[PATCH 0/3] arch/x86: Optionally flush L1D on context switch

[PATCH 0/3] arch/x86: Optionally flush L1D on context switch

[Balbir Singh]

Provide a mechanisn to flush the L1D cache on context switch.  The goal
is to allow tasks that are paranoid due to the recent snoop assisted data
sampling vulnerabilites, to flush their L1D on being switched out.
This protects their data from being snooped or leaked via side channels
after the task has context switched out.

The core of the patches is patch 3, the first two refactor the code so
that common bits can be reused.

Changelog:
 - Refactor the code and reuse cond_ibpb() - code bits provided by tglx
 - Merge mm state tracking for ibpb and l1d flush
 - Rename TIF_L1D_FLUSH to TIF_SPEC_FLUSH_L1D

Changelog RFC:
 - Reuse existing code for allocation and flush
 - Simplify the goto logic in the actual l1d_flush function
 - Optimize the code path with jump labels/static functions

The RFC patch was previously posted at

https://lore.kernel.org/lkml/20200325071101.29556-1-sblbir@amazon.com/

Balbir Singh (3):
  arch/x86/kvm: Refactor l1d flush lifecycle management
  arch/x86: Refactor tlbflush and l1d flush
  arch/x86: Optionally flush L1D on context switch

 arch/x86/include/asm/cacheflush.h  |  6 ++
 arch/x86/include/asm/thread_info.h |  6 +-
 arch/x86/include/asm/tlbflush.h    |  2 +-
 arch/x86/include/uapi/asm/prctl.h  |  3 +
 arch/x86/kernel/Makefile           |  1 +
 arch/x86/kernel/l1d_flush.c        | 85 +++++++++++++++++++++++++++
 arch/x86/kernel/process_64.c       | 10 +++-
 arch/x86/kvm/vmx/vmx.c             | 56 +++---------------
 arch/x86/mm/tlb.c                  | 92 +++++++++++++++++++++++-------
 9 files changed, 189 insertions(+), 72 deletions(-)
 create mode 100644 arch/x86/kernel/l1d_flush.c

-- 
2.17.1

[PATCH 1/3] arch/x86/kvm: Refactor l1d flush lifecycle management

[Balbir Singh]

Split out the allocation and free routines to be used in a follow
up set of patches (to reuse for L1D flushing).

Signed-off-by: Balbir Singh <sblbir@amazon.com>
---
 arch/x86/include/asm/cacheflush.h |  3 +++
 arch/x86/kernel/Makefile          |  1 +
 arch/x86/kernel/l1d_flush.c       | 36 +++++++++++++++++++++++++++++++
 arch/x86/kvm/vmx/vmx.c            | 25 +++------------------
 4 files changed, 43 insertions(+), 22 deletions(-)
 create mode 100644 arch/x86/kernel/l1d_flush.c

diff --git a/arch/x86/include/asm/cacheflush.h b/arch/x86/include/asm/cacheflush.h
index 63feaf2a5f93..6419a4cef0e8 100644
--- a/arch/x86/include/asm/cacheflush.h
+++ b/arch/x86/include/asm/cacheflush.h
@@ -6,6 +6,9 @@
 #include <asm-generic/cacheflush.h>
 #include <asm/special_insns.h>
 
+#define L1D_CACHE_ORDER 4
 void clflush_cache_range(void *addr, unsigned int size);
+void *alloc_l1d_flush_pages(void);
+void cleanup_l1d_flush_pages(void *l1d_flush_pages);
 
 #endif /* _ASM_X86_CACHEFLUSH_H */
diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile
index d6d61c4455fa..48f443e6c2de 100644
--- a/arch/x86/kernel/Makefile
+++ b/arch/x86/kernel/Makefile
@@ -160,3 +160,4 @@ ifeq ($(CONFIG_X86_64),y)
 endif
 
