presets: remove PAPI_FP_OPS from POWER9 & POWER10

The previous definition counted floating-point instructions instead of
operations. Added note to papi_events.csv that explains how to scale and
combine native events to measure all FLOPs with multiplexing.

These changes have been tested on the IBM POWER9 and POWER10
architectures.

src/configure

@@ -6827,6 +6827,7 @@ else
   with_sysdetect=yes
 fi
 
+# Enable sysdetect unless the user has explicitly told us not to.
 if test "$with_sysdetect" = "yes"; then
   { $as_echo "$as_me:${as_lineno-$LINENO}: result: yes" >&5
 $as_echo "yes" >&6; }
@@ -6916,7 +6917,6 @@ if test "$with_sysdetect" = "yes"; then
     fi
 fi
 
-
 PAPI_NUM_COMP=0
 for comp in $components; do
   idx=`echo "$comp" | sed -n "s/\/.*//p" | wc -c`

src/configure.in

@@ -1823,6 +1823,7 @@ AC_ARG_WITH(sysdetect,
                             [Build the sysdetect component (default: yes)])],
             [],
             [with_sysdetect=yes])
+# Enable sysdetect unless the user has explicitly told us not to.
 if test "$with_sysdetect" = "yes"; then
   AC_MSG_RESULT(yes)
 else

src/papi_events.csv

@@ -1844,7 +1844,10 @@ PRESET,PAPI_FMA_INS,NOT_DERIVED,PM_FMA_CMPL
 PRESET,PAPI_TOT_IIS,NOT_DERIVED,PM_INST_DISP
 PRESET,PAPI_TOT_INS,NOT_DERIVED,PM_INST_CMPL
 PRESET,PAPI_INT_INS,NOT_DERIVED,PM_FXU_FIN
-PRESET,PAPI_FP_OPS,NOT_DERIVED,PM_FLOP_CMPL
+# Note: PAPI_FP_OPS is not available on this architecture. The following combination is
+# equivalent to all FLOPs; however, these events cannot be added to the same event set.
+# If a user chooses, they can utilize the multiplexing feature with these events.
+# 8 * PM_8FLOP_CMPL + 4 * PM_4FLOP_CMPL + 2 * PM_2FLOP_CMPL + 1 * PM_1FLOP_CMPL
 PRESET,PAPI_FP_INS,NOT_DERIVED,PM_FLOP_CMPL
 PRESET,PAPI_DP_OPS,NOT_DERIVED,PM_DP_QP_FLOP_CMPL
 PRESET,PAPI_SP_OPS,NOT_DERIVED,PM_SP_FLOP_CMPL
@@ -1892,7 +1895,10 @@ PRESET,PAPI_FMA_INS,NOT_DERIVED,PM_FMA_CMPL
 PRESET,PAPI_TOT_IIS,NOT_DERIVED,PM_INST_DISP
 PRESET,PAPI_TOT_INS,NOT_DERIVED,PM_INST_CMPL
 PRESET,PAPI_INT_INS,NOT_DERIVED,PM_FXU_ISSUE
-PRESET,PAPI_FP_OPS,NOT_DERIVED,PM_FLOP_CMPL
+# Note: PAPI_FP_OPS is not available on this architecture. The following combination is
+# equivalent to all FLOPs; however, these events cannot be added to the same event set.
+# If a user chooses, they can utilize the multiplexing feature with these events.
+# 8 * PM_8FLOP_CMPL + 4 * PM_4FLOP_CMPL + 2 * PM_2FLOP_CMPL + 1 * PM_1FLOP_CMPL
 PRESET,PAPI_FP_INS,NOT_DERIVED,PM_FLOP_CMPL
 PRESET,PAPI_DP_OPS,NOT_DERIVED,PM_DPP_FLOP_CMPL
 PRESET,PAPI_SP_OPS,NOT_DERIVED,PM_SP_FLOP_CMPL