cpu_mitigations cpu_mitigations__ro_after_init =CPU_MITIGATIONS_AUTO ; + +static int __init mitigations_cmdline (char *arg) +{ +if (!strcmp (arg,"off" )) + cpu_mitigations =CPU_MITIGATIONS_OFF ; +else if (!strcmp (arg,"auto" )) + cpu_mitigations =CPU_MITIGATIONS_AUTO ; +else if (!strcmp (arg,"auto,nosmt" )) + cpu_mitigations =CPU_MITIGATIONS_AUTO_NOSMT ; + +return 0 ; +} +early_param ("mitigations" , mitigations_cmdline);
Signature · 1 line
2.17.2
PATCH v2 2/5 x86/speculation: Support 'mitigations=' cmdline option
In reply to Josh PoimboeufConfigure x86 runtime CPU speculation bug mitigations in accordance with the 'mitigations=' cmdline option. This affects Meltdown, Spectre v2, Speculative Store Bypass, and L1TF.
The default behavior is unchanged.
--- Documentation/admin-guide/kernel-parameters.txt | 16 +++++++++++----- arch/x86/kernel/cpu/bugs.c | 11 +++++++++-- arch/x86/mm/pti.c | 4 +++- 3 files changed, 23 insertions(+), 8 deletions(-)
Documentation/admin-guide/kernel-parameters.txt +11 -5
@@ -2545,15 +2545,20 @@ http://repo.or.cz/w/linux-2.6/mini2440.git mitigations= - Control optional mitigations for CPU vulnerabilities. - This is a set of curated, arch-independent options, each - of which is an aggregation of existing arch-specific - options. + [X86] Control optional mitigations for CPU + vulnerabilities. This is a set of curated, + arch-independent options, each of which is an + aggregation of existing arch-specific options. off Disable all optional CPU mitigations. This improves system performance, but it may also expose users to several CPU vulnerabilities. + Equivalent to: nopti [X86] + nospectre_v2 [X86] + spectre_v2_user=off [X86] + spec_store_bypass_disable=off [X86] + l1tf=off [X86] auto (default) Mitigate all CPU vulnerabilities, but leave SMT
@@ -2561,12 +2566,13 @@ users who don't want to be surprised by SMT getting disabled across kernel upgrades, or who have other ways of avoiding SMT-based attacks. - This is the default behavior. + Equivalent to: (default behavior) auto,nosmt Mitigate all CPU vulnerabilities, disabling SMT if needed. This is for users who always want to be fully mitigated, even if it means losing SMT. + Equivalent to: l1tf=flush,nosmt [X86] mminit_loglevel= [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this
arch/x86/kernel/cpu/bugs.c +9 -2
@@ -440,7 +440,8 @@ static enum spectre_v2_mitigation_cmd __init spectre_v2_parse_cmdline(void)char arg[20 ];int ret, i; -if (cmdline_find_option_bool (boot_command_line,"nospectre_v2" )) +if (cmdline_find_option_bool (boot_command_line,"nospectre_v2" ) || +cpu_mitigations_off ())return SPECTRE_V2_CMD_NONE ; ret =cmdline_find_option (boot_command_line,"spectre_v2" , arg,sizeof (arg));
@@ -672,7 +673,8 @@ static enum ssb_mitigation_cmd __init ssb_parse_cmdline(void)char arg[20 ];int ret, i; -if (cmdline_find_option_bool (boot_command_line,"nospec_store_bypass_disable" )) { +if (cmdline_find_option_bool (boot_command_line,"nospec_store_bypass_disable" ) || +cpu_mitigations_off ()) {return SPEC_STORE_BYPASS_CMD_NONE ; }else { ret = cmdline_find_option(boot_command_line, "spec_store_bypass_disable",
@@ -1008,6 +1010,11 @@ static void __init l1tf_select_mitigation(void)if (!boot_cpu_has_bug (X86_BUG_L1TF ))return ; +if (cpu_mitigations_off ()) + l1tf_mitigation =L1TF_MITIGATION_OFF ; +else if (cpu_mitigations_auto_nosmt ()) + l1tf_mitigation =L1TF_MITIGATION_FLUSH_NOSMT ; +override_cache_bits (&boot_cpu_data);switch (l1tf_mitigation) {
arch/x86/mm/pti.c +3 -1
@@ -35,6 +35,7 @@#include <linux/spinlock.h> #include <linux/mm.h> #include <linux/uaccess.h> +#include <linux/cpu.h> #include <asm/cpufeature.h> #include <asm/hypervisor.h>
@@ -115,7 +116,8 @@ void __init pti_check_boottime_disable(void) } } -if (cmdline_find_option_bool (boot_command_line,"nopti" )) { +if (cmdline_find_option_bool (boot_command_line,"nopti" ) || +cpu_mitigations_off ()) { pti_mode =PTI_FORCE_OFF ;pti_print_if_insecure ("disabled on command line." );return ;
Signature · 1 line
2.17.2
PATCH v2 3/5 powerpc/speculation: Support 'mitigations=' cmdline option
In reply to Josh PoimboeufConfigure powerpc CPU runtime speculation bug mitigations in accordance with the 'mitigations=' cmdline option. This affects Meltdown, Spectre v1, Spectre v2, and Speculative Store Bypass.
The default behavior is unchanged.
