diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml
index 3a766bf..7b89111 100644
--- a/.gitlab-ci.yml
+++ b/.gitlab-ci.yml
@@ -2,16 +2,50 @@ default:
   tags:
     - docker
 
+variables:
+  GIT_SUBMODULE_STRATEGY: recursive
+
+stages:
+  - build_toolchain
+  - build_sw
+  - build_hw
+  - test
+
+build-cc65:
+  stage: build_toolchain
+  image: gcc
+  script:
+   - cd sw
+   - make toolchain
+  artifacts:
+    paths:
+      - sw/cc65/bin
+      - sw/cc65/lib
+
 build-kernel:
-  stage: build
-  image: a2geek/cc65-pipeline
+  stage: build_sw
+  image: gcc
   script:
     - cd sw/kernel
     - make
 
+build-bootloader:
+  stage: build_sw
+  image: gcc
+  script:
+    - cd sw/bootloader
+    - make
+  artifacts:
+    paths:
+      - sw/bootloader/bootloader.hex
+    expire_in: 1 week
+
+
 
 build-fpga:
-  stage: build
+  stage: build_hw
+  dependencies:
+    - build-bootloader
   image: bslathi19/modelsim_18.1:lite
   script:
     - cd hw/fpga/
@@ -34,7 +68,7 @@ test_bb_spi:
 
 test-sw:
   stage: test
-  image: a2geek/cc65-pipeline
+  image: gcc
   script:
     - cd sw/kernel
     - make test
diff --git a/.gitmodules b/.gitmodules
new file mode 100644
index 0000000..21b2181
--- /dev/null
+++ b/.gitmodules
@@ -0,0 +1,3 @@
+[submodule "sw/cc65"]
+	path = sw/cc65
+	url = https://git.byronlathi.com/bslathi19/cc65
diff --git a/hw/fpga/rom.v b/hw/fpga/rom.v
index 24d9df9..c4f311c 100644
--- a/hw/fpga/rom.v
+++ b/hw/fpga/rom.v
@@ -84,13 +84,13 @@ module rom (
 		altsyncram_component.clock_enable_input_a = "BYPASS",
 		altsyncram_component.clock_enable_output_a = "BYPASS",
 `ifdef NO_PLI
-		altsyncram_component.init_file = "../../sw/bootrom.rif"
+		altsyncram_component.init_file = "../../sw/bootloader/bootloader.rif"
 `else
-		altsyncram_component.init_file = "../../sw/bootrom.hex"
+		altsyncram_component.init_file = "../../sw/bootloader/bootloader.hex"
 `endif
 ,
 		altsyncram_component.intended_device_family = "MAX 10",
-		altsyncram_component.lpm_hint = "ENABLE_RUNTIME_MOD=YES,INSTANCE_NAME=ROM0",
+		altsyncram_component.lpm_hint = "ENABLE_RUNTIME_MOD=NO",
 		altsyncram_component.lpm_type = "altsyncram",
 		altsyncram_component.numwords_a = 32768,
 		altsyncram_component.operation_mode = "ROM",
@@ -120,10 +120,10 @@ endmodule
 // Retrieval info: PRIVATE: INIT_FILE_LAYOUT STRING "PORT_A"
 // Retrieval info: PRIVATE: INIT_TO_SIM_X NUMERIC "0"
 // Retrieval info: PRIVATE: INTENDED_DEVICE_FAMILY STRING "MAX 10"
-// Retrieval info: PRIVATE: JTAG_ENABLED NUMERIC "1"
+// Retrieval info: PRIVATE: JTAG_ENABLED NUMERIC "0"
 // Retrieval info: PRIVATE: JTAG_ID STRING "ROM0"
 // Retrieval info: PRIVATE: MAXIMUM_DEPTH NUMERIC "0"
-// Retrieval info: PRIVATE: MIFfilename STRING "../../sw/bootrom.hex"
+// Retrieval info: PRIVATE: MIFfilename STRING "../../sw/bootloader/bootloader.hex"
 // Retrieval info: PRIVATE: NUMWORDS_A NUMERIC "32768"
 // Retrieval info: PRIVATE: RAM_BLOCK_TYPE NUMERIC "0"
 // Retrieval info: PRIVATE: RegAddr NUMERIC "1"
@@ -138,9 +138,9 @@ endmodule
 // Retrieval info: CONSTANT: ADDRESS_ACLR_A STRING "NONE"
 // Retrieval info: CONSTANT: CLOCK_ENABLE_INPUT_A STRING "BYPASS"
 // Retrieval info: CONSTANT: CLOCK_ENABLE_OUTPUT_A STRING "BYPASS"
-// Retrieval info: CONSTANT: INIT_FILE STRING "../../sw/bootrom.hex"
