[PATCH 01/18] tests/gem_storedw_loop: add new store_dword test to unify per-ring ones

[PATCH 01/18] tests/gem_storedw_loop: add new store_dword test to unify per-ring ones

[Jesse Barnes]

There was a lot of duplication going on...  Mark as basic while we're at
it as these should never fail.

Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 tests/Makefile.sources   |   1 +
 tests/gem_storedw_loop.c | 181 +++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 182 insertions(+)
 create mode 100644 tests/gem_storedw_loop.c

diff --git a/tests/Makefile.sources b/tests/Makefile.sources
index b9a4cb4..cdcee33 100644
--- a/tests/Makefile.sources
+++ b/tests/Makefile.sources
@@ -138,6 +138,7 @@ TESTS_progs = \
 	gem_seqno_wrap \
 	gem_set_tiling_vs_gtt \
 	gem_set_tiling_vs_pwrite \
+	gem_storedw_loop \
 	gem_storedw_loop_blt \
 	gem_storedw_loop_bsd \
 	gem_storedw_loop_render \
diff --git a/tests/gem_storedw_loop.c b/tests/gem_storedw_loop.c
new file mode 100644
index 0000000..e4ed35d
--- /dev/null
+++ b/tests/gem_storedw_loop.c
@@ -0,0 +1,181 @@
+/*
+ * Copyright © 2009 Intel Corporation
+ *
+ * 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:
+ *    Eric Anholt <eric@anholt.net>
+ *    Jesse Barnes <jbarnes@virtuousgeek.org> (based on gem_bad_blit.c)
+ *
+ */
+
+#include <stdlib.h>
+#include <stdio.h>
+#include <string.h>
+#include <fcntl.h>
+#include <inttypes.h>
+#include <errno.h>
+#include <sys/stat.h>
+#include <sys/time.h>
+#include "drm.h"
+#include "ioctl_wrappers.h"
+#include "drmtest.h"
+#include "intel_bufmgr.h"
+#include "intel_batchbuffer.h"
+#include "intel_io.h"
+#include "intel_chipset.h"
+
+IGT_TEST_DESCRIPTION("Basic CS check using MI_STORE_DATA_IMM.");
+
+#define LOCAL_I915_EXEC_VEBOX (4<<0)
+
+static drm_intel_bufmgr *bufmgr;
+struct intel_batchbuffer *batch;
+static drm_intel_bo *target_buffer;
+static int has_ppgtt = 0;
+static int devid;
+
+/*
+ * Testcase: Basic bsd MI check using MI_STORE_DATA_IMM
+ */
+
+static void
+emit_store_dword_imm(drm_intel_bo *dest, uint32_t val)
+{
+	int cmd;
+	cmd = MI_STORE_DWORD_IMM;
+	if (!has_ppgtt)
+		cmd |= MI_MEM_VIRTUAL;
+
+	BEGIN_BATCH(4, 0);
+	OUT_BATCH(cmd);
+	if (batch->gen >= 8) {
+		OUT_RELOC(dest, I915_GEM_DOMAIN_INSTRUCTION,
+			  I915_GEM_DOMAIN_INSTRUCTION, 0);
+		OUT_BATCH(val);
+	} else {
+		OUT_BATCH(0); /* reserved */
+		OUT_RELOC(dest, I915_GEM_DOMAIN_INSTRUCTION,
+			  I915_GEM_DOMAIN_INSTRUCTION, 0);
+		OUT_BATCH(val);
+	}
+	ADVANCE_BATCH();
+}
+
+static void
+store_dword_loop(int ring, int count, int divider)
+{
+	int i, val = 0;
+	uint32_t *buf;
+
+	igt_info("running storedw loop on render with stall every %i batch\n", divider);
+
+	for (i = 0; i < SLOW_QUICK(0x2000, 0x10); i++) {
+		emit_store_dword_imm(target_buffer, val);
+		intel_batchbuffer_flush_on_ring(batch, ring);
+
+		if (i % divider != 0)
+			goto cont;
+
+		drm_intel_bo_map(target_buffer, 0);
+
+		buf = target_buffer->virtual;
+		igt_assert_f(buf[0] == val,
+			     "value mismatch: cur 0x%08x, stored 0x%08x\n",
+			     buf[0], val);
+
+		drm_intel_bo_unmap(target_buffer);
+
+cont:
+		val++;
+	}
+
+	drm_intel_bo_map(target_buffer, 0);
+	buf = target_buffer->virtual;
+
+	igt_info("completed %d writes successfully, current value: 0x%08x\n", i,
+			buf[0]);
+	drm_intel_bo_unmap(target_buffer);
+}
+
+static void
+store_test(int ring, int count)
+{
+	drm_intel_bufmgr_gem_enable_reuse(bufmgr);
+
+	batch = intel_batchbuffer_alloc(bufmgr, devid);
+	igt_assert(batch);
+
+	target_buffer = drm_intel_bo_alloc(bufmgr, "target bo", 4096, 4096);
+	igt_assert(target_buffer);
+
+	store_dword_loop(ring, count, 1);
+	store_dword_loop(ring, count, 2);
+	if (!igt_run_in_simulation()) {
+		store_dword_loop(ring, count, 3);
+		store_dword_loop(ring, count, 5);
+	}
+
+	drm_intel_bo_unreference(target_buffer);
+	intel_batchbuffer_free(batch);
+	drm_intel_bufmgr_destroy(bufmgr);
+}
+
+struct ring {
+	const char *name;
+	int id;
+} rings[] = {
+	{ "bsd", I915_EXEC_BSD },
+	{ "render", I915_EXEC_RENDER },
+	{ "blt", I915_EXEC_BLT },
+	{ "vebox", I915_EXEC_VEBOX },
+};
+
+igt_main
+{
+	int fd, i;
+
+	igt_fixture {
+		fd = drm_open_any();
+		devid = intel_get_drm_devid(fd);
+
+		bufmgr = drm_intel_bufmgr_gem_init(fd, 4096);
+		igt_assert(bufmgr);
+
+		has_ppgtt = gem_uses_aliasing_ppgtt(fd);
+
+		igt_skip_on_f(intel_gen(devid) < 6,
+			      "MI_STORE_DATA can only use GTT address on gen4+/g33 and "
+			      "needs snoopable mem on pre-gen6\n");
+
+		/* This only works with ppgtt */
+		igt_require(has_ppgtt);
+	}
+
+	for (i = 0; i < ARRAY_SIZE(rings); i++) {
+		igt_subtest_f("basic-%s", rings[i].name)
+			store_test(rings[i].id, 16*1024);
+
+		igt_subtest_f("long-%s", rings[i].name)
+			store_test(rings[i].id, 1024*1024);
+	}
+
+	close(fd);
+}
-- 
1.9.1

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

[PATCH 02/18] tests/drv_module_reload: rename drv_module_reload to include in BATs

[Jesse Barnes]

Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 tests/Makefile.sources        |  2 +-
 tests/drv_module_reload       | 60 -------------------------------------------
 tests/drv_module_reload_basic | 60 +++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 61 insertions(+), 61 deletions(-)
 delete mode 100755 tests/drv_module_reload
 create mode 100755 tests/drv_module_reload_basic

diff --git a/tests/Makefile.sources b/tests/Makefile.sources
index cdcee33..4fe7d4c 100644
--- a/tests/Makefile.sources
+++ b/tests/Makefile.sources
@@ -178,7 +178,7 @@ TESTS_scripts = \
 	debugfs_emon_crash \
 	drv_debugfs_reader \
 	drv_missed_irq_hang \
-	drv_module_reload \
+	drv_module_reload_basic \
 	kms_sysfs_edid_timing \
 	sysfs_l3_parity \
 	test_rte_check \
diff --git a/tests/drv_module_reload b/tests/drv_module_reload
deleted file mode 100755
index bb29a64..0000000
--- a/tests/drv_module_reload
+++ /dev/null
@@ -1,60 +0,0 @@
-#!/bin/bash
-#
-# Testcase: Reload the drm module
-#
-# ... we've broken this way too often :(
-#
-
-SOURCE_DIR="$( dirname "${BASH_SOURCE[0]}" )"
-. $SOURCE_DIR/drm_lib.sh
-
-# no other drm service should be running, so we can just unbind
-
-# we must kick away fbcon (but only fbcon)
-for vtcon in /sys/class/vtconsole/vtcon*/ ; do
-	if grep "frame buffer device" $vtcon/name > /dev/null ; then
-		echo unbinding $vtcon: `cat $vtcon/name`
-		echo 0 > $vtcon/bind
-	fi
-done
-
-# The sound driver uses our power well
-pkill alsactl
-rmmod snd_hda_intel &> /dev/null
-
-#ignore errors in ips - gen5 only
-rmmod intel_ips &> /dev/null
-rmmod i915
-#ignore errors in intel-gtt, often built-in
-rmmod intel-gtt &> /dev/null
-# drm may be used by other devices (nouveau, radeon, udl, etc)
-rmmod drm_kms_helper &> /dev/null
-rmmod drm &> /dev/null
-
-if lsmod | grep i915 &> /dev/null ; then
-	echo WARNING: i915.ko still loaded!
-	exit 1
-else
-	echo module successfully unloaded
-fi
-
-modprobe i915
-echo 1 > /sys/class/vtconsole/vtcon1/bind
-
-modprobe snd_hda_intel
-
-# does the device exist?
-if $SOURCE_DIR/gem_alive > /dev/null ; then
-	echo "module successfully loaded again"
-else
-	echo "failed to reload module successfully"
-	exit 2
-fi
-
-# then try to run something
-if ! $SOURCE_DIR/gem_exec_nop > /dev/null ; then
-	echo "failed to execute a simple batch after reload"
-	exit 3
-fi
-
-exit 0
diff --git a/tests/drv_module_reload_basic b/tests/drv_module_reload_basic
new file mode 100755
index 0000000..bb29a64
--- /dev/null
+++ b/tests/drv_module_reload_basic
@@ -0,0 +1,60 @@
+#!/bin/bash
+#
+# Testcase: Reload the drm module
+#
+# ... we've broken this way too often :(
+#
+
+SOURCE_DIR="$( dirname "${BASH_SOURCE[0]}" )"
+. $SOURCE_DIR/drm_lib.sh
+
+# no other drm service should be running, so we can just unbind
+
+# we must kick away fbcon (but only fbcon)
+for vtcon in /sys/class/vtconsole/vtcon*/ ; do
+	if grep "frame buffer device" $vtcon/name > /dev/null ; then
+		echo unbinding $vtcon: `cat $vtcon/name`
+		echo 0 > $vtcon/bind
+	fi
+done
+
+# The sound driver uses our power well
+pkill alsactl
+rmmod snd_hda_intel &> /dev/null
+
+#ignore errors in ips - gen5 only
+rmmod intel_ips &> /dev/null
+rmmod i915
+#ignore errors in intel-gtt, often built-in
+rmmod intel-gtt &> /dev/null
+# drm may be used by other devices (nouveau, radeon, udl, etc)
+rmmod drm_kms_helper &> /dev/null
+rmmod drm &> /dev/null
+
+if lsmod | grep i915 &> /dev/null ; then
+	echo WARNING: i915.ko still loaded!
+	exit 1
+else
+	echo module successfully unloaded
+fi
+
+modprobe i915
+echo 1 > /sys/class/vtconsole/vtcon1/bind
+
+modprobe snd_hda_intel
+
+# does the device exist?
+if $SOURCE_DIR/gem_alive > /dev/null ; then
+	echo "module successfully loaded again"
+else
+	echo "failed to reload module successfully"
+	exit 2
+fi
+
+# then try to run something
+if ! $SOURCE_DIR/gem_exec_nop > /dev/null ; then
+	echo "failed to execute a simple batch after reload"
+	exit 3
+fi
+
+exit 0
-- 
1.9.1

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

[PATCH 03/18] tests/drv_module_reload_basic: use linear_blits after module_reload for sanity check

[Jesse Barnes]

Reduces runtime a lot...

Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 tests/drv_module_reload_basic | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tests/drv_module_reload_basic b/tests/drv_module_reload_basic
index bb29a64..cf7f2b8 100755
--- a/tests/drv_module_reload_basic
+++ b/tests/drv_module_reload_basic
@@ -52,7 +52,7 @@ else
 fi
 
 # then try to run something
-if ! $SOURCE_DIR/gem_exec_nop > /dev/null ; then
+if ! $SOURCE_DIR/gem_linear_blits --r basic > /dev/null ; then
 	echo "failed to execute a simple batch after reload"
 	exit 3
 fi
-- 
1.9.1

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

[PATCH 04/18] tests/drm_import_export: mark flink and prime tests as basic

[Jesse Barnes]

They're testing basic functionality and don't involve stress or race
induction.

Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 tests/drm_import_export.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/tests/drm_import_export.c b/tests/drm_import_export.c
index e24e0df..6a63ef9 100644
--- a/tests/drm_import_export.c
+++ b/tests/drm_import_export.c
@@ -265,7 +265,7 @@ igt_main {
 		test_import_close_race();
 	}
 
-	igt_subtest("flink") {
+	igt_subtest("basic-flink") {
 		use_flink = true;
 
 		pthread_create(&test_thread_id1, NULL, test_thread, NULL);
@@ -279,7 +279,7 @@ igt_main {
 		pthread_join(test_thread_id4, NULL);
 	}
 
-	igt_subtest("prime") {
+	igt_subtest("basic-prime") {
 		use_flink = false;
 
 		pthread_create(&test_thread_id1, NULL, test_thread, NULL);
-- 
1.9.1

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

[PATCH 05/18] tests/drv_getparams: mark EU and subslice fetch as basic

[Jesse Barnes]

Fundamental and simple functionality.

Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 tests/drv_getparams.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/tests/drv_getparams.c b/tests/drv_getparams.c
index 31382e9..5a5c8bb 100644
--- a/tests/drv_getparams.c
+++ b/tests/drv_getparams.c
@@ -157,9 +157,9 @@ igt_main
 		init();
 	}
 
-	igt_subtest("subslice-total")
+	igt_subtest("basic-subslice-total")
 		subslice_total();
 
-	igt_subtest("eu-total")
+	igt_subtest("basic-eu-total")
 		eu_total();
 }
-- 
1.9.1

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

[PATCH 06/18] tests/drv_suspend: mark sysfs tests as basic

[Jesse Barnes]

debugfs may not be mounted, but sysfs should always be restored after
suspend or hibernate.

Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 tests/drv_suspend.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/tests/drv_suspend.c b/tests/drv_suspend.c
index d67a794..60ca8e3 100644
--- a/tests/drv_suspend.c
+++ b/tests/drv_suspend.c
@@ -193,7 +193,7 @@ igt_main
 	igt_subtest("debugfs-reader")
 		test_debugfs_reader(false);
 
-	igt_subtest("sysfs-reader")
+	igt_subtest("basic-sysfs-reader")
 		test_sysfs_reader(false);
 
 	igt_subtest("forcewake")
@@ -208,7 +208,7 @@ igt_main
 	igt_subtest("debugfs-reader-hibernate")
 		test_debugfs_reader(true);
 
-	igt_subtest("sysfs-reader-hibernate")
+	igt_subtest("basic-sysfs-reader-hibernate")
 		test_sysfs_reader(true);
 
 	igt_subtest("forcewake-hibernate")
-- 
1.9.1

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

[PATCH 07/18] tests/gem_ctx_exec: mark lrc lite restore as basic

[Jesse Barnes]

Need some LRC tests in the 'basic' subset, and this is a good simple
one.

Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 tests/gem_ctx_exec.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tests/gem_ctx_exec.c b/tests/gem_ctx_exec.c
index 3df939c..ea0fb7f 100644
--- a/tests/gem_ctx_exec.c
+++ b/tests/gem_ctx_exec.c
@@ -216,7 +216,7 @@ igt_main
 		gem_context_destroy(fd, ctx_id);
 	}
 
-	igt_subtest("lrc-lite-restore") {
+	igt_subtest("basic-lrc-lite-restore") {
 		int i, j;
 
 		/*
-- 
1.9.1

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

[PATCH 08/18] tests/gem_mmap: mark basic object creation tests as basic

[Jesse Barnes]

We should be able to create small and moderate sized objects quickly and
without errors.

Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 tests/gem_mmap.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/tests/gem_mmap.c b/tests/gem_mmap.c
index 095f5b9..dd66ad6 100644
--- a/tests/gem_mmap.c
+++ b/tests/gem_mmap.c
@@ -118,7 +118,7 @@ igt_main
 		igt_assert(ret == -1 && errno == ENOENT);
 	}
 
-	igt_subtest("new-object") {
+	igt_subtest("basic") {
 		arg.handle = gem_create(fd, OBJECT_SIZE);
 		arg.offset = 0;
 		arg.size = OBJECT_SIZE;
@@ -154,7 +154,7 @@ igt_main
 		gem_close(fd, arg.handle);
 	}
 
-	igt_subtest("small-bo")
+	igt_subtest("basic-small-bo")
 		test_huge_bo(fd, -1);
 	igt_subtest("big-bo")
 		test_huge_bo(fd, 0);
-- 
1.9.1

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

[PATCH 09/18] tests/gem_mmap_gtt: mark basic access and copy tests as basic

[Jesse Barnes]

These ones should always pass and are fairly quick.

Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 tests/gem_mmap_gtt.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/tests/gem_mmap_gtt.c b/tests/gem_mmap_gtt.c
index a95b98a..f964b39 100644
--- a/tests/gem_mmap_gtt.c
+++ b/tests/gem_mmap_gtt.c
@@ -525,7 +525,7 @@ igt_main
 	igt_fixture
 		fd = drm_open_any();
 
-	igt_subtest("access")
+	igt_subtest("basic")
 		test_access(fd);
 	igt_subtest("short")
 		test_short(fd);
@@ -556,11 +556,11 @@ igt_main
 	igt_subtest("write-cpu-read-gtt")
 		test_write_cpu_read_gtt(fd);
 
-	igt_subtest("small-bo")
+	igt_subtest("basic-small-bo")
 		test_huge_bo(fd, -1, I915_TILING_NONE);
-	igt_subtest("small-bo-tiledX")
+	igt_subtest("basic-small-bo-tiledX")
 		test_huge_bo(fd, -1, I915_TILING_X);
-	igt_subtest("small-bo-tiledY")
+	igt_subtest("basic-small-bo-tiledY")
 		test_huge_bo(fd, -1, I915_TILING_Y);
 
 	igt_subtest("big-bo")
-- 
1.9.1

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

[PATCH 10/18] tests/gem_pread/pwrite: mark normal tests as basic

[Jesse Barnes]

These should always pass.

Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 tests/gem_pread.c  | 2 +-
 tests/gem_pwrite.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/tests/gem_pread.c b/tests/gem_pread.c
index cc83948..3ec5fb1 100644
--- a/tests/gem_pread.c
+++ b/tests/gem_pread.c
@@ -108,7 +108,7 @@ int main(int argc, char **argv)
 		src = malloc(object_size);
 	}
 
-	igt_subtest("normal") {
+	igt_subtest("basic") {
 		for (count = 1; count <= 1<<17; count <<= 1) {
 			struct timeval start, end;
 
diff --git a/tests/gem_pwrite.c b/tests/gem_pwrite.c
index 5b6a77f..1d867e7 100644
--- a/tests/gem_pwrite.c
+++ b/tests/gem_pwrite.c
@@ -169,7 +169,7 @@ int main(int argc, char **argv)
 		src = malloc(object_size);
 	}
 
-	igt_subtest("normal") {
+	igt_subtest("basic") {
 		for (count = 1; count <= 1<<17; count <<= 1) {
 			struct timeval start, end;
 
-- 
1.9.1

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

[PATCH 11/18] tests/gem_tiled_pread/pwrite: mark normal tests as basic

[Jesse Barnes]

These simple tests should always pass.

Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 tests/gem_tiled_pread.c        | 167 +++++++++++++++++++++--------------------
 tests/gem_tiled_pread_pwrite.c |  48 ++++++------
 2 files changed, 112 insertions(+), 103 deletions(-)

diff --git a/tests/gem_tiled_pread.c b/tests/gem_tiled_pread.c
index fdc5173..92bb649 100644
--- a/tests/gem_tiled_pread.c
+++ b/tests/gem_tiled_pread.c
@@ -112,7 +112,7 @@ calculate_expected(int offset)
 	return (base_y + tile_y) * WIDTH + base_x + tile_x;
 }
 
-igt_simple_main
+igt_main
 {
 	int fd;
 	int i, iter = 100;
@@ -120,96 +120,101 @@ igt_simple_main
 	uint32_t handle;
 	uint32_t devid;
 
-	fd = drm_open_any();
+	igt_fixture {
+		fd = drm_open_any();
 
-	handle = create_bo(fd);
-	gem_get_tiling(fd, handle, &tiling, &swizzle);
+		handle = create_bo(fd);
+		gem_get_tiling(fd, handle, &tiling, &swizzle);
 
-	devid = intel_get_drm_devid(fd);
-
-	if (IS_GEN2(devid)) {
-		tile_height = 16;
-		tile_width = 128;
-		tile_size = 2048;
-	} else {
-		tile_height = 8;
-		tile_width = 512;
-		tile_size = PAGE_SIZE;
+		devid = intel_get_drm_devid(fd);
 	}
 
-	/* Read a bunch of random subsets of the data and check that they come
-	 * out right.
-	 */
-	for (i = 0; i < iter; i++) {
-		int size = WIDTH * HEIGHT * 4;
-		int offset = (random() % size) & ~3;
-		int len = (random() % size) & ~3;
-		int j;
+	igt_subtest("basic") {
 
-		if (len == 0)
-			len = 4;
+		if (IS_GEN2(devid)) {
+			tile_height = 16;
+			tile_width = 128;
+			tile_size = 2048;
+		} else {
+			tile_height = 8;
+			tile_width = 512;
+			tile_size = PAGE_SIZE;
+		}
 
-		if (offset + len > size)
-			len = size - offset;
+		/* Read a bunch of random subsets of the data and check that they come
+		 * out right.
+		 */
+		for (i = 0; i < iter; i++) {
+			int size = WIDTH * HEIGHT * 4;
+			int offset = (random() % size) & ~3;
+			int len = (random() % size) & ~3;
+			int j;
 
-		if (i == 0) {
-			offset = 0;
-			len = size;
-		}
+			if (len == 0)
+				len = 4;
 
-		gem_read(fd, handle, offset, linear, len);
+			if (offset + len > size)
+				len = size - offset;
 
-		/* Translate from offsets in the read buffer to the swizzled
-		 * address that it corresponds to.  This is the opposite of
-		 * what Mesa does (calculate offset to be read given the linear
-		 * offset it's looking for).
-		 */
-		for (j = offset; j < offset + len; j += 4) {
-			uint32_t expected_val, found_val;
-			int swizzled_offset;
-			const char *swizzle_str;
-
-			switch (swizzle) {
-			case I915_BIT_6_SWIZZLE_NONE:
-				swizzled_offset = j;
-				swizzle_str = "none";
-				break;
-			case I915_BIT_6_SWIZZLE_9:
-				swizzled_offset = j ^
-					swizzle_bit(9, j);
-				swizzle_str = "bit9";
-				break;
-			case I915_BIT_6_SWIZZLE_9_10:
-				swizzled_offset = j ^
-					swizzle_bit(9, j) ^
-					swizzle_bit(10, j);
-				swizzle_str = "bit9^10";
-				break;
-			case I915_BIT_6_SWIZZLE_9_11:
-				swizzled_offset = j ^
-					swizzle_bit(9, j) ^
-					swizzle_bit(11, j);
-				swizzle_str = "bit9^11";
-				break;
-			case I915_BIT_6_SWIZZLE_9_10_11:
-				swizzled_offset = j ^
-					swizzle_bit(9, j) ^
-					swizzle_bit(10, j) ^
-					swizzle_bit(11, j);
-				swizzle_str = "bit9^10^11";
-				break;
-			default:
-				igt_assert_f(0, "Bad swizzle bits; %d\n",
-					     swizzle);
+			if (i == 0) {
+				offset = 0;
+				len = size;
+			}
+
+			gem_read(fd, handle, offset, linear, len);
+
+			/* Translate from offsets in the read buffer to the swizzled
+			 * address that it corresponds to.  This is the opposite of
+			 * what Mesa does (calculate offset to be read given the linear
+			 * offset it's looking for).
+			 */
+			for (j = offset; j < offset + len; j += 4) {
+				uint32_t expected_val, found_val;
+				int swizzled_offset;
+				const char *swizzle_str;
+
+				switch (swizzle) {
+				case I915_BIT_6_SWIZZLE_NONE:
+					swizzled_offset = j;
+					swizzle_str = "none";
+					break;
+				case I915_BIT_6_SWIZZLE_9:
+					swizzled_offset = j ^
+						swizzle_bit(9, j);
+					swizzle_str = "bit9";
+					break;
+				case I915_BIT_6_SWIZZLE_9_10:
+					swizzled_offset = j ^
+						swizzle_bit(9, j) ^
+						swizzle_bit(10, j);
+					swizzle_str = "bit9^10";
+					break;
+				case I915_BIT_6_SWIZZLE_9_11:
+					swizzled_offset = j ^
+						swizzle_bit(9, j) ^
+						swizzle_bit(11, j);
+					swizzle_str = "bit9^11";
+					break;
+				case I915_BIT_6_SWIZZLE_9_10_11:
+					swizzled_offset = j ^
+						swizzle_bit(9, j) ^
+						swizzle_bit(10, j) ^
+						swizzle_bit(11, j);
+					swizzle_str = "bit9^10^11";
+					break;
+				default:
+					igt_assert_f(0, "Bad swizzle bits; %d\n",
+						     swizzle);
+				}
+				expected_val = calculate_expected(swizzled_offset);
+				found_val = linear[(j - offset) / 4];
+				igt_assert_f(expected_val == found_val,
+					     "Bad read [%d]: %d instead of %d at 0x%08x "
+					     "for read from 0x%08x to 0x%08x, swizzle=%s\n",
+					     i, found_val, expected_val, j,
+					     offset, offset + len,
+					     swizzle_str);
 			}
-			expected_val = calculate_expected(swizzled_offset);
-			found_val = linear[(j - offset) / 4];
-			igt_assert_f(expected_val == found_val,
-				     "Bad read [%d]: %d instead of %d at 0x%08x "
-				     "for read from 0x%08x to 0x%08x, swizzle=%s\n",
-				     i, found_val, expected_val, j,
-				     offset, offset + len,
-				     swizzle_str);
 		}
 	}
 
diff --git a/tests/gem_tiled_pread_pwrite.c b/tests/gem_tiled_pread_pwrite.c
index 3d8fdc9..efb56d5 100644
--- a/tests/gem_tiled_pread_pwrite.c
+++ b/tests/gem_tiled_pread_pwrite.c
@@ -101,7 +101,7 @@ create_bo(int fd)
 	return handle;
 }
 
-igt_simple_main
+igt_main
 {
 	int fd;
 	uint32_t *data;
@@ -109,34 +109,38 @@ igt_simple_main
 	uint32_t tiling, swizzle;
 	uint32_t handle, handle_target;
 	int count;
-	
-	fd = drm_open_any();
-	count = SLOW_QUICK(intel_get_total_ram_mb() * 9 / 10, 8) ;
 
-	for (i = 0; i < count/2; i++) {
-		current_tiling_mode = I915_TILING_X;
+	igt_fixture {
+		fd = drm_open_any();
+		count = SLOW_QUICK(intel_get_total_ram_mb() * 9 / 10, 8);
+	}
+
+	igt_subtest("basic") {
+		for (i = 0; i < count/2; i++) {
+			current_tiling_mode = I915_TILING_X;
 
-		handle = create_bo_and_fill(fd);
-		gem_get_tiling(fd, handle, &tiling, &swizzle);
+			handle = create_bo_and_fill(fd);
+			gem_get_tiling(fd, handle, &tiling, &swizzle);
 
-		gem_read(fd, handle, 0, linear, sizeof(linear));
+			gem_read(fd, handle, 0, linear, sizeof(linear));
 
-		handle_target = create_bo(fd);
-		gem_write(fd, handle_target, 0, linear, sizeof(linear));
+			handle_target = create_bo(fd);
+			gem_write(fd, handle_target, 0, linear, sizeof(linear));
 
-		/* Check the target bo's contents. */
-		data = gem_mmap(fd, handle_target, sizeof(linear), PROT_READ | PROT_WRITE);
-		for (j = 0; j < WIDTH*HEIGHT; j++)
-			igt_assert_f(data[j] == j,
-				     "mismatch at %i: %i\n",
-				     j, data[j]);
-		munmap(data, sizeof(linear));
+			/* Check the target bo's contents. */
+			data = gem_mmap(fd, handle_target, sizeof(linear), PROT_READ | PROT_WRITE);
+			for (j = 0; j < WIDTH*HEIGHT; j++)
+				igt_assert_f(data[j] == j,
+					     "mismatch at %i: %i\n",
+					     j, data[j]);
+			munmap(data, sizeof(linear));
 
-		/* Leak both bos so that we use all of system mem! */
-		gem_madvise(fd, handle_target, I915_MADV_DONTNEED);
-		gem_madvise(fd, handle, I915_MADV_DONTNEED);
+			/* Leak both bos so that we use all of system mem! */
+			gem_madvise(fd, handle_target, I915_MADV_DONTNEED);
+			gem_madvise(fd, handle, I915_MADV_DONTNEED);
 
-		igt_progress("gem_tiled_pread_pwrite: ", i, count/2);
+			igt_progress("gem_tiled_pread_pwrite: ", i, count/2);
+		}
 	}
 
 	close(fd);
-- 
1.9.1

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

[PATCH 12/18] tests/kms_addfb: mark simple fb creation tests as basic

[Jesse Barnes]

We should always be able to create simple and tiled objects.

Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 tests/kms_addfb.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/tests/kms_addfb.c b/tests/kms_addfb.c
index 42ee632..f10e12b 100644
--- a/tests/kms_addfb.c
+++ b/tests/kms_addfb.c
@@ -121,7 +121,7 @@ static void pitch_tests(int fd)
 	}
 
 	f.handles[0] = gem_bo;
-	igt_subtest("normal") {
+	igt_subtest("basic") {
 		igt_assert(drmIoctl(fd, DRM_IOCTL_MODE_ADDFB2, &f) == 0);
 		igt_assert(drmIoctl(fd, DRM_IOCTL_MODE_RMFB, &f.fb_id) == 0);
 		f.fb_id = 0;
@@ -139,7 +139,7 @@ static void pitch_tests(int fd)
 		gem_set_tiling(fd, gem_bo, I915_TILING_X, 1024*4);
 	f.pitches[0] = 1024*4;
 
-	igt_subtest("X-tiled") {
+	igt_subtest("basic-X-tiled") {
 		igt_assert(drmIoctl(fd, DRM_IOCTL_MODE_ADDFB2, &f) == 0);
 		igt_assert(drmIoctl(fd, DRM_IOCTL_MODE_RMFB, &f.fb_id) == 0);
 		f.fb_id = 0;
@@ -162,7 +162,7 @@ static void pitch_tests(int fd)
 	igt_fixture
 		gem_set_tiling(fd, gem_bo, I915_TILING_Y, 1024*4);
 	f.pitches[0] = 1024*4;
-	igt_subtest("Y-tiled") {
+	igt_subtest("basic-Y-tiled") {
 		igt_assert(drmIoctl(fd, DRM_IOCTL_MODE_ADDFB2, &f) == -1 &&
 			   errno == EINVAL);
 	}
-- 
1.9.1

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

[PATCH 13/18] tests/kms_vblank: mark accuracy test as basic

[Jesse Barnes]

Need some simple vblank coverage in the BAT list.

Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 tests/kms_vblank.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tests/kms_vblank.c b/tests/kms_vblank.c
index 6946177..ef2f7ca 100644
--- a/tests/kms_vblank.c
+++ b/tests/kms_vblank.c
@@ -188,7 +188,7 @@ igt_main
 		igt_require(crtc0_active(fd));
 	}
 
-	igt_subtest("accuracy")
+	igt_subtest("basic-accuracy")
 		accuracy(fd);
 
 	igt_subtest("query-idle")
-- 
1.9.1

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

[PATCH 14/18] tests/pm_backlight: mark simple test as basic

[Jesse Barnes]

We should be able to adjust the backlight and observe changes in sysfs.

Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 tests/pm_backlight.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tests/pm_backlight.c b/tests/pm_backlight.c
index d02336d..cff2694 100644
--- a/tests/pm_backlight.c
+++ b/tests/pm_backlight.c
@@ -157,7 +157,7 @@ igt_main
 		igt_assert(backlight_read(&max, "max_brightness") > -1);
 	}
 
-	igt_subtest("brightness")
+	igt_subtest("basic-brightness")
 		test_brightness(max);
 	igt_subtest("bad-brightness")
 		test_bad_brightness(max);
-- 
1.9.1

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

[PATCH 15/18] tests/pm_rpm: mark RTE and D3 tests as basic

[Jesse Barnes]

These always need to pass for basic PM functionality.

Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 tests/pm_rpm.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/tests/pm_rpm.c b/tests/pm_rpm.c
index d509fa8..7ae5806 100644
--- a/tests/pm_rpm.c
+++ b/tests/pm_rpm.c
@@ -1832,11 +1832,11 @@ int main(int argc, char *argv[])
 			stay_subtest();
 
 	/* Essential things */
-	igt_subtest("rte")
+	igt_subtest("basic-rte")
 		basic_subtest();
 	igt_subtest("drm-resources-equal")
 		drm_resources_equal_subtest();
-	igt_subtest("pci-d3-state")
+	igt_subtest("basic-pci-d3-state")
 		pci_d3_state_subtest();
 
 	/* Basic modeset */
-- 
1.9.1

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

[PATCH 16/18] tests/kms_flip: add basic tests for flip, flip vs dpms, and flip modeset

[Jesse Barnes]

Simple variants that don't do multiple output or interruptible testing.

Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 tests/kms_flip.c | 11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)

diff --git a/tests/kms_flip.c b/tests/kms_flip.c
index a595d9f..a0e4112 100644
--- a/tests/kms_flip.c
+++ b/tests/kms_flip.c
@@ -80,6 +80,7 @@
 #define TEST_TS_CONT		(1 << 27)
 #define TEST_BO_TOOBIG		(1 << 28)
 #define TEST_HANG_ONCE		(1 << 29)
+#define TEST_BASIC		(1 << 30)
 
 #define EVENT_FLIP		(1 << 0)
 #define EVENT_VBLANK		(1 << 1)
@@ -1649,7 +1650,7 @@ int main(int argc, char **argv)
 					"blt-wf_vblank-vs-modeset" },
 		{ 60,  TEST_VBLANK | TEST_MODESET | TEST_WITH_DUMMY_RCS,
 					"rcs-wf_vblank-vs-modeset" },
-
+		{ 30, TEST_FLIP | TEST_BASIC, "basic-plain-flip" },
 		{ 30, TEST_FLIP , "plain-flip" },
 		{ 30, TEST_FLIP | TEST_EBUSY , "busy-flip" },
 		{ 30, TEST_FLIP | TEST_FENCE_STRESS , "flip-vs-fences" },
@@ -1657,12 +1658,14 @@ int main(int argc, char **argv)
 		{ 30, TEST_FLIP | TEST_CHECK_TS | TEST_FB_RECREATE,
 			"plain-flip-fb-recreate" },
 		{ 30, TEST_FLIP | TEST_RMFB | TEST_MODESET , "flip-vs-rmfb" },
+		{ 60, TEST_FLIP | TEST_DPMS | TEST_EINVAL | TEST_BASIC, "basic-flip-vs-dpms" },
 		{ 60, TEST_FLIP | TEST_DPMS | TEST_EINVAL, "flip-vs-dpms" },
 		{ 60, TEST_FLIP | TEST_DPMS | TEST_WITH_DUMMY_BCS, "blt-flip-vs-dpms" },
 		{ 60, TEST_FLIP | TEST_DPMS | TEST_WITH_DUMMY_RCS, "render-flip-vs-dpms" },
 		{ 30,  TEST_FLIP | TEST_PAN, "flip-vs-panning" },
 		{ 60, TEST_FLIP | TEST_PAN | TEST_WITH_DUMMY_BCS, "blt-flip-vs-panning" },
 		{ 60, TEST_FLIP | TEST_PAN | TEST_WITH_DUMMY_RCS, "render-flip-vs-panning" },
+		{ 60, TEST_FLIP | TEST_MODESET | TEST_EINVAL | TEST_BASIC, "basic-flip-vs-modeset" },
 		{ 60, TEST_FLIP | TEST_MODESET | TEST_EINVAL, "flip-vs-modeset" },
 		{ 60, TEST_FLIP | TEST_MODESET | TEST_WITH_DUMMY_BCS, "blt-flip-vs-modeset" },
 		{ 60, TEST_FLIP | TEST_MODESET | TEST_WITH_DUMMY_RCS, "render-flip-vs-modeset" },
@@ -1721,6 +1724,9 @@ int main(int argc, char **argv)
 		igt_subtest(tests[i].name)
 			run_test(tests[i].duration, tests[i].flags);
 
+		if (tests[i].flags & TEST_BASIC)
+			continue;
+
 		if (tests[i].flags & TEST_NO_2X_OUTPUT)
 			continue;
 
@@ -1740,6 +1746,9 @@ int main(int argc, char **argv)
 		    !(tests[i].flags & TEST_VBLANK_ABSOLUTE))
 			continue;
 
+		if (tests[i].flags & TEST_BASIC)
+			continue;
+
 		igt_subtest_f( "%s-interruptible", tests[i].name)
 			run_test(tests[i].duration, tests[i].flags);
 
-- 
1.9.1

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

[PATCH 17/18] tests/kms_setmode: mark simple clone test as basic

[Jesse Barnes]

Should cover simple, single CRTC mode sets.

Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 tests/kms_setmode.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tests/kms_setmode.c b/tests/kms_setmode.c
index 82769ab..2c0fe45 100644
--- a/tests/kms_setmode.c
+++ b/tests/kms_setmode.c
@@ -685,7 +685,7 @@ int main(int argc, char **argv)
 		const char *name;
 	} tests[] = {
 		{ TEST_CLONE | TEST_SINGLE_CRTC_CLONE,
-					"clone-single-crtc" },
+					"basic-clone-single-crtc" },
 		{ TEST_INVALID | TEST_CLONE | TEST_SINGLE_CRTC_CLONE,
 					"invalid-clone-single-crtc" },
 		{ TEST_INVALID | TEST_CLONE | TEST_EXCLUSIVE_CRTC_CLONE,
-- 
1.9.1

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

[PATCH 18/18] gitignore: ignore more files

[Jesse Barnes]

---
 .gitignore       |  3 +++
 tests/.gitignore | 13 +++++++++++++
 tools/.gitignore |  8 ++++++++
 3 files changed, 24 insertions(+)

diff --git a/.gitignore b/.gitignore
index a438c1c..533f6f1 100644
--- a/.gitignore
+++ b/.gitignore
@@ -90,3 +90,6 @@ gtk-doc.m4
 
 piglit
 results
+
+*.orig
+version.h
diff --git a/tests/.gitignore b/tests/.gitignore
index 31d35a5..e45cbb7 100644
--- a/tests/.gitignore
+++ b/tests/.gitignore
@@ -99,6 +99,7 @@ gem_set_tiling_vs_blt
 gem_set_tiling_vs_gtt
 gem_set_tiling_vs_pwrite
 gem_storedw_batches_loop
+gem_storedw_loop
 gem_storedw_loop_blt
 gem_storedw_loop_bsd
 gem_storedw_loop_render
@@ -169,3 +170,15 @@ prime_udl
 template
 test-list.txt
 testdisplay
+ddi_compute_wrpll
+gem_vmap_blits
+gem_wait_render_timeout
+igt_fork_helper
+igt_list_only
+igt_no_exit
+igt_no_exit_list_only
+igt_no_subtest
+igt_simulation
+pm_pc8
+pm_psr
+
diff --git a/tools/.gitignore b/tools/.gitignore
index 49bd24f..7bf91e3 100644
--- a/tools/.gitignore
+++ b/tools/.gitignore
@@ -37,3 +37,11 @@ intel_vga_write
 intel_watermark
 skl_compute_wrpll
 skl_ddb_allocation
+forcewaked
+intel_dpio_read
+intel_dpio_write
+intel_gpu_dump
+intel_nc_read
+intel_nc_write
+intel_punit_read
+intel_punit_write
-- 
1.9.1

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 18/18] gitignore: ignore more files

[Daniel Vetter]

On Thu, Aug 13, 2015 at 01:31:41PM -0700, Jesse Barnes wrote:

git clean fixes this all, at least over here git status is clean.
-Daniel

> ---
>  .gitignore       |  3 +++
>  tests/.gitignore | 13 +++++++++++++
>  tools/.gitignore |  8 ++++++++
>  3 files changed, 24 insertions(+)
> 
> diff --git a/.gitignore b/.gitignore
> index a438c1c..533f6f1 100644
> --- a/.gitignore
> +++ b/.gitignore
> @@ -90,3 +90,6 @@ gtk-doc.m4
>  
>  piglit
>  results
> +
> +*.orig
> +version.h
> diff --git a/tests/.gitignore b/tests/.gitignore
> index 31d35a5..e45cbb7 100644
> --- a/tests/.gitignore
> +++ b/tests/.gitignore
> @@ -99,6 +99,7 @@ gem_set_tiling_vs_blt
>  gem_set_tiling_vs_gtt
>  gem_set_tiling_vs_pwrite
>  gem_storedw_batches_loop
> +gem_storedw_loop
>  gem_storedw_loop_blt
>  gem_storedw_loop_bsd
>  gem_storedw_loop_render
> @@ -169,3 +170,15 @@ prime_udl
>  template
>  test-list.txt
>  testdisplay
> +ddi_compute_wrpll
> +gem_vmap_blits
> +gem_wait_render_timeout
> +igt_fork_helper
> +igt_list_only
> +igt_no_exit
> +igt_no_exit_list_only
> +igt_no_subtest
> +igt_simulation
> +pm_pc8
> +pm_psr
> +
> diff --git a/tools/.gitignore b/tools/.gitignore
> index 49bd24f..7bf91e3 100644
> --- a/tools/.gitignore
> +++ b/tools/.gitignore
> @@ -37,3 +37,11 @@ intel_vga_write
>  intel_watermark
>  skl_compute_wrpll
>  skl_ddb_allocation
> +forcewaked
> +intel_dpio_read
> +intel_dpio_write
> +intel_gpu_dump
> +intel_nc_read
> +intel_nc_write
> +intel_punit_read
> +intel_punit_write
> -- 
> 1.9.1
> 
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/intel-gfx

-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 01/18] tests/gem_storedw_loop: add new store_dword test to unify per-ring ones

[Daniel Vetter]

On Thu, Aug 13, 2015 at 01:31:24PM -0700, Jesse Barnes wrote:
> There was a lot of duplication going on...  Mark as basic while we're at
> it as these should never fail.
> 
> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
> ---
>  tests/Makefile.sources   |   1 +
>  tests/gem_storedw_loop.c | 181 +++++++++++++++++++++++++++++++++++++++++++++++
>  2 files changed, 182 insertions(+)
>  create mode 100644 tests/gem_storedw_loop.c
> 
> diff --git a/tests/Makefile.sources b/tests/Makefile.sources
> index b9a4cb4..cdcee33 100644
> --- a/tests/Makefile.sources
> +++ b/tests/Makefile.sources
> @@ -138,6 +138,7 @@ TESTS_progs = \
>  	gem_seqno_wrap \
>  	gem_set_tiling_vs_gtt \
>  	gem_set_tiling_vs_pwrite \
> +	gem_storedw_loop \
>  	gem_storedw_loop_blt \
>  	gem_storedw_loop_bsd \
>  	gem_storedw_loop_render \

Why not remove the old ones while at it? This just means more gunk in the
overall igt set. Also please update .gitignore here for these ...

> diff --git a/tests/gem_storedw_loop.c b/tests/gem_storedw_loop.c
> new file mode 100644
> index 0000000..e4ed35d
> --- /dev/null
> +++ b/tests/gem_storedw_loop.c
> @@ -0,0 +1,181 @@
> +/*
> + * Copyright © 2009 Intel Corporation
> + *
> + * 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:
> + *    Eric Anholt <eric@anholt.net>
> + *    Jesse Barnes <jbarnes@virtuousgeek.org> (based on gem_bad_blit.c)
> + *
> + */
> +
> +#include <stdlib.h>
> +#include <stdio.h>
> +#include <string.h>
> +#include <fcntl.h>
> +#include <inttypes.h>
> +#include <errno.h>
> +#include <sys/stat.h>
> +#include <sys/time.h>
> +#include "drm.h"
> +#include "ioctl_wrappers.h"
> +#include "drmtest.h"
> +#include "intel_bufmgr.h"
> +#include "intel_batchbuffer.h"
> +#include "intel_io.h"
> +#include "intel_chipset.h"
> +
> +IGT_TEST_DESCRIPTION("Basic CS check using MI_STORE_DATA_IMM.");
> +
> +#define LOCAL_I915_EXEC_VEBOX (4<<0)
> +
> +static drm_intel_bufmgr *bufmgr;
> +struct intel_batchbuffer *batch;
> +static drm_intel_bo *target_buffer;
> +static int has_ppgtt = 0;
> +static int devid;
> +
> +/*
> + * Testcase: Basic bsd MI check using MI_STORE_DATA_IMM
> + */
> +
> +static void
> +emit_store_dword_imm(drm_intel_bo *dest, uint32_t val)
> +{
> +	int cmd;
> +	cmd = MI_STORE_DWORD_IMM;
> +	if (!has_ppgtt)
> +		cmd |= MI_MEM_VIRTUAL;
> +
> +	BEGIN_BATCH(4, 0);
> +	OUT_BATCH(cmd);
> +	if (batch->gen >= 8) {
> +		OUT_RELOC(dest, I915_GEM_DOMAIN_INSTRUCTION,
> +			  I915_GEM_DOMAIN_INSTRUCTION, 0);
> +		OUT_BATCH(val);
> +	} else {
> +		OUT_BATCH(0); /* reserved */
> +		OUT_RELOC(dest, I915_GEM_DOMAIN_INSTRUCTION,
> +			  I915_GEM_DOMAIN_INSTRUCTION, 0);
> +		OUT_BATCH(val);
> +	}
> +	ADVANCE_BATCH();
> +}
> +
> +static void
> +store_dword_loop(int ring, int count, int divider)
> +{
> +	int i, val = 0;
> +	uint32_t *buf;
> +
> +	igt_info("running storedw loop on render with stall every %i batch\n", divider);
> +
> +	for (i = 0; i < SLOW_QUICK(0x2000, 0x10); i++) {
> +		emit_store_dword_imm(target_buffer, val);
> +		intel_batchbuffer_flush_on_ring(batch, ring);
> +
> +		if (i % divider != 0)
> +			goto cont;
> +
> +		drm_intel_bo_map(target_buffer, 0);
> +
> +		buf = target_buffer->virtual;
> +		igt_assert_f(buf[0] == val,
> +			     "value mismatch: cur 0x%08x, stored 0x%08x\n",
> +			     buf[0], val);
> +
> +		drm_intel_bo_unmap(target_buffer);
> +
> +cont:
> +		val++;
> +	}
> +
> +	drm_intel_bo_map(target_buffer, 0);
> +	buf = target_buffer->virtual;
> +
> +	igt_info("completed %d writes successfully, current value: 0x%08x\n", i,
> +			buf[0]);
> +	drm_intel_bo_unmap(target_buffer);
> +}
> +
> +static void
> +store_test(int ring, int count)
> +{
> +	drm_intel_bufmgr_gem_enable_reuse(bufmgr);
> +
> +	batch = intel_batchbuffer_alloc(bufmgr, devid);
> +	igt_assert(batch);
> +
> +	target_buffer = drm_intel_bo_alloc(bufmgr, "target bo", 4096, 4096);
> +	igt_assert(target_buffer);
> +
> +	store_dword_loop(ring, count, 1);
> +	store_dword_loop(ring, count, 2);
> +	if (!igt_run_in_simulation()) {
> +		store_dword_loop(ring, count, 3);
> +		store_dword_loop(ring, count, 5);
> +	}
> +
> +	drm_intel_bo_unreference(target_buffer);
> +	intel_batchbuffer_free(batch);
> +	drm_intel_bufmgr_destroy(bufmgr);
> +}
> +
> +struct ring {
> +	const char *name;
> +	int id;
> +} rings[] = {
> +	{ "bsd", I915_EXEC_BSD },
> +	{ "render", I915_EXEC_RENDER },
> +	{ "blt", I915_EXEC_BLT },
> +	{ "vebox", I915_EXEC_VEBOX },
> +};
> +
> +igt_main
> +{
> +	int fd, i;
> +
> +	igt_fixture {
> +		fd = drm_open_any();
> +		devid = intel_get_drm_devid(fd);
> +
> +		bufmgr = drm_intel_bufmgr_gem_init(fd, 4096);
> +		igt_assert(bufmgr);
> +
> +		has_ppgtt = gem_uses_aliasing_ppgtt(fd);
> +
> +		igt_skip_on_f(intel_gen(devid) < 6,
> +			      "MI_STORE_DATA can only use GTT address on gen4+/g33 and "
> +			      "needs snoopable mem on pre-gen6\n");
> +
> +		/* This only works with ppgtt */
> +		igt_require(has_ppgtt);
> +	}
> +
> +	for (i = 0; i < ARRAY_SIZE(rings); i++) {
> +		igt_subtest_f("basic-%s", rings[i].name)
> +			store_test(rings[i].id, 16*1024);
> +
> +		igt_subtest_f("long-%s", rings[i].name)
> +			store_test(rings[i].id, 1024*1024);
> +	}
> +
> +	close(fd);
> +}
> -- 
> 1.9.1
> 
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/intel-gfx

-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 02/18] tests/drv_module_reload: rename drv_module_reload to include in BATs

[Daniel Vetter]

On Thu, Aug 13, 2015 at 01:31:25PM -0700, Jesse Barnes wrote:
> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>

Absolutely-Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch>

btw this will cause lots of hilarity because of lockdep and vt and shit
like that. We probably need to grill gregkh a bit more with patches he's
not merging ...
-Daniel

