review: rename part_bounded to n_rs

Author: am17an

common/common.cpp

@@ -1422,7 +1422,7 @@ common_context_seq_rm_type common_context_can_seq_rm(llama_context * ctx) {
 
     if (llama_n_rs_seq(ctx) > 0) {
         LOG_INF("%s: the context supports bounded partial sequence removal\n", __func__);
-        res = COMMON_CONTEXT_SEQ_RM_TYPE_PART_BOUNDED;
+        res = COMMON_CONTEXT_SEQ_RM_TYPE_RS;
         goto done;
     }
 

common/common.h

@@ -160,7 +160,7 @@ enum common_speculative_type {
     COMMON_SPECULATIVE_TYPE_NONE,          // no speculative decoding
     COMMON_SPECULATIVE_TYPE_DRAFT_SIMPLE,  // standalone draft model speculative decoding
     COMMON_SPECULATIVE_TYPE_DRAFT_EAGLE3,  // Eagle3 speculative decoding
-    COMMON_SPECULATIVE_TYPE_DRAFT_MTP,           // multi-token prediction head loaded from the target GGUF
+    COMMON_SPECULATIVE_TYPE_DRAFT_MTP,     // Multi-token prediction
     COMMON_SPECULATIVE_TYPE_NGRAM_SIMPLE,  // simple self-speculative decoding based on n-grams
     COMMON_SPECULATIVE_TYPE_NGRAM_MAP_K,   // self-speculative decoding with n-gram keys only
     COMMON_SPECULATIVE_TYPE_NGRAM_MAP_K4V, // self-speculative decoding with n-gram keys and 4 m-gram values
@@ -884,7 +884,7 @@ enum common_context_seq_rm_type {
     COMMON_CONTEXT_SEQ_RM_TYPE_NO           = 0, // seq_rm not supported (e.g. no memory module)
     COMMON_CONTEXT_SEQ_RM_TYPE_PART         = 1, // can seq_rm partial sequences
     COMMON_CONTEXT_SEQ_RM_TYPE_FULL         = 2, // can seq_rm full sequences only
-    COMMON_CONTEXT_SEQ_RM_TYPE_PART_BOUNDED = 3, // can seq_rm partial sequences, bounded by n_rs_seq
+    COMMON_CONTEXT_SEQ_RM_TYPE_RS = 3, // can seq_rm partial sequences, bounded by n_rs_seq
 };
 
 // check if the llama_context can remove sequences

examples/speculative-simple/speculative-simple.cpp

@@ -83,7 +83,6 @@ int main(int argc, char ** argv) {
 
     // check if the context supports partial sequence removal
     const bool use_ckpt_tgt = (common_context_can_seq_rm(ctx_tgt)       == COMMON_CONTEXT_SEQ_RM_TYPE_FULL)
-        || (common_context_can_seq_rm(ctx_tgt) == COMMON_CONTEXT_SEQ_RM_TYPE_PART_BOUNDED);
     const bool use_ckpt_dft = (common_context_can_seq_rm(ctx_dft.get()) == COMMON_CONTEXT_SEQ_RM_TYPE_FULL);
 
     if (use_ckpt_tgt) {

tools/server/server-context.cpp

@@ -2696,8 +2696,8 @@ struct server_context_impl {
                     // - the model uses SWA (and we are not using `swa_full`)
                     // - the model supports partial sequence removal but only up to a fixed bound
                     do_checkpoint = do_checkpoint && (
-                            ctx_tgt_seq_rm_type == COMMON_CONTEXT_SEQ_RM_TYPE_FULL         ||
-                            ctx_tgt_seq_rm_type == COMMON_CONTEXT_SEQ_RM_TYPE_PART_BOUNDED ||
+                            ctx_tgt_seq_rm_type == COMMON_CONTEXT_SEQ_RM_TYPE_FULL ||
+                            ctx_tgt_seq_rm_type == COMMON_CONTEXT_SEQ_RM_TYPE_RS ||
                             n_swa > 0);
 
                     bool has_mtmd = false;