+// Retrieval info: CONSTANT: INIT_FILE STRING "../../sw/bootloader/bootloader.hex"
 // Retrieval info: CONSTANT: INTENDED_DEVICE_FAMILY STRING "MAX 10"
-// Retrieval info: CONSTANT: LPM_HINT STRING "ENABLE_RUNTIME_MOD=YES,INSTANCE_NAME=ROM0"
+// Retrieval info: CONSTANT: LPM_HINT STRING "ENABLE_RUNTIME_MOD=NO"
 // Retrieval info: CONSTANT: LPM_TYPE STRING "altsyncram"
 // Retrieval info: CONSTANT: NUMWORDS_A NUMERIC "32768"
 // Retrieval info: CONSTANT: OPERATION_MODE STRING "ROM"
diff --git a/sw/Makefile b/sw/Makefile
new file mode 100644
index 0000000..c379713
--- /dev/null
+++ b/sw/Makefile
@@ -0,0 +1,22 @@
+.PHONY: all install bootloader kernel clean
+
+all: toolchain bootloader kernel
+
+install: all
+	sh script/format_disk.sh
+	sh script/copy_files.sh
+
+toolchain:
+	@$(MAKE) -j4 -C cc65
+
+bootloader:
+	@$(MAKE) -C bootloader
+
+kernel:
+	@$(MAKE) -C kernel
+
+
+clean:
+	@$(MAKE) -C bootloader  --no-print-directory $@
+	@$(MAKE) -C kernel  --no-print-directory $@
+	@$(MAKE) -C cc65  --no-print-directory $@
\ No newline at end of file
diff --git a/sw/bootloader/Makefile b/sw/bootloader/Makefile
new file mode 100644
index 0000000..6f3db2c
--- /dev/null
+++ b/sw/bootloader/Makefile
@@ -0,0 +1,56 @@
+CC=../cc65/bin/cl65
+CFLAGS=-T -t none -I. --cpu "65C02"
+test: CFLAGS=-T -t sim65c02 -I.
+LDFLAGS=-C link.ld -m $(NAME).map
+SIM=sim65
+SIMARGS=-v -c -x 1000000
+
+NAME=bootloader
+
+TEST_BIN=test.bin
+BIN=$(NAME).bin
+HEX=$(NAME).hex
+
+LISTS=lists
+TESTS=tests
+
+SRCS=$(wildcard *.s) $(wildcard *.c)
+SRCS+=$(filter-out $(wildcard tests/*), $(wildcard **/*.s)) $(filter-out $(wildcard tests/*), $(wildcard **/*.c))
+OBJS+=$(patsubst %.s,%.o,$(filter %s,$(SRCS)))
+OBJS+=$(patsubst %.c,%.o,$(filter %c,$(SRCS)))
+
+TEST_SRCS=$(wildcard $(TESTS)/*.s) $(wildcard $(TESTS)/*.c)
+TEST_OBJS+=$(patsubst %.s,%.o,$(filter %s,$(TEST_SRCS)))
+TEST_OBJS+=$(patsubst %.c,%.o,$(filter %c,$(TEST_SRCS)))
+TEST_OBJS+=$(filter-out boot.o,$(filter-out main.o,$(filter-out vectors.o,$(OBJS))))
+
+all: $(HEX)
+
+test: $(TEST_BIN)
+	$(SIM) $(SIMARGS) $(TEST_BIN)
+
+$(TEST_BIN): $(OBJS) $(TEST_OBJS)
+	$(CC) $(CFLAGS) $(TEST_OBJS) -o $@	
+
+$(HEX): $(BIN)
+	objcopy --input-target=binary --output-target=ihex $(BIN) $(HEX)
+	
+
+$(BIN): $(OBJS)
+	$(CC) $(CFLAGS) $(LDFLAGS) $(OBJS) -o $@
+
+%.o: %.c $(LISTS)
+	$(CC) $(CFLAGS) -l $(LISTS)/$<.list -c $< -o $@
+
+%.o: %.s $(LISTS)
+	$(CC) $(CFLAGS) -l $(LISTS)/$<.list -c $< -o $@
+
+$(LISTS):
+	mkdir -p $(addprefix $(LISTS)/,$(sort $(dir $(SRCS))))
+	mkdir $(LISTS)/$(sort $(dir $(TEST_SRCS)))
+
+.PHONY: clean
+clean:
+	rm -rf $(OBJS) $(BIN) $(HEX) $(LISTS) $(NAME).map
+	rm -rf $(TEST_OBJS) $(TEST_BIN)
+
diff --git a/sw/bootloader/boot.s b/sw/bootloader/boot.s
new file mode 100644
index 0000000..00b87b1
--- /dev/null
+++ b/sw/bootloader/boot.s
@@ -0,0 +1,54 @@
+; ---------------------------------------------------------------------------
+; crt0.s
+; ---------------------------------------------------------------------------
+;
+; Startup code for cc65 (Single Board Computer version)
+
+.export   _init, _exit
+.import   _main
+
+.export   __STARTUP__ : absolute = 1        ; Mark as startup
+.import   __SDRAM_START__, __SDRAM_SIZE__       ; Linker generated
+
+.import    copydata, zerobss, initlib, donelib
+
+.include  "zeropage.inc"
+