> ---
>  tests/Makefile.sources        |  2 +-
>  tests/drv_module_reload       | 60 -------------------------------------------
>  tests/drv_module_reload_basic | 60 +++++++++++++++++++++++++++++++++++++++++++
>  3 files changed, 61 insertions(+), 61 deletions(-)
>  delete mode 100755 tests/drv_module_reload
>  create mode 100755 tests/drv_module_reload_basic
> 
> diff --git a/tests/Makefile.sources b/tests/Makefile.sources
> index cdcee33..4fe7d4c 100644
> --- a/tests/Makefile.sources
> +++ b/tests/Makefile.sources
> @@ -178,7 +178,7 @@ TESTS_scripts = \
>  	debugfs_emon_crash \
>  	drv_debugfs_reader \
>  	drv_missed_irq_hang \
> -	drv_module_reload \
> +	drv_module_reload_basic \
>  	kms_sysfs_edid_timing \
>  	sysfs_l3_parity \
>  	test_rte_check \
> diff --git a/tests/drv_module_reload b/tests/drv_module_reload
> deleted file mode 100755
> index bb29a64..0000000
> --- a/tests/drv_module_reload
> +++ /dev/null
> @@ -1,60 +0,0 @@
> -#!/bin/bash
> -#
> -# Testcase: Reload the drm module
> -#
> -# ... we've broken this way too often :(
> -#
> -
> -SOURCE_DIR="$( dirname "${BASH_SOURCE[0]}" )"
> -. $SOURCE_DIR/drm_lib.sh
> -
> -# no other drm service should be running, so we can just unbind
> -
> -# we must kick away fbcon (but only fbcon)
> -for vtcon in /sys/class/vtconsole/vtcon*/ ; do
> -	if grep "frame buffer device" $vtcon/name > /dev/null ; then
> -		echo unbinding $vtcon: `cat $vtcon/name`
> -		echo 0 > $vtcon/bind
> -	fi
> -done
> -
> -# The sound driver uses our power well
> -pkill alsactl
> -rmmod snd_hda_intel &> /dev/null
> -
> -#ignore errors in ips - gen5 only
> -rmmod intel_ips &> /dev/null
> -rmmod i915
> -#ignore errors in intel-gtt, often built-in
> -rmmod intel-gtt &> /dev/null
> -# drm may be used by other devices (nouveau, radeon, udl, etc)
> -rmmod drm_kms_helper &> /dev/null
> -rmmod drm &> /dev/null
> -
> -if lsmod | grep i915 &> /dev/null ; then
> -	echo WARNING: i915.ko still loaded!
> -	exit 1
> -else
> -	echo module successfully unloaded
> -fi
> -
> -modprobe i915
> -echo 1 > /sys/class/vtconsole/vtcon1/bind
> -
> -modprobe snd_hda_intel
> -
> -# does the device exist?
> -if $SOURCE_DIR/gem_alive > /dev/null ; then
> -	echo "module successfully loaded again"
> -else
> -	echo "failed to reload module successfully"
> -	exit 2
> -fi
> -
> -# then try to run something
> -if ! $SOURCE_DIR/gem_exec_nop > /dev/null ; then
> -	echo "failed to execute a simple batch after reload"
> -	exit 3
> -fi
> -
> -exit 0
> diff --git a/tests/drv_module_reload_basic b/tests/drv_module_reload_basic
> new file mode 100755
> index 0000000..bb29a64
> --- /dev/null
> +++ b/tests/drv_module_reload_basic
> @@ -0,0 +1,60 @@
> +#!/bin/bash
> +#
> +# Testcase: Reload the drm module
> +#
> +# ... we've broken this way too often :(
> +#
> +
> +SOURCE_DIR="$( dirname "${BASH_SOURCE[0]}" )"
> +. $SOURCE_DIR/drm_lib.sh
> +
> +# no other drm service should be running, so we can just unbind
> +
> +# we must kick away fbcon (but only fbcon)
> +for vtcon in /sys/class/vtconsole/vtcon*/ ; do
> +	if grep "frame buffer device" $vtcon/name > /dev/null ; then
> +		echo unbinding $vtcon: `cat $vtcon/name`
> +		echo 0 > $vtcon/bind
> +	fi
> +done
> +
> +# The sound driver uses our power well
> +pkill alsactl
> +rmmod snd_hda_intel &> /dev/null
> +
> +#ignore errors in ips - gen5 only
> +rmmod intel_ips &> /dev/null
> +rmmod i915
> +#ignore errors in intel-gtt, often built-in
> +rmmod intel-gtt &> /dev/null
> +# drm may be used by other devices (nouveau, radeon, udl, etc)
> +rmmod drm_kms_helper &> /dev/null
> +rmmod drm &> /dev/null
> +
> +if lsmod | grep i915 &> /dev/null ; then
> +	echo WARNING: i915.ko still loaded!
> +	exit 1
> +else
> +	echo module successfully unloaded
> +fi
> +
> +modprobe i915
> +echo 1 > /sys/class/vtconsole/vtcon1/bind
> +
> +modprobe snd_hda_intel
> +
> +# does the device exist?
> +if $SOURCE_DIR/gem_alive > /dev/null ; then
> +	echo "module successfully loaded again"
> +else
> +	echo "failed to reload module successfully"
> +	exit 2
> +fi
> +
> +# then try to run something
> +if ! $SOURCE_DIR/gem_exec_nop > /dev/null ; then
> +	echo "failed to execute a simple batch after reload"
> +	exit 3
> +fi
> +
> +exit 0
> -- 
> 1.9.1
> 
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/intel-gfx

-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 03/18] tests/drv_module_reload_basic: use linear_blits after module_reload for sanity check

[Daniel Vetter]

On Thu, Aug 13, 2015 at 01:31:26PM -0700, Jesse Barnes wrote:
> Reduces runtime a lot...
> 
> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
> ---
>  tests/drv_module_reload_basic | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/tests/drv_module_reload_basic b/tests/drv_module_reload_basic
> index bb29a64..cf7f2b8 100755
> --- a/tests/drv_module_reload_basic
> +++ b/tests/drv_module_reload_basic
> @@ -52,7 +52,7 @@ else
>  fi
>  
>  # then try to run something
> -if ! $SOURCE_DIR/gem_exec_nop > /dev/null ; then
> +if ! $SOURCE_DIR/gem_linear_blits --r basic > /dev/null ; then

Please use the full-lenght run-subtest to avoid lots of wtf when we add
another option starting with r eventually. With that fixed:

Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>

>  	echo "failed to execute a simple batch after reload"
>  	exit 3
>  fi
> -- 
> 1.9.1
> 
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/intel-gfx

-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 04/18] tests/drm_import_export: mark flink and prime tests as basic

[Daniel Vetter]

On Thu, Aug 13, 2015 at 01:31:27PM -0700, Jesse Barnes wrote:
> They're testing basic functionality and don't involve stress or race
> induction.
> 
> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>

These are more stress-tests in nature I think. For basic testscases of
prime I'd recommend instead

prime_self_import: Everything not marked with *-race

gem_flink: all of them.

instead of this patch here.
-Daniel

> ---
>  tests/drm_import_export.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/tests/drm_import_export.c b/tests/drm_import_export.c
> index e24e0df..6a63ef9 100644
> --- a/tests/drm_import_export.c
> +++ b/tests/drm_import_export.c
> @@ -265,7 +265,7 @@ igt_main {
>  		test_import_close_race();
>  	}
>  
> -	igt_subtest("flink") {
> +	igt_subtest("basic-flink") {
>  		use_flink = true;
>  
>  		pthread_create(&test_thread_id1, NULL, test_thread, NULL);
> @@ -279,7 +279,7 @@ igt_main {
>  		pthread_join(test_thread_id4, NULL);
>  	}
>  
> -	igt_subtest("prime") {
> +	igt_subtest("basic-prime") {
>  		use_flink = false;
>  
>  		pthread_create(&test_thread_id1, NULL, test_thread, NULL);
> -- 
> 1.9.1
> 
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/intel-gfx

-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 05/18] tests/drv_getparams: mark EU and subslice fetch as basic

[Daniel Vetter]

On Thu, Aug 13, 2015 at 01:31:28PM -0700, Jesse Barnes wrote:
> Fundamental and simple functionality.
> 
> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>

Mark entire testcase as basic instead to catch future extensions?
getparams should always complete super-fast I think.
-Daniel

> ---
>  tests/drv_getparams.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/tests/drv_getparams.c b/tests/drv_getparams.c
> index 31382e9..5a5c8bb 100644
> --- a/tests/drv_getparams.c
> +++ b/tests/drv_getparams.c
> @@ -157,9 +157,9 @@ igt_main
>  		init();
>  	}
>  
> -	igt_subtest("subslice-total")
> +	igt_subtest("basic-subslice-total")
>  		subslice_total();
>  
> -	igt_subtest("eu-total")
> +	igt_subtest("basic-eu-total")
>  		eu_total();
>  }
> -- 
> 1.9.1
> 
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/intel-gfx

-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 06/18] tests/drv_suspend: mark sysfs tests as basic

[Daniel Vetter]

On Thu, Aug 13, 2015 at 01:31:29PM -0700, Jesse Barnes wrote:
> debugfs may not be mounted, but sysfs should always be restored after
> suspend or hibernate.
> 
> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>

We already have a suspend/resume testcase in kms_pipc_crc_basic. Do we
have enough budget for this one?
> ---
>  tests/drv_suspend.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/tests/drv_suspend.c b/tests/drv_suspend.c
> index d67a794..60ca8e3 100644
> --- a/tests/drv_suspend.c
> +++ b/tests/drv_suspend.c
> @@ -193,7 +193,7 @@ igt_main
>  	igt_subtest("debugfs-reader")
>  		test_debugfs_reader(false);
>  
> -	igt_subtest("sysfs-reader")
> +	igt_subtest("basic-sysfs-reader")
>  		test_sysfs_reader(false);
>  
>  	igt_subtest("forcewake")
> @@ -208,7 +208,7 @@ igt_main
>  	igt_subtest("debugfs-reader-hibernate")
>  		test_debugfs_reader(true);
>  
> -	igt_subtest("sysfs-reader-hibernate")
> +	igt_subtest("basic-sysfs-reader-hibernate")
>  		test_sysfs_reader(true);

Hibernate is a giantic can of worms. We have a big pile of issues here
that we never fixed. Otoh not testing hibernate is definitely an
oversight. Assuming we have the testing budget:

Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch>
>  
>  	igt_subtest("forcewake-hibernate")
> -- 
> 1.9.1
> 
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/intel-gfx

-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 07/18] tests/gem_ctx_exec: mark lrc lite restore as basic

[Daniel Vetter]

On Thu, Aug 13, 2015 at 01:31:30PM -0700, Jesse Barnes wrote:
> Need some LRC tests in the 'basic' subset, and this is a good simple
> one.
> 
> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>

This is just a testcase for a very specific lrc corner case. We do already
exercise lrc with all the other execbuf testcases. Imo we're covered
enough already with what we have in the basic testset - testing for all 3
billion cornercases will make it grow out of scope I fear. I'd just drop
this one here as not needed for BAT.

If you want to extend execbuffer scope a bit then we should add a
concurrency test, i.e. one of the gem_concurrent_blt testcases as basic
ones. Unfortunately to be able to reliable trigger race conditions those
all take a few seconds. But inter-batch sync is a _big_ gap across all
archs, and something which is even more tricky with lrc (and scheduler).
Imo that would be a lot more useful than this test here.
-Daniel

> ---
>  tests/gem_ctx_exec.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/tests/gem_ctx_exec.c b/tests/gem_ctx_exec.c
> index 3df939c..ea0fb7f 100644
> --- a/tests/gem_ctx_exec.c
> +++ b/tests/gem_ctx_exec.c
> @@ -216,7 +216,7 @@ igt_main
>  		gem_context_destroy(fd, ctx_id);
>  	}
>  
> -	igt_subtest("lrc-lite-restore") {
> +	igt_subtest("basic-lrc-lite-restore") {
>  		int i, j;
>  
>  		/*
> -- 
> 1.9.1
> 
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/intel-gfx

-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 08/18] tests/gem_mmap: mark basic object creation tests as basic

[Daniel Vetter]

On Thu, Aug 13, 2015 at 01:31:31PM -0700, Jesse Barnes wrote:
> We should be able to create small and moderate sized objects quickly and
> without errors.
> 
> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>

They're all super-fast basic testcases for api cornercases. I'd vote to
rename the entire testcase to gem_mmap_basic.
-Daniel

> ---
>  tests/gem_mmap.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/tests/gem_mmap.c b/tests/gem_mmap.c
> index 095f5b9..dd66ad6 100644
> --- a/tests/gem_mmap.c
> +++ b/tests/gem_mmap.c
> @@ -118,7 +118,7 @@ igt_main
>  		igt_assert(ret == -1 && errno == ENOENT);
>  	}
>  
> -	igt_subtest("new-object") {
> +	igt_subtest("basic") {
>  		arg.handle = gem_create(fd, OBJECT_SIZE);
>  		arg.offset = 0;
>  		arg.size = OBJECT_SIZE;
> @@ -154,7 +154,7 @@ igt_main
>  		gem_close(fd, arg.handle);
>  	}
>  
> -	igt_subtest("small-bo")
> +	igt_subtest("basic-small-bo")
>  		test_huge_bo(fd, -1);
>  	igt_subtest("big-bo")
>  		test_huge_bo(fd, 0);
> -- 
> 1.9.1
> 
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/intel-gfx

-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 09/18] tests/gem_mmap_gtt: mark basic access and copy tests as basic

[Daniel Vetter]

On Thu, Aug 13, 2015 at 01:31:32PM -0700, Jesse Barnes wrote:
> These ones should always pass and are fairly quick.
> 
> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
> ---
>  tests/gem_mmap_gtt.c | 8 ++++----
>  1 file changed, 4 insertions(+), 4 deletions(-)
> 
> diff --git a/tests/gem_mmap_gtt.c b/tests/gem_mmap_gtt.c
> index a95b98a..f964b39 100644
> --- a/tests/gem_mmap_gtt.c
> +++ b/tests/gem_mmap_gtt.c
> @@ -525,7 +525,7 @@ igt_main
>  	igt_fixture
>  		fd = drm_open_any();
>  
> -	igt_subtest("access")
> +	igt_subtest("basic")
>  		test_access(fd);
>  	igt_subtest("short")
>  		test_short(fd);
> @@ -556,11 +556,11 @@ igt_main
>  	igt_subtest("write-cpu-read-gtt")
>  		test_write_cpu_read_gtt(fd);
>  
> -	igt_subtest("small-bo")
> +	igt_subtest("basic-small-bo")
>  		test_huge_bo(fd, -1, I915_TILING_NONE);
> -	igt_subtest("small-bo-tiledX")
> +	igt_subtest("basic-small-bo-tiledX")
>  		test_huge_bo(fd, -1, I915_TILING_X);
> -	igt_subtest("small-bo-tiledY")
> +	igt_subtest("basic-small-bo-tiledY")
>  		test_huge_bo(fd, -1, I915_TILING_Y);

All the testcases up to this one are super-basic sanity checks and
coherency checks. They should also run really fast with the exception of
fault-concurrent. I'd mark them all as basic, except fault-concurrent
since that's more a stress-test.
-Daniel

>  
>  	igt_subtest("big-bo")
> -- 
> 1.9.1
> 
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/intel-gfx

-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 10/18] tests/gem_pread/pwrite: mark normal tests as basic

[Daniel Vetter]

On Thu, Aug 13, 2015 at 01:31:33PM -0700, Jesse Barnes wrote:
> These should always pass.
> 
> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>

If they're fast enough ... Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch>
> ---
>  tests/gem_pread.c  | 2 +-
>  tests/gem_pwrite.c | 2 +-
>  2 files changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/tests/gem_pread.c b/tests/gem_pread.c
> index cc83948..3ec5fb1 100644
> --- a/tests/gem_pread.c
> +++ b/tests/gem_pread.c
> @@ -108,7 +108,7 @@ int main(int argc, char **argv)
>  		src = malloc(object_size);
>  	}
>  
> -	igt_subtest("normal") {
> +	igt_subtest("basic") {
>  		for (count = 1; count <= 1<<17; count <<= 1) {
>  			struct timeval start, end;
>  
> diff --git a/tests/gem_pwrite.c b/tests/gem_pwrite.c
> index 5b6a77f..1d867e7 100644
> --- a/tests/gem_pwrite.c
> +++ b/tests/gem_pwrite.c
> @@ -169,7 +169,7 @@ int main(int argc, char **argv)
>  		src = malloc(object_size);
>  	}
>  
> -	igt_subtest("normal") {
> +	igt_subtest("basic") {
>  		for (count = 1; count <= 1<<17; count <<= 1) {
>  			struct timeval start, end;
>  
> -- 
> 1.9.1
> 
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/intel-gfx

-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 08/18] tests/gem_mmap: mark basic object creation tests as basic

[Chris Wilson]

On Fri, Aug 14, 2015 at 02:33:25PM +0200, Daniel Vetter wrote:
> On Thu, Aug 13, 2015 at 01:31:31PM -0700, Jesse Barnes wrote:
> > We should be able to create small and moderate sized objects quickly and
> > without errors.
> > 
> > Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
> 
> They're all super-fast basic testcases for api cornercases. I'd vote to
> rename the entire testcase to gem_mmap_basic.

Hmm. Time to add another test, huger-bo.
-Chris

-- 
Chris Wilson, Intel Open Source Technology Centre
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 11/18] tests/gem_tiled_pread/pwrite: mark normal tests as basic

[Daniel Vetter]

On Thu, Aug 13, 2015 at 01:31:34PM -0700, Jesse Barnes wrote:
> These simple tests should always pass.
> 
> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>

Imo shouldn't be part of the basic set, they thrash the machine quite
badly. Especially gem_tiled_pread_pwrite thrashes all of memory, so nack
on that one from me. At least until we've implemented the speedup with
memlock that's been in JIRA since years ...

