mirror of
https://github.com/ggerganov/llama.cpp.git
synced 2024-12-25 10:54:36 +00:00
Added a single test function script and fix debug-test.sh to be more robust (#7279)
* run-single-test.sh: added a single test function script and fix debug-test.sh to be more robust * debug-test.sh: combined execute and gdb test mode via -g flag * debug-test.sh: refactor * debug-test: refactor for clarity * debug-test.sh: comment style changes * debug-test.sh: fix gdb
This commit is contained in:
parent
d273c1402b
commit
51e9d02599
@ -1,6 +1,6 @@
|
|||||||
# Debugging Tests Tips
|
# Debugging Tests Tips
|
||||||
|
|
||||||
## How to run & debug a specific test without anything else to keep the feedback loop short?
|
## How to run & execute or debug a specific test without anything else to keep the feedback loop short?
|
||||||
|
|
||||||
There is a script called debug-test.sh in the scripts folder whose parameter takes a REGEX and an optional test number.
|
There is a script called debug-test.sh in the scripts folder whose parameter takes a REGEX and an optional test number.
|
||||||
|
|
||||||
@ -10,13 +10,27 @@ For example, running the following command will output an interactive list from
|
|||||||
|
|
||||||
It will then build & run in the debugger for you.
|
It will then build & run in the debugger for you.
|
||||||
|
|
||||||
|
To just execute a test and get back a PASS or FAIL message run:
|
||||||
|
|
||||||
```bash
|
```bash
|
||||||
./scripts/debug-test.sh test-tokenizer
|
./scripts/debug-test.sh test-tokenizer
|
||||||
|
```
|
||||||
|
|
||||||
|
To test in GDB use the `-g` flag to enable gdb test mode.
|
||||||
|
|
||||||
|
```bash
|
||||||
|
./scripts/debug-test.sh -g test-tokenizer
|
||||||
|
|
||||||
# Once in the debugger, i.e. at the chevrons prompt, setting a breakpoint could be as follows:
|
# Once in the debugger, i.e. at the chevrons prompt, setting a breakpoint could be as follows:
|
||||||
>>> b main
|
>>> b main
|
||||||
```
|
```
|
||||||
|
|
||||||
|
To speed up the testing loop, if you know your test number you can just run it similar to below:
|
||||||
|
|
||||||
|
```bash
|
||||||
|
./scripts/debug-test.sh test 23
|
||||||
|
```
|
||||||
|
|
||||||
For further reference use `debug-test.sh -h` to print help.
|
For further reference use `debug-test.sh -h` to print help.
|
||||||
|
|
||||||
|
|
||||||
@ -41,7 +55,7 @@ cmake -DCMAKE_BUILD_TYPE=Debug -DLLAMA_CUDA=1 -DLLAMA_FATAL_WARNINGS=ON ..
|
|||||||
make -j
|
make -j
|
||||||
```
|
```
|
||||||
|
|
||||||
#### Step 3.1: Identify Test Command for Debugging
|
#### Step 3: Find all tests available that matches REGEX
|
||||||
|
|
||||||
The output of this command will give you the command & arguments needed to run GDB.
|
The output of this command will give you the command & arguments needed to run GDB.
|
||||||
|
|
||||||
@ -69,11 +83,13 @@ Labels: main
|
|||||||
...
|
...
|
||||||
```
|
```
|
||||||
|
|
||||||
So for test #1 we can tell these two pieces of relevant information:
|
#### Step 4: Identify Test Command for Debugging
|
||||||
|
|
||||||
|
So for test #1 above we can tell these two pieces of relevant information:
|
||||||
* Test Binary: `~/llama.cpp/build-ci-debug/bin/test-tokenizer-0`
|
* Test Binary: `~/llama.cpp/build-ci-debug/bin/test-tokenizer-0`
|
||||||
* Test GGUF Model: `~/llama.cpp/tests/../models/ggml-vocab-llama-spm.gguf`
|
* Test GGUF Model: `~/llama.cpp/tests/../models/ggml-vocab-llama-spm.gguf`
|
||||||
|
|
||||||
#### Step 3.2: Run GDB on test command
|
#### Step 5: Run GDB on test command
|
||||||
|
|
||||||
Based on the ctest 'test command' report above we can then run a gdb session via this command below:
|
Based on the ctest 'test command' report above we can then run a gdb session via this command below:
|
||||||
|
|
||||||
|
@ -1,117 +1,203 @@
|
|||||||
#!/bin/bash
|
#!/bin/bash
|
||||||
test_suite=${1:-}
|
|
||||||
test_number=${2:-}
|
|
||||||
|
|
||||||
PROG=${0##*/}
|
PROG=${0##*/}
|
||||||
build_dir="build-ci-debug"
|
build_dir="build-ci-debug"
|
||||||
|
|
||||||
if [ x"$1" = x"-h" ] || [ x"$1" = x"--help" ]; then
|
# Print Color Commands
|
||||||
echo "Usage: $PROG [OPTION]... <test_regex> (test_number)"
|
red=$(tput setaf 1)
|
||||||
echo "Debug specific ctest program."