+; ---------------------------------------------------------------------------
+; Place the startup code in a special segment
+
+.segment  "STARTUP"
+
+; ---------------------------------------------------------------------------
+; A little light 6502 housekeeping
+
+_init:    LDX     #$FF                 ; Initialize stack pointer to $01FF
+          TXS
+          CLD                          ; Clear decimal mode
+
+; ---------------------------------------------------------------------------
+; Set cc65 argument stack pointer
+
+          LDA     #<(__SDRAM_START__ + __SDRAM_SIZE__)
+          STA     sp
+          LDA     #>(__SDRAM_START__ + __SDRAM_SIZE__)
+          STA     sp+1
+
+; ---------------------------------------------------------------------------
+; Initialize memory storage
+
+          JSR     zerobss              ; Clear BSS segment
+          JSR     copydata             ; Initialize DATA segment
+          JSR     initlib              ; Run constructors
+
+; ---------------------------------------------------------------------------
+; Call main()
+          cli
+          JSR     _main
+          jmp     ($fffc)
+
+; ---------------------------------------------------------------------------
+; Back from main (this is also the _exit entry):  force a software break
+
+_exit:    JSR     donelib              ; Run destructors
+          BRK
diff --git a/sw/bootloader/devices/board_io.h b/sw/bootloader/devices/board_io.h
new file mode 100644
index 0000000..c55c7a5
--- /dev/null
+++ b/sw/bootloader/devices/board_io.h
@@ -0,0 +1,16 @@
+#ifndef _BOARD_IO_H
+#define _BOARD_IO_H
+
+#include <stdint.h>
+
+uint8_t hex_set_8(uint8_t val, uint8_t idx);
+uint8_t hex_set_16(uint16_t val);
+uint8_t hex_set_24(uint32_t val);
+
+void hex_enable(uint8_t mask);
+
+uint8_t sw_read();
+
+void led_set(uint8_t val);
+
+#endif
\ No newline at end of file
diff --git a/sw/bootloader/devices/board_io.s b/sw/bootloader/devices/board_io.s
new file mode 100644
index 0000000..95656fe
--- /dev/null
+++ b/sw/bootloader/devices/board_io.s
@@ -0,0 +1,71 @@
+.include "io.inc65"
+
+.importzp sp, sreg
+
+.export _hex_set_8
+.export _hex_set_16
+.export _hex_set_24
+.export _hex_enable
+.export _sw_read
+.export _led_set
+
+.autoimport	on
+
+.code
+
+; @in A: idx Stack[0]: val
+; @out A: 0 for success, 1 for failure.
+; Sets one of the 3 pairs of hex digits.
+_hex_set_8:
+        phx
+        cmp #$3         ; If idx >= 3 then fail
+        bcc @1
+        plx
+        lda #$1
+        rts
+@1:     tax             ; Move idx into x
+        jsr popa        ; put val into a
+        sta SEVEN_SEG,x ; write to val
+        lda #$0
+        plx
+        rts
+
+; @in A/X: val
+; @out A: 0 for success, 1 for failure
+; Sets the low 2 pairs of hex digits
+_hex_set_16:
+        sta SEVEN_SEG
+        stx SEVEN_SEG+1
+        lda #$0
+        rts
+
+; @in A/X/sreg: val
+; @out A: 0 for success, 1 for failure
+; Sets the 3 pairs of hex digits for a 24 bit value
+_hex_set_24:
+        sta SEVEN_SEG
+        stx SEVEN_SEG+1
+        lda sreg
+        sta SEVEN_SEG+2
+        lda #$0
+        rts
+
+; @in A: mask
+; Set the mask for seven seg enables
+_hex_enable:
+        sta SEVEN_SEG+3
+        rts
+
+; @out A: The Value of the switches
+; Reads the current values of the switches.
