platformio: speed version for ESP32-S3, minor updates

* added [env:esp32S3_8MB_S] with optimization for speed
* minor cleanups for MM environments

platformio.ini

@@ -64,8 +64,10 @@ default_envs =
   ; wemos_shield_esp32_16MB_LineIn_M
   esp32_pico_4MB_M
   esp32_4MB_PSRAM_S
+  ; esp32_4MB_PSRAM_REV3_S ;; experimental, optimized for WROVER-E with "revision3" chip
   esp32S3_8MB_M
   ;; esp32S3_8MB_PSRAM_M   ;; experiemental
+  ;; esp32S3_8MB_S         ;; experimental, optimized for speed
   ;; esp32s2_tinyUF2_PSRAM_S  ;; experimental - only for adafruit -S2 boards with tinyUF2 bootloader !!!
   esp32s2_PSRAM_M          ;; experimental
   esp32c3dev_4MB_M         ;; experimental
@@ -1071,8 +1073,8 @@ build_flags =
 
 [Speed_Flags]
 build_flags =
-  -O2 ;; optimize for performance instead of size
+  -O2 ;; optimize for performance instead 9of size
-  ;-O3 ;; try this to optimize even more
+  ;-ffast-math ;; gives a few  (2-5) percent sppedup on ESP32-S3, but causes slight slowdown on classic ESP32
   -mtarget-align -free -fipa-pta ;; these are very useful, too
   -fno-jump-tables -fno-tree-switch-conversion                 ;; needed
   -freorder-blocks -Wwrite-strings -fstrict-volatile-bitfields ;; needed
@@ -1463,17 +1465,16 @@ extends = esp32_4MB_V4_S_base
 board = lolin_d32_pro
 ;board_build.partitions = tools/WLED_ESP32_4MB_256KB_FS.csv   ;; Alternative for 4MB flash:   1.8MB firmware, 256KB filesystem (esptool erase_flash needed before changing)
 build_unflags = ${esp32_4MB_V4_S_base.build_unflags}
+  ;;${Speed_Flags.build_unflags}  ;; to override -Os
   -DARDUINO_EVENT_RUNNING_CORE=1  ;; we want to run wifi on core0, so remove the standard flag
   -mfix-esp32-psram-cache-issue    ;; this fix is not needed any more for revision 3
   -mfix-esp32-psram-cache-strategy=memw ;; same as above
-  ;;-Os ;; if you want to be even faster - use together with "-O2" in build_flags. Enable alternative default_partitions (above) to have more program space.
 
 build_flags = ${esp32_4MB_V4_S_base.build_flags}
   -DARDUINO_EVENT_RUNNING_CORE=0  ;; assign Wifi to core0, to have more CPU on core#1 (arduino loop)
   -DARDUINO_RUNNING_CORE=1        ;; should be default, but does not hurt
-  ;-DCONFIG_MBEDTLS_DYNAMIC_BUFFER=1 ;; optional - seems to move more buffers into PSRAM
+  -DCONFIG_MBEDTLS_DYNAMIC_BUFFER=1 ;; optional - seems to move more buffers into PSRAM
-  ;; enable the next line together with "-Os" in build_unflags (unfortunately --> 104.0%% Flash)
+  ;;${Speed_Flags.build_flags}  ;; optimize for speed instead of size --> over 100% flash, but works with 256KB filesystem (alternative partitions file)
-  ;;-O2 -fno-jump-tables -fno-tree-switch-conversion -Wwrite-strings -fstrict-volatile-bitfields
   -D WLEDMM_FASTPATH ; WLEDMM experimental option. Reduces audio lag (latency), and allows for faster LED framerates. May break compatibility with previous versions.
   -DBOARD_HAS_PSRAM -D WLED_USE_PSRAM_JSON ;; -D WLED_USE_PSRAM ;; WLED_USE_PSRAM causes major slow-down (slow LEDs) on some ESP32 boards
   ;;-D CONFIG_ESP32_REV_MIN=3  ;;  disables PSRAM bug workarounds in the core, reducing the code size and improving overall performance.
@@ -1491,8 +1492,8 @@ build_flags = ${esp32_4MB_V4_S_base.build_flags}
   ; -D SR_DEBUG
   ; -D MIC_LOGGER
 lib_ignore = IRremoteESP8266 ; use with WLED_DISABLE_INFRARED for faster compilation
-;; RAM:   [==        ]  16.8% (used 55124 bytes from 327680 bytes)
+;; RAM:   [====      ]  35.9% (used 117688 bytes from 327680 bytes)
-;; Flash: [========= ]  92.7% (used 1458729 bytes from 1572864 bytes)
+;; Flash: [========= ]  94.5% (used 1487097 bytes from 1572864 bytes)
 
 ;; PSRAM build env that only leaves 300Kb for filesystem (instead of 1MB), but adds 300kB for program space
 [env:esp32_4MB_PSRAM_M]
@@ -1518,6 +1519,7 @@ build_flags = ${esp32_4MB_V4_M_base.build_flags}
 ;; RAM:   [==        ]  24.9% (used 81484 bytes from 327680 bytes)
 ;; Flash: [========  ]  84.6% (used 1607857 bytes from 1900544 bytes)
 
+
 [env:esp32S3_8MB_M]
 extends = esp32_4MB_V4_M_base
 board = esp32-s3-devkitc-1
@@ -1604,6 +1606,51 @@ board_build.partitions = tools/WLED_ESP32_8MB.csv
 ; RAM:   [====      ]  36.3% (used 118828 bytes from 327680 bytes)
 ; Flash: [=======   ]  70.7% (used 1483465 bytes from 2097152 bytes)
 
+;; MM for ESP32-S3 PSRAM board - FASTPATH + optimize for speed
+[env:esp32S3_8MB_S]
+extends = esp32_4MB_V4_M_base
+board = esp32-s3-devkitc-1 ;; generic S3 dev board
+board_build.flash_mode = qio ;; use "dio" if your board gets unstable with "qio"
+build_unflags = ${env:esp32S3_8MB_M.build_unflags}   ;; use the same as "normal" S3 buildenv
+  ${Speed_Flags.build_unflags}  ;; to override -Os
+build_flags = ${common.build_flags} ${esp32s3.build_flags} -Wno-misleading-indentation -Wno-format-truncation
+  ${common_mm.build_flags_S}
+  -D WLED_RELEASE_NAME=esp32S3_8MB_PSRAM_S
+  -DARDUINO_USB_MODE=1 -DARDUINO_USB_CDC_ON_BOOT=0 -DARDUINO_USB_MSC_ON_BOOT=0 -DARDUINO_USB_DFU_ON_BOOT=0   ;; for Serial-to-USB chip
+  ;;-DARDUINO_USB_MODE=1 -DARDUINO_USB_CDC_ON_BOOT=1 -DARDUINO_USB_MSC_ON_BOOT=0 -DARDUINO_USB_DFU_ON_BOOT=1 ;; for Hardware-CDC USB mode
+  ;;-D WLED_DISABLE_ADALIGHT     ;; disables serial protocols - recommended for Hardware-CDC USB (Serial RX will receive junk commands when RX pin is unconnected, unless its pulled down by resistor)
+  ${Speed_Flags.build_flags}  ;; optimize for speed instead of size
+  -D WLEDMM_FASTPATH ; WLEDMM experimental option. Reduces audio lag (latency), and allows for faster LED framerates. May break compatibility with previous versions.
+  -D WLED_WATCHDOG_TIMEOUT=0 -D CONFIG_ASYNC_TCP_USE_WDT=0
+  -D WLED_DISABLE_LOXONE ; FLASH 1272 bytes
+  -D WLED_DISABLE_ALEXA ;  RAM 116 bytes; FLASH 13524 bytes
+  -D WLED_DISABLE_MQTT ;   RAM 216 bytes; FLASH 16496 bytes
+  -D WLED_DISABLE_HUESYNC ;RAM 122 bytes; FLASH 6308 bytes
+  -D WLED_DISABLE_INFRARED ;RAM 136 bytes; FLASH 24492 bytes
+  -D LEDPIN=4
+  ;-D STATUSLED=39
+  -D BTNPIN=48
+  -D RLYPIN=21
+  -D IRPIN=15
+  -D HW_PIN_SDA=42 ;; ESP32-S3 default: SDA = 8
+  -D HW_PIN_SCL=41 ;; ESP32-S3 default: SCL = 9
+  -D AUDIOPIN=-1
+  -D SR_DMTYPE=1 -D I2S_SDPIN=16 -D I2S_CKPIN=17 -D I2S_WSPIN=47 -D MCLK_PIN=40
+  -D ES7243_SDAPIN=8 -D ES7243_SCLPIN=18
+  ; -D WLED_DEBUG
+  ; -D SR_DEBUG
+  ; -D MIC_LOGGER
+lib_deps = ${esp32s3.lib_deps} ${common_mm.lib_deps_S}
+lib_ignore = IRremoteESP8266 ; use with WLED_DISABLE_INFRARED for faster compilation
+board_build.partitions = tools/WLED_ESP32_8MB.csv
+;; normal build
+; RAM:   [==        ]  24.9% (used 81656 bytes from 327680 bytes)
+; Flash: [=======   ]  66.1% (used 1386229 bytes from 2097152 bytes)
+;; optimized-for-speed build
+; RAM:   [==        ]  24.9% (used 81712 bytes from 327680 bytes)
+; Flash: [========  ]  75.2% (used 1576113 bytes from 2097152 bytes)
+
+
 ;; MM for Adafruit QT Py ESP32-S2 -> 4MB flash, PSRAM, and tinyUF2 bootloader
 ;; to ewowi - i'll optimize this entry later, as a few things can be inherited for sure. To softhack: sure ;-)
 [env:esp32s2_tinyUF2_PSRAM_S]
@@ -1704,6 +1751,7 @@ monitor_filters = esp32_exception_decoder
 ; RAM:   [===       ]  30.9% (used 101228 bytes from 327680 bytes)
 ; Flash: [========  ]  80.4% (used 1527278 bytes from 1900544 bytes)
 
+
 ;; MM environment for generic ESP32-C3 -> 4MB flash, no PSRAM
 [env:esp32c3dev_4MB_M]
 extends = esp32_4MB_V4_S_base