diff --git a/scrashme/control b/scrashme/control
index c9c420342..39f022729 100644
--- a/scrashme/control
+++ b/scrashme/control
@@ -1 +1,21 @@
+NAME='scrashme'
+AUTHOR='Yi Yang <yang.y.yi@gmail.com>'
+TEST_CATEGORY='Stress'
+TEST_CLASS='Kernel'
+TEST_TYPE='client'
+EXPERIMENTAL='True'
+DEPENDENCIES='GCC'
+TIME='MEDIUM'
+DOC='''\
+Runs the scrashme suite located at: 
+http://www.codemonkey.org.uk/projects/git-snapshots/scrashme/
+o Now tests all syscalls, not just 256
+o Certain syscalls made things get out of hand too quickly.
+  Forkbombing a box with this app kills it very quickly.
+  (An alternative would be to only run this with strict rlimits, but
+  there are a few other syscalls we also take care of..)
+o Do the syscall in a child process so if something hangs, we can continue.
+o Printing of the syscall name/return code.
+o Several extra modes of attack.
+'''
 job.run_test('scrashme')
diff --git a/selftest/control b/selftest/control
index e864260d5..353c6f215 100644
--- a/selftest/control
+++ b/selftest/control
@@ -1,9 +1,11 @@
-AUTHOR = "Autotest Team"
+AUTHOR = "Andy Whitcroft <apw@shadowen.org>"
+EXPERIMENTAL = "True"
 TIME = "MEDIUM"
 NAME = "Self Test"
 TEST_TYPE = "client"
 TEST_CATEGORY = "Functional"
-TEST_CLASS = "Stress"
+TEST_CLASS = "Software"
 DOC = """\
-Stub information needs to be updated!
+Run a selftest on the autotest installed branch
 """
+job.run_test('selftest', 1)
diff --git a/selftest/selftest.py b/selftest/selftest.py
index 825b9cfb1..15a9d7c3b 100644
--- a/selftest/selftest.py
+++ b/selftest/selftest.py
@@ -1,4 +1,4 @@
-import os
+import os, sys
 from autotest_lib.client.bin import test
 from autotest_lib.client.common_lib import error
 
diff --git a/signaltest/control b/signaltest/control
index dbeb3c397..2010962c8 100644
--- a/signaltest/control
+++ b/signaltest/control
@@ -1 +1,12 @@
+NAME='Signal Test'
+EXPERIMENTAL = "True"
+AUTHOR='Michal Piotrowski <michal.k.k.piotrowski@gmail.com>'
+TIME='SHORT'
+TEST_TYPE='client'
+TEST_CLASS='Kernel'
+TEST_CATEGORY='Functional'
+DEPENDENCIES='GCC'
+DOC='''\
+Test signal passing to processes
+'''
 job.run_test('signaltest')
diff --git a/sparse/control b/sparse/control
index 4a9e8120b..174917764 100755
--- a/sparse/control
+++ b/sparse/control
@@ -1,2 +1,24 @@
+NAME='Sparse Kernel Test'
+EXPERIMENTAL='True'
+TIME='MEDIUM'
+TEST_TYPE='client'
+TEST_CATEGORY='Functional'
+TEST_CLASS='Kernel'
+DEPENDENCIES='GCC'
+AUTHOR='Martin Bligh <mbligh@google.com>'
+DOC='''\
+Sparse is a semantic parser of source files: it's neither a compiler
+(although it could be used as a front-end for one) nor is it a
+preprocessor (although it contains as a part of it a preprocessing
+phase).
+
+It is meant to be a small - and simple - library.  Scanty and meager,
+and partly because of that easy to use.  It has one mission in life:
+create a semantic parse tree for some arbitrary user for further
+analysis.  It's not a tokenizer, nor is it some generic context-free
+parser.  In fact, context (semantics) is what it's all about - figuring
+out not just what the grouping of tokens are, but what the _types_ are
+that the grouping implies.
+'''
 job.run_test('sparse', '/usr/local/src/linux-2.6.14.tar.bz2', '/usr/local/src/patch-2.6.14-git6.bz2', 'http://ftp.kernel.org/pub/linux/kernel/people/mbligh/config/config.up')
 
diff --git a/spew/control b/spew/control
index 1cbb49a62..b3f35affc 100644
--- a/spew/control
+++ b/spew/control
@@ -1 +1,17 @@
+NAME='Spew'
+AUTHOR='Martin Bligh <mbligh@google.com>'
+EXPERIMENTAL='True'
+TEST_TYPE='client'
+DEPENDENCIES='GCC'
+TEST_CATEGORY='Benchmark'
+TEST_CLASS='Hardware'
+TIME='MEDIUM'
+DOC='''\
+The spew package is used to test I/O performance and to generate load
+on character devices, block devices, and file systems.  It is similar
+to the lmdd program found in the lmbench test-suite
+(http://www.bitmover.com/lmbench). It is a bit easier to use than lmdd
+and has some added functionality.  For example, it can test both
+random and sequential I/O.
+'''
 job.run_test('spew')
diff --git a/stress/control b/stress/control
index dd4c21391..2fa805ec7 100644
--- a/stress/control
+++ b/stress/control
@@ -1 +1,12 @@
+NAME='Stress'
+AUTHOR='Yi Yang <yang.y.yi@gmail.com>'
+EXPERIMENTAL='True'
+TEST_TYPE='client'
+TIME='MEDIUM'
+TEST_CATEGORY='Functional'
+DEPENDENCIES='GCC'
+TEST_CLASS='Software'
+DOC='''\
+stress is not a benchmark, but is rather a tool designed to put given subsytems under a specified load. Instances in which this is useful include those in which a system administrator wishes to perform tuning activities, a kernel or libc programmer wishes to evaluate denial of service possibilities, etc. 
+'''
 job.run_test('stress')