Skip to content

Commit

Permalink
Deprecation warning to assist with migration to new binary names (#8283)
Browse files Browse the repository at this point in the history
* Adding a simple program to provide a deprecation warning that can exist to help people notice the binary name change from #7809 and migrate to the new filenames.

* Build legacy replacement binaries only if they already exist. Check for their existence every time so that they are not ignored.
  • Loading branch information
HanClinto authored Jul 9, 2024
1 parent a03e8dd commit e500d61
Show file tree
Hide file tree
Showing 3 changed files with 160 additions and 4 deletions.
78 changes: 74 additions & 4 deletions Makefile
Original file line number Diff line number Diff line change
Expand Up @@ -64,10 +64,14 @@ TEST_TARGETS = \
tests/test-tokenizer-1-spm

# Legacy build targets that were renamed in #7809, but should still be removed when the project is cleaned
LEGACY_TARGETS = main quantize quantize-stats perplexity imatrix embedding vdot q8dot train-text-from-scratch convert-llama2c-to-ggml \
LEGACY_TARGETS_CLEAN = main quantize quantize-stats perplexity imatrix embedding vdot q8dot train-text-from-scratch convert-llama2c-to-ggml \
simple batched batched-bench save-load-state server gguf gguf-split eval-callback llama-bench libllava.a llava-cli baby-llama \
retrieval speculative infill tokenize benchmark-matmult parallel finetune export-lora lookahead lookup passkey gritlm

# Legacy build targets that were renamed in #7809, but we want to build binaries that for them that output a deprecation warning if people try to use them.
# We don't want to clutter things too much, so we only build replacements for the most commonly used binaries.
LEGACY_TARGETS_BUILD = main quantize perplexity embedding server finetune

# Deprecation aliases
ifdef LLAMA_CUBLAS
$(error LLAMA_CUBLAS is removed. Use GGML_CUDA instead.)
Expand Down Expand Up @@ -193,7 +197,7 @@ ifdef GGML_RPC
BUILD_TARGETS += rpc-server
endif

default: $(BUILD_TARGETS)
default: $(BUILD_TARGETS) $(LEGACY_TARGETS_BUILD)

test: $(TEST_TARGETS)
@failures=0; \
Expand Down Expand Up @@ -228,7 +232,7 @@ test: $(TEST_TARGETS)
fi
@echo 'All tests passed.'

all: $(BUILD_TARGETS) $(TEST_TARGETS)
all: $(BUILD_TARGETS) $(TEST_TARGETS) $(LEGACY_TARGETS_BUILD)

ifdef RISCV_CROSS_COMPILE
CC := riscv64-unknown-linux-gnu-gcc
Expand Down Expand Up @@ -1098,7 +1102,7 @@ clean:
rm -vrf ggml/src/ggml-cuda/template-instances/*.o
rm -rvf $(BUILD_TARGETS)
rm -rvf $(TEST_TARGETS)
rm -rvf $(LEGACY_TARGETS)
rm -rvf $(LEGACY_TARGETS_CLEAN)
find examples pocs -type f -name "*.o" -delete

#
Expand Down Expand Up @@ -1494,3 +1498,69 @@ llama-q8dot: pocs/vdot/q8dot.cpp ggml/src/ggml.o \
$(OBJ_GGML)
$(CXX) $(CXXFLAGS) -c $< -o $(call GET_OBJ_FILE, $<)
$(CXX) $(CXXFLAGS) $(filter-out $<,$^) $(call GET_OBJ_FILE, $<) -o $@ $(LDFLAGS)

#
# Deprecated binaries that we want to keep around long enough for people to migrate to the new filenames, then these can be removed.
#
# Mark legacy binary targets as .PHONY so that they are always checked.
.PHONY: main quantize perplexity embedding server finetune

main: examples/deprecation-warning/deprecation-warning.cpp
ifneq (,$(wildcard main))
$(CXX) $(CXXFLAGS) -c $< -o $(call GET_OBJ_FILE, $<)
$(CXX) $(CXXFLAGS) $(filter-out $<,$^) $(call GET_OBJ_FILE, $<) -o $@ $(LDFLAGS)
@echo "#########"
@echo "WARNING: The 'main' binary is deprecated. Please use 'llama-cli' instead."
@echo " Remove the 'main' binary to remove this warning."
@echo "#########"
endif

quantize: examples/deprecation-warning/deprecation-warning.cpp
ifneq (,$(wildcard quantize))
$(CXX) $(CXXFLAGS) -c $< -o $(call GET_OBJ_FILE, $<)
$(CXX) $(CXXFLAGS) $(filter-out %.h $<,$^) $(call GET_OBJ_FILE, $<) -o $@ $(LDFLAGS)
@echo "#########"
@echo "WARNING: The 'quantize' binary is deprecated. Please use 'llama-quantize' instead."
@echo " Remove the 'quantize' binary to remove this warning."
@echo "#########"
endif

perplexity: examples/deprecation-warning/deprecation-warning.cpp
ifneq (,$(wildcard perplexity))
$(CXX) $(CXXFLAGS) -c $< -o $(call GET_OBJ_FILE, $<)
$(CXX) $(CXXFLAGS) $(filter-out %.h $<,$^) $(call GET_OBJ_FILE, $<) -o $@ $(LDFLAGS)
@echo "#########"
@echo "WARNING: The 'perplexity' binary is deprecated. Please use 'llama-perplexity' instead."
@echo " Remove the 'perplexity' binary to remove this warning."
@echo "#########"
endif