 obj-$(CONFIG_IMA_SECURE_AND_OR_TRUSTED_BOOT)	+= ima_arch.o
+obj-y						+= l1d_flush.o
diff --git a/arch/x86/kernel/l1d_flush.c b/arch/x86/kernel/l1d_flush.c
new file mode 100644
index 000000000000..05f375c33423
--- /dev/null
+++ b/arch/x86/kernel/l1d_flush.c
@@ -0,0 +1,36 @@
+#include <linux/mm.h>
+#include <asm/cacheflush.h>
+
+void *alloc_l1d_flush_pages(void)
+{
+	struct page *page;
+	void *l1d_flush_pages = NULL;
+	int i;
+
+	/*
+	 * This allocation for l1d_flush_pages is not tied to a VM/task's
+	 * lifetime and so should not be charged to a memcg.
+	 */
+	page = alloc_pages(GFP_KERNEL, L1D_CACHE_ORDER);
+	if (!page)
+		return NULL;
+	l1d_flush_pages = page_address(page);
+
+	/*
+	 * Initialize each page with a different pattern in
+	 * order to protect against KSM in the nested
+	 * virtualization case.
+	 */
+	for (i = 0; i < 1u << L1D_CACHE_ORDER; ++i) {
+		memset(l1d_flush_pages + i * PAGE_SIZE, i + 1,
+				PAGE_SIZE);
+	}
+	return l1d_flush_pages;
+}
+EXPORT_SYMBOL_GPL(alloc_l1d_flush_pages);
+
+void cleanup_l1d_flush_pages(void *l1d_flush_pages)
+{
+	free_pages((unsigned long)l1d_flush_pages, L1D_CACHE_ORDER);
+}
+EXPORT_SYMBOL_GPL(cleanup_l1d_flush_pages);
diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c
index 9eaccf92d616..209e63798435 100644
--- a/arch/x86/kvm/vmx/vmx.c
+++ b/arch/x86/kvm/vmx/vmx.c
@@ -203,14 +203,10 @@ static const struct {
 	[VMENTER_L1D_FLUSH_NOT_REQUIRED] = {"not required", false},
 };
 
-#define L1D_CACHE_ORDER 4
 static void *vmx_l1d_flush_pages;
 
 static int vmx_setup_l1d_flush(enum vmx_l1d_flush_state l1tf)
 {
-	struct page *page;
-	unsigned int i;
-
 	if (!boot_cpu_has_bug(X86_BUG_L1TF)) {
 		l1tf_vmx_mitigation = VMENTER_L1D_FLUSH_NOT_REQUIRED;
 		return 0;
@@ -253,24 +249,9 @@ static int vmx_setup_l1d_flush(enum vmx_l1d_flush_state l1tf)
 
 	if (l1tf != VMENTER_L1D_FLUSH_NEVER && !vmx_l1d_flush_pages &&
 	    !boot_cpu_has(X86_FEATURE_FLUSH_L1D)) {
-		/*
-		 * This allocation for vmx_l1d_flush_pages is not tied to a VM
-		 * lifetime and so should not be charged to a memcg.
-		 */
-		page = alloc_pages(GFP_KERNEL, L1D_CACHE_ORDER);
-		if (!page)
+		vmx_l1d_flush_pages = alloc_l1d_flush_pages();
+		if (!vmx_l1d_flush_pages)
 			return -ENOMEM;
-		vmx_l1d_flush_pages = page_address(page);
-
-		/*
-		 * Initialize each page with a different pattern in
-		 * order to protect against KSM in the nested
-		 * virtualization case.
-		 */
-		for (i = 0; i < 1u << L1D_CACHE_ORDER; ++i) {
-			memset(vmx_l1d_flush_pages + i * PAGE_SIZE, i + 1,
-			       PAGE_SIZE);
-		}
 	}
 
 	l1tf_vmx_mitigation = l1tf;
@@ -7992,7 +7973,7 @@ static struct kvm_x86_ops vmx_x86_ops __ro_after_init = {
 static void vmx_cleanup_l1d_flush(void)
 {
 	if (vmx_l1d_flush_pages) {
-		free_pages((unsigned long)vmx_l1d_flush_pages, L1D_CACHE_ORDER);
+		cleanup_l1d_flush_pages(vmx_l1d_flush_pages);
 		vmx_l1d_flush_pages = NULL;
 	}
 	/* Restore state so sysfs ignores VMX */
-- 
2.17.1

[PATCH 2/3] arch/x86: Refactor tlbflush and l1d flush

[Balbir Singh]

Refactor the existing assembly bits into smaller helper functions
and also abstract L1D_FLUSH into a helper function. Use these
functions in kvm for L1D flushing.

Signed-off-by: Balbir Singh <sblbir@amazon.com>
---
 arch/x86/include/asm/cacheflush.h |  3 ++
 arch/x86/kernel/l1d_flush.c       | 49 +++++++++++++++++++++++++++++++
 arch/x86/kvm/vmx/vmx.c            | 31 ++++---------------
 3 files changed, 57 insertions(+), 26 deletions(-)

diff --git a/arch/x86/include/asm/cacheflush.h b/arch/x86/include/asm/cacheflush.h
index 6419a4cef0e8..66a46db7aadd 100644
--- a/arch/x86/include/asm/cacheflush.h
+++ b/arch/x86/include/asm/cacheflush.h
@@ -10,5 +10,8 @@
 void clflush_cache_range(void *addr, unsigned int size);
 void *alloc_l1d_flush_pages(void);
 void cleanup_l1d_flush_pages(void *l1d_flush_pages);
+void populate_tlb_with_flush_pages(void *l1d_flush_pages);
+void flush_l1d_cache_sw(void *l1d_flush_pages);
+int flush_l1d_cache_hw(void);
 
