[5/5] openmp: -foffload-memory=pinned
Commit Message
From: Andrew Stubbs <ams@codesourcery.com>
Implement the -foffload-memory=pinned option such that libgomp is
instructed to enable fully-pinned memory at start-up. The option is
intended to provide a performance boost to certain offload programs without
modifying the code.
This feature only works on Linux, at present, and simply calls mlockall to
enable always-on memory pinning. It requires that the ulimit feature is
set high enough to accommodate all the program's memory usage.
In this mode the ompx_pinned_memory_alloc feature is disabled as it is not
needed and may conflict.
gcc/ChangeLog:
* omp-low.cc (omp_enable_pinned_mode): New function.
(execute_lower_omp): Call omp_enable_pinned_mode.
libgomp/ChangeLog:
* config/linux/allocator.c (always_pinned_mode): New variable.
(GOMP_enable_pinned_mode): New function.
(linux_memspace_alloc): Disable pinning when always_pinned_mode set.
(linux_memspace_calloc): Likewise.
(linux_memspace_free): Likewise.
(linux_memspace_realloc): Likewise.
* libgomp.map (GOMP_5.1.1): New version space with
GOMP_enable_pinned_mode.
* testsuite/libgomp.c/alloc-pinned-7.c: New test.
gcc/testsuite/ChangeLog:
* c-c++-common/gomp/alloc-pinned-1.c: New test.
---
gcc/omp-low.cc | 68 +++++++++++++++++++
.../c-c++-common/gomp/alloc-pinned-1.c | 28 ++++++++
libgomp/config/linux/allocator.c | 26 +++++++
libgomp/libgomp.map | 5 ++
libgomp/testsuite/libgomp.c/alloc-pinned-7.c | 66 ++++++++++++++++++
5 files changed, 193 insertions(+)
create mode 100644 gcc/testsuite/c-c++-common/gomp/alloc-pinned-1.c
create mode 100644 libgomp/testsuite/libgomp.c/alloc-pinned-7.c
Comments
On 08/03/2022 11:30, Hafiz Abid Qadeer wrote:
> gcc/ChangeLog:
>
> * omp-low.cc (omp_enable_pinned_mode): New function.
> (execute_lower_omp): Call omp_enable_pinned_mode.
This worked for x86_64, but I needed to make the attached adjustment to
work on powerpc without a linker error.
I've committed it to OG11.
Andrew
openmp: BUILT_IN_GOMP_ENABLE_PINNED_MODE
Rework the GOMP_enable_pinned_mode call so that it works on powerpc where
the old way gave a local call.
gcc/ChangeLog:
* omp-builtins.def (BUILT_IN_GOMP_ENABLE_PINNED_MODE): New.
* omp-low.c (omp_enable_pinned_mode): Use
BUILT_IN_GOMP_ENABLE_PINNED_MODE.
diff --git a/gcc/omp-builtins.def b/gcc/omp-builtins.def
index c591d79fa07..e442b0b5c94 100644
--- a/gcc/omp-builtins.def
+++ b/gcc/omp-builtins.def
@@ -468,3 +468,6 @@ DEF_GOMP_BUILTIN (BUILT_IN_GOMP_ERROR, "GOMP_error",
DEF_GOMP_BUILTIN (BUILT_IN_GOMP_EVALUATE_TARGET_DEVICE, "GOMP_evaluate_target_device",
BT_FN_BOOL_INT_CONST_PTR_CONST_PTR_CONST_PTR,
ATTR_NOTHROW_LEAF_LIST)
+DEF_GOMP_BUILTIN (BUILT_IN_GOMP_ENABLE_PINNED_MODE,
+ "GOMP_enable_pinned_mode",
+ BT_FN_VOID, ATTR_NOTHROW_LIST)
diff --git a/gcc/omp-low.c b/gcc/omp-low.c
index f7ecfb52c73..4e8ab9e4ca0 100644
--- a/gcc/omp-low.c
+++ b/gcc/omp-low.c
@@ -15432,9 +15432,7 @@ omp_enable_pinned_mode ()
push_struct_function (decl);
init_tree_ssa (cfun);
- tree callname = get_identifier ("GOMP_enable_pinned_mode");