--- Documentation/admin-guide/kernel-parameters.txt | 9 +++++---- arch/powerpc/kernel/security.c | 6 +++--- arch/powerpc/kernel/setup_64.c | 2 +- 3 files changed, 9 insertions(+), 8 deletions(-)
Documentation/admin-guide/kernel-parameters.txt +5 -4
@@ -2545,7 +2545,7 @@ http://repo.or.cz/w/linux-2.6/mini2440.git mitigations= - [X86] Control optional mitigations for CPU + [X86,PPC] Control optional mitigations for CPU vulnerabilities. This is a set of curated, arch-independent options, each of which is an aggregation of existing arch-specific options.
@@ -2554,10 +2554,11 @@ Disable all optional CPU mitigations. This improves system performance, but it may also expose users to several CPU vulnerabilities. - Equivalent to: nopti [X86] - nospectre_v2 [X86] + Equivalent to: nopti [X86,PPC] + nospectre_v1 [PPC] + nospectre_v2 [X86,PPC] spectre_v2_user=off [X86] - spec_store_bypass_disable=off [X86] + spec_store_bypass_disable=off [X86,PPC] l1tf=off [X86] auto (default)
arch/powerpc/kernel/security.c +3 -3
@@ -57,7 +57,7 @@ void setup_barrier_nospec(void) enable =security_ftr_enabled (SEC_FTR_FAVOUR_SECURITY ) &&security_ftr_enabled (SEC_FTR_BNDS_CHK_SPEC_BAR ); -if (!no_nospec) +if (!no_nospec && !cpu_mitigations_off ())enable_barrier_nospec (enable); }
@@ -116,7 +116,7 @@ static int __init handle_nospectre_v2(char *p)early_param ("nospectre_v2" , handle_nospectre_v2);void setup_spectre_v2 (void ) { -if (no_spectrev2) +if (no_spectrev2 ||cpu_mitigations_off ())do_btb_flush_fixups ();else btb_flush_enabled = true;
@@ -300,7 +300,7 @@ void setup_stf_barrier(void) stf_enabled_flush_types = type; -if (!no_stf_barrier) +if (!no_stf_barrier && !cpu_mitigations_off ())stf_barrier_enable (enable); }
arch/powerpc/kernel/setup_64.c +1 -1
@@ -932,7 +932,7 @@ void setup_rfi_flush(enum l1d_flush_type types, bool enable) enabled_flush_types = types; -if (!no_rfi_flush) +if (!no_rfi_flush && !cpu_mitigations_off ())rfi_flush_enable (enable); }
Signature · 1 line
2.17.2
PATCH v2 4/5 s390/speculation: Support 'mitigations=' cmdline option
In reply to Josh PoimboeufConfigure s390 runtime CPU speculation bug mitigations in accordance with the 'mitigations=' cmdline option. This affects Spectre v1 and Spectre v2.
The default behavior is unchanged.
--- Documentation/admin-guide/kernel-parameters.txt | 5 +++-- arch/s390/kernel/nospec-branch.c | 3 ++- 2 files changed, 5 insertions(+), 3 deletions(-)
Documentation/admin-guide/kernel-parameters.txt +3 -2
@@ -2545,7 +2545,7 @@ http://repo.or.cz/w/linux-2.6/mini2440.git mitigations= - [X86,PPC] Control optional mitigations for CPU + [X86,PPC,S390] Control optional mitigations for CPU vulnerabilities. This is a set of curated, arch-independent options, each of which is an aggregation of existing arch-specific options.
@@ -2556,7 +2556,8 @@ expose users to several CPU vulnerabilities. Equivalent to: nopti [X86,PPC] nospectre_v1 [PPC] - nospectre_v2 [X86,PPC] + nobp=0 [S390] + nospectre_v2 [X86,PPC,S390] spectre_v2_user=off [X86] spec_store_bypass_disable=off [X86,PPC] l1tf=off [X86]
arch/s390/kernel/nospec-branch.c +2 -1
@@ -1,6 +1,7 @@// SPDX-License-Identifier: GPL-2.0 #include <linux/module.h> #include <linux/device.h> +#include <linux/cpu.h> #include <asm/nospec-branch.h> static int __init nobp_setup_early (char *str)
@@ -58,7 +59,7 @@ early_param("nospectre_v2", nospectre_v2_setup_early);void __init nospec_auto_detect (void ) { -if (test_facility (156 )) { +if (test_facility (156 ) ||cpu_mitigations_off ()) {/* * The machine supports etokens. * Disable expolines and disable nobp.
Signature · 1 line
2.17.2
PATCH v2 5/5 arm64/speculation: Support 'mitigations=' cmdline option
In reply to Josh PoimboeufConfigure arm64 runtime CPU speculation bug mitigations in accordance with the 'mitigations=' cmdline option. This affects Meltdown, Spectre v2, and Speculative Store Bypass.
The default behavior is unchanged.
---
NOTE: This is based on top of Jeremy Linton's patches:
https://lkml.kernel.org/r/[email protected]
Documentation/admin-guide/kernel-parameters.txt | 8 +++++--- arch/arm64/kernel/cpu_errata.c | 6 +++++- arch/arm64/kernel/cpufeature.c | 8 +++++++- 3 files changed, 17 insertions(+), 5 deletions(-)
Documentation/admin-guide/kernel-parameters.txt +5 -3
@@ -2545,8 +2545,8 @@ http://repo.or.cz/w/linux-2.6/mini2440.git mitigations= - [X86,PPC,S390] Control optional mitigations for CPU - vulnerabilities. This is a set of curated, + [X86,PPC,S390,ARM64] Control optional mitigations for + CPU vulnerabilities. This is a set of curated, arch-independent options, each of which is an aggregation of existing arch-specific options.