 #endif /* _ASM_X86_CACHEFLUSH_H */
diff --git a/arch/x86/kernel/l1d_flush.c b/arch/x86/kernel/l1d_flush.c
index 05f375c33423..60499f773046 100644
--- a/arch/x86/kernel/l1d_flush.c
+++ b/arch/x86/kernel/l1d_flush.c
@@ -34,3 +34,52 @@ void cleanup_l1d_flush_pages(void *l1d_flush_pages)
 	free_pages((unsigned long)l1d_flush_pages, L1D_CACHE_ORDER);
 }
 EXPORT_SYMBOL_GPL(cleanup_l1d_flush_pages);
+
+void populate_tlb_with_flush_pages(void *l1d_flush_pages)
+{
+	int size = PAGE_SIZE << L1D_CACHE_ORDER;
+
+	asm volatile(
+		/* First ensure the pages are in the TLB */
+		"xorl	%%eax, %%eax\n"
+		".Lpopulate_tlb:\n\t"
+		"movzbl	(%[flush_pages], %%" _ASM_AX "), %%ecx\n\t"
+		"addl	$4096, %%eax\n\t"
+		"cmpl	%%eax, %[size]\n\t"
+		"jne	.Lpopulate_tlb\n\t"
+		"xorl	%%eax, %%eax\n\t"
+		"cpuid\n\t"
+		:: [flush_pages] "r" (l1d_flush_pages),
+		    [size] "r" (size)
+		: "eax", "ebx", "ecx", "edx");
+}
+EXPORT_SYMBOL_GPL(populate_tlb_with_flush_pages);
+
+int flush_l1d_cache_hw(void)
+{
+	if (static_cpu_has(X86_FEATURE_FLUSH_L1D)) {
+		wrmsrl(MSR_IA32_FLUSH_CMD, L1D_FLUSH);
+		return 1;
+	}
+	return 0;
+}
+EXPORT_SYMBOL_GPL(flush_l1d_cache_hw);
+
+void flush_l1d_cache_sw(void *l1d_flush_pages)
+{
+	int size = PAGE_SIZE << L1D_CACHE_ORDER;
+
+	asm volatile(
+			/* Fill the cache */
+			"xorl	%%eax, %%eax\n"
+			".Lfill_cache:\n"
+			"movzbl	(%[flush_pages], %%" _ASM_AX "), %%ecx\n\t"
+			"addl	$64, %%eax\n\t"
+			"cmpl	%%eax, %[size]\n\t"
+			"jne	.Lfill_cache\n\t"
+			"lfence\n"
+			:: [flush_pages] "r" (l1d_flush_pages),
+			[size] "r" (size)
+			: "eax", "ecx");
+}
+EXPORT_SYMBOL_GPL(flush_l1d_cache_sw);
diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c
index 209e63798435..29dc5a5bb6ab 100644
--- a/arch/x86/kvm/vmx/vmx.c
+++ b/arch/x86/kvm/vmx/vmx.c
@@ -5956,8 +5956,6 @@ static int vmx_handle_exit(struct kvm_vcpu *vcpu,
  */
 static void vmx_l1d_flush(struct kvm_vcpu *vcpu)
 {
-	int size = PAGE_SIZE << L1D_CACHE_ORDER;
-
 	/*
 	 * This code is only executed when the the flush mode is 'cond' or
 	 * 'always'
@@ -5986,32 +5984,13 @@ static void vmx_l1d_flush(struct kvm_vcpu *vcpu)
 
 	vcpu->stat.l1d_flush++;
 
-	if (static_cpu_has(X86_FEATURE_FLUSH_L1D)) {
-		wrmsrl(MSR_IA32_FLUSH_CMD, L1D_FLUSH);
+	if (flush_l1d_cache_hw())
 		return;
-	}
 
-	asm volatile(
-		/* First ensure the pages are in the TLB */
-		"xorl	%%eax, %%eax\n"
-		".Lpopulate_tlb:\n\t"
-		"movzbl	(%[flush_pages], %%" _ASM_AX "), %%ecx\n\t"
-		"addl	$4096, %%eax\n\t"
-		"cmpl	%%eax, %[size]\n\t"
-		"jne	.Lpopulate_tlb\n\t"
-		"xorl	%%eax, %%eax\n\t"
-		"cpuid\n\t"
-		/* Now fill the cache */
-		"xorl	%%eax, %%eax\n"
-		".Lfill_cache:\n"
-		"movzbl	(%[flush_pages], %%" _ASM_AX "), %%ecx\n\t"
-		"addl	$64, %%eax\n\t"
-		"cmpl	%%eax, %[size]\n\t"
-		"jne	.Lfill_cache\n\t"
-		"lfence\n"
-		:: [flush_pages] "r" (vmx_l1d_flush_pages),
-		    [size] "r" (size)
-		: "eax", "ebx", "ecx", "edx");
+	preempt_disable();
+	populate_tlb_with_flush_pages(vmx_l1d_flush_pages);
+	flush_l1d_cache_sw(vmx_l1d_flush_pages);
+	preempt_enable();
 }
 
 static void update_cr8_intercept(struct kvm_vcpu *vcpu, int tpr, int irr)
-- 
2.17.1