For gem_tiled_pread, why not just rename to gem_tiled_pread_basic?
-Daniel

> ---
>  tests/gem_tiled_pread.c        | 167 +++++++++++++++++++++--------------------
>  tests/gem_tiled_pread_pwrite.c |  48 ++++++------
>  2 files changed, 112 insertions(+), 103 deletions(-)
> 
> diff --git a/tests/gem_tiled_pread.c b/tests/gem_tiled_pread.c
> index fdc5173..92bb649 100644
> --- a/tests/gem_tiled_pread.c
> +++ b/tests/gem_tiled_pread.c
> @@ -112,7 +112,7 @@ calculate_expected(int offset)
>  	return (base_y + tile_y) * WIDTH + base_x + tile_x;
>  }
>  
> -igt_simple_main
> +igt_main
>  {
>  	int fd;
>  	int i, iter = 100;
> @@ -120,96 +120,101 @@ igt_simple_main
>  	uint32_t handle;
>  	uint32_t devid;
>  
> -	fd = drm_open_any();
> +	igt_fixture {
> +		fd = drm_open_any();
>  
> -	handle = create_bo(fd);
> -	gem_get_tiling(fd, handle, &tiling, &swizzle);
> +		handle = create_bo(fd);
> +		gem_get_tiling(fd, handle, &tiling, &swizzle);
>  
> -	devid = intel_get_drm_devid(fd);
> -
> -	if (IS_GEN2(devid)) {
> -		tile_height = 16;
> -		tile_width = 128;
> -		tile_size = 2048;
> -	} else {
> -		tile_height = 8;
> -		tile_width = 512;
> -		tile_size = PAGE_SIZE;
> +		devid = intel_get_drm_devid(fd);
>  	}
>  
> -	/* Read a bunch of random subsets of the data and check that they come
> -	 * out right.
> -	 */
> -	for (i = 0; i < iter; i++) {
> -		int size = WIDTH * HEIGHT * 4;
> -		int offset = (random() % size) & ~3;
> -		int len = (random() % size) & ~3;
> -		int j;
> +	igt_subtest("basic") {
>  
> -		if (len == 0)
> -			len = 4;
> +		if (IS_GEN2(devid)) {
> +			tile_height = 16;
> +			tile_width = 128;
> +			tile_size = 2048;
> +		} else {
> +			tile_height = 8;
> +			tile_width = 512;
> +			tile_size = PAGE_SIZE;
> +		}
>  
> -		if (offset + len > size)
> -			len = size - offset;
> +		/* Read a bunch of random subsets of the data and check that they come
> +		 * out right.
> +		 */
> +		for (i = 0; i < iter; i++) {
> +			int size = WIDTH * HEIGHT * 4;
> +			int offset = (random() % size) & ~3;
> +			int len = (random() % size) & ~3;
> +			int j;
>  
> -		if (i == 0) {
> -			offset = 0;
> -			len = size;
> -		}
> +			if (len == 0)
> +				len = 4;
>  
> -		gem_read(fd, handle, offset, linear, len);
> +			if (offset + len > size)
> +				len = size - offset;
>  
> -		/* Translate from offsets in the read buffer to the swizzled
> -		 * address that it corresponds to.  This is the opposite of
> -		 * what Mesa does (calculate offset to be read given the linear
> -		 * offset it's looking for).
> -		 */
> -		for (j = offset; j < offset + len; j += 4) {
> -			uint32_t expected_val, found_val;
> -			int swizzled_offset;
> -			const char *swizzle_str;
> -
> -			switch (swizzle) {
> -			case I915_BIT_6_SWIZZLE_NONE:
> -				swizzled_offset = j;
> -				swizzle_str = "none";
> -				break;
> -			case I915_BIT_6_SWIZZLE_9:
> -				swizzled_offset = j ^
> -					swizzle_bit(9, j);
> -				swizzle_str = "bit9";
> -				break;
> -			case I915_BIT_6_SWIZZLE_9_10:
> -				swizzled_offset = j ^
> -					swizzle_bit(9, j) ^
> -					swizzle_bit(10, j);
> -				swizzle_str = "bit9^10";
> -				break;
> -			case I915_BIT_6_SWIZZLE_9_11:
> -				swizzled_offset = j ^
> -					swizzle_bit(9, j) ^
> -					swizzle_bit(11, j);
> -				swizzle_str = "bit9^11";
> -				break;
> -			case I915_BIT_6_SWIZZLE_9_10_11:
> -				swizzled_offset = j ^
> -					swizzle_bit(9, j) ^
> -					swizzle_bit(10, j) ^
> -					swizzle_bit(11, j);
> -				swizzle_str = "bit9^10^11";
> -				break;
> -			default:
> -				igt_assert_f(0, "Bad swizzle bits; %d\n",
> -					     swizzle);
> +			if (i == 0) {
> +				offset = 0;
> +				len = size;
> +			}
> +
> +			gem_read(fd, handle, offset, linear, len);
> +
> +			/* Translate from offsets in the read buffer to the swizzled
> +			 * address that it corresponds to.  This is the opposite of
> +			 * what Mesa does (calculate offset to be read given the linear
> +			 * offset it's looking for).
> +			 */
> +			for (j = offset; j < offset + len; j += 4) {
> +				uint32_t expected_val, found_val;
> +				int swizzled_offset;
> +				const char *swizzle_str;
> +
> +				switch (swizzle) {
> +				case I915_BIT_6_SWIZZLE_NONE:
> +					swizzled_offset = j;
> +					swizzle_str = "none";
> +					break;
> +				case I915_BIT_6_SWIZZLE_9:
> +					swizzled_offset = j ^
> +						swizzle_bit(9, j);
> +					swizzle_str = "bit9";
> +					break;
> +				case I915_BIT_6_SWIZZLE_9_10:
> +					swizzled_offset = j ^
> +						swizzle_bit(9, j) ^
> +						swizzle_bit(10, j);
> +					swizzle_str = "bit9^10";
> +					break;
> +				case I915_BIT_6_SWIZZLE_9_11:
> +					swizzled_offset = j ^
> +						swizzle_bit(9, j) ^
> +						swizzle_bit(11, j);
> +					swizzle_str = "bit9^11";
> +					break;
> +				case I915_BIT_6_SWIZZLE_9_10_11:
> +					swizzled_offset = j ^
> +						swizzle_bit(9, j) ^
> +						swizzle_bit(10, j) ^
> +						swizzle_bit(11, j);
> +					swizzle_str = "bit9^10^11";
> +					break;
> +				default:
> +					igt_assert_f(0, "Bad swizzle bits; %d\n",
> +						     swizzle);
> +				}
> +				expected_val = calculate_expected(swizzled_offset);
> +				found_val = linear[(j - offset) / 4];
> +				igt_assert_f(expected_val == found_val,
> +					     "Bad read [%d]: %d instead of %d at 0x%08x "
> +					     "for read from 0x%08x to 0x%08x, swizzle=%s\n",
> +					     i, found_val, expected_val, j,
> +					     offset, offset + len,
> +					     swizzle_str);
>  			}
> -			expected_val = calculate_expected(swizzled_offset);
> -			found_val = linear[(j - offset) / 4];
> -			igt_assert_f(expected_val == found_val,
> -				     "Bad read [%d]: %d instead of %d at 0x%08x "
> -				     "for read from 0x%08x to 0x%08x, swizzle=%s\n",
> -				     i, found_val, expected_val, j,
> -				     offset, offset + len,
> -				     swizzle_str);
>  		}
>  	}
>  
> diff --git a/tests/gem_tiled_pread_pwrite.c b/tests/gem_tiled_pread_pwrite.c
> index 3d8fdc9..efb56d5 100644
> --- a/tests/gem_tiled_pread_pwrite.c
> +++ b/tests/gem_tiled_pread_pwrite.c
> @@ -101,7 +101,7 @@ create_bo(int fd)
>  	return handle;
>  }
>  
> -igt_simple_main
> +igt_main
>  {
>  	int fd;
>  	uint32_t *data;
> @@ -109,34 +109,38 @@ igt_simple_main
>  	uint32_t tiling, swizzle;
>  	uint32_t handle, handle_target;
>  	int count;
> -	
> -	fd = drm_open_any();
> -	count = SLOW_QUICK(intel_get_total_ram_mb() * 9 / 10, 8) ;
>  
> -	for (i = 0; i < count/2; i++) {
> -		current_tiling_mode = I915_TILING_X;
> +	igt_fixture {
> +		fd = drm_open_any();
> +		count = SLOW_QUICK(intel_get_total_ram_mb() * 9 / 10, 8);
> +	}
> +
> +	igt_subtest("basic") {
> +		for (i = 0; i < count/2; i++) {
> +			current_tiling_mode = I915_TILING_X;
>  
> -		handle = create_bo_and_fill(fd);
> -		gem_get_tiling(fd, handle, &tiling, &swizzle);
> +			handle = create_bo_and_fill(fd);
> +			gem_get_tiling(fd, handle, &tiling, &swizzle);
>  
> -		gem_read(fd, handle, 0, linear, sizeof(linear));
> +			gem_read(fd, handle, 0, linear, sizeof(linear));
>  
> -		handle_target = create_bo(fd);
> -		gem_write(fd, handle_target, 0, linear, sizeof(linear));
> +			handle_target = create_bo(fd);
> +			gem_write(fd, handle_target, 0, linear, sizeof(linear));
>  
> -		/* Check the target bo's contents. */
> -		data = gem_mmap(fd, handle_target, sizeof(linear), PROT_READ | PROT_WRITE);
> -		for (j = 0; j < WIDTH*HEIGHT; j++)
> -			igt_assert_f(data[j] == j,
> -				     "mismatch at %i: %i\n",
> -				     j, data[j]);
> -		munmap(data, sizeof(linear));
> +			/* Check the target bo's contents. */
> +			data = gem_mmap(fd, handle_target, sizeof(linear), PROT_READ | PROT_WRITE);
> +			for (j = 0; j < WIDTH*HEIGHT; j++)
> +				igt_assert_f(data[j] == j,
> +					     "mismatch at %i: %i\n",
> +					     j, data[j]);
> +			munmap(data, sizeof(linear));
>  
> -		/* Leak both bos so that we use all of system mem! */
> -		gem_madvise(fd, handle_target, I915_MADV_DONTNEED);
> -		gem_madvise(fd, handle, I915_MADV_DONTNEED);
> +			/* Leak both bos so that we use all of system mem! */
> +			gem_madvise(fd, handle_target, I915_MADV_DONTNEED);
> +			gem_madvise(fd, handle, I915_MADV_DONTNEED);
>  
> -		igt_progress("gem_tiled_pread_pwrite: ", i, count/2);
> +			igt_progress("gem_tiled_pread_pwrite: ", i, count/2);
> +		}
>  	}
>  
>  	close(fd);
> -- 
> 1.9.1
> 
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/intel-gfx

-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 12/18] tests/kms_addfb: mark simple fb creation tests as basic

[Daniel Vetter]

On Thu, Aug 13, 2015 at 01:31:35PM -0700, Jesse Barnes wrote:
> We should always be able to create simple and tiled objects.
> 
> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>

There's are all super-fast abi tests which don't even do a modeset. Imo
add them all by renaming the testcase to kms_addfb_basic. The real
functional testcases are somewhere else (x/y-tiled flipping, rotation,
stuff like that).
-Daniel

> ---
>  tests/kms_addfb.c | 6 +++---
>  1 file changed, 3 insertions(+), 3 deletions(-)
> 
> diff --git a/tests/kms_addfb.c b/tests/kms_addfb.c
> index 42ee632..f10e12b 100644
> --- a/tests/kms_addfb.c
> +++ b/tests/kms_addfb.c
> @@ -121,7 +121,7 @@ static void pitch_tests(int fd)
>  	}
>  
>  	f.handles[0] = gem_bo;
> -	igt_subtest("normal") {
> +	igt_subtest("basic") {
>  		igt_assert(drmIoctl(fd, DRM_IOCTL_MODE_ADDFB2, &f) == 0);
>  		igt_assert(drmIoctl(fd, DRM_IOCTL_MODE_RMFB, &f.fb_id) == 0);
>  		f.fb_id = 0;
> @@ -139,7 +139,7 @@ static void pitch_tests(int fd)
>  		gem_set_tiling(fd, gem_bo, I915_TILING_X, 1024*4);
>  	f.pitches[0] = 1024*4;
>  
> -	igt_subtest("X-tiled") {
> +	igt_subtest("basic-X-tiled") {
>  		igt_assert(drmIoctl(fd, DRM_IOCTL_MODE_ADDFB2, &f) == 0);
>  		igt_assert(drmIoctl(fd, DRM_IOCTL_MODE_RMFB, &f.fb_id) == 0);
>  		f.fb_id = 0;
> @@ -162,7 +162,7 @@ static void pitch_tests(int fd)
>  	igt_fixture
>  		gem_set_tiling(fd, gem_bo, I915_TILING_Y, 1024*4);
>  	f.pitches[0] = 1024*4;
> -	igt_subtest("Y-tiled") {
> +	igt_subtest("basic-Y-tiled") {
>  		igt_assert(drmIoctl(fd, DRM_IOCTL_MODE_ADDFB2, &f) == -1 &&
>  			   errno == EINVAL);
>  	}
> -- 
> 1.9.1
> 
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/intel-gfx

-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 13/18] tests/kms_vblank: mark accuracy test as basic

[Daniel Vetter]

On Thu, Aug 13, 2015 at 01:31:36PM -0700, Jesse Barnes wrote:
> Need some simple vblank coverage in the BAT list.
> 
> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>

This testcase relies upon fbcon to have enabled pipe 0, which means it
spuriously skips (when fbcon is disabled or when it's blanked the screen)
until that's fixed. Same problem happens with drm_read.

Until that's addressed imo Nack for basic igt testcase list. I'll do a
jira about this issue.
-Daniel
> ---
>  tests/kms_vblank.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/tests/kms_vblank.c b/tests/kms_vblank.c
> index 6946177..ef2f7ca 100644
> --- a/tests/kms_vblank.c
> +++ b/tests/kms_vblank.c
> @@ -188,7 +188,7 @@ igt_main
>  		igt_require(crtc0_active(fd));
>  	}
>  
> -	igt_subtest("accuracy")
> +	igt_subtest("basic-accuracy")
>  		accuracy(fd);
>  
>  	igt_subtest("query-idle")
> -- 
> 1.9.1
> 
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/intel-gfx

-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 13/18] tests/kms_vblank: mark accuracy test as basic

[Daniel Vetter]

On Fri, Aug 14, 2015 at 02:44:14PM +0200, Daniel Vetter wrote:
> On Thu, Aug 13, 2015 at 01:31:36PM -0700, Jesse Barnes wrote:
> > Need some simple vblank coverage in the BAT list.
> > 
> > Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
> 
> This testcase relies upon fbcon to have enabled pipe 0, which means it
> spuriously skips (when fbcon is disabled or when it's blanked the screen)
> until that's fixed. Same problem happens with drm_read.
> 
> Until that's addressed imo Nack for basic igt testcase list. I'll do a
> jira about this issue.

Meanwhile for a basic vblank testcase I think we should pick one of the
kms_flip ones.
-Daniel

> -Daniel
> > ---
> >  tests/kms_vblank.c | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> > 
> > diff --git a/tests/kms_vblank.c b/tests/kms_vblank.c
> > index 6946177..ef2f7ca 100644
> > --- a/tests/kms_vblank.c
> > +++ b/tests/kms_vblank.c
> > @@ -188,7 +188,7 @@ igt_main
> >  		igt_require(crtc0_active(fd));
> >  	}
> >  
> > -	igt_subtest("accuracy")
> > +	igt_subtest("basic-accuracy")
> >  		accuracy(fd);
> >  
> >  	igt_subtest("query-idle")
> > -- 
> > 1.9.1
> > 
> > _______________________________________________
> > Intel-gfx mailing list
> > Intel-gfx@lists.freedesktop.org
> > http://lists.freedesktop.org/mailman/listinfo/intel-gfx
> 
> -- 
> Daniel Vetter
> Software Engineer, Intel Corporation
> http://blog.ffwll.ch

-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 14/18] tests/pm_backlight: mark simple test as basic

[Daniel Vetter]

On Thu, Aug 13, 2015 at 01:31:37PM -0700, Jesse Barnes wrote:
> We should be able to adjust the backlight and observe changes in sysfs.
> 
> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>

Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>

> ---
>  tests/pm_backlight.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/tests/pm_backlight.c b/tests/pm_backlight.c
> index d02336d..cff2694 100644
> --- a/tests/pm_backlight.c
> +++ b/tests/pm_backlight.c
> @@ -157,7 +157,7 @@ igt_main
>  		igt_assert(backlight_read(&max, "max_brightness") > -1);
>  	}
>  
> -	igt_subtest("brightness")
> +	igt_subtest("basic-brightness")
>  		test_brightness(max);
>  	igt_subtest("bad-brightness")
>  		test_bad_brightness(max);
> -- 
> 1.9.1
> 
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/intel-gfx

-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 15/18] tests/pm_rpm: mark RTE and D3 tests as basic

[Daniel Vetter]

On Thu, Aug 13, 2015 at 01:31:38PM -0700, Jesse Barnes wrote:
> These always need to pass for basic PM functionality.
> 
> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
> ---
>  tests/pm_rpm.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/tests/pm_rpm.c b/tests/pm_rpm.c
> index d509fa8..7ae5806 100644
> --- a/tests/pm_rpm.c
> +++ b/tests/pm_rpm.c
> @@ -1832,11 +1832,11 @@ int main(int argc, char *argv[])
>  			stay_subtest();
>  
>  	/* Essential things */
> -	igt_subtest("rte")
> +	igt_subtest("basic-rte")

I thought our BAT criteria would be basic|rte?

rte is runtime environment check and more along the lines of "did QA set
up their machine correctly", so imo good to keep separate.
-Daniel

>  		basic_subtest();
>  	igt_subtest("drm-resources-equal")
>  		drm_resources_equal_subtest();
> -	igt_subtest("pci-d3-state")
> +	igt_subtest("basic-pci-d3-state")
>  		pci_d3_state_subtest();
>  
>  	/* Basic modeset */
> -- 
> 1.9.1
> 
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/intel-gfx

-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 08/18] tests/gem_mmap: mark basic object creation tests as basic

[Chris Wilson]

On Fri, Aug 14, 2015 at 01:37:46PM +0100, Chris Wilson wrote:
> On Fri, Aug 14, 2015 at 02:33:25PM +0200, Daniel Vetter wrote:
> > On Thu, Aug 13, 2015 at 01:31:31PM -0700, Jesse Barnes wrote:
> > > We should be able to create small and moderate sized objects quickly and
> > > without errors.
> > > 
> > > Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
> > 
> > They're all super-fast basic testcases for api cornercases. I'd vote to
> > rename the entire testcase to gem_mmap_basic.
> 
> Hmm. Time to add another test, huger-bo.