|
green=$(tput setaf 2)
|
||||||
echo
|
yellow=$(tput setaf 3)
|
||||||
echo "Options:"
|
blue=$(tput setaf 4)
|
||||||
echo " -h, --help Display this help and exit"
|
magenta=$(tput setaf 5)
|
||||||
echo
|
cyan=$(tput setaf 6)
|
||||||
echo "Arguments:"
|
normal=$(tput sgr0)
|
||||||
echo " <test_regex> (Mandatory) Supply one regex to the script to filter tests"
|
|
||||||
echo " (test_number) (Optional) Test number to run a specific test"
|
|
||||||
echo
|
|
||||||
echo "Example:"
|
|
||||||
echo " $PROG test-tokenizer"
|
|
||||||
echo " $PROG test-tokenizer 3"
|
|
||||||
echo
|
|
||||||
exit 0
|
|
||||||
fi
|
|
||||||
|
|
||||||
# Function to select and debug a test
|
|
||||||
function select_test() {
|
|
||||||
test_suite=${1:-test}
|
|
||||||
test_number=${2:-}
|
|
||||||
|
|
||||||
# Sanity Check If Tests Is Detected
|
# Print Help Message
|
||||||
printf "\n\nGathering tests that fit REGEX: ${test_suite} ...\n"
|
####################
|
||||||
tests=($(ctest -R ${test_suite} -V -N | grep -E " +Test +#[0-9]+*" | cut -d':' -f2 | awk '{$1=$1};1'))
|
|
||||||
if [ ${#tests[@]} -eq 0 ]
|
|
||||||
then
|
|
||||||
echo "No tests avaliable... check your compliation process..."
|
|
||||||
echo "Exiting."
|
|
||||||
exit 1
|
|
||||||
fi
|
|
||||||
|
|
||||||
if [ -z $test_number ]
|
print_full_help() {
|
||||||
then
|
cat << EOF
|
||||||
# List out avaliable tests
|
Usage: $PROG [OPTION]... <test_regex> (test_number)
|
||||||
printf "Which test would you like to debug?\n"
|
Debug specific ctest program.
|
||||||
id=0
|
|
||||||
for s in "${tests[@]}"
|
|
||||||
do
|
|
||||||
echo "Test# ${id}"
|
|
||||||
echo " $s"
|
|
||||||
((id++))
|
|
||||||
done
|
|
||||||
|
|
||||||
# Prompt user which test they wanted to run
|
Options:
|
||||||
printf "\nRun test#? "
|
-h, --help display this help and exit
|
||||||
read test_number
|
-g run in gdb mode
|
||||||
else
|
|
||||||
printf "\nUser Already Requested #${test_number}"
|
|
||||||
fi
|
|
||||||
|
|
||||||
# Start GDB with the requested test binary and arguments
|
Arguments:
|
||||||
printf "Debugging(GDB) test: ${tests[test_number]}\n"
|
<test_regex> (Mandatory) Supply one regex to the script to filter tests
|
||||||
# Change IFS (Internal Field Separator)
|
(test_number) (Optional) Test number to run a specific test
|
||||||
sIFS=$IFS
|
|
||||||
IFS=$'\n'
|
|
||||||
|
|
||||||
# Get test args
|
Example:
|
||||||
gdb_args=($(ctest -R ${test_suite} -V -N | grep "Test command" | cut -d':' -f3 | awk '{$1=$1};1' ))
|
$PROG test-tokenizer
|
||||||
IFS=$sIFS
|
$PROG test-tokenizer 3
|
||||||
printf "Debug arguments: ${gdb_args[test_number]}\n\n"
|
EOF
|
||||||
|
|
||||||
# Expand paths if needed
|
|
||||||
args=()
|
|
||||||
for x in $(echo ${gdb_args[test_number]} | sed -e 's/"\/\<//' -e 's/\>"//')
|
|
||||||
do
|
|
||||||
args+=($(echo $x | sed -e 's/.*\/..\//..\//'))
|
|
||||||
done
|
|
||||||
|
|
||||||
# Execute debugger
|
|
||||||
echo "gdb args: ${args[@]}"
|
|
||||||
gdb --args ${args[@]}
|
|
||||||
}
|
}
|
||||||
|
|
||||||
|
abort() {
|
||||||
|
echo "Error: $1" >&2
|
||||||
|
cat << EOF >&2
|
||||||
|
Usage: $PROG [OPTION]... <test_regex> (test_number)
|
||||||
|
Debug specific ctest program.
