[PATCH 3/3] x86/vdso: Document vgtod_ts better

[Andy Lutomirski <luto@kernel.org>]

After reading do_hres() and do_course() and scratching my head a
bit, I figured out why the arithmetic is strange.  Document it.

Signed-off-by: Andy Lutomirski <luto@kernel.org>
---
 arch/x86/include/asm/vgtod.h | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/arch/x86/include/asm/vgtod.h b/arch/x86/include/asm/vgtod.h
index d17b092b9f1b..69d05c6d47f5 100644
--- a/arch/x86/include/asm/vgtod.h
+++ b/arch/x86/include/asm/vgtod.h
@@ -13,6 +13,15 @@ typedef u64 gtod_long_t;
 typedef unsigned long gtod_long_t;
 #endif
 
+/*
+ * There is one of these objects in the vvar page for each
+ * vDSO-accelerated clockid.  For high-resolution clocks, this encodes
+ * the time corresponding to vsyscall_gtod_data.cycle_last.  For coarse
+ * clocks, this encodes the actual time.
+ *
+ * To confuse the reader, for high-resolution clocks, nsec is left-shifted
+ * by vsyscall_gtod_data.shift.
+ */
 struct vgtod_ts {
 	u64		sec;
 	u64		nsec;
-- 
2.17.1