+_sw_read:
+        lda SW
+        ldx #$0
+        rts
+
+; @in A: val
+; @out A: 0 for success, 1 for failure
+; Sets the LEDs
+_led_set:
+        sta LED
+        rts
\ No newline at end of file
diff --git a/sw/bootloader/devices/conio.s b/sw/bootloader/devices/conio.s
new file mode 100644
index 0000000..30b4d6a
--- /dev/null
+++ b/sw/bootloader/devices/conio.s
@@ -0,0 +1,60 @@
+.importzp sp, sreg
+
+.import _uart_txb_block
+.import _lastchar
+
+.export _cputc
+.export gotoxy
+.export _clrscr
+.export _cgetc
+
+.autoimport	on
+
+.code
+
+; void __fastcall__ cputc (char c);
+_cputc:
+        jsr _uart_txb_block
+        cmp #$0a
+        bne @1
+        lda #$0d
+        jsr _uart_txb_block
+@1:     rts
+
+; void __fastcall__ gotoxy (unsigned char x, unsigned char y);
+gotoxy:
+        phx
+        phy
+        tay                         ; Move y position to y
+        lda (sp)
+        tax                         ; Move x position to x
+        lda #$1b
+        jsr _uart_txb_block
+        lda #'['
+        jsr _uart_txb_block
+        tya
+        jsr _uart_txb_block
+        lda #';'
+        jsr _uart_txb_block
+        txa
+        jsr _uart_txb_block
+        lda #'H'
+        jsr _uart_txb_block
+        ply
+        plx
+        rts
+
+_clrscr:
+        phx
+        lda #$1b
+        jsr _uart_txb_block
+        lda #'c'
+        jsr _uart_txb_block
+        pla
+        rts
+
+_cgetc:
+@2:     lda _lastchar
+        beq @2
+        stz _lastchar
+        rts
\ No newline at end of file
diff --git a/sw/bootloader/devices/interrupt.h b/sw/bootloader/devices/interrupt.h
new file mode 100644
index 0000000..d0231aa
--- /dev/null
+++ b/sw/bootloader/devices/interrupt.h
@@ -0,0 +1,15 @@
+#ifndef _INTERRUPT_H
+#define _INTERRUPT_H
+
+#include <stdint.h>
+
+#define BUTTON  (1 << 0)
+#define UART    (1 << 1)
+
+void irq_int();
+void nmi_int();
+
+uint8_t irq_get_status();
+void irq_set_status(uint8_t);
+
+#endif
\ No newline at end of file
diff --git a/sw/bootloader/devices/interrupt.s b/sw/bootloader/devices/interrupt.s
new file mode 100644
index 0000000..f87f81d
--- /dev/null
+++ b/sw/bootloader/devices/interrupt.s
@@ -0,0 +1,58 @@
+; ---------------------------------------------------------------------------
+; interrupt.s
+; ---------------------------------------------------------------------------
+;
+; Interrupt handler.
+;
+; Checks for a BRK instruction and returns from all valid interrupts.
+
+.import   _handle_irq
+
+.export   _irq_int, _nmi_int
+.export   _irq_get_status, _irq_set_status
+
+.include "io.inc65"
+
+.segment  "CODE"
+
+.PC02                             ; Force 65C02 assembly mode
+
+; ---------------------------------------------------------------------------
+; Non-maskable interrupt (NMI) service routine
+
+_nmi_int:  RTI                    ; Return from all NMI interrupts
+
+; ---------------------------------------------------------------------------
+; Maskable interrupt (IRQ) service routine
+
+_irq_int:  PHX                    ; Save X register contents to stack
+           TSX                    ; Transfer stack pointer to X
+           PHA                    ; Save accumulator contents to stack
+           INX                    ; Increment X so it points to the status
+           INX                    ;   register value saved on the stack
+           LDA $100,X             ; Load status register contents
+           AND #$10               ; Isolate B status bit
+           BNE break              ; If B = 1, BRK detected
+
+; ---------------------------------------------------------------------------
+; IRQ detected, return
+
+irq:       PLA                    ; Restore accumulator contents
+           PLX                    ; Restore X register contents
+           jsr _handle_irq        ; Handle the IRQ