Actually, the problem is that I broke the test...
-Chris

-- 
Chris Wilson, Intel Open Source Technology Centre
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 16/18] tests/kms_flip: add basic tests for flip, flip vs dpms, and flip modeset

[Daniel Vetter]

On Thu, Aug 13, 2015 at 01:31:39PM -0700, Jesse Barnes wrote:
> Simple variants that don't do multiple output or interruptible testing.
> 
> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
> ---
>  tests/kms_flip.c | 11 ++++++++++-
>  1 file changed, 10 insertions(+), 1 deletion(-)
> 
> diff --git a/tests/kms_flip.c b/tests/kms_flip.c
> index a595d9f..a0e4112 100644
> --- a/tests/kms_flip.c
> +++ b/tests/kms_flip.c
> @@ -80,6 +80,7 @@
>  #define TEST_TS_CONT		(1 << 27)
>  #define TEST_BO_TOOBIG		(1 << 28)
>  #define TEST_HANG_ONCE		(1 << 29)
> +#define TEST_BASIC		(1 << 30)
>  
>  #define EVENT_FLIP		(1 << 0)
>  #define EVENT_VBLANK		(1 << 1)
> @@ -1649,7 +1650,7 @@ int main(int argc, char **argv)
>  					"blt-wf_vblank-vs-modeset" },
>  		{ 60,  TEST_VBLANK | TEST_MODESET | TEST_WITH_DUMMY_RCS,
>  					"rcs-wf_vblank-vs-modeset" },
> -
> +		{ 30, TEST_FLIP | TEST_BASIC, "basic-plain-flip" },

I think for better coverage we should pick the timestamp checking ones
here. Also we need one to cover vblank since kms_vblank can't be used. And
we can do that all in one testcases to save testing time!

>  		{ 30, TEST_FLIP , "plain-flip" },
>  		{ 30, TEST_FLIP | TEST_EBUSY , "busy-flip" },
>  		{ 30, TEST_FLIP | TEST_FENCE_STRESS , "flip-vs-fences" },
> @@ -1657,12 +1658,14 @@ int main(int argc, char **argv)
>  		{ 30, TEST_FLIP | TEST_CHECK_TS | TEST_FB_RECREATE,
>  			"plain-flip-fb-recreate" },
>  		{ 30, TEST_FLIP | TEST_RMFB | TEST_MODESET , "flip-vs-rmfb" },
> +		{ 60, TEST_FLIP | TEST_DPMS | TEST_EINVAL | TEST_BASIC, "basic-flip-vs-dpms" },
>  		{ 60, TEST_FLIP | TEST_DPMS | TEST_EINVAL, "flip-vs-dpms" },
>  		{ 60, TEST_FLIP | TEST_DPMS | TEST_WITH_DUMMY_BCS, "blt-flip-vs-dpms" },
>  		{ 60, TEST_FLIP | TEST_DPMS | TEST_WITH_DUMMY_RCS, "render-flip-vs-dpms" },
>  		{ 30,  TEST_FLIP | TEST_PAN, "flip-vs-panning" },
>  		{ 60, TEST_FLIP | TEST_PAN | TEST_WITH_DUMMY_BCS, "blt-flip-vs-panning" },
>  		{ 60, TEST_FLIP | TEST_PAN | TEST_WITH_DUMMY_RCS, "render-flip-vs-panning" },
> +		{ 60, TEST_FLIP | TEST_MODESET | TEST_EINVAL | TEST_BASIC, "basic-flip-vs-modeset" },
>  		{ 60, TEST_FLIP | TEST_MODESET | TEST_EINVAL, "flip-vs-modeset" },
>  		{ 60, TEST_FLIP | TEST_MODESET | TEST_WITH_DUMMY_BCS, "blt-flip-vs-modeset" },
>  		{ 60, TEST_FLIP | TEST_MODESET | TEST_WITH_DUMMY_RCS, "render-flip-vs-modeset" },
> @@ -1721,6 +1724,9 @@ int main(int argc, char **argv)
>  		igt_subtest(tests[i].name)
>  			run_test(tests[i].duration, tests[i].flags);
>  
> +		if (tests[i].flags & TEST_BASIC)
> +			continue;

This means you remove the tests you marked as BASIC from the 2x and
interruptible sets. What about instead adding the basic prefix at runtime,
i.e. for all my requests:
diff --git a/tests/kms_flip.c b/tests/kms_flip.c
index a595d9f1d69f..aaf03b07df87 100644
--- a/tests/kms_flip.c
+++ b/tests/kms_flip.c
@@ -80,6 +80,7 @@
 #define TEST_TS_CONT		(1 << 27)
 #define TEST_BO_TOOBIG		(1 << 28)
 #define TEST_HANG_ONCE		(1 << 29)
+#define TEST_BASIC		(1 << 31)
 
 #define EVENT_FLIP		(1 << 0)
 #define EVENT_VBLANK		(1 << 1)
@@ -1650,20 +1651,20 @@ int main(int argc, char **argv)
 		{ 60,  TEST_VBLANK | TEST_MODESET | TEST_WITH_DUMMY_RCS,
 					"rcs-wf_vblank-vs-modeset" },
 
-		{ 30, TEST_FLIP , "plain-flip" },
+		{ 30, TEST_FLIP, "plain-flip" },
 		{ 30, TEST_FLIP | TEST_EBUSY , "busy-flip" },
 		{ 30, TEST_FLIP | TEST_FENCE_STRESS , "flip-vs-fences" },
 		{ 30, TEST_FLIP | TEST_CHECK_TS, "plain-flip-ts-check" },
 		{ 30, TEST_FLIP | TEST_CHECK_TS | TEST_FB_RECREATE,
 			"plain-flip-fb-recreate" },
 		{ 30, TEST_FLIP | TEST_RMFB | TEST_MODESET , "flip-vs-rmfb" },
-		{ 60, TEST_FLIP | TEST_DPMS | TEST_EINVAL, "flip-vs-dpms" },
+		{ 60, TEST_FLIP | TEST_DPMS | TEST_EINVAL | TEST_BASIC, "flip-vs-dpms" },
 		{ 60, TEST_FLIP | TEST_DPMS | TEST_WITH_DUMMY_BCS, "blt-flip-vs-dpms" },
 		{ 60, TEST_FLIP | TEST_DPMS | TEST_WITH_DUMMY_RCS, "render-flip-vs-dpms" },
 		{ 30,  TEST_FLIP | TEST_PAN, "flip-vs-panning" },
 		{ 60, TEST_FLIP | TEST_PAN | TEST_WITH_DUMMY_BCS, "blt-flip-vs-panning" },
 		{ 60, TEST_FLIP | TEST_PAN | TEST_WITH_DUMMY_RCS, "render-flip-vs-panning" },
-		{ 60, TEST_FLIP | TEST_MODESET | TEST_EINVAL, "flip-vs-modeset" },
+		{ 60, TEST_FLIP | TEST_MODESET | TEST_EINVAL | TEST_BASIC, "flip-vs-modeset" },
 		{ 60, TEST_FLIP | TEST_MODESET | TEST_WITH_DUMMY_BCS, "blt-flip-vs-modeset" },
 		{ 60, TEST_FLIP | TEST_MODESET | TEST_WITH_DUMMY_RCS, "render-flip-vs-modeset" },
 		{ 30,  TEST_FLIP | TEST_VBLANK_EXPIRED_SEQ,
@@ -1671,7 +1672,7 @@ int main(int argc, char **argv)
 
 		{ 30, TEST_FLIP | TEST_VBLANK | TEST_VBLANK_ABSOLUTE |
 		      TEST_CHECK_TS, "flip-vs-absolute-wf_vblank" },
-		{ 30, TEST_FLIP | TEST_VBLANK | TEST_CHECK_TS,
+		{ 30, TEST_FLIP | TEST_VBLANK | TEST_CHECK_TS | TEST_BASIC,
 					"flip-vs-wf_vblank" },
 		{ 30, TEST_FLIP | TEST_VBLANK | TEST_VBLANK_BLOCK |
 			TEST_CHECK_TS, "flip-vs-blocking-wf-vblank" },
@@ -1718,7 +1719,9 @@ int main(int argc, char **argv)
 		test_nonblocking_read(drm_fd);
 
 	for (i = 0; i < sizeof(tests) / sizeof (tests[0]); i++) {
-		igt_subtest(tests[i].name)
+		igt_subtest_f("%s%s",
+			      tests[i].flags & TEST_BASIC ? "basic-" : "",
+			      tests[i].name)
 			run_test(tests[i].duration, tests[i].flags);
 
 		if (tests[i].flags & TEST_NO_2X_OUTPUT)


Totally untested ;-)

Cheers, Daniel
-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 17/18] tests/kms_setmode: mark simple clone test as basic

[Daniel Vetter]

On Thu, Aug 13, 2015 at 01:31:40PM -0700, Jesse Barnes wrote:
> Should cover simple, single CRTC mode sets.
> 
> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
> ---
>  tests/kms_setmode.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/tests/kms_setmode.c b/tests/kms_setmode.c
> index 82769ab..2c0fe45 100644
> --- a/tests/kms_setmode.c
> +++ b/tests/kms_setmode.c
> @@ -685,7 +685,7 @@ int main(int argc, char **argv)
>  		const char *name;
>  	} tests[] = {
>  		{ TEST_CLONE | TEST_SINGLE_CRTC_CLONE,
> -					"clone-single-crtc" },
> +					"basic-clone-single-crtc" },

This isn't all that simple, it covers cloning. But since kms_flip doesn't
do that yet definitely makes sense to include (assuming we still have
budget for kms tests).

Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch>

>  		{ TEST_INVALID | TEST_CLONE | TEST_SINGLE_CRTC_CLONE,
>  					"invalid-clone-single-crtc" },
>  		{ TEST_INVALID | TEST_CLONE | TEST_EXCLUSIVE_CRTC_CLONE,
> -- 
> 1.9.1
> 
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/intel-gfx

-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 15/18] tests/pm_rpm: mark RTE and D3 tests as basic

[Paulo Zanoni]

2015-08-14 9:50 GMT-03:00 Daniel Vetter <daniel@ffwll.ch>:
> On Thu, Aug 13, 2015 at 01:31:38PM -0700, Jesse Barnes wrote:
>> These always need to pass for basic PM functionality.
>>
>> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
>> ---
>>  tests/pm_rpm.c | 4 ++--
>>  1 file changed, 2 insertions(+), 2 deletions(-)
>>
>> diff --git a/tests/pm_rpm.c b/tests/pm_rpm.c
>> index d509fa8..7ae5806 100644
>> --- a/tests/pm_rpm.c
>> +++ b/tests/pm_rpm.c
>> @@ -1832,11 +1832,11 @@ int main(int argc, char *argv[])
>>                       stay_subtest();
>>
>>       /* Essential things */
>> -     igt_subtest("rte")
>> +     igt_subtest("basic-rte")
>
> I thought our BAT criteria would be basic|rte?
>
> rte is runtime environment check and more along the lines of "did QA set
> up their machine correctly", so imo good to keep separate.

When I originally named the test, I thought the test would catch bad
machine setups more often than the other case. But after QA managed to
properly setup their machines, we got tons and tons of Kernel
regressions that broke pm_rpm/rte. So "basic-rte" wouldn't be a bad
name, since we can actually catch both bad setups and fundamental
regressions here.

> -Daniel
>
>>               basic_subtest();
>>       igt_subtest("drm-resources-equal")
>>               drm_resources_equal_subtest();
>> -     igt_subtest("pci-d3-state")
>> +     igt_subtest("basic-pci-d3-state")
>>               pci_d3_state_subtest();
>>
>>       /* Basic modeset */
>> --
>> 1.9.1
>>
>> _______________________________________________
>> Intel-gfx mailing list
>> Intel-gfx@lists.freedesktop.org
>> http://lists.freedesktop.org/mailman/listinfo/intel-gfx
>
> --
> Daniel Vetter
> Software Engineer, Intel Corporation
> http://blog.ffwll.ch
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/intel-gfx



-- 
Paulo Zanoni
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 18/18] gitignore: ignore more files

[Jesse Barnes]

git clean updates the .gitignore file?  Not having to run git clean is the whole point of this patch...

On 08/14/2015 01:09 AM, Daniel Vetter wrote:
> On Thu, Aug 13, 2015 at 01:31:41PM -0700, Jesse Barnes wrote:
> 
> git clean fixes this all, at least over here git status is clean.
> -Daniel
> 
>> ---
>>  .gitignore       |  3 +++
>>  tests/.gitignore | 13 +++++++++++++
>>  tools/.gitignore |  8 ++++++++
>>  3 files changed, 24 insertions(+)
>>
>> diff --git a/.gitignore b/.gitignore
>> index a438c1c..533f6f1 100644
>> --- a/.gitignore
>> +++ b/.gitignore
>> @@ -90,3 +90,6 @@ gtk-doc.m4
>>  
>>  piglit
>>  results
>> +
>> +*.orig
>> +version.h
>> diff --git a/tests/.gitignore b/tests/.gitignore
>> index 31d35a5..e45cbb7 100644
>> --- a/tests/.gitignore
>> +++ b/tests/.gitignore
>> @@ -99,6 +99,7 @@ gem_set_tiling_vs_blt
>>  gem_set_tiling_vs_gtt
>>  gem_set_tiling_vs_pwrite
>>  gem_storedw_batches_loop
>> +gem_storedw_loop
>>  gem_storedw_loop_blt
>>  gem_storedw_loop_bsd
>>  gem_storedw_loop_render
>> @@ -169,3 +170,15 @@ prime_udl
>>  template
>>  test-list.txt
>>  testdisplay
>> +ddi_compute_wrpll
>> +gem_vmap_blits
>> +gem_wait_render_timeout
>> +igt_fork_helper
>> +igt_list_only
>> +igt_no_exit
>> +igt_no_exit_list_only
>> +igt_no_subtest
>> +igt_simulation
>> +pm_pc8
>> +pm_psr
>> +
>> diff --git a/tools/.gitignore b/tools/.gitignore
>> index 49bd24f..7bf91e3 100644
>> --- a/tools/.gitignore
>> +++ b/tools/.gitignore
>> @@ -37,3 +37,11 @@ intel_vga_write
>>  intel_watermark
>>  skl_compute_wrpll
>>  skl_ddb_allocation
>> +forcewaked
>> +intel_dpio_read
>> +intel_dpio_write
>> +intel_gpu_dump
>> +intel_nc_read
>> +intel_nc_write
>> +intel_punit_read
>> +intel_punit_write
>> -- 
>> 1.9.1
>>
>> _______________________________________________
>> Intel-gfx mailing list
>> Intel-gfx@lists.freedesktop.org
>> http://lists.freedesktop.org/mailman/listinfo/intel-gfx
> 

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 01/18] tests/gem_storedw_loop: add new store_dword test to unify per-ring ones

[Jesse Barnes]

On 08/14/2015 05:19 AM, Daniel Vetter wrote:
> On Thu, Aug 13, 2015 at 01:31:24PM -0700, Jesse Barnes wrote:
>> There was a lot of duplication going on...  Mark as basic while we're at
>> it as these should never fail.
>>
>> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
>> ---
>>  tests/Makefile.sources   |   1 +
>>  tests/gem_storedw_loop.c | 181 +++++++++++++++++++++++++++++++++++++++++++++++
>>  2 files changed, 182 insertions(+)
>>  create mode 100644 tests/gem_storedw_loop.c
>>
>> diff --git a/tests/Makefile.sources b/tests/Makefile.sources
>> index b9a4cb4..cdcee33 100644
>> --- a/tests/Makefile.sources
>> +++ b/tests/Makefile.sources
>> @@ -138,6 +138,7 @@ TESTS_progs = \
>>  	gem_seqno_wrap \
>>  	gem_set_tiling_vs_gtt \
>>  	gem_set_tiling_vs_pwrite \
>> +	gem_storedw_loop \
>>  	gem_storedw_loop_blt \
>>  	gem_storedw_loop_bsd \
>>  	gem_storedw_loop_render \
> 
> Why not remove the old ones while at it? This just means more gunk in the
> overall igt set. Also please update .gitignore here for these ...

Yeah figured that would be a separate patch assuming this one looked ok.  I added it to the .gitignore in a later patch to update it all at once.

Jesse

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 03/18] tests/drv_module_reload_basic: use linear_blits after module_reload for sanity check

[Jesse Barnes]

On 08/14/2015 05:22 AM, Daniel Vetter wrote:
> On Thu, Aug 13, 2015 at 01:31:26PM -0700, Jesse Barnes wrote:
>> Reduces runtime a lot...
>>
>> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
>> ---
>>  tests/drv_module_reload_basic | 2 +-
>>  1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/tests/drv_module_reload_basic b/tests/drv_module_reload_basic
>> index bb29a64..cf7f2b8 100755
>> --- a/tests/drv_module_reload_basic
>> +++ b/tests/drv_module_reload_basic
>> @@ -52,7 +52,7 @@ else
>>  fi
>>  
>>  # then try to run something
>> -if ! $SOURCE_DIR/gem_exec_nop > /dev/null ; then
>> +if ! $SOURCE_DIR/gem_linear_blits --r basic > /dev/null ; then
> 
> Please use the full-lenght run-subtest to avoid lots of wtf when we add
> another option starting with r eventually. With that fixed:
> 
> Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>

Done, thanks.