@@ -2555,11 +2555,13 @@ improves system performance, but it may also expose users to several CPU vulnerabilities. Equivalent to: nopti [X86,PPC] + kpti=0 [ARM64] nospectre_v1 [PPC] nobp=0 [S390] - nospectre_v2 [X86,PPC,S390] + nospectre_v2 [X86,PPC,S390,ARM64] spectre_v2_user=off [X86] spec_store_bypass_disable=off [X86,PPC] + ssbd=force-off [ARM64] l1tf=off [X86] auto (default)
arch/arm64/kernel/cpu_errata.c +5 -1
@@ -19,6 +19,7 @@#include <linux/arm-smccc.h> #include <linux/psci.h> #include <linux/types.h> +#include <linux/cpu.h> #include <asm/cpu.h> #include <asm/cputype.h> #include <asm/cpufeature.h>
@@ -405,6 +406,9 @@ static bool has_ssbd_mitigation(const struct arm64_cpu_capabilities *entry, this_cpu_safe = true; } +if (cpu_mitigations_off ()) + ssbd_state =ARM64_SSBD_FORCE_DISABLE ; +if (psci_ops.smccc_version ==SMCCC_VERSION_1_0 ) { ssbd_state =ARM64_SSBD_UNKNOWN ;if (!this_cpu_safe)
@@ -599,7 +603,7 @@ check_branch_predictor(const struct arm64_cpu_capabilities *entry, int scope) }/* forced off */ -if (__nospectre_v2) { +if (__nospectre_v2 ||cpu_mitigations_off ()) {pr_info_once ("spectrev2 mitigation disabled by command line option\n" ); __hardenbp_enab = false;return false;
arch/arm64/kernel/cpufeature.c +7 -1
@@ -25,6 +25,7 @@#include <linux/stop_machine.h> #include <linux/types.h> #include <linux/mm.h> +#include <linux/cpu.h> #include <asm/cpu.h> #include <asm/cpufeature.h> #include <asm/cpu_ops.h>
@@ -966,7 +967,7 @@ static bool unmap_kernel_at_el0(const struct arm64_cpu_capabilities *entry, MIDR_ALL_VERSIONS(MIDR_HISI_TSV110), {/* sentinel */ } }; -char const *str ="command line option" ; +char const *str ="kpti command line option" ;bool meltdown_safe; meltdown_safe =is_midr_in_range_list (read_cpuid_id (), kpti_safe_list);
@@ -988,6 +989,11 @@ static bool unmap_kernel_at_el0(const struct arm64_cpu_capabilities *entry, __kpti_forced =-1 ; } +if (cpu_mitigations_off () && !__kpti_forced) { + str ="mitigations=off" ; + __kpti_forced =-1 ; + } +/* Useful for KASLR robustness */ if (IS_ENABLED (CONFIG_RANDOMIZE_BASE ) &&kaslr_offset () >0 ) {if (!__kpti_forced) {
Signature · 1 line
2.17.2
Randy Dunlap
reviewerRe: [PATCH v2 5/5] arm64/speculation: Support 'mitigations=' cmdline option
Quoted 36 lines · Josh Poimboeuf
Configure arm64 runtime CPU speculation bug mitigations in accordance with the 'mitigations=' cmdline option. This affects Meltdown, Spectre v2, and Speculative Store Bypass.
The default behavior is unchanged.
Signed-off-by Josh Poimboeuf <[email protected]>---
NOTE: This is based on top of Jeremy Linton's patches:https://lkml.kernel.org/r/[email protected]Documentation/admin-guide/kernel-parameters.txt | 8 +++++--- arch/arm64/kernel/cpu_errata.c | 6 +++++- arch/arm64/kernel/cpufeature.c | 8 +++++++- 3 files changed, 17 insertions(+), 5 deletions(-)Documentation/admin-guide/kernel-parameters.txt +3
-2@@ -2545,8 +2545,8 @@ http://repo.or.cz/w/linux-2.6/mini2440.git mitigations= - [X86,PPC,S390] Control optional mitigations for CPU - vulnerabilities. This is a set of curated, + [X86,PPC,S390,ARM64] Control optional mitigations for + CPU vulnerabilities. This is a set of curated, arch-independent options, each of which is an aggregation of existing arch-specific options.@@ -2555,11 +2555,13 @@ improves system performance, but it may also expose users to several CPU vulnerabilities. Equivalent to: nopti [X86,PPC] + kpti=0 [ARM64]
Documentation/admin-guide/kernel-parameters.txt line 2567
Quoted 10 lines
Documentation/admin-guide/kernel-parameters.txt +2
-1nospectre_v1 [PPC] nobp=0 [S390] - nospectre_v2 [X86,PPC,S390] + nospectre_v2 [X86,PPC,S390,ARM64] spectre_v2_user=off [X86] spec_store_bypass_disable=off [X86,PPC] + ssbd=force-off [ARM64] l1tf=off [X86] auto (default)
Hi,
Do we need to add "ARM64" to Documentation/admin-guide/kernel-parameters.rst?
Signature · 1 line
~Randy
PATCH 1/1 Documentation: Add ARM64 to kernel-parameters.rst
Add ARM64 to the legend of architectures. It's already used in several places in kernel-parameters.txt.
--- Documentation/admin-guide/kernel-parameters.rst | 1 + 1 file changed, 1 insertion(+)
Documentation/admin-guide/kernel-parameters.rst +1 -0
@@ -88,6 +88,7 @@ parameter is applicable:: APIC APIC support is enabled. APM Advanced Power Management support is enabled. ARM ARM architecture is enabled. + ARM64 ARM64 architecture is enabled. AX25 Appropriate AX.25 support is enabled. CLK Common clock infrastructure is enabled. CMA Contiguous Memory Area support is enabled.
