fix missing n_past in various places

this is actually a revert of ggml-org@cda0e4b

Author: ngxson

examples/llama-bench/llama-bench.cpp

@@ -1427,7 +1427,7 @@ struct sql_printer : public printer {
     }
 };
 
-static void test_prompt(llama_context * ctx, int n_prompt, int n_batch, int n_threads) {
+static void test_prompt(llama_context * ctx, int n_prompt, int n_past, int n_batch, int n_threads) {
     llama_set_n_threads(ctx, n_threads, n_threads);
 
     const llama_model * model   = llama_get_model(ctx);
@@ -1444,15 +1444,15 @@ static void test_prompt(llama_context * ctx, int n_prompt, int n_batch, int n_th
         for (int i = 1; i < n_tokens; i++) {
             tokens[i] = std::rand() % n_vocab;
         }
-        llama_batch_ext_ptr batch(llama_batch_ext_init_from_text(tokens.data(), n_tokens, 0, 0, true));
+        llama_batch_ext_ptr batch(llama_batch_ext_init_from_text(tokens.data(), n_tokens, n_past + n_processed, 0, true));
         llama_decode_ext(ctx, batch.get());
         n_processed += n_tokens;
     }
 
     llama_synchronize(ctx);
 }
 
-static void test_gen(llama_context * ctx, int n_gen, int n_threads) {
+static void test_gen(llama_context * ctx, int n_gen, int n_past, int n_threads) {
     llama_set_n_threads(ctx, n_threads, n_threads);
 
     const llama_model * model   = llama_get_model(ctx);
@@ -1462,7 +1462,7 @@ static void test_gen(llama_context * ctx, int n_gen, int n_threads) {
     llama_token token = llama_vocab_get_add_bos(vocab) ? llama_vocab_bos(vocab) : std::rand() % n_vocab;
 
     for (int i = 0; i < n_gen; i++) {
-        llama_batch_ext_ptr batch(llama_batch_ext_init_from_text(&token, 1, 0, 0, true));
+        llama_batch_ext_ptr batch(llama_batch_ext_init_from_text(&token, 1, n_past + i, 0, true));
         llama_decode_ext(ctx, batch.get());
         llama_synchronize(ctx);
         token = std::rand() % n_vocab;
@@ -1610,13 +1610,13 @@ int main(int argc, char ** argv) {
                 fprintf(stderr, "llama-bench: benchmark %d/%zu: warmup prompt run\n", params_idx, params_count);
             }
             //test_prompt(ctx, std::min(t.n_batch, std::min(t.n_prompt, 32)), 0, t.n_batch, t.n_threads);
-            test_prompt(ctx, t.n_prompt, t.n_batch, t.n_threads);
+            test_prompt(ctx, t.n_prompt, 0, t.n_batch, t.n_threads);
         }
         if (t.n_gen > 0) {
             if (params.progress) {
                 fprintf(stderr, "llama-bench: benchmark %d/%zu: warmup generation run\n", params_idx, params_count);
             }
-            test_gen(ctx, 1, t.n_threads);
+            test_gen(ctx, 1, 0, t.n_threads);
         }
 
         for (int i = 0; i < params.reps; i++) {
@@ -1629,14 +1629,14 @@ int main(int argc, char ** argv) {
                     fprintf(stderr, "llama-bench: benchmark %d/%zu: prompt run %d/%d\n", params_idx, params_count,
                             i + 1, params.reps);
                 }
-                test_prompt(ctx, t.n_prompt, t.n_batch, t.n_threads);
+                test_prompt(ctx, t.n_prompt, 0, t.n_batch, t.n_threads);
             }
             if (t.n_gen > 0) {
                 if (params.progress) {
                     fprintf(stderr, "llama-bench: benchmark %d/%zu: generation run %d/%d\n", params_idx, params_count,
                             i + 1, params.reps);
                 }
-                test_gen(ctx, t.n_gen, t.n_threads);
+                test_gen(ctx, t.n_gen, t.n_prompt, t.n_threads);
             }
 
             uint64_t t_ns = get_time_ns() - t_start;

examples/lookahead/lookahead.cpp

@@ -92,8 +92,8 @@ int main(int argc, char ** argv) {
     const auto t_enc_start = ggml_time_us();
 