[PATCH 3/3] arch/x86: Optionally flush L1D on context switch

[Balbir Singh]

Implement a mechanism to selectively flush the L1D cache. The goal is to
allow tasks that are paranoid due to the recent snoop assisted data sampling
vulnerabilites, to flush their L1D on being switched out.  This protects
their data from being snooped or leaked via side channels after the task
has context switched out.

There are two scenarios we might want to protect against, a task leaving
the CPU with data still in L1D (which is the main concern of this patch),
the second scenario is a malicious task coming in (not so well trusted)
for which we want to clean up the cache before it starts. Only the case
for the former is addressed.

Add arch specific prctl()'s to opt-in to the L1D cache on context switch
out, the existing mechanisms of tracking prev_mm via cpu_tlbstate is
reused. cond_ibpb() is refactored and renamed into cond_mitigation().

A new thread_info flag TIF_SPEC_FLUSH_L1D is added to track tasks which
opt-into L1D flushing. cpu_tlbstate.last_user_mm_ibpb is renamed to
cpu_tlbstate.last_user_mm_spec, this is used to convert the TIF flags
into mm state (per cpu via last_user_mm_spec) in cond_mitigation(),
which then used to do decide when to call flush_l1d().

The current version benefited from discussions with Kees and Thomas.
Thomas suggested and provided the code snippet for refactoring the
existing cond_ibpb() code.

Signed-off-by: Balbir Singh <sblbir@amazon.com>
---
 arch/x86/include/asm/thread_info.h |  6 +-
 arch/x86/include/asm/tlbflush.h    |  2 +-
 arch/x86/include/uapi/asm/prctl.h  |  3 +
 arch/x86/kernel/process_64.c       | 10 +++-
 arch/x86/mm/tlb.c                  | 92 +++++++++++++++++++++++-------
 5 files changed, 89 insertions(+), 24 deletions(-)

diff --git a/arch/x86/include/asm/thread_info.h b/arch/x86/include/asm/thread_info.h
index 8de8ceccb8bc..5cb250872643 100644
--- a/arch/x86/include/asm/thread_info.h
+++ b/arch/x86/include/asm/thread_info.h
@@ -84,7 +84,7 @@ struct thread_info {
 #define TIF_SYSCALL_AUDIT	7	/* syscall auditing active */
 #define TIF_SECCOMP		8	/* secure computing */
 #define TIF_SPEC_IB		9	/* Indirect branch speculation mitigation */
-#define TIF_SPEC_FORCE_UPDATE	10	/* Force speculation MSR update in context switch */
+#define TIF_SPEC_FLUSH_L1D	10	/* Flush L1D on mm switches (processes) */
 #define TIF_USER_RETURN_NOTIFY	11	/* notify kernel of userspace return */
 #define TIF_UPROBE		12	/* breakpointed or singlestepping */
 #define TIF_PATCH_PENDING	13	/* pending live patching update */
@@ -96,6 +96,7 @@ struct thread_info {
 #define TIF_MEMDIE		20	/* is terminating due to OOM killer */
 #define TIF_POLLING_NRFLAG	21	/* idle is polling for TIF_NEED_RESCHED */
 #define TIF_IO_BITMAP		22	/* uses I/O bitmap */
+#define TIF_SPEC_FORCE_UPDATE	23	/* Force speculation MSR update in context switch */
 #define TIF_FORCED_TF		24	/* true if TF in eflags artificially */
 #define TIF_BLOCKSTEP		25	/* set when we want DEBUGCTLMSR_BTF */
 #define TIF_LAZY_MMU_UPDATES	27	/* task is updating the mmu lazily */
@@ -132,6 +133,7 @@ struct thread_info {
 #define _TIF_ADDR32		(1 << TIF_ADDR32)
 #define _TIF_X32		(1 << TIF_X32)
 #define _TIF_FSCHECK		(1 << TIF_FSCHECK)
+#define _TIF_SPEC_FLUSH_L1D	(1 << TIF_SPEC_FLUSH_L1D)
 