Signature · 1 line
2.17.2
Randy Dunlap
reviewerRe: [PATCH] Documentation: Add ARM64 to kernel-parameters.rst
Quoted 15 lines · Josh Poimboeuf
Add ARM64 to the legend of architectures. It's already used in several places in kernel-parameters.txt.
Suggested-by Randy Dunlap <[email protected]>Signed-off-by Josh Poimboeuf <[email protected]>--- Documentation/admin-guide/kernel-parameters.rst | 1 + 1 file changed, 1 insertion(+)Documentation/admin-guide/kernel-parameters.rst +0
-0
Documentation/admin-guide/kernel-parameters.rst parameter is applicable:: · line 94
Quoted 7 lines
Documentation/admin-guide/kernel-parameters.rst +1
-0@@ -88,6 +88,7 @@ parameter is applicable:: APIC APIC support is enabled. APM Advanced Power Management support is enabled. ARM ARM architecture is enabled. + ARM64 ARM64 architecture is enabled. AX25 Appropriate AX.25 support is enabled. CLK Common clock infrastructure is enabled. CMA Contiguous Memory Area support is enabled.
Thanks.
Signature · 1 line
~Randy
Quoted 39 lines · Josh Poimboeuf
v2:
- docs improvements: [Randy, Michael]
- Rename to "mitigations=" [Michael]
- Addcpu_mitigations_off()function wrapper [Michael]
- x86: Simplify logic [Boris]
- powerpc: Fixno_rfi_flushchecking bug (use '&&' instead of '||')
- arm64: Rebase onto Jeremy Linton's v7 patches [Will]
- arm64: "kpti command line option" [Steve P]
- arm64: Addnospectre_v2support-----------
Keeping track of the number of mitigations for all the CPU speculation bugs has become overwhelming for many users. It's getting more and more complicated to decide which mitigations are needed for a given architecture. Complicating matters is the fact that each arch tends to have its own custom way to mitigate the same vulnerability.
Most users fall into a few basic categories:
a) they want all mitigations off;
b) they want all reasonable mitigations on, with SMT enabled even if it's vulnerable; or
c) they want all reasonable mitigations on, with SMT disabled if vulnerable.
Define a set of curated, arch-independent options, each of which is an aggregation of existing options:
- mitigations=off: Disable all mitigations.
- mitigations=auto: [default] Enable all the default mitigations, but leave SMT enabled, even if it's vulnerable.
- mitigations=auto,nosmt: Enable all the default mitigations, disabling SMT if needed by a mitigation.
Quoted 6 lines
Josh Poimboeuf (5):
cpu/speculation: Add 'mitigations=' cmdline option x86/speculation: Support 'mitigations=' cmdline option powerpc/speculation: Support 'mitigations=' cmdline option s390/speculation: Support 'mitigations=' cmdline option arm64/speculation: Support 'mitigations=' cmdline option
Thanks,
Signature · 2 lines
Jiri Kosina SUSE Labs
Borislav Petkov
maintainerkernel/cpu.c void __init boot_cpu_hotplug_init(void) · line 2326
Quoted 12 lines · Josh Poimboeuf
kernel/cpu.c +4
-0@@ -2320,3 +2320,18 @@ void __init boot_cpu_hotplug_init(void)#endif this_cpu_write (cpuhp_state.state,CPUHP_ONLINE ); } + +enum cpu_mitigations cpu_mitigations__ro_after_init =CPU_MITIGATIONS_AUTO ; + +static int __init mitigations_cmdline (char *arg)
Forgot the verb: "mitigations_parse_cmdline".
Signature · 4 lines
Regards/Gruss,
Boris.
Good mailing practices for 400: avoid top-posting and trim the reply.Quoted 15 lines · Borislav Petkov
kernel/cpu.c void __init boot_cpu_hotplug_init(void) · line 2326
Quoted 12 lines · Josh Poimboeuf
kernel/cpu.c +4
-0@@ -2320,3 +2320,18 @@ void __init boot_cpu_hotplug_init(void)#endif this_cpu_write (cpuhp_state.state,CPUHP_ONLINE ); } + +enum cpu_mitigations cpu_mitigations__ro_after_init =CPU_MITIGATIONS_AUTO ; + +static int __init mitigations_cmdline (char *arg)Forgot the verb: "
mitigations_parse_cmdline".
Sure.
kernel/cpu.c +2 -2
@@ -2323,7 +2323,7 @@ void __init boot_cpu_hotplug_init(void)enum cpu_mitigations cpu_mitigations__ro_after_init =CPU_MITIGATIONS_AUTO ; -static int __init mitigations_cmdline (char *arg) +static int __init mitigations_parse_cmdline (char *arg) {if (!strcmp (arg,"off" )) cpu_mitigations =CPU_MITIGATIONS_OFF ;
@@ -2334,4 +2334,4 @@ static int __init mitigations_cmdline(char *arg)return 0 ; } -early_param ("mitigations" , mitigations_cmdline); +early_param ("mitigations" , mitigations_parse_cmdline);
Re: [PATCH v2 5/5] arm64/speculation: Support 'mitigations=' cmdline option
In reply to Josh PoimboeufQuoted 10 lines · Josh Poimboeuf
Configure arm64 runtime CPU speculation bug mitigations in accordance with the 'mitigations=' cmdline option. This affects Meltdown, Spectre v2, and Speculative Store Bypass.
The default behavior is unchanged.
