freedreno: import libdrm_freedreno + redesign submit

In the pursuit of lowering driver overhead, it became clear that some
amount of redesign of how libdrm_freedreno constructs the submit ioctl
would be needed.  In particular, as the gallium driver is starting to
make heavier use of CP_SET_DRAW_STATE state groups/objects, the over-
head of tracking cmd buffers and relocs becomes too much.  And for
"streaming" state, which isn't ever reused (like uniform uploads) the
overhead of allocating/freeing ringbuffer[1] objects is too high.

This redesign makes two main changes:

 1) Introduces a fd_submit object for tracking bos and cmds table
    for the submit ioctl, making ringbuffer objects more light-
    weight.  This was previously done in the ringbuffer.  But we
    have many ringbuffer instances involved in a submit (gmem +
    draw + potentially 1000's of state-group rbs), and only need
    a single bos and cmds table.  (Reloc table is still per-rb)

    The submit is also a convenient place for a slab allocator for
    ringbuffer objects.  Other options would have required locking
    because, while we can guarantee allocations will only happen on
    a single thread, free's could happen either on the application
    thread or the flush_queue thread.  With the slab allocator in
    the submit object, any frees that happen on the flush_queue
    thread happen after we know that the application thread is done
    with the submit.

 2) Introduce a new "softpin" msm_ringbuffer_sp implementation that
    does not use relocs and only has cmds table entries for IB1 (ie.
    the cmdstream buffers that kernel needs to CP_INDIRECT_BUFFER
    to from the RB).  To do this properly will require some updates
    on the kernel side, so whether you get the softpin or legacy
    submit/ringbuffer implementation at runtime depends on your
    kernel version.

To make all these changes in libdrm would basically require adding a
libdrm_freedreno2, so this is a good point to just pull the libdrm code
into mesa.  Plus it allows for using mesa's hashtable, slab allocator,
etc.  And it lets us have asserts enabled for debug mesa buids but
omitted for release builds.  And it makes life easier if further API
changes become necessary.

At this point I haven't tried to pull in the kgsl backend.  Although
I left the level of vfunc indirection which would make it possible
to have other backends.  (And this was convenient to keep to allow
for the "softpin" ringbuffer to coexist.)

NOTE: if bisecting a build error takes you here, try a clean build.
There are a bunch of ways things can go wrong if you still have
libdrm_freedreno cflags.

[1] "ringbuffer" is probably a bad name, the only level of cmdstream
    buffer that is actually a ring is RB managed by kernel.  User-
    space cmdstream is all IB1/IB2 and state-groups.

Reviewed-by: Kristian H. Kristensen <[email protected]>
Reviewed-by: Eric Engestrom <[email protected]>
Signed-off-by: Rob Clark <[email protected]>

1 files changed, 156 insertions, 0 deletions

diff --git a/src/gallium/drivers/freedreno/drm/freedreno_device.c b/src/gallium/drivers/freedreno/drm/freedreno_device.c
new file mode 100644
index 00000000000..b2f6c981963
--- /dev/null
+++ b/src/gallium/drivers/freedreno/drm/freedreno_device.c
@@ -0,0 +1,156 @@
+/*
+ * Copyright (C) 2012-2018 Rob Clark <[email protected]>
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ *
+ * Authors:
+ *    Rob Clark <[email protected]>
+ */
+
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <unistd.h>
+
+#include "freedreno_drmif.h"
+#include "freedreno_priv.h"
+
+static pthread_mutex_t table_lock = PTHREAD_MUTEX_INITIALIZER;
+
+static uint32_t
+u32_hash(const void *key)
+{
+	return _mesa_hash_data(key, sizeof(uint32_t));
+}
+
+static bool
+u32_equals(const void *key1, const void *key2)
+{
+	return *(const uint32_t *)key1 == *(const uint32_t *)key2;
+}
+
+
+struct fd_device * kgsl_device_new(int fd);
+struct fd_device * msm_device_new(int fd);
+
+struct fd_device * fd_device_new(int fd)
+{
+	struct fd_device *dev;
+	drmVersionPtr version;
+
+	/* figure out if we are kgsl or msm drm driver: */
+	version = drmGetVersion(fd);
+	if (!version) {
+		ERROR_MSG("cannot get version: %s", strerror(errno));
+		return NULL;
+	}
+
+	if (!strcmp(version->name, "msm")) {
+		DEBUG_MSG("msm DRM device");
+		if (version->version_major != 1) {
+			ERROR_MSG("unsupported version: %u.%u.%u", version->version_major,
+				version->version_minor, version->version_patchlevel);
+			dev = NULL;
+			goto out;
+		}
+
+		dev = msm_device_new(fd);
+		dev->version = version->version_minor;
+#if HAVE_FREEDRENO_KGSL
+	} else if (!strcmp(version->name, "kgsl")) {
+		DEBUG_MSG("kgsl DRM device");
+		dev = kgsl_device_new(fd);
+#endif
+	} else {
+		ERROR_MSG("unknown device: %s", version->name);
+		dev = NULL;
+	}
+
+out:
+	drmFreeVersion(version);
+
+	if (!dev)
+		return NULL;
+
+	p_atomic_set(&dev->refcnt, 1);
+	dev->fd = fd;
+	dev->handle_table = _mesa_hash_table_create(NULL, u32_hash, u32_equals);
+	dev->name_table = _mesa_hash_table_create(NULL, u32_hash, u32_equals);
+	fd_bo_cache_init(&dev->bo_cache, FALSE);
+	fd_bo_cache_init(&dev->ring_cache, TRUE);
+
+	return dev;
+}
+
+/* like fd_device_new() but creates it's own private dup() of the fd
+ * which is close()d when the device is finalized.
+ */
+struct fd_device * fd_device_new_dup(int fd)
+{
+	int dup_fd = dup(fd);
+	struct fd_device *dev = fd_device_new(dup_fd);
+	if (dev)
+		dev->closefd = 1;
+	else
+		close(dup_fd);
+	return dev;
+}
+
+struct fd_device * fd_device_ref(struct fd_device *dev)
+{
+	p_atomic_inc(&dev->refcnt);
+	return dev;
+}
+
+static void fd_device_del_impl(struct fd_device *dev)
+{
+	int close_fd = dev->closefd ? dev->fd : -1;
+	fd_bo_cache_cleanup(&dev->bo_cache, 0);
+	_mesa_hash_table_destroy(dev->handle_table, NULL);
+	_mesa_hash_table_destroy(dev->name_table, NULL);
+	dev->funcs->destroy(dev);
+	if (close_fd >= 0)
+		close(close_fd);
+}
+
+void fd_device_del_locked(struct fd_device *dev)
+{
+	if (!atomic_dec_and_test(&dev->refcnt))
+		return;
+	fd_device_del_impl(dev);
+}
+
+void fd_device_del(struct fd_device *dev)
+{
+	if (!atomic_dec_and_test(&dev->refcnt))
+		return;
+	pthread_mutex_lock(&table_lock);
+	fd_device_del_impl(dev);
+	pthread_mutex_unlock(&table_lock);
+}
+
+int fd_device_fd(struct fd_device *dev)
+{
+	return dev->fd;
+}
+
+enum fd_version fd_device_version(struct fd_device *dev)
+{
+	return dev->version;
+}