x86/fsgsbase/64: Fix ptrace() to read the FS/GS base accurately

On 64-bit kernels ptrace can read the FS/GS base using the register access
APIs (PTRACE_PEEKUSER, etc.) or PTRACE_ARCH_PRCTL.

Make both of these mechanisms return the actual FS/GS base.

This will improve debuggability by providing the correct information
to ptracer such as GDB.

[ chang: Rebased and revised patch description. ]
[ mingo: Revised the changelog some more. ]

Signed-off-by: Andy Lutomirski <[email protected]>
Signed-off-by: Chang S. Bae <[email protected]>
Cc: Andy Lutomirski <[email protected]>
Cc: Borislav Petkov <[email protected]>
Cc: Brian Gerst <[email protected]>
Cc: Dave Hansen <[email protected]>
Cc: Denys Vlasenko <[email protected]>
Cc: H. Peter Anvin <[email protected]>
Cc: Linus Torvalds <[email protected]>
Cc: Markus T Metzger <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Ravi Shankar <[email protected]>
Cc: Rik van Riel <[email protected]>
Cc: Thomas Gleixner <[email protected]>
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Ingo Molnar <[email protected]>

1 files changed, 52 insertions, 10 deletions

diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c
index e2ee403865eb..3acbf45cb7fb 100644
--- a/arch/x86/kernel/ptrace.c
+++ b/arch/x86/kernel/ptrace.c
@@ -39,6 +39,7 @@
 #include <asm/hw_breakpoint.h>
 #include <asm/traps.h>
 #include <asm/syscall.h>
+#include <asm/mmu_context.h>
 
 #include "tls.h"
 
@@ -342,6 +343,49 @@ 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)
@@ -435,18 +479,16 @@ static unsigned long getreg(struct task_struct *task, unsigned long offset)
 
 #ifdef CONFIG_X86_64
 	case offsetof(struct user_regs_struct, fs_base): {
-		/*
-		 * XXX: This will not behave as expected if called on
-		 * current or if fsindex != 0.
-		 */
-		return task->thread.fsbase;
+		if (task->thread.fsindex == 0)
+			return task->thread.fsbase;
+		else
+			return task_seg_base(task, task->thread.fsindex);
 	}
 	case offsetof(struct user_regs_struct, gs_base): {
-		/*
-		 * XXX: This will not behave as expected if called on
-		 * current or if fsindex != 0.
-		 */
-		return task->thread.gsbase;
+		if (task->thread.gsindex == 0)
+			return task->thread.gsbase;
+		else
+			return task_seg_base(task, task->thread.gsindex);
 	}
 #endif
 	}