Signed-off-by Josh Poimboeuf <[email protected]>---
NOTE: This is based on top of Jeremy Linton's patches:https://lkml.kernel.org/r/[email protected]
So I keep that out and we have to revisit that once the ARM64 stuff hits a tree, right? I can have a branch with just the 4 first patches applied which ARM64 folks can pull in when they apply Jeremy's patches before te merge window.
Thanks,
tglx
Re: [PATCH v2 5/5] arm64/speculation: Support 'mitigations=' cmdline option
Quoted 13 lines · Thomas Gleixner
Quoted 10 lines · Josh Poimboeuf
Configure arm64 runtime CPU speculation bug mitigations in accordance with the 'mitigations=' cmdline option. This affects Meltdown, Spectre v2, and Speculative Store Bypass.
The default behavior is unchanged.
Signed-off-by Josh Poimboeuf <[email protected]>---
NOTE: This is based on top of Jeremy Linton's patches:https://lkml.kernel.org/r/[email protected]
Quoted 4 lines
So I keep that out and we have to revisit that once the ARM64 stuff hits a tree, right? I can have a branch with just the 4 first patches applied which ARM64 folks can pull in when they apply Jeremy's patches before te merge window.
Sounds good to me (though I guess it's up to the arm64 maintainers how they want to handle the dependencies).
Signature · 1 line
Josh
Re: [PATCH v2 5/5] arm64/speculation: Support 'mitigations=' cmdline option
In reply to Thomas GleixnerQuoted 13 lines · Thomas Gleixner
Quoted 10 lines · Josh Poimboeuf
Configure arm64 runtime CPU speculation bug mitigations in accordance with the 'mitigations=' cmdline option. This affects Meltdown, Spectre v2, and Speculative Store Bypass.
The default behavior is unchanged.
Signed-off-by Josh Poimboeuf <[email protected]>---
NOTE: This is based on top of Jeremy Linton's patches:https://lkml.kernel.org/r/[email protected]
Quoted 4 lines
So I keep that out and we have to revisit that once the ARM64 stuff hits a tree, right? I can have a branch with just the 4 first patches applied which ARM64 folks can pull in when they apply Jeremy's patches before te merge window.
Yes, that would work for us, cheers. I should get to Jeremy's latest version next week and I'm certainly planning to get them queued up for 5.2.
Will
tip-bot for Josh Poimboeuf
merge botcpu/speculation: Add 'mitigations=' cmdline option
Commit-ID: 98af8452945c55652de68536afdde3b520fec429 Gitweb: https://git.kernel.org/tip/98af8452945c55652de68536afdde3b520fec429 Author: Josh Poimboeuf <[email protected]> AuthorDate: Fri, 12 Apr 2019 15:39:28 -0500 Committer: Thomas Gleixner <[email protected]> CommitDate: Wed, 17 Apr 2019 21:37:28 +0200
cpu/speculation: Add 'mitigations=' cmdline option
Keeping track of the number of mitigations for all the CPU speculation bugs has become overwhelming for many users. It's getting more and more complicated to decide which mitigations are needed for a given architecture. Complicating matters is the fact that each arch tends to have its own custom way to mitigate the same vulnerability.
Most users fall into a few basic categories:
a) they want all mitigations off;
b) they want all reasonable mitigations on, with SMT enabled even if it's vulnerable; or
c) they want all reasonable mitigations on, with SMT disabled if vulnerable.
Define a set of curated, arch-independent options, each of which is an aggregation of existing options:
- mitigations=off: Disable all mitigations.
- mitigations=auto: [default] Enable all the default mitigations, but leave SMT enabled, even if it's vulnerable.
- mitigations=auto,nosmt: Enable all the default mitigations, disabling SMT if needed by a mitigation.
Currently, these options are placeholders which don't actually do anything. They will be fleshed out in upcoming patches.
--- Documentation/admin-guide/kernel-parameters.txt | 24 ++++++++++++++++++++++++ include/linux/cpu.h | 24 ++++++++++++++++++++++++ kernel/cpu.c | 15 +++++++++++++++ 3 files changed, 63 insertions(+)
Documentation/admin-guide/kernel-parameters.txt +24 -0
@@ -2513,6 +2513,30 @@ in the "bleeding edge" mini2440 support kernel at http://repo.or.cz/w/linux-2.6/mini2440.git + mitigations= + Control optional mitigations for CPU vulnerabilities. + This is a set of curated, arch-independent options, each + of which is an aggregation of existing arch-specific + options. + + off + Disable all optional CPU mitigations. This + improves system performance, but it may also + expose users to several CPU vulnerabilities. + + auto (default) + Mitigate all CPU vulnerabilities, but leave SMT + enabled, even if it's vulnerable. This is for + users who don't want to be surprised by SMT + getting disabled across kernel upgrades, or who + have other ways of avoiding SMT-based attacks. + This is the default behavior. + + auto,nosmt + Mitigate all CPU vulnerabilities, disabling SMT + if needed. This is for users who always want to + be fully mitigated, even if it means losing SMT. + mminit_loglevel= [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this parameter allows control of the logging verbosity for
include/linux/cpu.h +24 -0
@@ -187,4 +187,28 @@ static inline void cpu_smt_disable(bool force) { }static inline void cpu_smt_check_topology (void ) { }#endif +/* +* These are used for a global "mitigations=" cmdline option for toggling +* optional CPU mitigations. +*/ +enum cpu_mitigations { +CPU_MITIGATIONS_OFF , +CPU_MITIGATIONS_AUTO , +CPU_MITIGATIONS_AUTO_NOSMT , +}; + +extern enum cpu_mitigations cpu_mitigations; + +/* mitigations=off */ +static inline bool cpu_mitigations_off (void ) +{ +return cpu_mitigations ==CPU_MITIGATIONS_OFF ; +} + +/* mitigations=auto,nosmt */ +static inline bool cpu_mitigations_auto_nosmt (void ) +{ +return cpu_mitigations ==CPU_MITIGATIONS_AUTO_NOSMT ; +} +#endif /* _LINUX_CPU_H_ */
