mirror/esphome
1
0
Fork 0
mirror of https://github.com/esphome/esphome.git synced 2026-03-01 02:14:19 -07:00
Code Issues Packages Projects Releases Wiki Activity

[web_server] Flatten deq_push_back_with_dedup_ to inline vector realloc

Browse Source 
On ESP8266 (GCC 10.3), std::vector::push_back() emits a separate
_M_realloc_insert<DeferredEvent> function (198 bytes) that is only
called from one site. Adding __attribute__((flatten)) forces the
compiler to inline it, eliminating the out-of-line function's
prologue/epilogue and call overhead.

This matches what GCC 14.2 (ESP-IDF toolchain) already does
naturally for the equivalent code in web_server_idf.cpp.

Saves 80 bytes of flash on ESP8266.
This commit is contained in:
J. Nick Koston
2026-02-12 16:07:19 -06:00
parent 136d17366f
commit 8cf3cbf7ea
1 changed files with 2 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
esphome/components/web_server/web_server.cpp
View File

@@ -198,7 +198,8 @@ EntityMatchResult UrlMatch::match_entity(EntityBase *entity) const {
#if !defined(USE_ESP32) && defined(USE_ARDUINO)
// helper for allowing only unique entries in the queue
void DeferredUpdateEventSource::deq_push_back_with_dedup_(void *source, message_generator_t *message_generator) {
void __attribute__((flatten))
DeferredUpdateEventSource::deq_push_back_with_dedup_(void *source, message_generator_t *message_generator) {
DeferredEvent item(source, message_generator);
// Use range-based for loop instead of std::find_if to reduce template instantiation overhead and binary size