| /linux-6.12.1/Documentation/admin-guide/hw-vuln/ |
| D | gather_data_sampling.rst | 29 Without mitigation, GDS can infer stale data across virtually all
38 Because of this, it is important to ensure that the mitigation stays enabled in
41 The hardware enforces the mitigation for SGX. Likewise, VMMs should ensure
42 that guests are not allowed to disable the GDS mitigation. If a host erred and
43 allowed this, a guest could theoretically disable GDS mitigation, mount an
46 Mitigation mechanism
53 and mitigation support.
55 IA32_MCU_OPT_CTRL[GDS_MITG_DIS] R/W Disables the mitigation
71 Mitigation control on the kernel command line
73 The mitigation can be disabled by setting "gather_data_sampling=off" or
[all …]
|
| D | srso.rst | 6 This is a mitigation for the speculative return stack overflow (SRSO)
39 The sysfs file showing SRSO mitigation status is:
60 The "Safe RET" mitigation (see below) has been applied to protect the
70 Note that User->User mitigation is controlled by how the IBPB aspect in
71 the Spectre v2 mitigation is selected:
85 * 'Mitigation: Safe RET':
87 Combined microcode/software mitigation. It complements the
93 * 'Mitigation: IBPB':
100 * 'Mitigation: IBPB on VMEXIT':
102 Mitigation addressing the cloud provider scenario - the Guest->Host
[all …]
|
| D | mds.rst | 26 Not all processors are affected by all variants of MDS, but the mitigation
103 - The processor is vulnerable, but no mitigation enabled
106 mitigation is enabled on a best effort basis.
109 based mitigation mechanism is not advertised via CPUID, the kernel
110 selects a best effort mitigation mode. This mode invokes the mitigation
118 * - 'Mitigation: Clear CPU buffers'
119 - The processor is vulnerable and the CPU buffer clearing mitigation is
132 Mitigation mechanism
139 enables the mitigation by default. The mitigation can be controlled at boot
148 The mitigation for MDS clears the affected CPU buffers on return to user
[all …]
|
| D | special-register-buffer-data-sampling.rst | 62 Mitigation mechanism
87 the mitigation for RDRAND and RDSEED instructions executed outside of Intel
89 disable the mitigation using this opt-out mechanism, RDRAND and RDSEED do not
97 Along with the mitigation for this issue, Intel added a new thread-scope
103 disables the mitigation for RDRAND and RDSEED executed outside of an Intel SGX
104 enclave on that logical processor. Opting out of the mitigation for a
108 Note that inside of an Intel SGX enclave, the mitigation is applied regardless
111 Mitigation control on the kernel command line
113 The kernel command line allows control over the SRBDS mitigation at boot time
117 off This option disables SRBDS mitigation for RDRAND and RDSEED on
[all …]
|
| D | reg-file-data-sampling.rst | 40 Mitigation chapter
44 mitigation strategy to force the CPU to clear the affected buffers before an
50 Mitigation points
60 vulnerability and mitigation capability:
66 Mitigation control on the kernel command line
68 The kernel command line allows to control RFDS mitigation at boot time with the
72 on If the CPU is vulnerable, enable mitigation; CPU buffer clearing
74 off Disables mitigation.
77 Mitigation default is selected by CONFIG_MITIGATION_RFDS.
79 Mitigation status information
[all …]
|
| D | processor_mmio_stale_data.rst | 10 provided to untrusted guests may need mitigation. These vulnerabilities are
13 by an unmitigated transient execution attack. Mitigation for these
110 section, mitigation largely remains the same for all the variants, i.e. to
117 specific variants of Processor MMIO Stale Data vulnerabilities and mitigation
150 Mitigation chapter
153 same mitigation strategy to force the CPU to clear the affected buffers before
166 additional mitigation is needed on such CPUs.