Jesse

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 05/18] tests/drv_getparams: mark EU and subslice fetch as basic

[Jesse Barnes]

On 08/14/2015 05:27 AM, Daniel Vetter wrote:
> On Thu, Aug 13, 2015 at 01:31:28PM -0700, Jesse Barnes wrote:
>> Fundamental and simple functionality.
>>
>> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
> 
> Mark entire testcase as basic instead to catch future extensions?
> getparams should always complete super-fast I think.

Sure, that works too.

Jesse

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 04/18] tests/drm_import_export: mark flink and prime tests as basic

[Jesse Barnes]

On 08/14/2015 05:26 AM, Daniel Vetter wrote:
> On Thu, Aug 13, 2015 at 01:31:27PM -0700, Jesse Barnes wrote:
>> They're testing basic functionality and don't involve stress or race
>> induction.
>>
>> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
> 
> These are more stress-tests in nature I think. For basic testscases of
> prime I'd recommend instead
> 
> prime_self_import: Everything not marked with *-race
> 
> gem_flink: all of them.
> 
> instead of this patch here.

Sounds good, done.

Jesse

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 06/18] tests/drv_suspend: mark sysfs tests as basic

[Jesse Barnes]

On 08/14/2015 05:29 AM, Daniel Vetter wrote:
> On Thu, Aug 13, 2015 at 01:31:29PM -0700, Jesse Barnes wrote:
>> debugfs may not be mounted, but sysfs should always be restored after
>> suspend or hibernate.
>>
>> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
> 
> We already have a suspend/resume testcase in kms_pipc_crc_basic. Do we
> have enough budget for this one?

Yeah I thought about getting rid of the suspend/resume ones in pipe_crc (marking them as non-basic), since I really just want the one set of tests.  Any preference?

Jesse

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 07/18] tests/gem_ctx_exec: mark lrc lite restore as basic

[Jesse Barnes]

On 08/14/2015 05:32 AM, Daniel Vetter wrote:
> On Thu, Aug 13, 2015 at 01:31:30PM -0700, Jesse Barnes wrote:
>> Need some LRC tests in the 'basic' subset, and this is a good simple
>> one.
>>
>> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
> 
> This is just a testcase for a very specific lrc corner case. We do already
> exercise lrc with all the other execbuf testcases. Imo we're covered
> enough already with what we have in the basic testset - testing for all 3
> billion cornercases will make it grow out of scope I fear. I'd just drop
> this one here as not needed for BAT.
> 
> If you want to extend execbuffer scope a bit then we should add a
> concurrency test, i.e. one of the gem_concurrent_blt testcases as basic
> ones. Unfortunately to be able to reliable trigger race conditions those
> all take a few seconds. But inter-batch sync is a _big_ gap across all
> archs, and something which is even more tricky with lrc (and scheduler).
> Imo that would be a lot more useful than this test here.

Yeah that's a good point; I just saw 'lrc' and though "I want that", but you're right we should already be covered.  Definitely open to adding some concurrency stuff (maybe just a few seconds worth) as we get things in place.

I'll drop this one.

Jesse

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 08/18] tests/gem_mmap: mark basic object creation tests as basic

[Jesse Barnes]

On 08/14/2015 05:33 AM, Daniel Vetter wrote:
> On Thu, Aug 13, 2015 at 01:31:31PM -0700, Jesse Barnes wrote:
>> We should be able to create small and moderate sized objects quickly and
>> without errors.
>>
>> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
> 
> They're all super-fast basic testcases for api cornercases. I'd vote to
> rename the entire testcase to gem_mmap_basic.

Not quite, the huge ones were pretty slow in my measurements...  Not as bad as suspend/resume, but not super fast either.

Jesse

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 11/18] tests/gem_tiled_pread/pwrite: mark normal tests as basic

[Jesse Barnes]

On 08/14/2015 05:41 AM, Daniel Vetter wrote:
> On Thu, Aug 13, 2015 at 01:31:34PM -0700, Jesse Barnes wrote:
>> These simple tests should always pass.
>>
>> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
> 
> Imo shouldn't be part of the basic set, they thrash the machine quite
> badly. Especially gem_tiled_pread_pwrite thrashes all of memory, so nack
> on that one from me. At least until we've implemented the speedup with
> memlock that's been in JIRA since years ...
> 
> For gem_tiled_pread, why not just rename to gem_tiled_pread_basic?

Yeah sounds good.  Fixed.

Jesse

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 13/18] tests/kms_vblank: mark accuracy test as basic

[Jesse Barnes]

On 08/14/2015 05:44 AM, Daniel Vetter wrote:
> On Thu, Aug 13, 2015 at 01:31:36PM -0700, Jesse Barnes wrote:
>> Need some simple vblank coverage in the BAT list.
>>
>> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
> 
> This testcase relies upon fbcon to have enabled pipe 0, which means it
> spuriously skips (when fbcon is disabled or when it's blanked the screen)
> until that's fixed. Same problem happens with drm_read.
> 
> Until that's addressed imo Nack for basic igt testcase list. I'll do a
> jira about this issue.

Ok dropped.

Jesse

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 15/18] tests/pm_rpm: mark RTE and D3 tests as basic

[Jesse Barnes]

On 08/14/2015 05:50 AM, Daniel Vetter wrote:
> On Thu, Aug 13, 2015 at 01:31:38PM -0700, Jesse Barnes wrote:
>> These always need to pass for basic PM functionality.
>>
>> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
>> ---
>>  tests/pm_rpm.c | 4 ++--
>>  1 file changed, 2 insertions(+), 2 deletions(-)
>>
>> diff --git a/tests/pm_rpm.c b/tests/pm_rpm.c
>> index d509fa8..7ae5806 100644
>> --- a/tests/pm_rpm.c
>> +++ b/tests/pm_rpm.c
>> @@ -1832,11 +1832,11 @@ int main(int argc, char *argv[])
>>  			stay_subtest();
>>  
>>  	/* Essential things */
>> -	igt_subtest("rte")
>> +	igt_subtest("basic-rte")
> 
> I thought our BAT criteria would be basic|rte?
> 
> rte is runtime environment check and more along the lines of "did QA set
> up their machine correctly", so imo good to keep separate.

I was hoping to keep the BAT in the 'basic' namespace only, so rte tests would be included as well, but we'd need to prefix them with basic- too.  Sounds like Paulo is ok with that, so I'll keep it here.  Makes test running that much simpler. :)

Jesse

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 06/18] tests/drv_suspend: mark sysfs tests as basic

[Daniel Vetter]

On Fri, Aug 14, 2015 at 08:29:40AM -0700, Jesse Barnes wrote:
> On 08/14/2015 05:29 AM, Daniel Vetter wrote:
> > On Thu, Aug 13, 2015 at 01:31:29PM -0700, Jesse Barnes wrote:
> >> debugfs may not be mounted, but sysfs should always be restored after
> >> suspend or hibernate.
> >>
> >> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
> > 
> > We already have a suspend/resume testcase in kms_pipc_crc_basic. Do we
> > have enough budget for this one?
> 
> Yeah I thought about getting rid of the suspend/resume ones in pipe_crc
> (marking them as non-basic), since I really just want the one set of
> tests.  Any preference?

Imo crc is such a basic (hah!) validation tool that we really should try
to run all the corner cases. Otherwise we have ugly situations where some
tests randomyl fail, depending upon when they've been run after system
suspend/resume or not. And with all our troubles we need to reboot a few
times for a full run, so this is likely.

This actually happened (that's why we have these tests) and resulted in
lots and lots of wtf until resolved. So yeah I think for crc we really
want them all, even if it's a bit expensive, since if we don't there's a
good chance for lots of flaky test results. And that's the prime problem
we have with igt.
-Daniel
-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 07/18] tests/gem_ctx_exec: mark lrc lite restore as basic

[Daniel Vetter]

On Fri, Aug 14, 2015 at 08:31:01AM -0700, Jesse Barnes wrote:
> On 08/14/2015 05:32 AM, Daniel Vetter wrote:
> > On Thu, Aug 13, 2015 at 01:31:30PM -0700, Jesse Barnes wrote:
> >> Need some LRC tests in the 'basic' subset, and this is a good simple
> >> one.
> >>
> >> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
> > 
> > This is just a testcase for a very specific lrc corner case. We do already
> > exercise lrc with all the other execbuf testcases. Imo we're covered
> > enough already with what we have in the basic testset - testing for all 3
> > billion cornercases will make it grow out of scope I fear. I'd just drop
> > this one here as not needed for BAT.
> > 
> > If you want to extend execbuffer scope a bit then we should add a
> > concurrency test, i.e. one of the gem_concurrent_blt testcases as basic
> > ones. Unfortunately to be able to reliable trigger race conditions those
> > all take a few seconds. But inter-batch sync is a _big_ gap across all
> > archs, and something which is even more tricky with lrc (and scheduler).
> > Imo that would be a lot more useful than this test here.
> 
> Yeah that's a good point; I just saw 'lrc' and though "I want that", but
> you're right we should already be covered.  Definitely open to adding
> some concurrency stuff (maybe just a few seconds worth) as we get things
> in place.

For concurrency mabye we can do that in a 2nd round - we need to go over
gem_concurrent_* anyway and reshuffle that one into basic, full and
subtests that we're only going to expose through cmdline options for
manual testing. Atm there's way too many of those tests.
-Daniel
-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 08/18] tests/gem_mmap: mark basic object creation tests as basic

[Daniel Vetter]

On Fri, Aug 14, 2015 at 08:31:53AM -0700, Jesse Barnes wrote:
> On 08/14/2015 05:33 AM, Daniel Vetter wrote:
> > On Thu, Aug 13, 2015 at 01:31:31PM -0700, Jesse Barnes wrote:
> >> We should be able to create small and moderate sized objects quickly and
> >> without errors.
> >>
> >> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
> > 
> > They're all super-fast basic testcases for api cornercases. I'd vote to
> > rename the entire testcase to gem_mmap_basic.
> 
> Not quite, the huge ones were pretty slow in my measurements...  Not as
> bad as suspend/resume, but not super fast either.

Right I missed those two at the bottom. But everything above small-bo
looks very basic, so perhaps just include them all with a basic-* prefix?
And keep the first one as basic-new-object?
-Daniel
-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 15/18] tests/pm_rpm: mark RTE and D3 tests as basic

[Daniel Vetter]

On Fri, Aug 14, 2015 at 08:48:42AM -0700, Jesse Barnes wrote:
> On 08/14/2015 05:50 AM, Daniel Vetter wrote:
> > On Thu, Aug 13, 2015 at 01:31:38PM -0700, Jesse Barnes wrote:
> >> These always need to pass for basic PM functionality.
> >>
> >> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
> >> ---
> >>  tests/pm_rpm.c | 4 ++--
> >>  1 file changed, 2 insertions(+), 2 deletions(-)
> >>
> >> diff --git a/tests/pm_rpm.c b/tests/pm_rpm.c
> >> index d509fa8..7ae5806 100644
> >> --- a/tests/pm_rpm.c
> >> +++ b/tests/pm_rpm.c
> >> @@ -1832,11 +1832,11 @@ int main(int argc, char *argv[])
> >>  			stay_subtest();
> >>  
> >>  	/* Essential things */
> >> -	igt_subtest("rte")
> >> +	igt_subtest("basic-rte")
> > 
> > I thought our BAT criteria would be basic|rte?
> > 
> > rte is runtime environment check and more along the lines of "did QA set
> > up their machine correctly", so imo good to keep separate.
> 
> I was hoping to keep the BAT in the 'basic' namespace only, so rte tests
> would be included as well, but we'd need to prefix them with basic- too.
> Sounds like Paulo is ok with that, so I'll keep it here.  Makes test
> running that much simpler. :)

Yeah makes sense, ok with me if we go with just "basic".
-Daniel
-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 16/18] tests/kms_flip: add basic tests for flip, flip vs dpms, and flip modeset

[Jesse Barnes]

On 08/14/2015 05:56 AM, Daniel Vetter wrote:
> On Thu, Aug 13, 2015 at 01:31:39PM -0700, Jesse Barnes wrote:
>> Simple variants that don't do multiple output or interruptible testing.
>>
>> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
>> ---
>>  tests/kms_flip.c | 11 ++++++++++-
>>  1 file changed, 10 insertions(+), 1 deletion(-)
>>
>> diff --git a/tests/kms_flip.c b/tests/kms_flip.c
>> index a595d9f..a0e4112 100644
>> --- a/tests/kms_flip.c
>> +++ b/tests/kms_flip.c
>> @@ -80,6 +80,7 @@
>>  #define TEST_TS_CONT		(1 << 27)
>>  #define TEST_BO_TOOBIG		(1 << 28)
>>  #define TEST_HANG_ONCE		(1 << 29)
>> +#define TEST_BASIC		(1 << 30)
>>  
>>  #define EVENT_FLIP		(1 << 0)
>>  #define EVENT_VBLANK		(1 << 1)
>> @@ -1649,7 +1650,7 @@ int main(int argc, char **argv)
>>  					"blt-wf_vblank-vs-modeset" },
>>  		{ 60,  TEST_VBLANK | TEST_MODESET | TEST_WITH_DUMMY_RCS,
>>  					"rcs-wf_vblank-vs-modeset" },
>> -
>> +		{ 30, TEST_FLIP | TEST_BASIC, "basic-plain-flip" },
> 
> I think for better coverage we should pick the timestamp checking ones
> here. Also we need one to cover vblank since kms_vblank can't be used. And
> we can do that all in one testcases to save testing time!
> 
>>  		{ 30, TEST_FLIP , "plain-flip" },
>>  		{ 30, TEST_FLIP | TEST_EBUSY , "busy-flip" },
>>  		{ 30, TEST_FLIP | TEST_FENCE_STRESS , "flip-vs-fences" },
>> @@ -1657,12 +1658,14 @@ int main(int argc, char **argv)
>>  		{ 30, TEST_FLIP | TEST_CHECK_TS | TEST_FB_RECREATE,
>>  			"plain-flip-fb-recreate" },
>>  		{ 30, TEST_FLIP | TEST_RMFB | TEST_MODESET , "flip-vs-rmfb" },
>> +		{ 60, TEST_FLIP | TEST_DPMS | TEST_EINVAL | TEST_BASIC, "basic-flip-vs-dpms" },
>>  		{ 60, TEST_FLIP | TEST_DPMS | TEST_EINVAL, "flip-vs-dpms" },
>>  		{ 60, TEST_FLIP | TEST_DPMS | TEST_WITH_DUMMY_BCS, "blt-flip-vs-dpms" },
>>  		{ 60, TEST_FLIP | TEST_DPMS | TEST_WITH_DUMMY_RCS, "render-flip-vs-dpms" },
>>  		{ 30,  TEST_FLIP | TEST_PAN, "flip-vs-panning" },
>>  		{ 60, TEST_FLIP | TEST_PAN | TEST_WITH_DUMMY_BCS, "blt-flip-vs-panning" },
>>  		{ 60, TEST_FLIP | TEST_PAN | TEST_WITH_DUMMY_RCS, "render-flip-vs-panning" },
>> +		{ 60, TEST_FLIP | TEST_MODESET | TEST_EINVAL | TEST_BASIC, "basic-flip-vs-modeset" },
>>  		{ 60, TEST_FLIP | TEST_MODESET | TEST_EINVAL, "flip-vs-modeset" },
>>  		{ 60, TEST_FLIP | TEST_MODESET | TEST_WITH_DUMMY_BCS, "blt-flip-vs-modeset" },
>>  		{ 60, TEST_FLIP | TEST_MODESET | TEST_WITH_DUMMY_RCS, "render-flip-vs-modeset" },
>> @@ -1721,6 +1724,9 @@ int main(int argc, char **argv)
>>  		igt_subtest(tests[i].name)
>>  			run_test(tests[i].duration, tests[i].flags);
>>  
>> +		if (tests[i].flags & TEST_BASIC)
>> +			continue;
> 
> This means you remove the tests you marked as BASIC from the 2x and
> interruptible sets. What about instead adding the basic prefix at runtime,
> i.e. for all my requests:
> diff --git a/tests/kms_flip.c b/tests/kms_flip.c
> index a595d9f1d69f..aaf03b07df87 100644
> --- a/tests/kms_flip.c
> +++ b/tests/kms_flip.c
> @@ -80,6 +80,7 @@
>  #define TEST_TS_CONT		(1 << 27)
>  #define TEST_BO_TOOBIG		(1 << 28)
>  #define TEST_HANG_ONCE		(1 << 29)
> +#define TEST_BASIC		(1 << 31)
>  
>  #define EVENT_FLIP		(1 << 0)
>  #define EVENT_VBLANK		(1 << 1)
> @@ -1650,20 +1651,20 @@ int main(int argc, char **argv)
>  		{ 60,  TEST_VBLANK | TEST_MODESET | TEST_WITH_DUMMY_RCS,
>  					"rcs-wf_vblank-vs-modeset" },
>  
> -		{ 30, TEST_FLIP , "plain-flip" },
> +		{ 30, TEST_FLIP, "plain-flip" },
>  		{ 30, TEST_FLIP | TEST_EBUSY , "busy-flip" },
>  		{ 30, TEST_FLIP | TEST_FENCE_STRESS , "flip-vs-fences" },
>  		{ 30, TEST_FLIP | TEST_CHECK_TS, "plain-flip-ts-check" },
>  		{ 30, TEST_FLIP | TEST_CHECK_TS | TEST_FB_RECREATE,
>  			"plain-flip-fb-recreate" },
>  		{ 30, TEST_FLIP | TEST_RMFB | TEST_MODESET , "flip-vs-rmfb" },
> -		{ 60, TEST_FLIP | TEST_DPMS | TEST_EINVAL, "flip-vs-dpms" },
> +		{ 60, TEST_FLIP | TEST_DPMS | TEST_EINVAL | TEST_BASIC, "flip-vs-dpms" },
>  		{ 60, TEST_FLIP | TEST_DPMS | TEST_WITH_DUMMY_BCS, "blt-flip-vs-dpms" },
>  		{ 60, TEST_FLIP | TEST_DPMS | TEST_WITH_DUMMY_RCS, "render-flip-vs-dpms" },
>  		{ 30,  TEST_FLIP | TEST_PAN, "flip-vs-panning" },
>  		{ 60, TEST_FLIP | TEST_PAN | TEST_WITH_DUMMY_BCS, "blt-flip-vs-panning" },
>  		{ 60, TEST_FLIP | TEST_PAN | TEST_WITH_DUMMY_RCS, "render-flip-vs-panning" },
> -		{ 60, TEST_FLIP | TEST_MODESET | TEST_EINVAL, "flip-vs-modeset" },
> +		{ 60, TEST_FLIP | TEST_MODESET | TEST_EINVAL | TEST_BASIC, "flip-vs-modeset" },
>  		{ 60, TEST_FLIP | TEST_MODESET | TEST_WITH_DUMMY_BCS, "blt-flip-vs-modeset" },
>  		{ 60, TEST_FLIP | TEST_MODESET | TEST_WITH_DUMMY_RCS, "render-flip-vs-modeset" },
>  		{ 30,  TEST_FLIP | TEST_VBLANK_EXPIRED_SEQ,
> @@ -1671,7 +1672,7 @@ int main(int argc, char **argv)
>  
>  		{ 30, TEST_FLIP | TEST_VBLANK | TEST_VBLANK_ABSOLUTE |
>  		      TEST_CHECK_TS, "flip-vs-absolute-wf_vblank" },
> -		{ 30, TEST_FLIP | TEST_VBLANK | TEST_CHECK_TS,
> +		{ 30, TEST_FLIP | TEST_VBLANK | TEST_CHECK_TS | TEST_BASIC,
>  					"flip-vs-wf_vblank" },
>  		{ 30, TEST_FLIP | TEST_VBLANK | TEST_VBLANK_BLOCK |
>  			TEST_CHECK_TS, "flip-vs-blocking-wf-vblank" },
> @@ -1718,7 +1719,9 @@ int main(int argc, char **argv)
>  		test_nonblocking_read(drm_fd);
>  
>  	for (i = 0; i < sizeof(tests) / sizeof (tests[0]); i++) {
> -		igt_subtest(tests[i].name)
> +		igt_subtest_f("%s%s",
> +			      tests[i].flags & TEST_BASIC ? "basic-" : "",
> +			      tests[i].name)
>  			run_test(tests[i].duration, tests[i].flags);
>  
>  		if (tests[i].flags & TEST_NO_2X_OUTPUT)
> 
> 
> Totally untested ;-)