embedding: examples/deprecation-warning/deprecation-warning.cpp
ifneq (,$(wildcard embedding))
$(CXX) $(CXXFLAGS) -c $< -o $(call GET_OBJ_FILE, $<)
$(CXX) $(CXXFLAGS) $(filter-out %.h $<,$^) $(call GET_OBJ_FILE, $<) -o $@ $(LDFLAGS)
@echo "#########"
@echo "WARNING: The 'embedding' binary is deprecated. Please use 'llama-embedding' instead."
@echo " Remove the 'embedding' binary to remove this warning."
@echo "#########"
endif

server: examples/deprecation-warning/deprecation-warning.cpp
ifneq (,$(wildcard server))
$(CXX) $(CXXFLAGS) -c $< -o $(call GET_OBJ_FILE, $<)
$(CXX) $(CXXFLAGS) $(filter-out %.h $<,$^) $(call GET_OBJ_FILE, $<) -o $@ $(LDFLAGS)
@echo "#########"
@echo "WARNING: The 'server' binary is deprecated. Please use 'llama-server' instead."
@echo " Remove the 'server' binary to remove this warning."
@echo "#########"
endif

finetune: examples/deprecation-warning/deprecation-warning.cpp
ifneq (,$(wildcard finetune))
$(CXX) $(CXXFLAGS) -c $< -o $(call GET_OBJ_FILE, $<)
$(CXX) $(CXXFLAGS) $(filter-out %.h $<,$^) $(call GET_OBJ_FILE, $<) -o $@ $(LDFLAGS)
@echo "#########"
@echo "WARNING: The 'finetune' binary is deprecated. Please use 'llama-finetune' instead."
@echo " Remove the 'finetune' binary to remove this warning."
@echo "#########"
endif
51 changes: 51 additions & 0 deletions examples/deprecation-warning/README.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,51 @@
# Migration notice for binary filenames

> [!IMPORTANT]
[2024 Jun 12] Binaries have been renamed w/ a `llama-` prefix. `main` is now `llama-cli`, `server` is `llama-server`, etc (https://github.com/ggerganov/llama.cpp/pull/7809)

This migration was important, but it is a breaking change that may not always be immediately obvious to users.

Please update all scripts and workflows to use the new binary names.

| Old Filename | New Filename |
| ---- | ---- |
| main | llama-cli |
| server | llama-server |
| llama-bench | llama-bench |
| embedding | llama-embedding |
| finetune | llama-finetune |
| quantize | llama-quantize |
| tokenize | llama-tokenize |
| export-lora | llama-export-lora |
| libllava.a | libllava.a |
| baby-llama | llama-baby-llama |
| batched | llama-batched |
| batched-bench | llama-batched-bench |
| benchmark-matmult | llama-benchmark-matmult |
| convert-llama2c-to-ggml | llama-convert-llama2c-to-ggml |
| eval-callback | llama-eval-callback |
| gbnf-validator | llama-gbnf-validator |
| gguf | llama-gguf |
| gguf-split | llama-gguf-split |
| gritlm | llama-gritlm |
| imatrix | llama-imatrix |
| infill | llama-infill |
| llava-cli | llama-llava-cli |
| lookahead | llama-lookahead |
| lookup | llama-lookup |
| lookup-create | llama-lookup-create |
| lookup-merge | llama-lookup-merge |
| lookup-stats | llama-lookup-stats |
| parallel | llama-parallel |
| passkey | llama-passkey |
| perplexity | llama-perplexity |
| q8dot | llama-q8dot |
| quantize-stats | llama-quantize-stats |
| retrieval | llama-retrieval |
| save-load-state | llama-save-load-state |
| simple | llama-simple |
| speculative | llama-speculative |
| train-text-from-scratch | llama-train-text-from-scratch |
| vdot | llama-vdot |
| tests/test-c.o | tests/test-c.o |

35 changes: 35 additions & 0 deletions examples/deprecation-warning/deprecation-warning.cpp
Original file line number Diff line number Diff line change
@@ -0,0 +1,35 @@
// Warns users that this filename was deprecated, and provides a link for more information.

#include <cstdio>
#include <string>
#include <unordered_map>

// Main
int main(int argc, char** argv) {
std::string filename = "main";
if (argc >= 1) {
filename = argv[0];
}

// Get only the program name from the full path
auto pos = filename.find_last_of('/');
if (pos != std::string::npos) {
filename = filename.substr(pos+1);
}

// Append "llama-" to the beginning of filename to get the replacemnt filename
auto replacement_filename = "llama-" + filename;

// The exception is if the filename is "main", then our replacement filename is "llama-cli"
if (filename == "main") {
replacement_filename = "llama-cli";
}

fprintf(stdout, "\n");
fprintf(stdout, "WARNING: The binary '%s' is deprecated.\n", filename.c_str());
fprintf(stdout, " Please use '%s' instead.\n", replacement_filename.c_str());
fprintf(stdout, " See https://github.com/ggerganov/llama.cpp/tree/master/examples/deprecation-warning/README.md for more information.\n");
fprintf(stdout, "\n");

return EXIT_FAILURE;
}

0 comments on commit e500d61

Please sign in to comment.