Move w_lgamma to libm-compat-calls-auto

This patch adds the suffix "_compat2" to the wrappers for lgamma,
which use _LIB_VERSION / matherr / __kernel_standard functionality.
The suffix "_compat2" is used because the suffix "_compat" is already
used.

Tested for powerpc64le, s390, and x86_64.

	* math/Makefile (libm-calls): Move w_lgammaF...
	(libm-compat-calls-auto): Here.

	* math/w_lgamma.c: Add suffix "_compat2" to filename.
	* math/w_lgammaf.c: Likewise.
	* math/w_lgammal.c: Likewise.

	* math/w_lgamma_compat2.c: New file, copied from above.
	* math/w_lgammaf_compat2.c: Likewise.
	* math/w_lgammal_compat2.c: Likewise.

ChangeLog

@@ -1,3 +1,16 @@
+2017-02-08  Gabriel F. T. Gomes  <gftg@linux.vnet.ibm.com>
+
+	* math/Makefile (libm-calls): Move w_lgammaF...
+	(libm-compat-calls-auto): Here.
+
+	* math/w_lgamma.c: Add suffix "_compat2" to filename.
+	* math/w_lgammaf.c: Likewise.
+	* math/w_lgammal.c: Likewise.
+
+	* math/w_lgamma_compat2.c: New file, copied from above.
+	* math/w_lgammaf_compat2.c: Likewise.
+	* math/w_lgammal_compat2.c: Likewise.
+
 2017-02-08  Gabriel F. T. Gomes  <gftg@linux.vnet.ibm.com>
 
 	* math/Makefile (libm-calls): Move w_lgammaF_r...

math/Makefile

@@ -67,7 +67,6 @@ libm-calls =								  \
 	s_nextafterF s_nexttowardF s_rintF s_scalblnF			  \
 	s_significandF s_sinF s_tanF s_tanhF				  \
 	w_expF								  \
-	w_lgammaF							  \
 	s_fpclassifyF s_truncF						  \
 	s_remquoF e_log2F e_exp2F s_roundF s_nearbyintF s_sincosF	  \
 	s_fmaF s_lrintF s_llrintF s_lroundF s_llroundF e_exp10F		  \
@@ -84,7 +83,7 @@ libm-compat-calls-auto =						\
 	w_jnF_compat w_log2F_compat w_log10F_compat w_logF_compat	\
 	w_powF_compat w_remainderF_compat w_scalbF_compat		\
 	w_sinhF_compat w_sqrtF_compat w_tgammaF_compat			\
-	w_lgammaF_r_compat
+	w_lgammaF_r_compat w_lgammaF_compat2
 
 libm-compat-calls-ldouble-yes = w_lgamma_compatl k_standardl
 libm-compat-calls = w_lgamma_compatf w_lgamma_compat k_standard k_standardf \