168 For CPUs not affected by MDS or TAA, mitigation is needed only for the attacker
173 Mitigation points
177 Same mitigation as MDS when affected by MDS/TAA, otherwise no mitigation
[all …]
|
| D | tsx_async_abort.rst | 99 …- The CPU is affected by this vulnerability and the microcode and kernel mitigation are not applie…
102 mitigation is enabled on a best effort basis.
105 based mitigation mechanism is not advertised via CPUID, the kernel
106 selects a best effort mitigation mode. This mode invokes the mitigation
114 * - 'Mitigation: Clear CPU buffers'
116 * - 'Mitigation: TSX disabled'
121 Mitigation mechanism
126 enables the mitigation by default.
129 The mitigation can be controlled at boot time via a kernel command line option.
132 Virtualization mitigation
[all …]
|
| D | multihit.rst | 81 * - KVM: Mitigation: Split huge pages
83 * - KVM: Mitigation: VMX unsupported
85 * - KVM: Mitigation: VMX disabled
88 - The processor is vulnerable, but no mitigation enabled
104 Mitigation mechanism
122 Mitigation control on the kernel command line and KVM - module parameter
125 The KVM hypervisor mitigation mechanism for marking huge pages as
133 force Mitigation is enabled. In this case, the mitigation implements
139 off Mitigation is disabled.
141 auto Enable mitigation only if the platform is affected and the kernel
[all …]
|
| D | spectre.rst | 143 For a full mitigation against BHB attacks it is recommended to set BHI_DIS_S or
330 mitigation status of the system for Spectre: whether the system is
333 The sysfs file showing Spectre variant 1 mitigation status is:
347 * - 'Mitigation: usercopy/swapgs barriers and __user pointer sanitization'
357 retpoline mitigation or if the CPU has hardware mitigation, and if the
358 CPU has support for additional process-specific mitigation.
371 The sysfs file showing Spectre variant 2 mitigation status is:
381 'Mitigation: None' Vulnerable, no mitigation
382 'Mitigation: Retpolines' Use Retpoline thunks
383 'Mitigation: LFENCE' Use LFENCE instructions
[all …]
|
| D | l1tf.rst | 78 The Linux kernel contains a mitigation for this attack vector, PTE
92 PTE inversion mitigation for L1TF, to attack physical host memory.
132 'Mitigation: PTE Inversion' The host protection is active
136 information is appended to the 'Mitigation: PTE Inversion' part:
158 Host mitigation mechanism
165 Guest mitigation mechanisms
282 of other mitigation solutions like confining guests to dedicated cores.
351 Disabling EPT for virtual machines provides full mitigation for L1TF even
354 significant performance impact especially when the Meltdown mitigation
359 There is ongoing research and development for new mitigation mechanisms to
[all …]
|
| D | cross-thread-rsb.rst | 65 Mitigation mechanism
77 Mitigation control on the kernel command line
82 Mitigation control for KVM - module parameter
88 mitigation that covers this path is not enabled by default.
90 The mitigation for the KVM_CAP_X86_DISABLE_EXITS capability can be turned on
|
| D | l1d_flush.rst | 31 Mitigation chapter
39 mechanism is used, software fallback for the mitigation, is not supported.
