[tip:x86/asm] x86/fsgsbase/64: Introduce FS/GS base helper functions

[tip-bot for Chang S. Bae]

Commit-ID:  b1378a561fd16afdd96ef0bc912b1bcd2b85a68e
Gitweb:     https://git.kernel.org/tip/b1378a561fd16afdd96ef0bc912b1bcd2b85a68e
Author:     Chang S. Bae <chang.seok.bae@xxxxxxxxx>
AuthorDate: Tue, 18 Sep 2018 16:08:53 -0700
Committer:  Ingo Molnar <mingo@xxxxxxxxxx>
CommitDate: Mon, 8 Oct 2018 10:41:08 +0200

x86/fsgsbase/64: Introduce FS/GS base helper functions

Introduce FS/GS base access functionality via <asm/fsgsbase.h>,
not yet used by anything directly.

Factor out task_seg_base() from x86/ptrace.c and rename it to
x86_fsgsbase_read_task() to make it part of the new helpers.

This will allow us to enhance FSGSBASE support and eventually enable
the FSBASE/GSBASE instructions.

An "inactive" GS base refers to a base saved at kernel entry
and being part of an inactive, non-running/stopped user-task.
(The typical ptrace model.)

Here are the new functions:

  x86_fsbase_read_task()
  x86_gsbase_read_task()
  x86_fsbase_write_task()
  x86_gsbase_write_task()
  x86_fsbase_read_cpu()
  x86_fsbase_write_cpu()
  x86_gsbase_read_cpu_inactive()
  x86_gsbase_write_cpu_inactive()

As an advantage of the unified namespace we can now see all FS/GSBASE
API use in the kernel via the following 'git grep' pattern:

  $ git grep x86_.*sbase

[ mingo: Wrote new changelog. ]

Based-on-code-from: Andy Lutomirski <luto@xxxxxxxxxx>
Suggested-by: Ingo Molnar <mingo@xxxxxxxxxx>
Signed-off-by: Chang S. Bae <chang.seok.bae@xxxxxxxxx>
Cc: Andy Lutomirski <luto@xxxxxxxxxxxxxx>
Cc: Andy Lutomirski <luto@xxxxxxxxxx>
Cc: Borislav Petkov <bp@xxxxxxxxx>
Cc: Brian Gerst <brgerst@xxxxxxxxx>
Cc: Dave Hansen <dave.hansen@xxxxxxxxxxxxxxx>
Cc: Denys Vlasenko <dvlasenk@xxxxxxxxxx>
Cc: H. Peter Anvin <hpa@xxxxxxxxx>
Cc: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx>
Cc: Markus T Metzger <markus.t.metzger@xxxxxxxxx>
Cc: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
Cc: Ravi Shankar <ravi.v.shankar@xxxxxxxxx>
Cc: Rik van Riel <riel@xxxxxxxxxxx>
Cc: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
Link: http://lkml.kernel.org/r/1537312139-5580-3-git-send-email-chang.seok.bae@xxxxxxxxx
Signed-off-by: Ingo Molnar <mingo@xxxxxxxxxx>
---
 arch/x86/include/asm/fsgsbase.h |  50 ++++++++++++++++
 arch/x86/kernel/process_64.c    | 124 ++++++++++++++++++++++++++++++++++++++++
 arch/x86/kernel/ptrace.c        |  51 ++---------------
 3 files changed, 179 insertions(+), 46 deletions(-)