kernel/cpu.c +15 -0
@@ -2279,3 +2279,18 @@ void __init boot_cpu_hotplug_init(void)#endif this_cpu_write (cpuhp_state.state,CPUHP_ONLINE ); } + +enum cpu_mitigations cpu_mitigations__ro_after_init =CPU_MITIGATIONS_AUTO ; + +static int __init mitigations_parse_cmdline (char *arg) +{ +if (!strcmp (arg,"off" )) + cpu_mitigations =CPU_MITIGATIONS_OFF ; +else if (!strcmp (arg,"auto" )) + cpu_mitigations =CPU_MITIGATIONS_AUTO ; +else if (!strcmp (arg,"auto,nosmt" )) + cpu_mitigations =CPU_MITIGATIONS_AUTO_NOSMT ; + +return 0 ; +} +early_param ("mitigations" , mitigations_parse_cmdline);
tip-bot for Josh Poimboeuf
merge bot[tip:core/speculation] x86/speculation: Support 'mitigations=' cmdline option
In reply to Josh Poimboeufx86/speculation: Support 'mitigations=' cmdline option
Commit-ID: d68be4c4d31295ff6ae34a8ddfaa4c1a8ff42812 Gitweb: https://git.kernel.org/tip/d68be4c4d31295ff6ae34a8ddfaa4c1a8ff42812 Author: Josh Poimboeuf <[email protected]> AuthorDate: Fri, 12 Apr 2019 15:39:29 -0500 Committer: Thomas Gleixner <[email protected]> CommitDate: Wed, 17 Apr 2019 21:37:28 +0200
x86/speculation: Support 'mitigations=' cmdline option
Configure x86 runtime CPU speculation bug mitigations in accordance with the 'mitigations=' cmdline option. This affects Meltdown, Spectre v2, Speculative Store Bypass, and L1TF.
The default behavior is unchanged.
--- Documentation/admin-guide/kernel-parameters.txt | 16 +++++++++++----- arch/x86/kernel/cpu/bugs.c | 11 +++++++++-- arch/x86/mm/pti.c | 4 +++- 3 files changed, 23 insertions(+), 8 deletions(-)
Documentation/admin-guide/kernel-parameters.txt +11 -5
@@ -2514,15 +2514,20 @@ http://repo.or.cz/w/linux-2.6/mini2440.git mitigations= - Control optional mitigations for CPU vulnerabilities. - This is a set of curated, arch-independent options, each - of which is an aggregation of existing arch-specific - options. + [X86] Control optional mitigations for CPU + vulnerabilities. This is a set of curated, + arch-independent options, each of which is an + aggregation of existing arch-specific options. off Disable all optional CPU mitigations. This improves system performance, but it may also expose users to several CPU vulnerabilities. + Equivalent to: nopti [X86] + nospectre_v2 [X86] + spectre_v2_user=off [X86] + spec_store_bypass_disable=off [X86] + l1tf=off [X86] auto (default) Mitigate all CPU vulnerabilities, but leave SMT
@@ -2530,12 +2535,13 @@ users who don't want to be surprised by SMT getting disabled across kernel upgrades, or who have other ways of avoiding SMT-based attacks. - This is the default behavior. + Equivalent to: (default behavior) auto,nosmt Mitigate all CPU vulnerabilities, disabling SMT if needed. This is for users who always want to be fully mitigated, even if it means losing SMT. + Equivalent to: l1tf=flush,nosmt [X86] mminit_loglevel= [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this
arch/x86/kernel/cpu/bugs.c +9 -2
@@ -440,7 +440,8 @@ static enum spectre_v2_mitigation_cmd __init spectre_v2_parse_cmdline(void)char arg[20 ];int ret, i; -if (cmdline_find_option_bool (boot_command_line,"nospectre_v2" )) +if (cmdline_find_option_bool (boot_command_line,"nospectre_v2" ) || +cpu_mitigations_off ())return SPECTRE_V2_CMD_NONE ; ret =cmdline_find_option (boot_command_line,"spectre_v2" , arg,sizeof (arg));
@@ -672,7 +673,8 @@ static enum ssb_mitigation_cmd __init ssb_parse_cmdline(void)char arg[20 ];int ret, i; -if (cmdline_find_option_bool (boot_command_line,"nospec_store_bypass_disable" )) { +if (cmdline_find_option_bool (boot_command_line,"nospec_store_bypass_disable" ) || +cpu_mitigations_off ()) {return SPEC_STORE_BYPASS_CMD_NONE ; }else { ret = cmdline_find_option(boot_command_line, "spec_store_bypass_disable",
@@ -996,6 +998,11 @@ static void __init l1tf_select_mitigation(void)if (!boot_cpu_has_bug (X86_BUG_L1TF ))return ; +if (cpu_mitigations_off ()) + l1tf_mitigation =L1TF_MITIGATION_OFF ; +else if (cpu_mitigations_auto_nosmt ()) + l1tf_mitigation =L1TF_MITIGATION_FLUSH_NOSMT ; +override_cache_bits (&boot_cpu_data);switch (l1tf_mitigation) {
arch/x86/mm/pti.c +3 -1
@@ -35,6 +35,7 @@#include <linux/spinlock.h> #include <linux/mm.h> #include <linux/uaccess.h> +#include <linux/cpu.h> #include <asm/cpufeature.h> #include <asm/hypervisor.h>
@@ -115,7 +116,8 @@ void __init pti_check_boottime_disable(void) } } -if (cmdline_find_option_bool (boot_command_line,"nopti" )) { +if (cmdline_find_option_bool (boot_command_line,"nopti" ) || +cpu_mitigations_off ()) { pti_mode =PTI_FORCE_OFF ;pti_print_if_insecure ("disabled on command line." );return ;
tip-bot for Josh Poimboeuf
merge bot[tip:core/speculation] powerpc/speculation: Support 'mitigations=' cmdline option
In reply to Josh Poimboeufpowerpc/speculation: Support 'mitigations=' cmdline option
Commit-ID: 782e69efb3dfed6e8360bc612e8c7827a901a8f9 Gitweb: https://git.kernel.org/tip/782e69efb3dfed6e8360bc612e8c7827a901a8f9 Author: Josh Poimboeuf <[email protected]> AuthorDate: Fri, 12 Apr 2019 15:39:30 -0500 Committer: Thomas Gleixner <[email protected]> CommitDate: Wed, 17 Apr 2019 21:37:29 +0200
powerpc/speculation: Support 'mitigations=' cmdline option
Configure powerpc CPU runtime speculation bug mitigations in accordance with the 'mitigations=' cmdline option. This affects Meltdown, Spectre v1, Spectre v2, and Speculative Store Bypass.
The default behavior is unchanged.
--- Documentation/admin-guide/kernel-parameters.txt | 9 +++++---- arch/powerpc/kernel/security.c | 6 +++--- arch/powerpc/kernel/setup_64.c | 2 +- 3 files changed, 9 insertions(+), 8 deletions(-)
Documentation/admin-guide/kernel-parameters.txt +5 -4
@@ -2514,7 +2514,7 @@ http://repo.or.cz/w/linux-2.6/mini2440.git mitigations= - [X86] Control optional mitigations for CPU + [X86,PPC] Control optional mitigations for CPU vulnerabilities. This is a set of curated, arch-independent options, each of which is an aggregation of existing arch-specific options.
@@ -2523,10 +2523,11 @@ Disable all optional CPU mitigations. This improves system performance, but it may also expose users to several CPU vulnerabilities. - Equivalent to: nopti [X86] - nospectre_v2 [X86] + Equivalent to: nopti [X86,PPC] + nospectre_v1 [PPC] + nospectre_v2 [X86,PPC] spectre_v2_user=off [X86] - spec_store_bypass_disable=off [X86] + spec_store_bypass_disable=off [X86,PPC] l1tf=off [X86] auto (default)
arch/powerpc/kernel/security.c +3 -3
@@ -57,7 +57,7 @@ void setup_barrier_nospec(void) enable =security_ftr_enabled (SEC_FTR_FAVOUR_SECURITY ) &&security_ftr_enabled (SEC_FTR_BNDS_CHK_SPEC_BAR ); -if (!no_nospec) +if (!no_nospec && !cpu_mitigations_off ())enable_barrier_nospec (enable); }
@@ -116,7 +116,7 @@ static int __init handle_nospectre_v2(char *p)early_param ("nospectre_v2" , handle_nospectre_v2);void setup_spectre_v2 (void ) { -if (no_spectrev2) +if (no_spectrev2 ||cpu_mitigations_off ())do_btb_flush_fixups ();else btb_flush_enabled = true;
@@ -307,7 +307,7 @@ void setup_stf_barrier(void) stf_enabled_flush_types = type; -if (!no_stf_barrier) +if (!no_stf_barrier && !cpu_mitigations_off ())stf_barrier_enable (enable); }
arch/powerpc/kernel/setup_64.c +1 -1
@@ -958,7 +958,7 @@ void setup_rfi_flush(enum l1d_flush_type types, bool enable) enabled_flush_types = types; -if (!no_rfi_flush) +if (!no_rfi_flush && !cpu_mitigations_off ())rfi_flush_enable (enable); }
tip-bot for Josh Poimboeuf
merge bot[tip:core/speculation] s390/speculation: Support 'mitigations=' cmdline option
In reply to Josh Poimboeufs390/speculation: Support 'mitigations=' cmdline option
Commit-ID: 0336e04a6520bdaefdb0769d2a70084fa52e81ed Gitweb: https://git.kernel.org/tip/0336e04a6520bdaefdb0769d2a70084fa52e81ed Author: Josh Poimboeuf <[email protected]> AuthorDate: Fri, 12 Apr 2019 15:39:31 -0500 Committer: Thomas Gleixner <[email protected]> CommitDate: Wed, 17 Apr 2019 21:37:29 +0200
s390/speculation: Support 'mitigations=' cmdline option
Configure s390 runtime CPU speculation bug mitigations in accordance with the 'mitigations=' cmdline option. This affects Spectre v1 and Spectre v2.
The default behavior is unchanged.
--- Documentation/admin-guide/kernel-parameters.txt | 5 +++-- arch/s390/kernel/nospec-branch.c | 3 ++- 2 files changed, 5 insertions(+), 3 deletions(-)
Documentation/admin-guide/kernel-parameters.txt +3 -2
@@ -2514,7 +2514,7 @@ http://repo.or.cz/w/linux-2.6/mini2440.git mitigations= - [X86,PPC] Control optional mitigations for CPU + [X86,PPC,S390] Control optional mitigations for CPU vulnerabilities. This is a set of curated, arch-independent options, each of which is an aggregation of existing arch-specific options.
@@ -2525,7 +2525,8 @@ expose users to several CPU vulnerabilities. Equivalent to: nopti [X86,PPC] nospectre_v1 [PPC] - nospectre_v2 [X86,PPC] + nobp=0 [S390] + nospectre_v2 [X86,PPC,S390] spectre_v2_user=off [X86] spec_store_bypass_disable=off [X86,PPC] l1tf=off [X86]
arch/s390/kernel/nospec-branch.c +2 -1
@@ -1,6 +1,7 @@// SPDX-License-Identifier: GPL-2.0 #include <linux/module.h> #include <linux/device.h> +#include <linux/cpu.h> #include <asm/nospec-branch.h> static int __init nobp_setup_early (char *str)
@@ -58,7 +59,7 @@ early_param("nospectre_v2", nospectre_v2_setup_early);void __init nospec_auto_detect (void ) { -if (test_facility (156 )) { +if (test_facility (156 ) ||cpu_mitigations_off ()) {/* * The machine supports etokens. * Disable expolines and disable nobp.
Re: [PATCH v2 5/5] arm64/speculation: Support 'mitigations=' cmdline option
In reply to Thomas GleixnerHi Thomas,
Quoted 13 lines · Thomas Gleixner
Quoted 10 lines · Josh Poimboeuf
Configure arm64 runtime CPU speculation bug mitigations in accordance with the 'mitigations=' cmdline option. This affects Meltdown, Spectre v2, and Speculative Store Bypass.
The default behavior is unchanged.
Signed-off-by Josh Poimboeuf <[email protected]>---
NOTE: This is based on top of Jeremy Linton's patches:https://lkml.kernel.org/r/[email protected]
Quoted 4 lines
So I keep that out and we have to revisit that once the ARM64 stuff hits a tree, right? I can have a branch with just the 4 first patches applied which ARM64 folks can pull in when they apply Jeremy's patches before te merge window.
I'm assuming that this refers to the core/speculation branch in tip:
https://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git/log/?h=core/speculation
but please can you confirm that I'm good to pull that into arm64?
Cheers,
Will
Re: [PATCH v2 5/5] arm64/speculation: Support 'mitigations=' cmdline option
Quoted 21 lines · Will Deacon
Hi Thomas,
Quoted 13 lines · Thomas Gleixner
Quoted 10 lines · Josh Poimboeuf
Configure arm64 runtime CPU speculation bug mitigations in accordance with the 'mitigations=' cmdline option. This affects Meltdown, Spectre v2, and Speculative Store Bypass.
The default behavior is unchanged.
Signed-off-by Josh Poimboeuf <[email protected]>---
NOTE: This is based on top of Jeremy Linton's patches:https://lkml.kernel.org/r/[email protected]Quoted 4 lines
So I keep that out and we have to revisit that once the ARM64 stuff hits a tree, right? I can have a branch with just the 4 first patches applied which ARM64 folks can pull in when they apply Jeremy's patches before te merge window.
Quoted 5 lines
I'm assuming that this refers to the core/speculation branch in tip:
https://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git/log/?h=core/speculation
but please can you confirm that I'm good to pull that into arm64?
Yes. It's all yours :)
Thanks,
tglx
Jonathan Corbet
maintainerRe: [PATCH] Documentation: Add ARM64 to kernel-parameters.rst
In reply to Josh PoimboeufQuoted 5 lines · On Fri, 12 Apr 2019 22:56:21 -0500 Josh Poimboeuf <[email protected]> wrote:
Add ARM64 to the legend of architectures. It's already used in several places in kernel-parameters.txt.
Suggested-by Randy Dunlap <[email protected]>Signed-off-by Josh Poimboeuf <[email protected]>
It looks like nobody has picked this up...so I've applied it.
Thanks,
jon
Re: [PATCH] Documentation: Add ARM64 to kernel-parameters.rst
Quoted 10 lines · Jonathan Corbet
Quoted 5 lines · On Fri, 12 Apr 2019 22:56:21 -0500 Josh Poimboeuf <[email protected]> wrote:
Add ARM64 to the legend of architectures. It's already used in several places in kernel-parameters.txt.
Suggested-by Randy Dunlap <[email protected]>Signed-off-by Josh Poimboeuf <[email protected]>It looks like nobody has picked this up...so I've applied it.
It's queued and tagged in the arm64 tree, which should also be in next!
Will
Jonathan Corbet
maintainerRe: [PATCH] Documentation: Add ARM64 to kernel-parameters.rst
Quoted 3 lines · On Fri, 3 May 2019 13:39:40 +0100 Will Deacon <[email protected]> wrote:
Quoted 1 line
It looks like nobody has picked this up...so I've applied it.
It's queued and tagged in the arm64 tree, which should also be in next!
Just looked again, I still don't see it there. Josh's mitigations= change is there, but not this one. In any case, I've unapplied it, so it's all yours.
Thanks,
jon
Re: [PATCH] Documentation: Add ARM64 to kernel-parameters.rst
Quoted 10 lines · Jonathan Corbet
Quoted 3 lines · On Fri, 3 May 2019 13:39:40 +0100 Will Deacon <[email protected]> wrote:
Quoted 1 line
It looks like nobody has picked this up...so I've applied it.
It's queued and tagged in the arm64 tree, which should also be in next!
Just looked again, I still don't see it there. Josh's mitigations= change is there, but not this one. In any case, I've unapplied it, so it's all yours.
Weird... I see it in -next as 4ad499c94264:
Will