Merge changes from topic "kernel419"

* changes:
  Regenerate headers after .in file change.
  xlat: workaround V4L2_CID_USER_IMX_BASE Linux kernel ABI breakage
diff --git a/xlat/v4l2_control_id_bases.h b/xlat/v4l2_control_id_bases.h
index e668d94..60e9cb5 100644
--- a/xlat/v4l2_control_id_bases.h
+++ b/xlat/v4l2_control_id_bases.h
@@ -31,12 +31,16 @@
 #else
 # define V4L2_CID_USER_TI_VPE_BASE (V4L2_CID_BASE + 0x1050)
 #endif
+#ifndef STRACE_WORKAROUND_FOR_V4L2_CID_USER_IMX_BASE
+# define STRACE_WORKAROUND_FOR_V4L2_CID_USER_IMX_BASE
+# undef V4L2_CID_USER_IMX_BASE
+#endif
 #if defined(V4L2_CID_USER_IMX_BASE) || (defined(HAVE_DECL_V4L2_CID_USER_IMX_BASE) && HAVE_DECL_V4L2_CID_USER_IMX_BASE)
 DIAG_PUSH_IGNORE_TAUTOLOGICAL_COMPARE
-static_assert((V4L2_CID_USER_IMX_BASE) == ((V4L2_CID_BASE + 0x1090)), "V4L2_CID_USER_IMX_BASE != (V4L2_CID_BASE + 0x1090)");
+static_assert((V4L2_CID_USER_IMX_BASE) == ((V4L2_CID_BASE + 0x10b0)), "V4L2_CID_USER_IMX_BASE != (V4L2_CID_BASE + 0x10b0)");
 DIAG_POP_IGNORE_TAUTOLOGICAL_COMPARE
 #else
-# define V4L2_CID_USER_IMX_BASE (V4L2_CID_BASE + 0x1090)
+# define V4L2_CID_USER_IMX_BASE (V4L2_CID_BASE + 0x10b0)
 #endif
 #if defined(V4L2_CID_MPEG_BASE) || (defined(HAVE_DECL_V4L2_CID_MPEG_BASE) && HAVE_DECL_V4L2_CID_MPEG_BASE)
 DIAG_PUSH_IGNORE_TAUTOLOGICAL_COMPARE
@@ -146,6 +150,21 @@
  XLAT(V4L2_CID_USER_MEYE_BASE),
  XLAT(V4L2_CID_USER_BTTV_BASE),
  XLAT(V4L2_CID_USER_TI_VPE_BASE),
+#ifndef STRACE_WORKAROUND_FOR_V4L2_CID_USER_IMX_BASE
+# define STRACE_WORKAROUND_FOR_V4L2_CID_USER_IMX_BASE
+/*
+* Linux kernel commit v4.18-rc2-106-g421860b9d47053badce4b247576fa48df9ab4c48
+* has changed the value of V4L2_CID_USER_IMX_BASE constant introduced
+* by commit v4.13-rc1~141^2~121 because the old value was already used
+* by V4L2_CID_USER_MAX217X_BASE.
+* This is of course an ABI breakage that affects Linux kernels starting
+* with 4.13 and up to 4.18, as well as their LTS derivatives.
+* Since the imx driver didn't provide any public control ID definitions,
+* it looks like the best way to handle this situation is to pretend that
+* the old value of V4L2_CID_USER_IMX_BASE didn't exist.
+*/
+# undef V4L2_CID_USER_IMX_BASE
+#endif
  XLAT(V4L2_CID_USER_IMX_BASE),
  XLAT(V4L2_CID_MPEG_BASE),
  XLAT(V4L2_CID_MPEG_CX2341X_BASE),
diff --git a/xlat/v4l2_control_id_bases.in b/xlat/v4l2_control_id_bases.in
index a0359be..f3fd925 100644
--- a/xlat/v4l2_control_id_bases.in
+++ b/xlat/v4l2_control_id_bases.in
@@ -2,7 +2,22 @@
 V4L2_CID_USER_MEYE_BASE			(V4L2_CID_BASE + 0x1000)
 V4L2_CID_USER_BTTV_BASE			(V4L2_CID_BASE + 0x1010)
 V4L2_CID_USER_TI_VPE_BASE		(V4L2_CID_BASE + 0x1050)
-V4L2_CID_USER_IMX_BASE			(V4L2_CID_BASE + 0x1090)
+#ifndef STRACE_WORKAROUND_FOR_V4L2_CID_USER_IMX_BASE
+# define STRACE_WORKAROUND_FOR_V4L2_CID_USER_IMX_BASE
+/*
+ * Linux kernel commit v4.18-rc2-106-g421860b9d47053badce4b247576fa48df9ab4c48
+ * has changed the value of V4L2_CID_USER_IMX_BASE constant introduced
+ * by commit v4.13-rc1~141^2~121 because the old value was already used
+ * by V4L2_CID_USER_MAX217X_BASE.
+ * This is of course an ABI breakage that affects Linux kernels starting
+ * with 4.13 and up to 4.18, as well as their LTS derivatives.
+ * Since the imx driver didn't provide any public control ID definitions,
+ * it looks like the best way to handle this situation is to pretend that
+ * the old value of V4L2_CID_USER_IMX_BASE didn't exist.
+ */
+# undef V4L2_CID_USER_IMX_BASE
+#endif
+V4L2_CID_USER_IMX_BASE			(V4L2_CID_BASE + 0x10b0)
 V4L2_CID_MPEG_BASE			(V4L2_CTRL_CLASS_MPEG | 0x900)
 V4L2_CID_MPEG_CX2341X_BASE		(V4L2_CTRL_CLASS_MPEG | 0x1000)
 V4L2_CID_MPEG_MFC51_BASE		(V4L2_CTRL_CLASS_MPEG | 0x1100)