Yeah looks almost right.  Fixed.

Jesse

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 18/18] gitignore: ignore more files

[Daniel Vetter]

On Fri, Aug 14, 2015 at 08:20:22AM -0700, Jesse Barnes wrote:
> git clean updates the .gitignore file?  Not having to run git clean is the whole point of this patch...

I looked at this patch first, but later noticed that you have a few
renames where you don't update the .gitignore.
> 
> On 08/14/2015 01:09 AM, Daniel Vetter wrote:
> > On Thu, Aug 13, 2015 at 01:31:41PM -0700, Jesse Barnes wrote:
> > 
> > git clean fixes this all, at least over here git status is clean.
> > -Daniel
> > 
> >> ---
> >>  .gitignore       |  3 +++
> >>  tests/.gitignore | 13 +++++++++++++
> >>  tools/.gitignore |  8 ++++++++
> >>  3 files changed, 24 insertions(+)
> >>
> >> diff --git a/.gitignore b/.gitignore
> >> index a438c1c..533f6f1 100644
> >> --- a/.gitignore
> >> +++ b/.gitignore
> >> @@ -90,3 +90,6 @@ gtk-doc.m4
> >>  
> >>  piglit
> >>  results
> >> +
> >> +*.orig
> >> +version.h
> >> diff --git a/tests/.gitignore b/tests/.gitignore
> >> index 31d35a5..e45cbb7 100644
> >> --- a/tests/.gitignore
> >> +++ b/tests/.gitignore
> >> @@ -99,6 +99,7 @@ gem_set_tiling_vs_blt
> >>  gem_set_tiling_vs_gtt
> >>  gem_set_tiling_vs_pwrite
> >>  gem_storedw_batches_loop
> >> +gem_storedw_loop
> >>  gem_storedw_loop_blt
> >>  gem_storedw_loop_bsd
> >>  gem_storedw_loop_render
> >> @@ -169,3 +170,15 @@ prime_udl
> >>  template
> >>  test-list.txt
> >>  testdisplay
> >> +ddi_compute_wrpll
> >> +gem_vmap_blits
> >> +gem_wait_render_timeout
> >> +igt_fork_helper
> >> +igt_list_only
> >> +igt_no_exit
> >> +igt_no_exit_list_only
> >> +igt_no_subtest
> >> +igt_simulation

But all the igt_* have moved to lib/tests/ so you definitely have a pile
of old build system gunk that needs to be cleaned out first with git
clean.

> >> +pm_pc8
> >> +pm_psr
> >> +
> >> diff --git a/tools/.gitignore b/tools/.gitignore
> >> index 49bd24f..7bf91e3 100644
> >> --- a/tools/.gitignore
> >> +++ b/tools/.gitignore
> >> @@ -37,3 +37,11 @@ intel_vga_write
> >>  intel_watermark
> >>  skl_compute_wrpll
> >>  skl_ddb_allocation
> >> +forcewaked
> >> +intel_dpio_read
> >> +intel_dpio_write
> >> +intel_gpu_dump
> >> +intel_nc_read
> >> +intel_nc_write
> >> +intel_punit_read
> >> +intel_punit_write

Same with those, Jani killed them all iirc.
-Daniel

> >> -- 
> >> 1.9.1
> >>
> >> _______________________________________________
> >> Intel-gfx mailing list
> >> Intel-gfx@lists.freedesktop.org
> >> http://lists.freedesktop.org/mailman/listinfo/intel-gfx
> > 
> 

-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 01/18] tests/gem_storedw_loop: add new store_dword test to unify per-ring ones

[Daniel Vetter]

On Fri, Aug 14, 2015 at 08:21:12AM -0700, Jesse Barnes wrote:
> On 08/14/2015 05:19 AM, Daniel Vetter wrote:
> > On Thu, Aug 13, 2015 at 01:31:24PM -0700, Jesse Barnes wrote:
> >> There was a lot of duplication going on...  Mark as basic while we're at
> >> it as these should never fail.
> >>
> >> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
> >> ---
> >>  tests/Makefile.sources   |   1 +
> >>  tests/gem_storedw_loop.c | 181 +++++++++++++++++++++++++++++++++++++++++++++++
> >>  2 files changed, 182 insertions(+)
> >>  create mode 100644 tests/gem_storedw_loop.c
> >>
> >> diff --git a/tests/Makefile.sources b/tests/Makefile.sources
> >> index b9a4cb4..cdcee33 100644
> >> --- a/tests/Makefile.sources
> >> +++ b/tests/Makefile.sources
> >> @@ -138,6 +138,7 @@ TESTS_progs = \
> >>  	gem_seqno_wrap \
> >>  	gem_set_tiling_vs_gtt \
> >>  	gem_set_tiling_vs_pwrite \
> >> +	gem_storedw_loop \
> >>  	gem_storedw_loop_blt \
> >>  	gem_storedw_loop_bsd \
> >>  	gem_storedw_loop_render \
> > 
> > Why not remove the old ones while at it? This just means more gunk in the
> > overall igt set. Also please update .gitignore here for these ...
> 
> Yeah figured that would be a separate patch assuming this one looked ok.
> I added it to the .gitignore in a later patch to update it all at once.

At least in the past when we renamed tests we've done it in one go. That
way QA has an easier time finding the rename in git logs, otherwise
they're just "oh it's gone" and don't carry over existing bug reports.
-Daniel
-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 06/18] tests/drv_suspend: mark sysfs tests as basic

[Jesse Barnes]

On 08/14/2015 09:01 AM, Daniel Vetter wrote:
> On Fri, Aug 14, 2015 at 08:29:40AM -0700, Jesse Barnes wrote:
>> On 08/14/2015 05:29 AM, Daniel Vetter wrote:
>>> On Thu, Aug 13, 2015 at 01:31:29PM -0700, Jesse Barnes wrote:
>>>> debugfs may not be mounted, but sysfs should always be restored after
>>>> suspend or hibernate.
>>>>
>>>> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
>>>
>>> We already have a suspend/resume testcase in kms_pipc_crc_basic. Do we
>>> have enough budget for this one?
>>
>> Yeah I thought about getting rid of the suspend/resume ones in pipe_crc
>> (marking them as non-basic), since I really just want the one set of
>> tests.  Any preference?
> 
> Imo crc is such a basic (hah!) validation tool that we really should try
> to run all the corner cases. Otherwise we have ugly situations where some
> tests randomyl fail, depending upon when they've been run after system
> suspend/resume or not. And with all our troubles we need to reboot a few
> times for a full run, so this is likely.
> 
> This actually happened (that's why we have these tests) and resulted in
> lots and lots of wtf until resolved. So yeah I think for crc we really
> want them all, even if it's a bit expensive, since if we don't there's a
> good chance for lots of flaky test results. And that's the prime problem
> we have with igt.

Ok sounds good, I'll drop the drv_suspend test then.

Jesse

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 18/18] gitignore: ignore more files

[Jesse Barnes]

On 08/14/2015 09:07 AM, Daniel Vetter wrote:
> On Fri, Aug 14, 2015 at 08:20:22AM -0700, Jesse Barnes wrote:
>> git clean updates the .gitignore file?  Not having to run git clean is the whole point of this patch...
> 
> I looked at this patch first, but later noticed that you have a few
> renames where you don't update the .gitignore.
>>
>> On 08/14/2015 01:09 AM, Daniel Vetter wrote:
>>> On Thu, Aug 13, 2015 at 01:31:41PM -0700, Jesse Barnes wrote:
>>>
>>> git clean fixes this all, at least over here git status is clean.
>>> -Daniel
>>>
>>>> ---
>>>>  .gitignore       |  3 +++
>>>>  tests/.gitignore | 13 +++++++++++++
>>>>  tools/.gitignore |  8 ++++++++
>>>>  3 files changed, 24 insertions(+)
>>>>
>>>> diff --git a/.gitignore b/.gitignore
>>>> index a438c1c..533f6f1 100644
>>>> --- a/.gitignore
>>>> +++ b/.gitignore
>>>> @@ -90,3 +90,6 @@ gtk-doc.m4
>>>>  
>>>>  piglit
>>>>  results
>>>> +
>>>> +*.orig
>>>> +version.h
>>>> diff --git a/tests/.gitignore b/tests/.gitignore
>>>> index 31d35a5..e45cbb7 100644
>>>> --- a/tests/.gitignore
>>>> +++ b/tests/.gitignore
>>>> @@ -99,6 +99,7 @@ gem_set_tiling_vs_blt
>>>>  gem_set_tiling_vs_gtt
>>>>  gem_set_tiling_vs_pwrite
>>>>  gem_storedw_batches_loop
>>>> +gem_storedw_loop
>>>>  gem_storedw_loop_blt
>>>>  gem_storedw_loop_bsd
>>>>  gem_storedw_loop_render
>>>> @@ -169,3 +170,15 @@ prime_udl
>>>>  template
>>>>  test-list.txt
>>>>  testdisplay
>>>> +ddi_compute_wrpll
>>>> +gem_vmap_blits
>>>> +gem_wait_render_timeout
>>>> +igt_fork_helper
>>>> +igt_list_only
>>>> +igt_no_exit
>>>> +igt_no_exit_list_only
>>>> +igt_no_subtest
>>>> +igt_simulation
> 
> But all the igt_* have moved to lib/tests/ so you definitely have a pile
> of old build system gunk that needs to be cleaned out first with git
> clean.
> 
>>>> +pm_pc8
>>>> +pm_psr
>>>> +
>>>> diff --git a/tools/.gitignore b/tools/.gitignore
>>>> index 49bd24f..7bf91e3 100644
>>>> --- a/tools/.gitignore
>>>> +++ b/tools/.gitignore
>>>> @@ -37,3 +37,11 @@ intel_vga_write
>>>>  intel_watermark
>>>>  skl_compute_wrpll
>>>>  skl_ddb_allocation
>>>> +forcewaked
>>>> +intel_dpio_read
>>>> +intel_dpio_write
>>>> +intel_gpu_dump
>>>> +intel_nc_read
>>>> +intel_nc_write
>>>> +intel_punit_read
>>>> +intel_punit_write
> 
> Same with those, Jani killed them all iirc.

ah ok, I can drop those.  I'll clean and such and just add the minimal set here.

Jesse

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 18/18] gitignore: ignore more files

[Jani Nikula]

On Fri, 14 Aug 2015, Jesse Barnes <jbarnes@virtuousgeek.org> wrote:
> On 08/14/2015 09:07 AM, Daniel Vetter wrote:
>> On Fri, Aug 14, 2015 at 08:20:22AM -0700, Jesse Barnes wrote:
>>> git clean updates the .gitignore file?  Not having to run git clean is the whole point of this patch...
>> 
>> I looked at this patch first, but later noticed that you have a few
>> renames where you don't update the .gitignore.
>>>
>>> On 08/14/2015 01:09 AM, Daniel Vetter wrote:
>>>> On Thu, Aug 13, 2015 at 01:31:41PM -0700, Jesse Barnes wrote:
>>>>
>>>> git clean fixes this all, at least over here git status is clean.
>>>> -Daniel
>>>>
>>>>> ---
>>>>>  .gitignore       |  3 +++
>>>>>  tests/.gitignore | 13 +++++++++++++
>>>>>  tools/.gitignore |  8 ++++++++
>>>>>  3 files changed, 24 insertions(+)
>>>>>
>>>>> diff --git a/.gitignore b/.gitignore
>>>>> index a438c1c..533f6f1 100644
>>>>> --- a/.gitignore
>>>>> +++ b/.gitignore
>>>>> @@ -90,3 +90,6 @@ gtk-doc.m4
>>>>>  
>>>>>  piglit
>>>>>  results
>>>>> +
>>>>> +*.orig
>>>>> +version.h
>>>>> diff --git a/tests/.gitignore b/tests/.gitignore
>>>>> index 31d35a5..e45cbb7 100644
>>>>> --- a/tests/.gitignore
>>>>> +++ b/tests/.gitignore
>>>>> @@ -99,6 +99,7 @@ gem_set_tiling_vs_blt
>>>>>  gem_set_tiling_vs_gtt
>>>>>  gem_set_tiling_vs_pwrite
>>>>>  gem_storedw_batches_loop
>>>>> +gem_storedw_loop
>>>>>  gem_storedw_loop_blt
>>>>>  gem_storedw_loop_bsd
>>>>>  gem_storedw_loop_render
>>>>> @@ -169,3 +170,15 @@ prime_udl
>>>>>  template
>>>>>  test-list.txt
>>>>>  testdisplay
>>>>> +ddi_compute_wrpll
>>>>> +gem_vmap_blits
>>>>> +gem_wait_render_timeout
>>>>> +igt_fork_helper
>>>>> +igt_list_only
>>>>> +igt_no_exit
>>>>> +igt_no_exit_list_only
>>>>> +igt_no_subtest
>>>>> +igt_simulation
>> 
>> But all the igt_* have moved to lib/tests/ so you definitely have a pile
>> of old build system gunk that needs to be cleaned out first with git
>> clean.
>> 
>>>>> +pm_pc8
>>>>> +pm_psr
>>>>> +
>>>>> diff --git a/tools/.gitignore b/tools/.gitignore
>>>>> index 49bd24f..7bf91e3 100644
>>>>> --- a/tools/.gitignore
>>>>> +++ b/tools/.gitignore
>>>>> @@ -37,3 +37,11 @@ intel_vga_write
>>>>>  intel_watermark
>>>>>  skl_compute_wrpll
>>>>>  skl_ddb_allocation
>>>>> +forcewaked
>>>>> +intel_dpio_read
>>>>> +intel_dpio_write
>>>>> +intel_gpu_dump
>>>>> +intel_nc_read
>>>>> +intel_nc_write
>>>>> +intel_punit_read
>>>>> +intel_punit_write
>> 
>> Same with those, Jani killed them all iirc.
>
> ah ok, I can drop those.  I'll clean and such and just add the minimal
> set here.

Ooops, I had sent the patch to the list, but never pushed. Done now.

BR,
Jani.



>
> Jesse
>
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/intel-gfx

-- 
Jani Nikula, Intel Open Source Technology Center
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx