From 045912cd6abbfd411e29db000b4f945a73d3753a Mon Sep 17 00:00:00 2001
From: Marc Herbert <marc.herbert@intel.com>
Date: Tue, 24 May 2022 16:21:40 -0700
Subject: [PATCH] hijack: add a quick write test to monitor storage perf. after
 each test

We've experienced a fair number of test failures that seem to point to
storage performance issues:

https://github.com/thesofproject/linux/issues/3387
https://github.com/thesofproject/linux/issues/3669

This (temporary?) addition runs a quick write test after each audio test
to monitor storage sanity.

As a bonus feature the "sync" could help us collect more logs.

On our (slowest) BYT devices the test adds 3s per test; much less on
newer devices.

Sample output:

```
2022-05-24 23:20:41 UTC [INFO] pkill -TERM sof-logger
2022-05-24 23:20:42 UTC [INFO] nlines=1132 /home/mherber2/SOF/sof-test/logs/BOGUS-check-playback/2022-05-24-16:20:35-3288/slogger.txt
+ timeout -s CONT 5 sudo sync

real	0m0.062s
user	0m0.005s
sys	0m0.019s
+ timeout -s CONT 10 dd if=/dev/zero of=/home/mherber2/HD_TEST_DELETE_ME bs=1M count=200 conv=fsync
200+0 records in
200+0 records out
209715200 bytes (210 MB, 200 MiB) copied, 2.0893 s, 100 MB/s
+ timeout -s CONT 5 sudo sync

real	0m0.037s
user	0m0.004s
sys	0m0.018s
2022-05-24 23:20:44 UTC [INFO] Test Result: PASS!
```

Signed-off-by: Marc Herbert <marc.herbert@intel.com>
---
 case-lib/hijack.sh |  2 ++
 case-lib/lib.sh    | 19 +++++++++++++++++++
 2 files changed, 21 insertions(+)

diff --git a/case-lib/hijack.sh b/case-lib/hijack.sh
index 65a6e7ff3..6b0f3f1bd 100644
--- a/case-lib/hijack.sh
+++ b/case-lib/hijack.sh
@@ -100,6 +100,8 @@ function func_exit_handler()
 
     fi
 
+    storage_checks || exit_status=1
+
     if [[ "$KERNEL_CHECKPOINT" =~ ^[0-9]{10} ]]; then
         # Do not collect the entire duration of the test but only the
         # last iteration.
diff --git a/case-lib/lib.sh b/case-lib/lib.sh
index c9ea3e355..eda3cf85d 100644
--- a/case-lib/lib.sh
+++ b/case-lib/lib.sh
@@ -75,6 +75,25 @@ poll_wait_for()
     $pass
 }
 
+
+storage_checks()
+{
+    local max_sync_duration=5
+
+    ( set -x
+      # Thanks to CONT this does not actually timeout; it only returns a
+      # non-zero exit status when taking too long.
+      time timeout -s CONT "$max_sync_duration" sudo sync || return $?
+      # Spend a few seconds to test and show the current write speed
+      timeout -s CONT 6 dd if=/dev/zero of=~/HD_TEST_DELETE_ME bs=1M count=200 conv=fsync ||
+          return $?
+      time timeout -s CONT "$max_sync_duration" sudo sync
+    ) || return $?
+
+    rm ~/HD_TEST_DELETE_ME
+}
+
+
 setup_kernel_check_point()
 {
     # Make the check point $SOF_TEST_INTERVAL second(s) earlier to avoid