Fix flash attention long argument for mainloine compatibility

Iwan Kawrakow · Iwan Kawrakow · commit be1a8cb9d8b7 · 2025-11-13T19:22:16.000+02:00
diff --git a/common/common.cpp b/common/common.cpp
@@ -1135,7 +1135,7 @@ bool gpt_params_find_arg(int argc, char ** argv, const std::string & arg, gpt_pa
         params.flash_attn = false;
         return true;
     }
-    if (arg == "-fa" || arg == "--flash-attention") {
+    if (arg == "-fa" || arg == "--flash-attn") {
         CHECK_ARG
         std::string next_arg{argv[i]};
         for (auto& c : next_arg) c = std::tolower(c);

Original file line number	Diff line number	Diff line change
`@@ -1135,7 +1135,7 @@ bool gpt_params_find_arg(int argc, char ** argv, const std::string & arg, gpt_pa`
`1135`	`1135`	`params.flash_attn = false;`
`1136`	`1136`	`return true;`
`1137`	`1137`	`}`
`1138`		`- if (arg == "-fa" \|\| arg == "--flash-attention") {`
	`1138`	`+ if (arg == "-fa" \|\| arg == "--flash-attn") {`
`1139`	`1139`	`CHECK_ARG`
`1140`	`1140`	`std::string next_arg{argv[i]};`
`1141`	`1141`	`for (auto& c : next_arg) c = std::tolower(c);`