 /* Work to do before invoking the actual syscall. */
 #define _TIF_WORK_SYSCALL_ENTRY	\
@@ -239,6 +241,8 @@ extern void arch_task_cache_init(void);
 extern int arch_dup_task_struct(struct task_struct *dst, struct task_struct *src);
 extern void arch_release_task_struct(struct task_struct *tsk);
 extern void arch_setup_new_exec(void);
+extern int enable_l1d_flush_for_task(struct task_struct *tsk);
+extern int disable_l1d_flush_for_task(struct task_struct *tsk);
 #define arch_setup_new_exec arch_setup_new_exec
 #endif	/* !__ASSEMBLY__ */
 
diff --git a/arch/x86/include/asm/tlbflush.h b/arch/x86/include/asm/tlbflush.h
index 6f66d841262d..69e6ea20679c 100644
--- a/arch/x86/include/asm/tlbflush.h
+++ b/arch/x86/include/asm/tlbflush.h
@@ -172,7 +172,7 @@ struct tlb_state {
 	/* Last user mm for optimizing IBPB */
 	union {
 		struct mm_struct	*last_user_mm;
-		unsigned long		last_user_mm_ibpb;
+		unsigned long		last_user_mm_spec;
 	};
 
 	u16 loaded_mm_asid;
diff --git a/arch/x86/include/uapi/asm/prctl.h b/arch/x86/include/uapi/asm/prctl.h
index 5a6aac9fa41f..1361e5e25791 100644
--- a/arch/x86/include/uapi/asm/prctl.h
+++ b/arch/x86/include/uapi/asm/prctl.h
@@ -14,4 +14,7 @@
 #define ARCH_MAP_VDSO_32	0x2002
 #define ARCH_MAP_VDSO_64	0x2003
 
+#define ARCH_SET_L1D_FLUSH	0x3001
+#define ARCH_GET_L1D_FLUSH	0x3002
+
 #endif /* _ASM_X86_PRCTL_H */
diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
index 5ef9d8f25b0e..ecf542f13572 100644
--- a/arch/x86/kernel/process_64.c
+++ b/arch/x86/kernel/process_64.c
@@ -699,7 +699,15 @@ long do_arch_prctl_64(struct task_struct *task, int option, unsigned long arg2)
 	case ARCH_MAP_VDSO_64:
 		return prctl_map_vdso(&vdso_image_64, arg2);
 #endif
-
+	case ARCH_GET_L1D_FLUSH:
+		return test_ti_thread_flag(&task->thread_info, TIF_SPEC_FLUSH_L1D);
+	case ARCH_SET_L1D_FLUSH: {
+		if (arg2 >= 1)
+			return enable_l1d_flush_for_task(task);
+		else
+			return disable_l1d_flush_for_task(task);
+		break;
+	}
 	default:
 		ret = -EINVAL;
 		break;
diff --git a/arch/x86/mm/tlb.c b/arch/x86/mm/tlb.c
index 66f96f21a7b6..c8861b1c1ce4 100644
--- a/arch/x86/mm/tlb.c
+++ b/arch/x86/mm/tlb.c
@@ -13,6 +13,7 @@
 #include <asm/mmu_context.h>
 #include <asm/nospec-branch.h>
 #include <asm/cache.h>
+#include <asm/cacheflush.h>
 #include <asm/apic.h>
 #include <asm/uv/uv.h>
 
@@ -33,10 +34,12 @@
  */
 
 /*
- * Use bit 0 to mangle the TIF_SPEC_IB state into the mm pointer which is
- * stored in cpu_tlb_state.last_user_mm_ibpb.
+ * Bits to mangle the TIF_SPEC_* state into the mm pointer which is
+ * stored in cpu_tlb_state.last_user_mm_spec.
  */
 #define LAST_USER_MM_IBPB	0x1UL
+#define LAST_USER_MM_FLUSH_L1D	0x2UL
+#define LAST_USER_MM_SPEC_MASK	(LAST_USER_MM_IBPB | LAST_USER_MM_FLUSH_L1D)
 
 /*
  * We get here when we do something requiring a TLB invalidation
@@ -151,6 +154,50 @@ void leave_mm(int cpu)
 }
 EXPORT_SYMBOL_GPL(leave_mm);
 
+static void *l1d_flush_pages;
+static DEFINE_MUTEX(l1d_flush_mutex);
+
+int enable_l1d_flush_for_task(struct task_struct *tsk)
+{
+	struct page *page;
+	int ret = 0;
+
+	if (static_cpu_has(X86_FEATURE_FLUSH_L1D))
+		goto done;
+
+	page = READ_ONCE(l1d_flush_pages);
+	if (unlikely(!page)) {
+		mutex_lock(&l1d_flush_mutex);
+		if (!l1d_flush_pages) {
+			l1d_flush_pages = alloc_l1d_flush_pages();
+			if (!l1d_flush_pages)
+				return -ENOMEM;
+		}
+		mutex_unlock(&l1d_flush_mutex);
+	}
+	/* I don't think we need to worry about KSM */
+done:
+	set_ti_thread_flag(&tsk->thread_info, TIF_SPEC_FLUSH_L1D);
+	return ret;
+}
+
+int disable_l1d_flush_for_task(struct task_struct *tsk)
+{
+	clear_ti_thread_flag(&tsk->thread_info, TIF_SPEC_FLUSH_L1D);
+	return 0;
+}
+
+/*
+ * Flush the L1D cache for this CPU. We want to this at switch mm time,
+ * this is a pessimistic security measure and an opt-in for those tasks
+ * that host sensitive information.
+ */
+static void flush_l1d(void)
+{
+	if (!flush_l1d_cache_hw())
+		flush_l1d_cache_sw(l1d_flush_pages);
+}
+
 void switch_mm(struct mm_struct *prev, struct mm_struct *next,
 	       struct task_struct *tsk)
 {
@@ -189,19 +236,26 @@ static void sync_current_stack_to_mm(struct mm_struct *mm)
 	}
 }
 