+           RTI                    ; Return from all IRQ interrupts
+
+; ---------------------------------------------------------------------------
+; BRK detected, stop
+
+break:     JMP break              ; If BRK is detected, something very bad
+                                  ;   has happened, so stop running
+
+_irq_get_status:
+           lda IRQ_STATUS
+           ldx #$00
+           rts
+
+_irq_set_status:
+           sta IRQ_STATUS
+           rts
\ No newline at end of file
diff --git a/sw/bootloader/devices/io.inc65 b/sw/bootloader/devices/io.inc65
new file mode 100644
index 0000000..bad732c
--- /dev/null
+++ b/sw/bootloader/devices/io.inc65
@@ -0,0 +1,18 @@
+SEVEN_SEG   = $7ff0
+
+UART        = $7ff4
+UART_TXB    = UART
+UART_RXB    = UART
+UART_STATUS = UART + 1
+
+LED         = $7ff6
+SW          = LED
+
+MM_CTRL		= $7ff7
+MM_DATA		= $7fe0
+
+SD_ARG      = $7ff8
+SD_CMD      = $7ffc
+SD_DATA     = $7ffd
+
+IRQ_STATUS  = $7fff
diff --git a/sw/bootloader/devices/mapper.h b/sw/bootloader/devices/mapper.h
new file mode 100644
index 0000000..8276b60
--- /dev/null
+++ b/sw/bootloader/devices/mapper.h
@@ -0,0 +1,12 @@
+#ifndef _MAPPER_H
+#define _MAPPER_H
+
+#include <stdint.h>
+
+void mapper_enable(uint8_t en);
+
+uint8_t mapper_read(uint8_t addr);
+void mapper_write(uint8_t data, uint8_t addr);
+
+#endif
+
diff --git a/sw/bootloader/devices/mapper.s b/sw/bootloader/devices/mapper.s
new file mode 100644
index 0000000..2ed1bc0
--- /dev/null
+++ b/sw/bootloader/devices/mapper.s
@@ -0,0 +1,32 @@
+.include "io.inc65"
+
+.importzp sp, sreg
+
+.export _mapper_enable
+.export _mapper_read, _mapper_write
+
+.autoimport on
+
+.code
+
+
+; void mapper_enable(uint8_t en)
+_mapper_enable:
+	sta MM_CTRL
+	rts
+
+_mapper_read:
+	phx
+	tax
+	lda MM_DATA,x
+	ldx #$00
+	rts
+
+_mapper_write:
+	phx
+	tax
+	jsr popa
+	sta MM_DATA,x
+	plx
+	rts
+
diff --git a/sw/bootloader/devices/sd_card.c b/sw/bootloader/devices/sd_card.c
new file mode 100644
index 0000000..752cd0e
--- /dev/null
+++ b/sw/bootloader/devices/sd_card.c
@@ -0,0 +1,75 @@
+#include <stdint.h>
+#include <conio.h>
+
+#include "devices/sd_card.h"
+
+void sd_init() {
+	uint32_t resp;
+	sd_card_command(0, 0);
+
+	sd_card_command(0x000001aa, 8);
+	sd_card_resp(&resp);
+	cprintf("CMD8: %lx\n", resp);
+
+	sd_card_command(0, 55);
+	sd_card_command(0x40180000, 41);
+	sd_card_resp(&resp);
+	cprintf("CMD41: %lx\n", resp);
+
+	sd_card_command(0, 55);
+	sd_card_command(0x40180000, 41);
+	sd_card_resp(&resp);
+	cprintf("CMD41: %lx\n", resp);
+
+	sd_card_command(0, 2);
+	sd_card_resp(&resp);
+	cprintf("CMD2: %lx\n", resp);
+}
+
+uint16_t sd_get_rca() {
+	uint32_t resp;
+
+	sd_card_command(0, 3);
+	resp = 0;
+	sd_card_resp(&resp);
+
+	//cprintf("CMD3: %lx\n", resp);
+
+	return resp >> 16;
+}
+
+uint16_t sd_select_card(uint16_t rca) {
+	uint32_t resp;
+
+	sd_card_command((uint32_t)rca << 16, 7);
+	sd_card_resp(&resp);
+
+	return (uint16_t) resp;
+}
+
+uint16_t sd_get_status(uint16_t rca) {
+	uint32_t resp;
+
+	sd_card_command((uint32_t)rca << 16, 13);
+	sd_card_resp(&resp);
+
+	return (uint16_t) resp;
+}
+
+void sd_readblock(uint32_t addr, void* buf) {
+	uint32_t resp;
+	int i;
+
+	sd_card_command(addr, 17);
+	sd_card_resp(&resp);
+	//cprintf("CMD17: %lx\n", resp);
+
+	sd_card_wait_for_data();
+
+	//cprintf("Read data: \n");
+	for (i = 0; i < 512; i++){
+		((uint8_t*)buf)[i] = sd_card_read_byte();
+	}
+
+	//cprintf("\n");
+}
diff --git a/sw/bootloader/devices/sd_card.h b/sw/bootloader/devices/sd_card.h
new file mode 100644
index 0000000..7e3cb00
--- /dev/null
+++ b/sw/bootloader/devices/sd_card.h
@@ -0,0 +1,18 @@
+#ifndef _SD_CARD_H
+#define _SD_CARD_H
+
+#include <stdint.h>
+
+void sd_init();
+uint16_t sd_get_rca();
+uint16_t sd_select_card(uint16_t rca);
+uint16_t sd_get_status(uint16_t rca);
+void sd_readblock(uint32_t addr, void* buf);
+
+void sd_card_command(uint32_t arg, uint8_t cmd);
