linux-kernel - [PATCH v5] x86/speculation/l1tf: Add KConfig for setting the L1D cache flush mode

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Message-Id: <20200708194715.4073300-1-abhishekbh@google.com>
Date:   Wed,  8 Jul 2020 12:47:15 -0700
From:   Abhishek Bhardwaj <abhishekbh@...gle.com>
To:     LKML <linux-kernel@...r.kernel.org>
Cc:     Abhishek Bhardwaj <abhishekbh@...gle.com>,
        Anthony Steinhauser <asteinhauser@...gle.com>,
        Borislav Petkov <bp@...en8.de>,
        "H. Peter Anvin" <hpa@...or.com>, Ingo Molnar <mingo@...hat.com>,
        Jim Mattson <jmattson@...gle.com>,
        Joerg Roedel <joro@...tes.org>,
        Josh Poimboeuf <jpoimboe@...hat.com>,
        Mark Gross <mgross@...ux.intel.com>,
        Paolo Bonzini <pbonzini@...hat.com>,
        Pawan Gupta <pawan.kumar.gupta@...ux.intel.com>,
        Peter Zijlstra <peterz@...radead.org>,
        Sean Christopherson <sean.j.christopherson@...el.com>,
        Thomas Gleixner <tglx@...utronix.de>,
        Tony Luck <tony.luck@...el.com>,
        Vitaly Kuznetsov <vkuznets@...hat.com>,
        Waiman Long <longman@...hat.com>,
        Wanpeng Li <wanpengli@...cent.com>, kvm@...r.kernel.org,
        x86@...nel.org
Subject: [PATCH v5] x86/speculation/l1tf: Add KConfig for setting the L1D
 cache flush mode

This change adds a new kernel configuration that sets the l1d cache
flush setting at compile time rather than at run time.

The reasons for this change are as follows -

 - Kernel command line arguments are getting unwieldy. These parameters
 are not a scalable way to set the kernel config. They're intended as a
 super limited way for the bootloader to pass info to the kernel and
 also as a way for end users who are not compiling the kernel themselves
 to tweak the kernel behavior.

 - Also, if a user wants this setting from the start. It's a definite
 smell that it deserves to be a compile time thing rather than adding
 extra code plus whatever miniscule time at runtime to pass an
 extra argument.

 - Finally, it doesn't preclude the runtime / kernel command line way.
 Users are free to use those as well.

Signed-off-by: Abhishek Bhardwaj <abhishekbh@...gle.com>

---

Changes in v5:
- Fix indentation of the help text in the KConfig.

Changes in v4:
- Add motivation for the change in the commit message.

Changes in v3:
- Change depends on to only x86_64.
- Remove copy paste errors at the end of the KConfig.

Changes in v2:
- Fix typo in the help of the new KConfig.

 arch/x86/kernel/cpu/bugs.c |  8 ++++++++
 arch/x86/kvm/Kconfig       | 13 +++++++++++++
 2 files changed, 21 insertions(+)

diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c
index 0b71970d2d3d2..1dcc875cf5547 100644
--- a/arch/x86/kernel/cpu/bugs.c
+++ b/arch/x86/kernel/cpu/bugs.c
@@ -1406,7 +1406,15 @@ enum l1tf_mitigations l1tf_mitigation __ro_after_init = L1TF_MITIGATION_FLUSH;
 #if IS_ENABLED(CONFIG_KVM_INTEL)
 EXPORT_SYMBOL_GPL(l1tf_mitigation);
 #endif
+#if (CONFIG_KVM_VMENTRY_L1D_FLUSH == 1)
+enum vmx_l1d_flush_state l1tf_vmx_mitigation = VMENTER_L1D_FLUSH_NEVER;
+#elif (CONFIG_KVM_VMENTRY_L1D_FLUSH == 2)
+enum vmx_l1d_flush_state l1tf_vmx_mitigation = VMENTER_L1D_FLUSH_COND;
+#elif (CONFIG_KVM_VMENTRY_L1D_FLUSH == 3)
+enum vmx_l1d_flush_state l1tf_vmx_mitigation = VMENTER_L1D_FLUSH_ALWAYS;
+#else
 enum vmx_l1d_flush_state l1tf_vmx_mitigation = VMENTER_L1D_FLUSH_AUTO;
+#endif
 EXPORT_SYMBOL_GPL(l1tf_vmx_mitigation);
 
 /*
diff --git a/arch/x86/kvm/Kconfig b/arch/x86/kvm/Kconfig
index b277a2db62676..cb2639dfb6da1 100644
--- a/arch/x86/kvm/Kconfig
+++ b/arch/x86/kvm/Kconfig
@@ -107,4 +107,17 @@ config KVM_MMU_AUDIT
 	 This option adds a R/W kVM module parameter 'mmu_audit', which allows
 	 auditing of KVM MMU events at runtime.
 
+config KVM_VMENTRY_L1D_FLUSH
+	int "L1D cache flush settings (1-3)"
+	range 1 3
+	default "2"
+	depends on KVM && X86_64
+	help
+	  This setting determines the L1D cache flush behavior before a VMENTER.
+	  This is similar to setting the option / parameter to
+	  kvm-intel.vmentry_l1d_flush.
+	  1 - Never flush.
+	  2 - Conditionally flush.
+	  3 - Always flush.
+
 endif # VIRTUALIZATION
-- 
2.27.0.383.g050319c2ae-goog