-static inline unsigned long mm_mangle_tif_spec_ib(struct task_struct *next)
+static inline unsigned long mm_mangle_tif_spec_bits(struct task_struct *next)
 {
 	unsigned long next_tif = task_thread_info(next)->flags;
-	unsigned long ibpb = (next_tif >> TIF_SPEC_IB) & LAST_USER_MM_IBPB;
+	unsigned long spec_bits = (next_tif >> TIF_SPEC_IB) & LAST_USER_MM_SPEC_MASK;
+
+	BUILD_BUG_ON(TIF_SPEC_FLUSH_L1D != TIF_SPEC_IB + 1);
 
-	return (unsigned long)next->mm | ibpb;
+	return (unsigned long)next->mm | spec_bits;
 }
 
-static void cond_ibpb(struct task_struct *next)
+static void cond_mitigation(struct task_struct *next)
 {
+	unsigned long prev_mm, next_mm;
+
 	if (!next || !next->mm)
 		return;
 
+	next_mm = mm_mangle_tif_spec_bits(next);
+	prev_mm = this_cpu_read(cpu_tlbstate.last_user_mm_spec);
+
 	/*
 	 * Both, the conditional and the always IBPB mode use the mm
 	 * pointer to avoid the IBPB when switching between tasks of the
@@ -212,8 +266,6 @@ static void cond_ibpb(struct task_struct *next)
 	 * exposed data is not really interesting.
 	 */
 	if (static_branch_likely(&switch_mm_cond_ibpb)) {
-		unsigned long prev_mm, next_mm;
-
 		/*
 		 * This is a bit more complex than the always mode because
 		 * it has to handle two cases:
@@ -243,20 +295,14 @@ static void cond_ibpb(struct task_struct *next)
 		 * Optimize this with reasonably small overhead for the
 		 * above cases. Mangle the TIF_SPEC_IB bit into the mm
 		 * pointer of the incoming task which is stored in
-		 * cpu_tlbstate.last_user_mm_ibpb for comparison.
-		 */
-		next_mm = mm_mangle_tif_spec_ib(next);
-		prev_mm = this_cpu_read(cpu_tlbstate.last_user_mm_ibpb);
-
-		/*
+		 * cpu_tlbstate.last_user_mm_spec for comparison.
+		 *
 		 * Issue IBPB only if the mm's are different and one or
 		 * both have the IBPB bit set.
 		 */
 		if (next_mm != prev_mm &&
 		    (next_mm | prev_mm) & LAST_USER_MM_IBPB)
 			indirect_branch_prediction_barrier();
-
-		this_cpu_write(cpu_tlbstate.last_user_mm_ibpb, next_mm);
 	}
 
 	if (static_branch_unlikely(&switch_mm_always_ibpb)) {
@@ -265,11 +311,15 @@ static void cond_ibpb(struct task_struct *next)
 		 * different context than the user space task which ran
 		 * last on this CPU.
 		 */
-		if (this_cpu_read(cpu_tlbstate.last_user_mm) != next->mm) {
+		if ((prev_mm & ~LAST_USER_MM_SPEC_MASK) !=
+					(unsigned long)next->mm)
 			indirect_branch_prediction_barrier();
-			this_cpu_write(cpu_tlbstate.last_user_mm, next->mm);
-		}
 	}
+
+	if (prev_mm & LAST_USER_MM_FLUSH_L1D)
+		flush_l1d();
+
+	this_cpu_write(cpu_tlbstate.last_user_mm_spec, next_mm);
 }
 
 void switch_mm_irqs_off(struct mm_struct *prev, struct mm_struct *next,
@@ -375,7 +425,7 @@ void switch_mm_irqs_off(struct mm_struct *prev, struct mm_struct *next,
 		 * predictor when switching between processes. This stops
 		 * one process from doing Spectre-v2 attacks on another.
 		 */
-		cond_ibpb(tsk);
+		cond_mitigation(tsk);
 
 		if (IS_ENABLED(CONFIG_VMAP_STACK)) {
 			/*
@@ -501,7 +551,7 @@ void initialize_tlbstate_and_flush(void)
 	write_cr3(build_cr3(mm->pgd, 0));
 