+
+void sd_card_resp(uint32_t* resp);
+uint8_t sd_card_read_byte();
+void sd_card_wait_for_data();
+
+#endif
diff --git a/sw/bootloader/devices/sd_card_asm.s b/sw/bootloader/devices/sd_card_asm.s
new file mode 100644
index 0000000..fe4f4e2
--- /dev/null
+++ b/sw/bootloader/devices/sd_card_asm.s
@@ -0,0 +1,66 @@
+.include "io.inc65"
+
+.importzp sp, sreg, ptr1
+
+.export _sd_card_command
+.export _sd_card_resp
+.export _sd_card_read_byte
+.export _sd_card_wait_for_data
+
+.autoimport on
+
+.code
+
+; Send sd card command.
+; command is in A register, the args are on the stack
+; I think the order is high byte first?
+_sd_card_command:
+    pha
+
+    jsr popeax
+    sta SD_ARG
+    stx SD_ARG+1
+    lda sreg
+    sta SD_ARG+2
+    lda sreg+1
+    sta SD_ARG+3
+
+    pla
+    sta SD_CMD
+    rts
+
+; void sd_card_resp(uint32_t* resp);
+_sd_card_resp:
+        phy
+        sta ptr1        ; store pointer
+        stx ptr1+1
+@1:     lda SD_CMD      ; wait for status flag
+        and #$01
+        beq @1
+        lda SD_ARG
+        ldy #$0
+        sta (ptr1),y
+        lda SD_ARG+1
+        iny
+        sta (ptr1),y
+        lda SD_ARG+2
+        iny
+        sta (ptr1),y
+        lda SD_ARG+3
+        iny
+        sta (ptr1),y
+        ply
+        rts
+
+_sd_card_read_byte:
+        lda SD_DATA
+        ldx #$00
+        rts
+
+_sd_card_wait_for_data:
+        pha
+@1:     lda SD_CMD      ; wait for status flag
+        and #$02
+        beq @1
+        pla
+        rts
\ No newline at end of file
diff --git a/sw/bootloader/devices/uart.h b/sw/bootloader/devices/uart.h
new file mode 100644
index 0000000..e8f06db
--- /dev/null
+++ b/sw/bootloader/devices/uart.h
@@ -0,0 +1,13 @@
+#ifndef _UART_H
+#define _UART_H
+
+#include <stdint.h>
+
+void uart_txb(uint8_t val);
+void uart_txb_block(uint8_t val);
+
+uint8_t uart_rxb();
+
+uint8_t uart_status();
+
+#endif
\ No newline at end of file
diff --git a/sw/bootloader/devices/uart.s b/sw/bootloader/devices/uart.s
new file mode 100644
index 0000000..c852f7e
--- /dev/null
+++ b/sw/bootloader/devices/uart.s
@@ -0,0 +1,36 @@
+.include "io.inc65"
+
+.importzp sp, sreg
+
+.export _uart_txb, _uart_txb_block
+.export _uart_rxb
+.export _uart_status
+
+.autoimport	on
+
+.code
+
+; @in A: byte to transmit
+; Transmits a byte over the UART
+_uart_txb:
+        sta UART_TXB        ; Just write value, don't wait
+        rts
+
+_uart_txb_block:
+        pha
+        sta UART_TXB        ; Write value
+@1:     lda UART_STATUS     ; Wait for status[0] to be 0
+        bit #$01
+        bne @1
+        pla
+        rts
+
+_uart_rxb:
+        lda UART_RXB        ; Read value
+        ldx #$00
+        rts
+
+_uart_status:
+        lda UART_STATUS
+        ldx #$00
+        rts
diff --git a/sw/bootloader/filesystem/fat.c b/sw/bootloader/filesystem/fat.c
new file mode 100644
index 0000000..fbbd50f
--- /dev/null
+++ b/sw/bootloader/filesystem/fat.c
@@ -0,0 +1,208 @@
+#include <conio.h>
+#include <string.h>
+#include <stdlib.h>
+
+#include "fat.h"
+#include "devices/sd_card.h"
+
+uint8_t fat_buf[512];
+
+static uint32_t fat_end_of_chain;
+
+static full_bpb_t bpb;
+
+static uint32_t data_region_start;
+
+void fat_init(){
+	int i;
+
+    sd_readblock(0, fat_buf);
+
+    memcpy(&bpb, &fat_buf[11], sizeof(full_bpb_t));
+
+	sd_readblock(1, fat_buf);
+	sd_readblock(32, fat_buf);
+
+	data_region_start = bpb.reserved_sectors + bpb.fat_count*bpb.sectors_per_fat_32;
+
+	sd_readblock(bpb.reserved_sectors, fat_buf);
+
+	//uncomment to view start of FAT
+
+	/*
+	for (i = 0; i < FAT_CLUSTERS_PER_SECTOR; i++) {
+		cprintf("%lx ", ((uint32_t*)fat_buf)[i]);
+	}
+	cprintf("\n\n");
+	*/
+
+	fat_end_of_chain = ((uint32_t*)fat_buf)[1] & FAT_EOC_CLUSTERMASK;
+	cprintf("End of chain indicator: %lx\n", fat_end_of_chain);
+}
+
+
+// make sure you have enough space.