|
||||||
|
Refer to --help for full instructions.
|
||||||
|
EOF
|
||||||
|
exit 1
|
||||||
|
}
|
||||||
|
|
||||||
|
|
||||||
|
# Dependency Sanity Check
|
||||||
|
#########################
|
||||||
|
|
||||||
|
check_dependency() {
|
||||||
|
command -v "$1" >/dev/null 2>&1 || {
|
||||||
|
abort "$1 is required but not found. Please install it and try again."
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
check_dependency ctest
|
||||||
|
check_dependency cmake
|
||||||
|
|
||||||
|
|
||||||
# Step 0: Check the args
|
# Step 0: Check the args
|
||||||
if [ -z "$test_suite" ]
|
########################
|
||||||
then
|
|
||||||
echo "Usage: $PROG [OPTION]... <test_regex> (test_number)"
|
if [ x"$1" = x"-h" ] || [ x"$1" = x"--help" ]; then
|
||||||
echo "Supply one regex to the script to filter tests,"
|
print_full_help >&2
|
||||||
echo "and optionally a test number to run a specific test."
|
exit 0
|
||||||
echo "Use --help flag for full instructions"
|
|
||||||
exit 1
|
|
||||||
fi
|
fi
|
||||||
|
|
||||||
|
# Parse command-line options
|
||||||
|
gdb_mode=false
|
||||||
|
while getopts "g" opt; do
|
||||||
|
case $opt in
|
||||||
|
g)
|
||||||
|
gdb_mode=true
|
||||||
|
echo "gdb_mode Mode Enabled"
|
||||||
|
;;
|
||||||
|
esac
|
||||||
|
done
|
||||||
|
|
||||||
|
# Shift the option parameters
|
||||||
|
shift $((OPTIND - 1))
|
||||||
|
|
||||||
|
# Positionial Argument Processing : <test_regex>
|
||||||
|
if [ -z "${1}" ]; then
|
||||||
|
abort "Test regex is required"
|
||||||
|
else
|
||||||
|
test_suite=${1:-}
|
||||||
|
fi
|
||||||
|
|
||||||
|
# Positionial Argument Processing : (test_number)
|
||||||
|
test_number=${2:-}
|
||||||
|
|
||||||
|
|
||||||
# Step 1: Reset and Setup folder context
|
# Step 1: Reset and Setup folder context
|
||||||
|
########################################
|
||||||
|
|
||||||
## Sanity check that we are actually in a git repo
|
## Sanity check that we are actually in a git repo
|
||||||
repo_root=$(git rev-parse --show-toplevel)
|
repo_root=$(git rev-parse --show-toplevel)
|
||||||
if [ ! -d "$repo_root" ]; then
|
if [ ! -d "$repo_root" ]; then
|
||||||
echo "Error: Not in a Git repository."
|
abort "Not in a Git repository."
|
||||||
exit 1
|
|
||||||
fi
|
fi
|
||||||
|
|
||||||
## Reset folder to root context of git repo
|
## Reset folder to root context of git repo and Create and enter build directory
|
||||||
pushd "$repo_root" || exit 1
|
pushd "$repo_root"
|
||||||
|
rm -rf "$build_dir" && mkdir "$build_dir" || abort "Failed to make $build_dir"
|
||||||
|
|
||||||
## Create and enter build directory
|
|
||||||
rm -rf "$build_dir" && mkdir "$build_dir" || exit 1
|
|
||||||
|
|
||||||
# Step 2: Setup Build Environment and Compile Test Binaries
|
# Step 2: Setup Build Environment and Compile Test Binaries
|
||||||
cmake -B "./$build_dir" -DCMAKE_BUILD_TYPE=Debug -DLLAMA_CUDA=1 -DLLAMA_FATAL_WARNINGS=ON || exit 1
|
###########################################################
|
||||||
pushd "$build_dir" && make -j || exit 1
|
|
||||||
|
|
||||||
# Step 3: Debug the Test
|
# Note: test-eval-callback requires -DLLAMA_CURL
|
||||||
select_test "$test_suite" "$test_number"
|
cmake -B "./$build_dir" -DCMAKE_BUILD_TYPE=Debug -DLLAMA_CUDA=1 -DLLAMA_CURL=1 || abort "Failed to build enviroment"
|
||||||
|
pushd "$build_dir"
|
||||||
|
make -j || abort "Failed to compile"