- tree calldecl = build_decl (UNKNOWN_LOCATION, FUNCTION_DECL, callname,
- voidfntype);
+ tree calldecl = builtin_decl_explicit (BUILT_IN_GOMP_ENABLE_PINNED_MODE);
gcall *call = gimple_build_call (calldecl, 0);
gimple_seq seq = NULL;
@@ -14441,6 +14441,70 @@ lower_omp (gimple_seq *body, omp_context *ctx)
input_location = saved_location;
}
+/* Emit a constructor function to enable -foffload-memory=pinned
+ at runtime. Libgomp handles the OS mode setting, but we need to trigger
+ it by calling GOMP_enable_pinned mode before the program proper runs. */
+
+static void
+omp_enable_pinned_mode ()
+{
+ static bool visited = false;
+ if (visited)
+ return;
+ visited = true;
+
+ /* Create a new function like this:
+
+ static void __attribute__((constructor))
+ __set_pinned_mode ()
+ {
+ GOMP_enable_pinned_mode ();
+ }
+ */
+
+ tree name = get_identifier ("__set_pinned_mode");
+ tree voidfntype = build_function_type_list (void_type_node, NULL_TREE);
+ tree decl = build_decl (UNKNOWN_LOCATION, FUNCTION_DECL, name, voidfntype);
+
+ TREE_STATIC (decl) = 1;
+ TREE_USED (decl) = 1;
+ DECL_ARTIFICIAL (decl) = 1;
+ DECL_IGNORED_P (decl) = 0;
+ TREE_PUBLIC (decl) = 0;
+ DECL_UNINLINABLE (decl) = 1;
+ DECL_EXTERNAL (decl) = 0;
+ DECL_CONTEXT (decl) = NULL_TREE;
+ DECL_INITIAL (decl) = make_node (BLOCK);
+ BLOCK_SUPERCONTEXT (DECL_INITIAL (decl)) = decl;
+ DECL_STATIC_CONSTRUCTOR (decl) = 1;
+ DECL_ATTRIBUTES (decl) = tree_cons (get_identifier ("constructor"),
+ NULL_TREE, NULL_TREE);
+
+ tree t = build_decl (UNKNOWN_LOCATION, RESULT_DECL, NULL_TREE,
+ void_type_node);
+ DECL_ARTIFICIAL (t) = 1;
+ DECL_IGNORED_P (t) = 1;
+ DECL_CONTEXT (t) = decl;
+ DECL_RESULT (decl) = t;
+
+ push_struct_function (decl);
+ init_tree_ssa (cfun);
+
+ tree callname = get_identifier ("GOMP_enable_pinned_mode");
+ tree calldecl = build_decl (UNKNOWN_LOCATION, FUNCTION_DECL, callname,
+ voidfntype);
+ gcall *call = gimple_build_call (calldecl, 0);
+
+ gimple_seq seq = NULL;
+ gimple_seq_add_stmt (&seq, call);
+ gimple_set_body (decl, gimple_build_bind (NULL_TREE, seq, NULL));
+
+ cfun->function_end_locus = UNKNOWN_LOCATION;
+ cfun->curr_properties |= PROP_gimple_any;
+ pop_cfun ();
+ cgraph_node::add_new_function (decl, true);
+}
+
/* Main entry point. */
static unsigned int
@@ -14497,6 +14561,10 @@ execute_lower_omp (void)
for (auto task_stmt : task_cpyfns)
finalize_task_copyfn (task_stmt);
task_cpyfns.release ();
+
+ if (flag_offload_memory == OFFLOAD_MEMORY_PINNED)
+ omp_enable_pinned_mode ();
+
return 0;
}
new file mode 100644
@@ -0,0 +1,28 @@
+/* { dg-do run } */
+/* { dg-additional-options "-foffload-memory=pinned" } */
+/* { dg-xfail-run-if "Pinning not implemented on this host" { ! *-*-linux-gnu } } */
+
+#if __cplusplus
+#define EXTERNC extern "C"
+#else
+#define EXTERNC
+#endif
+
+/* Intercept the libgomp initialization call to check it happens. */
+
+int good = 0;
+
+EXTERNC void
+GOMP_enable_pinned_mode ()
+{
+ good = 1;
+}
+
+int
+main ()
+{
+ if (!good)
+ __builtin_exit (1);