+void fat_read_cluster(uint16_t cluster, uint8_t* buf) {
+	uint8_t i;
+
+	for (i = 0; i < bpb.sectors_per_cluster; i++) {
+		sd_readblock(data_region_start + i + (cluster - 2) * 8, buf+i*bpb.bytes_per_sector);
+	}
+}
+
+uint32_t fat_get_chain_value(uint16_t cluster) {
+	sd_readblock(bpb.reserved_sectors, fat_buf);
+	return ((uint32_t*)fat_buf)[cluster];
+}
+
+//the dentry is a double pointer because we need to increment it.
+void fat_parse_vfat_filenamename(vfat_dentry_t** vfat_dentry, char* name, uint32_t cluster) {
+	uint8_t i;
+	uint8_t overflows;
+	uint8_t done;
+	char* shift_name;
+	uint8_t sequence_number = (*vfat_dentry)->sequence_number;
+	overflows = 0;
+
+	for (;;){
+		shift_name = name + 13*((sequence_number & FAT_LFN_ENTRY_MASK) - 1);
+
+		done = 0;
+		for(i = 0; i < 5; i++) {
+			shift_name[i] = (*vfat_dentry)->filename0[i];
+			if (!shift_name[i]) {
+				done = 1;
+				break;
+			}
+		}
+
+		if (!done) {
+			for(i = 0; i < 6; i++) {
+				shift_name[i+5] = (*vfat_dentry)->filename1[i];
+				if (!shift_name[i+5]) {
+					done = 1;
+					break;
+				}
+			}
+		}
+
+		if (!done) {
+			for(i = 0; i < 2; i++) {
+				shift_name[i+11] = (*vfat_dentry)->filename2[i];
+				if (!shift_name[i+11]) {
+					done = 1;
+					break;
+				}
+			}
+		}
+
+		if ((sequence_number & FAT_LFN_ENTRY_MASK) == 1) {
+			break;
+		} else {
+			do {
+				(*vfat_dentry)++;
+				if ((uint8_t*)*vfat_dentry >= fat_buf + sizeof(fat_buf)) {
+					overflows++;
+					if (overflows == bpb.sectors_per_cluster) {
+						cprintf("Too many overflows, go back to fat!\n");		//TODO this
+						return;
+					}
+					sd_readblock(data_region_start + (cluster - 2) * 8 + overflows, fat_buf);
+					*vfat_dentry = (vfat_dentry_t*)fat_buf;
+				}
+			} while((*vfat_dentry)->sequence_number == 0xe5);
+			sequence_number = (*vfat_dentry)->sequence_number;
+		}
+	}
+
+}
+
+uint32_t fat_find_cluster_num(char* name, uint32_t cluster) {
+	vfat_dentry_t* vfat_dentry;
+	dos_dentry_t* dos_dentry;
+	char* vfat_name;
+
+	cprintf("Looking for file %s\n", name);
+
+	sd_readblock(data_region_start + (cluster - 2) * 8, fat_buf);
+	vfat_dentry = (vfat_dentry_t*)fat_buf;
+
+	vfat_name = (char*)malloc(FAT_MAX_FILE_NAME);
+
+	while(vfat_dentry->sequence_number == 0xe5)
+		vfat_dentry++;
+
+	vfat_name[0] = '\0';
+
+	while(vfat_dentry->sequence_number) {
+		fat_parse_vfat_filenamename(&vfat_dentry, vfat_name, cluster);
+		cprintf("Parsed filename: %s\n", vfat_name);
+
+		if (!strcmp(vfat_name, name)) {				//TODO this is probably unsafe, use strncmp
+			cprintf("Found file %s\n", vfat_name);
+			break;
+		} else {
+			vfat_dentry += 2;
+			while(vfat_dentry->sequence_number == 0xe5)
+				vfat_dentry++;
+		}
+	}
+
+	free(vfat_name);
+
+	if (!vfat_dentry->sequence_number) {
+		cprintf("File not found.\n");
+		return -1;