|
||||||
|
popd > /dev/null || exit 1
|
||||||
|
|
||||||
# Step 4: Return to the directory from which the user ran the command.
|
|
||||||
popd || exit 1
|
# Step 3: Find all tests available that matches REGEX
|
||||||
popd || exit 1
|
####################################################
|
||||||
popd || exit 1
|
|
||||||
|
# Ctest Gather Tests
|
||||||
|
# `-R test-tokenizer` : looks for all the test files named `test-tokenizer*` (R=Regex)
|
||||||
|
# `-N` : "show-only" disables test execution & shows test commands that you can feed to GDB.
|
||||||
|
# `-V` : Verbose Mode
|
||||||
|
printf "\n\nGathering tests that fit REGEX: ${test_suite} ...\n"
|
||||||
|
pushd "$build_dir"
|
||||||
|
tests=($(ctest -R ${test_suite} -V -N | grep -E " +Test +#[0-9]+*" | cut -d':' -f2 | awk '{$1=$1};1'))
|
||||||
|
if [ ${#tests[@]} -eq 0 ]; then
|
||||||
|
abort "No tests avaliable... check your compliation process..."
|
||||||
|
fi
|
||||||
|
popd > /dev/null || exit 1
|
||||||
|
|
||||||
|
|
||||||
|
# Step 4: Identify Test Command for Debugging
|
||||||
|
#############################################
|
||||||
|
|
||||||
|
# Select test number
|
||||||
|
if [ -z $test_number ]; then
|
||||||
|
# List out avaliable tests
|
||||||
|
printf "Which test would you like to debug?\n"
|
||||||
|
id=0
|
||||||
|
for s in "${tests[@]}"
|
||||||
|
do
|
||||||
|
echo "Test# ${id}"
|
||||||
|
echo " $s"
|
||||||
|
((id++))
|
||||||
|
done
|
||||||
|
|
||||||
|
# Prompt user which test they wanted to run
|
||||||
|
printf "\nRun test#? "
|
||||||
|
read test_number
|
||||||
|
|
||||||
|
else
|
||||||
|
printf "\nUser Already Requested #${test_number}\n"
|
||||||
|
|
||||||
|
fi
|
||||||
|
|
||||||
|
# Grab all tests commands
|
||||||
|
pushd "$build_dir"
|
||||||
|
sIFS=$IFS # Save Initial IFS (Internal Field Separator)
|
||||||
|
IFS=$'\n' # Change IFS (Internal Field Separator) (So we split ctest output by newline rather than by spaces)
|
||||||
|
test_args=($(ctest -R ${test_suite} -V -N | grep "Test command" | cut -d':' -f3 | awk '{$1=$1};1' )) # Get test args
|
||||||
|
IFS=$sIFS # Reset IFS (Internal Field Separator)
|
||||||
|
popd > /dev/null || exit 1
|
||||||
|
|
||||||
|
# Grab specific test command
|
||||||
|
single_test_name="${tests[test_number]}"
|
||||||
|
single_test_command="${test_args[test_number]}"
|
||||||
|
|
||||||
|
|
||||||
|
# Step 5: Execute or GDB Debug
|
||||||
|
##############################
|
||||||
|
|
||||||
|
printf "${magenta}Running Test #${test_number}: ${single_test_name}${normal}\n"
|
||||||
|
printf "${cyan}single_test_command: ${single_test_command}${normal}\n"
|
||||||
|
|
||||||
|
if [ "$gdb_mode" = "true" ]; then
|
||||||
|
# Execute debugger
|
||||||
|
pushd "$repo_root" || exit 1
|
||||||
|
eval "gdb --args ${single_test_command}"
|
||||||
|
popd > /dev/null || exit 1
|
||||||
|
|
||||||
|
else
|
||||||
|
# Execute Test
|
||||||
|
pushd "$repo_root" || exit 1
|
||||||
|
eval "${single_test_command}"
|
||||||
|
exit_code=$?
|
||||||
|
popd > /dev/null || exit 1
|
||||||
|
|
||||||
|
# Print Result
|
||||||
|
printf "${blue}Ran Test #${test_number}: ${single_test_name}${normal}\n"
|
||||||
|
printf "${yellow}Command: ${single_test_command}${normal}\n"
|
||||||
|
if [ $exit_code -eq 0 ]; then
|
||||||
|
printf "${green}TEST PASS${normal}\n"
|
||||||
|
else
|
||||||
|
printf "${red}TEST FAIL${normal}\n"
|
||||||
|
fi
|
||||||
|
|
||||||
|
fi
|
||||||
|
|
||||||
|
# Return to the directory from which the user ran the command.
|
||||||
|
popd > /dev/null || exit 1
|
||||||
|
Loading…
Reference in New Issue
Block a user