Got QEMU testing for resume working, took a bunch of changes to test module

changes to test module:

- removed "-no-reboot" parameter, needed for full resume test, and added
  a check for kernel panic messages to stop QEMU from rebooting
  indefinitely

- changed default size of test_rootfs from 16M to 256M to allow 128M of swap for
  testing (left 128M for disk as it's necessary for certain filesystem,
  like btrfs which enforces minimum partition size of 40M)

- `test_resume` parameter (off by default), enables most of the extra
  functionality for resume testing

- wrote a `resume_tests` function for the `test_image` with script to
  very hibernate/reboot returns to the same point

- rearranged functions in the `test_image` module so that the
  `test_flag` is printed in the `init_final` stage in the `complete_tests`
  function, allows tests in `init_main` to be run before passing

- changed shebang for `test_image` module to include "-l" to load
  `/etc/profile` in the test image

- added the `swapon` binary to the `test_image`, so the resume swap file
  can be mounted before hibernation

Author: truffle0

src/ugrd/base/test.py

@@ -163,7 +163,10 @@ def test_image(self):
             if self["test_flag"] in line:
                 self.logger.info("Test flag found in output: %s", c_(line, "green"))
                 break
-            elif line.endswith("exitcode=0x00000000"):
+            elif (
+                line.endswith("exitcode=0x00000000")
+                or "---[ end Kernel panic - not syncing: sysrq triggered crash ]---" in line
+            ):
                 failed = True
                 break
             elif "press space" in line.lower():

src/ugrd/base/test.toml

@@ -1,16 +1,15 @@
 test_rootfs_name = 'ugrd-test-rootfs'
 test_rootfs_build_dir = 'initramfs_test_rootfs'
-test_image_size = 16
+test_image_size = 256
 
 test_copy_config = ["_mounts", "mounts", "out_dir", "tmpdir", "clean", "test_image_size", "test_no_rootfs", "test_flag", "cryptsetup"]
 kmod_init = ["ata_piix", "sd_mod"]
 
 test_memory = '256M'
 test_cpu = 'host'
 test_arch = 'x86_64'
-test_timeout = 15
-test_cmdline = 'console=ttyS0,115200 panic=1'
-#qemu_bool_args = ['nographic', 'no-reboot', 'enable-kvm']
+test_timeout = 30
+test_cmdline = 'console=ttyS0,115200 panic=0'
 qemu_bool_args = ['nographic', 'enable-kvm' ]
 
 [imports.build_pre]

src/ugrd/fs/resume.py

@@ -27,6 +27,7 @@ def resume(self) -> str:
             eerror "Cannot safely resume with mounted block devices:\n$(lsblk -Q MOUNTPOINT -no PATH)"
             return 1
         fi
+
         [ -b "$1" ] || (ewarn "\'$1\' is not a valid block device!" ; return 1)
         einfo "Attempting resume from: $1"
         # TODO: This could maybe be printf?
@@ -66,15 +67,18 @@ def handle_late_resume(self) -> None:
     self.logger.warning(
             "[late_resume] enabled, this can result in data loss if filesystems are modified before resuming. Read the docs for more info."
     )
-    return handle_early_resume(self)  # At the moment it's the same code but delayed, will change when more features are added
+
+    # At the moment it's the same code but delayed, will change when more features are added
+    return handle_early_resume(self)
+
 
 @contains("test_resume")
 def test_init_swap_uuid(self):
     if "test_cpu" in self:
         from uuid import uuid4
 
-        self["test_swap_uuid"] = swap_uuid = uuid4()
+        if not self["test_swap_uuid"]:
+            self["test_swap_uuid"] = swap_uuid = uuid4()
 
         # append to test kernel cmdline and adjust planned image size to allow enough space
         self["test_cmdline"] = f"{self.get('test_cmdline')} resume=UUID={swap_uuid}"
-        self["test_image_size"] = 256 + self.get("test_image_size")

src/ugrd/fs/resume.toml

@@ -19,8 +19,5 @@ handle_late_resume = ["crypt_init", "init_lvm"]
 