     // eval the prompt
-    llama_batch_ext_ptr batch0(llama_batch_ext_init_from_text( inp.data(), n_input - 1, 0, 0, true));
-    llama_batch_ext_ptr batch1(llama_batch_ext_init_from_text(&inp.back(),           1, 0, 0, true));
+    llama_batch_ext_ptr batch0(llama_batch_ext_init_from_text( inp.data(), n_input - 1, 0,           0, true));
+    llama_batch_ext_ptr batch1(llama_batch_ext_init_from_text(&inp.back(),           1, n_input - 1, 0, true));
     llama_decode_ext(ctx, batch0.get());
     llama_decode_ext(ctx, batch1.get());
 

examples/lookup/lookup.cpp

@@ -91,8 +91,8 @@ int main(int argc, char ** argv){
 
     const auto t_enc_start = ggml_time_us();
 
-    llama_batch_ext_ptr batch0(llama_batch_ext_init_from_text( inp.data(), n_input - 1, 0, 0, true));
-    llama_batch_ext_ptr batch1(llama_batch_ext_init_from_text(&inp.back(),           1, 0, 0, true));
+    llama_batch_ext_ptr batch0(llama_batch_ext_init_from_text( inp.data(), n_input - 1, 0,           0, true));
+    llama_batch_ext_ptr batch1(llama_batch_ext_init_from_text(&inp.back(),           1, n_input - 1, 0, true));
     llama_decode_ext(ctx, batch0.get());
     llama_decode_ext(ctx, batch1.get());
 

examples/save-load-state/save-load-state.cpp

@@ -133,7 +133,7 @@ int main(int argc, char ** argv) {
         result1 += next_token_str;
 
         llama_batch_ext_clear(batch);
-        llama_seq_id seq_id = 1;
+        llama_seq_id seq_id = 0;
         llama_batch_ext_add_text(batch, next_token, 0, &seq_id, 1, true);
 
         if (llama_decode_ext(ctx2, batch)) {
@@ -215,7 +215,7 @@ int main(int argc, char ** argv) {
         result2 += next_token_str;
 
         llama_batch_ext_clear(batch);
-        llama_seq_id seq_id = 1;
+        llama_seq_id seq_id = 1; // seq 1 instead of 0
         llama_batch_ext_add_text(batch, next_token, 0, &seq_id, 1, true);
 
         if (llama_decode_ext(ctx3, batch)) {

examples/simple/simple.cpp

@@ -182,7 +182,7 @@ int main(int argc, char ** argv) {
             // prepare the next batch with the sampled token
             llama_batch_ext_clear(batch);
             llama_seq_id seq_id = 0;
-            llama_batch_ext_add_text(batch, new_token_id, 0, &seq_id, 1, true);
+            llama_batch_ext_add_text(batch, new_token_id, n_pos, &seq_id, 1, true);
 
             n_decode += 1;
         }

examples/speculative/speculative.cpp

@@ -166,9 +166,9 @@ int main(int argc, char ** argv) {
     const auto t_enc_start = ggml_time_us();
 
     // eval the prompt with both models
-    llama_batch_ext_ptr batch0(llama_batch_ext_init_from_text( inp.data(), n_input - 1, 0, 0, true));
-    llama_batch_ext_ptr batch1(llama_batch_ext_init_from_text(&inp.back(),           1, 0, 0, true));
-    llama_batch_ext_ptr batch2(llama_batch_ext_init_from_text( inp.data(), n_input    , 0, 0, true));
+    llama_batch_ext_ptr batch0(llama_batch_ext_init_from_text( inp.data(), n_input - 1, 0,           0, true));
+    llama_batch_ext_ptr batch1(llama_batch_ext_init_from_text(&inp.back(),           1, n_input - 1, 0, true));
+    llama_batch_ext_ptr batch2(llama_batch_ext_init_from_text( inp.data(), n_input    , 0,           0, true));
     llama_decode_ext(ctx_tgt, batch0);
     llama_decode_ext(ctx_tgt, batch1);
     llama_decode_ext(ctx_dft, batch2);