 	/* Reinitialize tlbstate. */
-	this_cpu_write(cpu_tlbstate.last_user_mm_ibpb, LAST_USER_MM_IBPB);
+	this_cpu_write(cpu_tlbstate.last_user_mm_spec, LAST_USER_MM_IBPB);
 	this_cpu_write(cpu_tlbstate.loaded_mm_asid, 0);
 	this_cpu_write(cpu_tlbstate.next_asid, 1);
 	this_cpu_write(cpu_tlbstate.ctxs[0].ctx_id, mm->context.ctx_id);
-- 
2.17.1

Re: [PATCH 0/3] arch/x86: Optionally flush L1D on context switch

[Josh Poimboeuf]

On Thu, Apr 02, 2020 at 05:23:58PM +1100, Balbir Singh wrote:
> Provide a mechanisn to flush the L1D cache on context switch.  The goal
> is to allow tasks that are paranoid due to the recent snoop assisted data
> sampling vulnerabilites, to flush their L1D on being switched out.

Hi Balbir,

Just curious, is it really vulnerabilities, plural?  I thought there was
only one: CVE-2020-0550 (Snoop-assisted L1 Data Sampling).

(There was a similar one without the "snoop": L1D Eviction Sampling, but
it's supposed to get fixed in microcode).

-- 
Josh

Re: [PATCH 0/3] arch/x86: Optionally flush L1D on context switch

[Singh, Balbir]

On Thu, 2020-04-02 at 15:13 -0500, Josh Poimboeuf wrote:
> CAUTION: This email originated from outside of the organization. Do not
> click links or open attachments unless you can confirm the sender and know
> the content is safe.
> 
> 
> 
> On Thu, Apr 02, 2020 at 05:23:58PM +1100, Balbir Singh wrote:
> > Provide a mechanisn to flush the L1D cache on context switch.  The goal
> > is to allow tasks that are paranoid due to the recent snoop assisted data
> > sampling vulnerabilites, to flush their L1D on being switched out.
> 
> Hi Balbir,
> 
> Just curious, is it really vulnerabilities, plural?  I thought there was
> only one: CVE-2020-0550 (Snoop-assisted L1 Data Sampling).
> 
> (There was a similar one without the "snoop": L1D Eviction Sampling, but
> it's supposed to get fixed in microcode).
> 

Hi, Josh

Yes, that CVE the motivation, the mitigation for CVE-2020-0550 does suggest
flushing the cache on context switch. But in general, as we begin to find more
ways of evicting data or snopping data, a generic mechanism is more useful and
that is why I am making it an opt-in.

Balbir Singh.

Re: [PATCH 0/3] arch/x86: Optionally flush L1D on context switch

[Josh Poimboeuf]

On Thu, Apr 02, 2020 at 08:35:46PM +0000, Singh, Balbir wrote:
> On Thu, 2020-04-02 at 15:13 -0500, Josh Poimboeuf wrote:
> > CAUTION: This email originated from outside of the organization. Do not
> > click links or open attachments unless you can confirm the sender and know
> > the content is safe.
> > 
> > 
> > 
> > On Thu, Apr 02, 2020 at 05:23:58PM +1100, Balbir Singh wrote:
> > > Provide a mechanisn to flush the L1D cache on context switch.  The goal
> > > is to allow tasks that are paranoid due to the recent snoop assisted data
> > > sampling vulnerabilites, to flush their L1D on being switched out.
> > 
> > Hi Balbir,
> > 
> > Just curious, is it really vulnerabilities, plural?  I thought there was
> > only one: CVE-2020-0550 (Snoop-assisted L1 Data Sampling).
> > 
> > (There was a similar one without the "snoop": L1D Eviction Sampling, but
> > it's supposed to get fixed in microcode).
> > 
> 
> Hi, Josh
> 
> Yes, that CVE the motivation, the mitigation for CVE-2020-0550 does suggest
> flushing the cache on context switch. But in general, as we begin to find more
> ways of evicting data or snopping data, a generic mechanism is more useful and
> that is why I am making it an opt-in.

Ok.  I think it would be a good idea to expand on that justification
more precisely in the commit message.  That would help both reviewers of
the code and users of the new option understand what level of paranoia
they're opting in to :-)

-- 
Josh

Re: [PATCH 0/3] arch/x86: Optionally flush L1D on context switch

[Thomas Gleixner]

Josh Poimboeuf <jpoimboe@redhat.com> writes:
> On Thu, Apr 02, 2020 at 08:35:46PM +0000, Singh, Balbir wrote:
>> Yes, that CVE the motivation, the mitigation for CVE-2020-0550 does suggest
>> flushing the cache on context switch. But in general, as we begin to find more
>> ways of evicting data or snopping data, a generic mechanism is more useful and
>> that is why I am making it an opt-in.
>
> Ok.  I think it would be a good idea to expand on that justification
> more precisely in the commit message.  That would help both reviewers of
> the code and users of the new option understand what level of paranoia
> they're opting in to :-)

The commit message is mostly useful for reviewers and people who have to
do code archeaology. 