41 Mitigation control on the kernel command line
63 cores or by disabling SMT. See the relevant chapter in the L1TF mitigation
|
| /linux-6.12.1/arch/x86/kernel/cpu/ |
| D | bugs.c | 199 * If the host has SSBD mitigation enabled, force it in the host's in x86_virt_spec_ctrl()
235 /* Default mitigation for MDS-affected CPUs */
242 [MDS_MITIGATION_FULL] = "Mitigation: Clear CPU buffers",
296 /* Default mitigation for TAA-affected CPUs */
304 [TAA_MITIGATION_VERW] = "Mitigation: Clear CPU buffers",
305 [TAA_MITIGATION_TSX_DISABLED] = "Mitigation: TSX disabled",
327 * TAA mitigation via VERW is turned off if both in taa_select_mitigation()
353 * TSX is enabled, select alternate mitigation for TAA which is in taa_select_mitigation()
357 * present on host, enable the mitigation for UCODE_NEEDED as well. in taa_select_mitigation()
395 /* Default mitigation for Processor MMIO Stale Data vulnerabilities */
[all …]
|
| /linux-6.12.1/Documentation/arch/x86/ |
| D | mds.rst | 1 Microarchitectural Data Sampling (MDS) mitigation
70 Mitigation strategy
73 All variants have the same mitigation strategy at least for the single CPU
82 command. The latter is issued when L1TF mitigation is enabled so the extra
101 The mitigation is invoked on kernel/userspace, hypervisor/guest and C-state
115 Kernel internal mitigation modes
119 off Mitigation is disabled. Either the CPU is not affected or
122 full Mitigation is enabled. CPU is affected and MD_CLEAR is
125 vmwerv Mitigation is enabled. CPU is affected and MD_CLEAR is not
133 line then the kernel selects the appropriate mitigation mode depending on
[all …]
|
| D | tsx_async_abort.rst | 3 TSX Async Abort (TAA) mitigation
21 Mitigation strategy
33 Kernel internal mitigation modes
37 off Mitigation is disabled. Either the CPU is not affected or
40 tsx disabled Mitigation is enabled. TSX feature is disabled by default at
43 verw Mitigation is enabled. CPU is affected and MD_CLEAR is
46 ucode needed Mitigation is enabled. CPU is affected and MD_CLEAR is not
54 not provided then the kernel selects an appropriate mitigation depending on the
58 TAA mitigation, VERW behavior and TSX feature for various combinations of
66 …A_NO MDS_NO TSX_CTRL_MSR TSX state VERW can clear TAA mitigation TAA mitigation
[all …]
|
| /linux-6.12.1/arch/arm64/kernel/ |
| D | proton-pack.c | 3 * Handle detection, reporting and mitigation of Spectre v1, v2, v3a and v4, as
36 * We try to ensure that the mitigation state can never change as the result of
64 return sprintf(buf, "Mitigation: __user pointer sanitization\n"); in cpu_show_spectre_v1()
97 pr_info_once("spectre-v2 mitigation disabled by command line option\n"); in spectre_v2_mitigations_off()
146 return sprintf(buf, "Mitigation: %s%s\n", v2_str, bhb_str); in cpu_show_spectre_v2()
298 * still rely on firmware for the mitigation at EL2. in spectre_v2_enable_fw_mitigation()
357 * A major source of headaches is that the software mitigation is enabled both
364 * all of the currently onlined CPUs are safelisted, as the mitigation tends to
367 * The only good part is that if the firmware mitigation is present, then it is
369 * vulnerable CPU if one of the boot CPUs is using the firmware mitigation.
[all …]
|
| /linux-6.12.1/tools/testing/selftests/powerpc/security/ |
| D | mitigation-patching.sh | 9 local mitigation="$1"
14 orig=$(cat "$mitigation")
21 echo 0 > "$mitigation"
22 echo 1 > "$mitigation"
27 echo "$orig" > "$mitigation"
|
| D | spectre_v2.c | 140 printf("Error: couldn't determine spectre_v2 mitigation state?\n"); in spectre_v2_test()
201 printf("Possible mismatch between reported & actual mitigation\n"); in spectre_v2_test()
210 printf("Possible mismatch between reported & actual mitigation\n"); in spectre_v2_test()
217 printf("Possible mismatch between reported & actual mitigation\n"); in spectre_v2_test()
227 printf("OK - Measured branch prediction rates match reported spectre v2 mitigation.\n"); in spectre_v2_test()
|
| /linux-6.12.1/drivers/thermal/ |
| D | thermal_debugfs.c | 94 * @trip_temp: trip temperature at mitigation start
95 * @trip_hyst: trip hysteresis at mitigation start
111 * struct tz_episode - A mitigation episode information
113 * The tz_episode structure describes a mitigation episode. A
114 * mitigation episode begins the trip point with the lower temperature
121 * @duration: total duration of the mitigation episode
135 * struct tz_debugfs - Store all mitigation episodes for a thermal zone
137 * The tz_debugfs structure contains the list of the mitigation
139 * order to handle correctly nested trip point mitigation episodes.