+
+ return 0;
+}
@@ -39,9 +39,26 @@
#include <string.h>
#include "libgomp.h"
+static bool always_pinned_mode = false;
+
+/* This function is called by the compiler when -foffload-memory=pinned
+ is used. */
+
+void
+GOMP_enable_pinned_mode ()
+{
+ if (mlockall (MCL_CURRENT | MCL_FUTURE) != 0)
+ gomp_error ("failed to pin all memory (ulimit too low?)");
+ else
+ always_pinned_mode = true;
+}
+
static void *
linux_memspace_alloc (omp_memspace_handle_t memspace, size_t size, int pin)
{
+ /* Explicit pinning may not be required. */
+ pin = pin && !always_pinned_mode;
+
if (memspace == ompx_unified_shared_mem_space)
{
return gomp_usm_alloc (size, GOMP_DEVICE_ICV);
@@ -69,6 +86,9 @@ linux_memspace_alloc (omp_memspace_handle_t memspace, size_t size, int pin)
static void *
linux_memspace_calloc (omp_memspace_handle_t memspace, size_t size, int pin)
{
+ /* Explicit pinning may not be required. */
+ pin = pin && !always_pinned_mode;
+
if (memspace == ompx_unified_shared_mem_space)
{
void *ret = gomp_usm_alloc (size, GOMP_DEVICE_ICV);
@@ -86,6 +106,9 @@ static void
linux_memspace_free (omp_memspace_handle_t memspace, void *addr, size_t size,
int pin)
{
+ /* Explicit pinning may not be required. */
+ pin = pin && !always_pinned_mode;
+
if (memspace == ompx_unified_shared_mem_space)
gomp_usm_free (addr, GOMP_DEVICE_ICV);
else if (pin)
@@ -98,6 +121,9 @@ static void *
linux_memspace_realloc (omp_memspace_handle_t memspace, void *addr,
size_t oldsize, size_t size, int oldpin, int pin)
{
+ /* Explicit pinning may not be required. */
+ pin = pin && !always_pinned_mode;
+
if (memspace == ompx_unified_shared_mem_space)
goto manual_realloc;
else if (oldpin && pin)
@@ -402,6 +402,11 @@ GOMP_5.1 {
GOMP_teams4;
} GOMP_5.0.1;
+GOMP_5.1.1 {
+ global:
+ GOMP_enable_pinned_mode;
+} GOMP_5.1;
+
OACC_2.0 {
global:
acc_get_num_devices;
new file mode 100644
@@ -0,0 +1,66 @@
+/* { dg-do run } */
+/* { dg-additional-options "-foffload-memory=pinned" } */
+
+/* { dg-xfail-run-if "Pinning not implemented on this host" { ! *-*-linux-gnu } } */
+
+/* Test that pinned memory works. */
+
+#ifdef __linux__
+#include <sys/types.h>
+#include <unistd.h>
+#include <stdio.h>
+#include <stdlib.h>
+
+#include <sys/mman.h>
+
+int
+get_pinned_mem ()
+{
+ int pid = getpid ();
+ char buf[100];
+ sprintf (buf, "/proc/%d/status", pid);
+
+ FILE *proc = fopen (buf, "r");
+ if (!proc)
+ abort ();
+ while (fgets (buf, 100, proc))
+ {
+ int val;
+ if (sscanf (buf, "VmLck: %d", &val))
+ {
+ printf ("lock %d\n", val);
+ fclose (proc);
+ return val;
+ }
+ }
+ abort ();
+}
+#else
+int
+get_pinned_mem ()
+{
+ return 0;
+}
+
+#define mlockall(...) 0
+#endif
+
+#include <omp.h>
+
+/* Allocate more than a page each time, but stay within the ulimit. */
+#define SIZE 10*1024
+
+int
+main ()
+{
+ // Sanity check
+ if (get_pinned_mem () == 0)
+ {
+ /* -foffload-memory=pinned has failed, but maybe that's because
+ isufficient pinned memory was available. */
+ if (mlockall (MCL_CURRENT | MCL_FUTURE) == 0)
+ abort ();
+ }
+
+ return 0;
+}