Documentation/admin-guide/hw-vuln/ has plenty of space to host a
document with explanations. paranoia.rst comes to my mind. :)

Thanks,

        tglx
 

Re: [PATCH 3/3] arch/x86: Optionally flush L1D on context switch

[Borislav Petkov]

On Thu, Apr 02, 2020 at 05:24:01PM +1100, Balbir Singh wrote:
> +int enable_l1d_flush_for_task(struct task_struct *tsk)
> +{
> +	struct page *page;
> +	int ret = 0;
> +
> +	if (static_cpu_has(X86_FEATURE_FLUSH_L1D))
> +		goto done;
> +
> +	page = READ_ONCE(l1d_flush_pages);
> +	if (unlikely(!page)) {
> +		mutex_lock(&l1d_flush_mutex);
> +		if (!l1d_flush_pages) {
> +			l1d_flush_pages = alloc_l1d_flush_pages();
> +			if (!l1d_flush_pages)
> +				return -ENOMEM;

Do I see it correctly that you'll return here with l1d_flush_mutex still
held?

> +		}
> +		mutex_unlock(&l1d_flush_mutex);
> +	}
> +	/* I don't think we need to worry about KSM */

Pls use passive voice: no "we" or "I", etc.

-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette

Re: [PATCH 3/3] arch/x86: Optionally flush L1D on context switch

[Singh, Balbir]

On Fri, 2020-04-03 at 00:11 +0200, Borislav Petkov wrote:
> CAUTION: This email originated from outside of the organization. Do not
> click links or open attachments unless you can confirm the sender and know
> the content is safe.
> 
> 
> 
> On Thu, Apr 02, 2020 at 05:24:01PM +1100, Balbir Singh wrote:
> > +int enable_l1d_flush_for_task(struct task_struct *tsk)
> > +{
> > +     struct page *page;
> > +     int ret = 0;
> > +
> > +     if (static_cpu_has(X86_FEATURE_FLUSH_L1D))
> > +             goto done;
> > +
> > +     page = READ_ONCE(l1d_flush_pages);
> > +     if (unlikely(!page)) {
> > +             mutex_lock(&l1d_flush_mutex);
> > +             if (!l1d_flush_pages) {
> > +                     l1d_flush_pages = alloc_l1d_flush_pages();
> > +                     if (!l1d_flush_pages)
> > +                             return -ENOMEM;
> 
> Do I see it correctly that you'll return here with l1d_flush_mutex still
> held?

Darn.. :(

I had goto done to release the mutex earlier, let me fix that :)

Thanks,
Balbir Singh

> 
> > +             }
> > +             mutex_unlock(&l1d_flush_mutex);
> > +     }
> > +     /* I don't think we need to worry about KSM */
> 
> Pls use passive voice: no "we" or "I", etc.
> 
> --
> Regards/Gruss,
>     Boris.
> 
> https://people.kernel.org/tglx/notes-about-netiquette

Re: [PATCH 0/3] arch/x86: Optionally flush L1D on context switch

[Singh, Balbir]

On Thu, 2020-04-02 at 23:45 +0200, Thomas Gleixner wrote:
> CAUTION: This email originated from outside of the organization. Do not
> click links or open attachments unless you can confirm the sender and know
> the content is safe.
> 
> 
> 
> Josh Poimboeuf <jpoimboe@redhat.com> writes:
> > On Thu, Apr 02, 2020 at 08:35:46PM +0000, Singh, Balbir wrote:
> > > Yes, that CVE the motivation, the mitigation for CVE-2020-0550 does
> > > suggest
> > > flushing the cache on context switch. But in general, as we begin to
> > > find more
> > > ways of evicting data or snopping data, a generic mechanism is more
> > > useful and
> > > that is why I am making it an opt-in.
> > 
> > Ok.  I think it would be a good idea to expand on that justification
> > more precisely in the commit message.  That would help both reviewers of
> > the code and users of the new option understand what level of paranoia
> > they're opting in to :-)
> 
> The commit message is mostly useful for reviewers and people who have to
> do code archeaology.
> 
> Documentation/admin-guide/hw-vuln/ has plenty of space to host a
> document with explanations. paranoia.rst comes to my mind. :)

I hope people don't go looking for aliens in there :) I'll write up some
documentation if that helps, starting with something simple.

Balbir

> 
> Thanks,
> 
>         tglx
>