diff --git a/arch/x86/include/asm/fsgsbase.h b/arch/x86/include/asm/fsgsbase.h
new file mode 100644
index 000000000000..1ab465ee23fe
--- /dev/null
+++ b/arch/x86/include/asm/fsgsbase.h
@@ -0,0 +1,50 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _ASM_FSGSBASE_H
+#define _ASM_FSGSBASE_H 1
+
+#ifndef __ASSEMBLY__
+
+#ifdef CONFIG_X86_64
+
+#include <asm/msr-index.h>
+
+unsigned long x86_fsgsbase_read_task(struct task_struct *task,
+				     unsigned short selector);
+
+/*
+ * Read/write a task's fsbase or gsbase. This returns the value that
+ * the FS/GS base would have (if the task were to be resumed). These
+ * work on current or on a different non-running task.
+ */
+unsigned long x86_fsbase_read_task(struct task_struct *task);
+unsigned long x86_gsbase_read_task(struct task_struct *task);
+int x86_fsbase_write_task(struct task_struct *task, unsigned long fsbase);
+int x86_gsbase_write_task(struct task_struct *task, unsigned long gsbase);
+
+/* Helper functions for reading/writing FS/GS base */
+
+static inline unsigned long x86_fsbase_read_cpu(void)
+{
+	unsigned long fsbase;
+
+	rdmsrl(MSR_FS_BASE, fsbase);
+	return fsbase;
+}
+
+void x86_fsbase_write_cpu(unsigned long fsbase);
+
+static inline unsigned long x86_gsbase_read_cpu_inactive(void)
+{
+	unsigned long gsbase;
+
+	rdmsrl(MSR_KERNEL_GS_BASE, gsbase);
+	return gsbase;
+}
+
+void x86_gsbase_write_cpu_inactive(unsigned long gsbase);
+
+#endif /* CONFIG_X86_64 */
+
+#endif /* __ASSEMBLY__ */
+
+#endif /* _ASM_FSGSBASE_H */
diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
index ea5ea850348d..2a53ff8d1baf 100644
--- a/arch/x86/kernel/process_64.c
+++ b/arch/x86/kernel/process_64.c
@@ -54,6 +54,7 @@
 #include <asm/vdso.h>
 #include <asm/intel_rdt_sched.h>
 #include <asm/unistd.h>
+#include <asm/fsgsbase.h>
 #ifdef CONFIG_IA32_EMULATION
 /* Not included via unistd.h */
 #include <asm/unistd_32_ia32.h>
@@ -286,6 +287,129 @@ static __always_inline void load_seg_legacy(unsigned short prev_index,
 	}
 }
 
+unsigned long x86_fsgsbase_read_task(struct task_struct *task,
+				     unsigned short selector)
+{
+	unsigned short idx = selector >> 3;
+	unsigned long base;
+
+	if (likely((selector & SEGMENT_TI_MASK) == 0)) {
+		if (unlikely(idx >= GDT_ENTRIES))
+			return 0;
+
+		/*
+		 * There are no user segments in the GDT with nonzero bases
+		 * other than the TLS segments.
+		 */
+		if (idx < GDT_ENTRY_TLS_MIN || idx > GDT_ENTRY_TLS_MAX)
+			return 0;
+
+		idx -= GDT_ENTRY_TLS_MIN;
+		base = get_desc_base(&task->thread.tls_array[idx]);
+	} else {
+#ifdef CONFIG_MODIFY_LDT_SYSCALL
+		struct ldt_struct *ldt;
+
+		/*
+		 * If performance here mattered, we could protect the LDT
+		 * with RCU.  This is a slow path, though, so we can just
+		 * take the mutex.
+		 */
+		mutex_lock(&task->mm->context.lock);
+		ldt = task->mm->context.ldt;
+		if (unlikely(idx >= ldt->nr_entries))
+			base = 0;
+		else
+			base = get_desc_base(ldt->entries + idx);
+		mutex_unlock(&task->mm->context.lock);
+#else
+		base = 0;
+#endif
+	}
+
+	return base;
+}
+
+void x86_fsbase_write_cpu(unsigned long fsbase)
+{
+	/*
+	 * Set the selector to 0 as a notion, that the segment base is
+	 * overwritten, which will be checked for skipping the segment load
+	 * during context switch.
+	 */
+	loadseg(FS, 0);
+	wrmsrl(MSR_FS_BASE, fsbase);
+}
+
+void x86_gsbase_write_cpu_inactive(unsigned long gsbase)
+{
+	/* Set the selector to 0 for the same reason as %fs above. */
+	loadseg(GS, 0);
+	wrmsrl(MSR_KERNEL_GS_BASE, gsbase);
+}
+
+unsigned long x86_fsbase_read_task(struct task_struct *task)
+{
+	unsigned long fsbase;
+
+	if (task == current)
+		fsbase = x86_fsbase_read_cpu();
+	else if (task->thread.fsindex == 0)
+		fsbase = task->thread.fsbase;
+	else
+		fsbase = x86_fsgsbase_read_task(task, task->thread.fsindex);
+
+	return fsbase;
+}
+
+unsigned long x86_gsbase_read_task(struct task_struct *task)
+{
+	unsigned long gsbase;
+
+	if (task == current)
+		gsbase = x86_gsbase_read_cpu_inactive();
+	else if (task->thread.gsindex == 0)
+		gsbase = task->thread.gsbase;
+	else
+		gsbase = x86_fsgsbase_read_task(task, task->thread.gsindex);
+
+	return gsbase;
+}
+
+int x86_fsbase_write_task(struct task_struct *task, unsigned long fsbase)
+{
+	/*
+	 * Not strictly needed for %fs, but do it for symmetry
+	 * with %gs
+	 */
+	if (unlikely(fsbase >= TASK_SIZE_MAX))
+		return -EPERM;
+
+	preempt_disable();
+	task->thread.fsbase = fsbase;
+	if (task == current)
+		x86_fsbase_write_cpu(fsbase);
+	task->thread.fsindex = 0;
+	preempt_enable();
+
+	return 0;
+}
+
+int x86_gsbase_write_task(struct task_struct *task, unsigned long gsbase)
+{
+	if (unlikely(gsbase >= TASK_SIZE_MAX))
+		return -EPERM;
+
+	preempt_disable();
+	task->thread.gsbase = gsbase;
+	if (task == current)
+		x86_gsbase_write_cpu_inactive(gsbase);
+	task->thread.gsindex = 0;
+	preempt_enable();
+
+	return 0;
+}
+
 int copy_thread_tls(unsigned long clone_flags, unsigned long sp,
 		unsigned long arg, struct task_struct *p, unsigned long tls)
 {
diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c
index 3acbf45cb7fb..fbde2a7ce377 100644
--- a/arch/x86/kernel/ptrace.c
+++ b/arch/x86/kernel/ptrace.c
@@ -39,7 +39,7 @@
 #include <asm/hw_breakpoint.h>
 #include <asm/traps.h>
 #include <asm/syscall.h>
-#include <asm/mmu_context.h>
+#include <asm/fsgsbase.h>
 