 [custom_parameters]
 late_resume = "bool"
-test_resume = "bool"
-test_swap_uuid = "str"
-
-[import_order.before]
-handle_resume = "mount_fstab"
+test_resume = "bool" # Test hibernation/resume pathways when test is enabled
+test_swap_uuid = "str" # Define the uuid of the swap partition on the test image

src/ugrd/fs/test_image.py

@@ -9,37 +9,43 @@
 
 MIN_FS_SIZES = {"btrfs": 110, "f2fs": 50}
 
-@contains("test_flag", "A test flag must be set to create a test image", raise_exception=True)
+
 def init_banner(self):
     """Initialize the test image banner, set a random flag if not set."""
-    self["banner"] = f"echo {self['test_flag']}"
+    self["banner"] = "echo ugRD Test Image"
 
 
 @contains("test_resume")
 def resume_tests(self):
-    return [
-        'if [ "$(</sys/power/resume)" != "0:0" ] ; then',
-        '   [ -e "/resumed" ] && (rm /resumed ; echo c > /proc/sysrq-trigger)',
-        # Set correct resume parameters
-        "   echo reboot > /sys/power/disk",
-        # trigger resume
-        "   echo disk > /sys/power/state",
-        '   [ -e "/resume" ] || echo c > /proc/sysrq-trigger',
-        # if we reach this point, resume was successful
-        # reset environment in case resume needs to be rerun
-        "   rm /resumed",
-        '   echo "Resume completed without error.',
-        "else",
-        '   echo "No resume device found! Resume test not possible!',
-        "fi",
-    ]
+    return """
+        echo "Begin resume testing."
+        if [ "$(</sys/power/resume)" != "0:0" ] ; then
+            echo reboot > /sys/power/disk
+            echo 1 > /sys/power/pm_debug_messages
+
+            local SWAPDEV=/dev/$(source "/sys/dev/block/$(</sys/power/resume)/uevent" && echo $DEVNAME)
+            echo "Activating swap device ${SWAPDEV}..."
+            swapon ${SWAPDEV}
+
+            echo "Triggering test hibernation..."
+            echo disk > /sys/power/state || (echo "Suspend to disk failed!" ; echo c > /proc/sysrq-trigger)
 
+            # Assume at this point system has hibernated then resumed again
+            echo "Resume test completed without error."
+        else
+            echo "No resume device found! Resume test not possible!"
+            echo c > /proc/sysrq-trigger
+        fi
+    """
 
+
+@contains("test_flag", "A test flag must be set to create a test image", raise_exception=True)
 def complete_tests(self):
-    return [
-        "echo s > /proc/sysrq-trigger",
-        "echo o > /proc/sysrq-trigger",
-    ]
+    return f"""
+        echo {self["test_flag"]}
+        echo s > /proc/sysrq-trigger
+        echo o > /proc/sysrq-trigger
+    """
 
 
 def _allocate_image(self, image_path, padding=0):
@@ -65,6 +71,7 @@ def _allocate_image(self, image_path, padding=0):
         self.logger.debug(f"[{f.name}] Total bytes: { c_(total_size, 'green')}")
         f.write(b"\0" * total_size)
 
+
 def _copy_fs_contents(self, image_path, build_dir):
     """ Mount and copy the filesystem contents into the image,
     for filesystems which cannot be created directly from a directory"""
@@ -151,9 +158,7 @@ def make_test_image(self):
     loopback = None
     if self.get("test_resume"):
         try:
-            self._run(["sgdisk", "-og", image_path])
-            self._run(["sgdisk", "-n", "1:0:+256", image_path])
-            self._run(["sgdisk", "-n", "2:0", image_path])
+            self._run(["sgdisk", "-og", "-n", "1:0:+128M", "-n", "2:0:0", image_path])
         except RuntimeError as e:
             raise RuntimeError("Failed to partition test disk: %s", e)
 

src/ugrd/fs/test_image.toml

@@ -1,8 +1,9 @@
 modules = [ 'ugrd.base.banner' ]  # get_archive_path
 paths = ['dev', 'sys', 'proc']
+binaries = ['swapon']
 
-test_image_size = 16
-shebang = "#!/bin/sh"  # Don't use -l, no profile is generated
+test_image_size = 256
+shebang = "#!/bin/sh -l"
 _cryptsetup_root = "root"
 
 
@@ -27,3 +28,4 @@ _cryptsetup_root = "str"  # Define the root device for cryptsetup
 test_image_size = "int"  # Define the size of the test image in MiB
 test_flag = "str"  # Define the success flag used to determine if the test was successful
 test_resume = "bool" # Enable code to test the suspend/resume pathways
+test_swap_uuid = "str" # Define the UUID of the swap partition