145 * @tz_episodes: a list of thermal mitigation episodes
[all …]
|
| /linux-6.12.1/arch/s390/kernel/ |
| D | nospec-sysfs.c | 10 return sprintf(buf, "Mitigation: __user pointer sanitization\n"); in cpu_show_spectre_v1()
17 return sprintf(buf, "Mitigation: etokens\n"); in cpu_show_spectre_v2()
19 return sprintf(buf, "Mitigation: execute trampolines\n"); in cpu_show_spectre_v2()
21 return sprintf(buf, "Mitigation: limited branch prediction\n"); in cpu_show_spectre_v2()
|
| /linux-6.12.1/drivers/net/wireless/broadcom/b43/ |
| D | sysfs.c | 56 count = sysfs_emit(buf, "0 (No Interference Mitigation)\n"); in b43_attr_interfmode_show()
60 "1 (Non-WLAN Interference Mitigation)\n"); in b43_attr_interfmode_show()
63 count = sysfs_emit(buf, "2 (WLAN Interference Mitigation)\n"); in b43_attr_interfmode_show()
108 b43err(wldev->wl, "Interference Mitigation not " in b43_attr_interfmode_store()
|
| /linux-6.12.1/Documentation/driver-api/thermal/ |
| D | cpu-idle-cooling.rst | 70 performance penalty and a fixed latency. Mitigation can be increased
90 the duty cycle percentage. When no mitigation is happening the cooling
93 When the mitigation begins, depending on the governor's policy, a
133 mitigation begins. It is platform dependent and will depend on the
138 for thermal mitigation, otherwise we end up consuming more energy.
194 potentially invert the mitigation effect
|
| /linux-6.12.1/Documentation/userspace-api/ |
| D | spec_ctrl.rst | 9 The kernel provides mitigation for such vulnerabilities in various
34 0 PR_SPEC_PRCTL Mitigation can be controlled per task by
36 1 PR_SPEC_ENABLE The speculation feature is enabled, mitigation is
38 2 PR_SPEC_DISABLE The speculation feature is disabled, mitigation is
48 If PR_SPEC_PRCTL is set, then the per-task control of the mitigation is
|
| /linux-6.12.1/drivers/net/wireless/broadcom/b43legacy/ |
| D | sysfs.c | 78 count = sysfs_emit(buf, "0 (No Interference Mitigation)\n"); in b43legacy_attr_interfmode_show()
82 "1 (Non-WLAN Interference Mitigation)\n"); in b43legacy_attr_interfmode_show()
85 count = sysfs_emit(buf, "2 (WLAN Interference Mitigation)\n"); in b43legacy_attr_interfmode_show()
131 b43legacyerr(wldev->wl, "Interference Mitigation not " in b43legacy_attr_interfmode_store()
|
| /linux-6.12.1/arch/x86/kvm/mmu/ |
| D | spte.c | 185 * For simplicity, enforce the NX huge page mitigation even if not in make_spte()
186 * strictly necessary. KVM could ignore the mitigation if paging is in make_spte()
188 * abuse. But to safely ignore the mitigation, KVM would have to in make_spte()
192 * when CR0.PG is toggled, but leveraging that to ignore the mitigation in make_spte()
313 * the page executable as the NX hugepage mitigation no longer in make_huge_page_split_spte()
384 * are used to hold the relocated GFN when the L1TF mitigation is in kvm_mmu_set_mmio_spte_mask()
465 * to achieve more effective mitigation, e.g. if system RAM overlaps in kvm_mmu_reset_all_pte_masks()
|