 #include "tls.h"
 
@@ -343,49 +343,6 @@ static int set_segment_reg(struct task_struct *task,
 	return 0;
 }
 
-static unsigned long task_seg_base(struct task_struct *task,
-				   unsigned short selector)
-{
-	unsigned short idx = selector >> 3;
-	unsigned long base;
-
-	if (likely((selector & SEGMENT_TI_MASK) == 0)) {
-		if (unlikely(idx >= GDT_ENTRIES))
-			return 0;
-
-		/*
-		 * There are no user segments in the GDT with nonzero bases
-		 * other than the TLS segments.
-		 */
-		if (idx < GDT_ENTRY_TLS_MIN || idx > GDT_ENTRY_TLS_MAX)
-			return 0;
-
-		idx -= GDT_ENTRY_TLS_MIN;
-		base = get_desc_base(&task->thread.tls_array[idx]);
-	} else {
-#ifdef CONFIG_MODIFY_LDT_SYSCALL
-		struct ldt_struct *ldt;
-
-		/*
-		 * If performance here mattered, we could protect the LDT
-		 * with RCU.  This is a slow path, though, so we can just
-		 * take the mutex.
-		 */
-		mutex_lock(&task->mm->context.lock);
-		ldt = task->mm->context.ldt;
-		if (unlikely(idx >= ldt->nr_entries))
-			base = 0;
-		else
-			base = get_desc_base(ldt->entries + idx);
-		mutex_unlock(&task->mm->context.lock);
-#else
-		base = 0;
-#endif
-	}
-
-	return base;
-}
-
 #endif	/* CONFIG_X86_32 */
 
 static unsigned long get_flags(struct task_struct *task)
@@ -482,13 +439,15 @@ static unsigned long getreg(struct task_struct *task, unsigned long offset)
 		if (task->thread.fsindex == 0)
 			return task->thread.fsbase;
 		else
-			return task_seg_base(task, task->thread.fsindex);
+			return x86_fsgsbase_read_task(task,
+						      task->thread.fsindex);
 	}
 	case offsetof(struct user_regs_struct, gs_base): {
 		if (task->thread.gsindex == 0)
 			return task->thread.gsbase;
 		else
-			return task_seg_base(task, task->thread.gsindex);
+			return x86_fsgsbase_read_task(task,
+						      task->thread.gsindex);
 	}
 #endif
 	}