remove the old .nqp-based benchmarks. Replace with a series of pir-ba…

…sed benchmarks with a factored library file

benchmarks/benchmarker.pir

@@ -0,0 +1,95 @@
+.sub push_shift_benchmarks
+    .param string targettype
+    .param int numtrials
+    .param int preallocate
+
+    $P0 = new [targettype]
+    $P1 = new ['Integer']
+    $P1 = 5
+    $I0 = numtrials
+    $N0 = time
+
+    if preallocate goto preallocate_storage
+    goto storage_is_ok
+  preallocate_storage:
+    $P0 = numtrials
+  storage_is_ok:
+
+  push_loop_top:
+    push $P0, $P1
+    $I0 = $I0 - 1
+    if $I0 == 0 goto push_loop_bottom
+    goto push_loop_top
+  push_loop_bottom:
+    $N1 = time
+    $I0 = numtrials
+  shift_loop_top:
+    $P2 = shift $P0
+    $I0 = $I0 - 1
+    if $I0 == 0 goto shift_loop_bottom
+    goto shift_loop_top
+  shift_loop_bottom:
+    $N2 = time
+    $N3 = $N1 - $N0
+    $N4 = $N2 - $N1
+    $N5 = $N2 - $N0
+    print_partial_result(targettype, "push ", numtrials, $N3)
+    print_partial_result(targettype, "shift", numtrials, $N4)
+    print_partial_result(targettype, "total", numtrials, $N5)
+.end
+
+.sub push_pop_benchmarks
+    .param string targettype
+    .param int numtrials
+    .param int preallocate
+
+    $P0 = new [targettype]
+    $P1 = new ['Integer']
+    $P1 = 5
+    $I0 = numtrials
+    $N0 = time
+
+    if preallocate goto preallocate_storage
+    goto storage_is_ok
+  preallocate_storage:
+    $P0 = numtrials
+  storage_is_ok:
+
+  push_loop_top:
+    push $P0, $P1
+    $I0 = $I0 - 1
+    if $I0 == 0 goto push_loop_bottom
+    goto push_loop_top
+  push_loop_bottom:
+    $N1 = time
+    $I0 = numtrials
+  shift_loop_top:
+    $P2 = pop $P0
+    $I0 = $I0 - 1
+    if $I0 == 0 goto shift_loop_bottom
+    goto shift_loop_top
+  shift_loop_bottom:
+    $N2 = time
+    $N3 = $N1 - $N0
+    $N4 = $N2 - $N1
+    $N5 = $N2 - $N0
+    print_partial_result(targettype, "push ", numtrials, $N3)
+    print_partial_result(targettype, "pop  ", numtrials, $N4)
+    print_partial_result(targettype, "total", numtrials, $N5)
+.end
+
+.sub print_partial_result
+    .param string targettype
+    .param string operation
+    .param int numtrials
+    .param num totaltime
+
+    print "("
+    print targettype
+    print ") Time to "
+    print operation
+    print " "
+    print numtrials
+    print ": "
+    say totaltime
+.end

benchmarks/fixedpmcqueue.pir

@@ -0,0 +1,11 @@
+.sub main :main
+    .local pmc pds
+    pds = loadlib "./dynext/pds_group"
+    if pds goto has_pds_group
+    exit 1
+  has_pds_group:
+    push_shift_benchmarks("FixedPMCQueue", 100000, 1)
+    say ""
+.end
+
+.include "benchmarks/benchmarker.pir"

benchmarks/fixedpmcstack.pir

@@ -0,0 +1,11 @@
+.sub main :main
+    .local pmc pds
+    pds = loadlib "./dynext/pds_group"
+    if pds goto has_pds_group
+    exit 1
+  has_pds_group:
+    push_pop_benchmarks("FixedPMCStack", 1000000, 1)
+    say ""
+.end
+
+.include "benchmarks/benchmarker.pir"

benchmarks/resizablepmcarray.pir

@@ -0,0 +1,13 @@
+.sub main :main
+    .local pmc pds
+    pds = loadlib "./dynext/pds_group"
+    if pds goto has_pds_group
+    exit 1
+  has_pds_group:
+    push_pop_benchmarks("ResizablePMCArray", 1000000, 0)
+    say ""
+    push_shift_benchmarks("ResizablePMCArray", 100000, 0)
+    say ""
+.end
+
+.include "benchmarks/benchmarker.pir"