+	}
+
+	dos_dentry = (dos_dentry_t*) vfat_dentry + 1;	//dos entry follows last vfat entry
+
+	cluster = ((uint32_t)dos_dentry->first_cluster_h << 16) + dos_dentry->first_cluster_l;
+	cprintf("Cluster: %ld\n", cluster);
+
+	return cluster;
+}
+
+uint16_t fat_parse_path_to_cluster(char* filename) {
+	//basically start at the root folder and search through it
+	int i;
+	int len;
+	uint8_t dirs = 0;
+
+	char* spaced_filename;
+	char* fragment;
+
+	uint32_t cluster = 2;	//root chain is chain 2
+
+	if (filename[0] != '/') {
+		cprintf("Filename does not begin with '/'\n");
+		return 0;
+	}
+
+	filename++;
+	len = strlen(filename);
+	spaced_filename = (char*)malloc(len+1);	//need to account for null byte
+
+	for (i = 0; i <= len; i++) {
+		if (filename[i] == '/') {
+			spaced_filename[i] = '\0';
+			dirs++;
+		} else {
+			spaced_filename[i] = filename[i];
+		}
+	}
+
+	fragment = spaced_filename;
+
+	cprintf("Dirs: %d\n", dirs);
+
+	for (i = 0; i <= dirs; i++) {
+		cprintf("Fragment: %s\n", fragment);
+		cluster = fat_find_cluster_num(fragment, cluster);
+		fragment = spaced_filename + strlen(fragment) + 1;
+	}
+
+	free(spaced_filename);
+
+	return cluster;
+}
diff --git a/sw/bootloader/filesystem/fat.h b/sw/bootloader/filesystem/fat.h
new file mode 100644
index 0000000..2f7a640
--- /dev/null
+++ b/sw/bootloader/filesystem/fat.h
@@ -0,0 +1,122 @@
+#ifndef _FAT_H
+#define _FAT_H
+
+#include <stdint.h>
+
+extern uint8_t fat_buf[];
+
+#define FAT_MAX_FILE_NAME 255
+#define FAT_CLUSTERS_PER_SECTOR 128
+
+#define FAT_CLUSTERMASK 0x0fffffff
+#define FAT_EOC_CLUSTERMASK 0x0ffffff8
+
+#define FAT_LAST_LFN_MASK (1 << 6)
+#define FAT_LFN_ENTRY_MASK 0x1f
+
+typedef struct {
+    uint16_t bytes_per_sector;
+    uint8_t sectors_per_cluster;
+    uint16_t reserved_sectors;
+    uint8_t fat_count;
+    uint16_t max_dir_entries;
+    uint16_t total_sector_count;
+    uint8_t media_descriptor;
+    uint16_t sectors_per_fat;
+} dos_2_bpb_t;
+
+typedef struct {
+    dos_2_bpb_t bpb2;
+    uint16_t sectors_per_track;
+    uint16_t head_count;
+    uint32_t hidden_sector_count;
+    uint32_t logical_sector_count;
+    uint32_t sectors_per_fat;
+    uint16_t extended_flags;
+    uint16_t version;
+    uint32_t root_cluster;
+    uint16_t system_information;
+    uint16_t backup_boot_sector;
+    uint8_t reserved[12];
+} dos_3_bpb_t;
+
+typedef struct {
+    dos_3_bpb_t bpb3;
+    uint8_t drive_num;
+    uint8_t reserved;
+    uint8_t extended_signature;
+    uint32_t volume_id;
+    uint8_t partition_label[11];
+    uint8_t filesystem_type[8];
+} ebpb_t;
+
+typedef struct {
+    uint16_t bytes_per_sector;
+    uint8_t sectors_per_cluster;
+    uint16_t reserved_sectors;
+    uint8_t fat_count;
+    uint16_t max_dir_entries;
+    uint16_t total_sector_count;
+    uint8_t media_descriptor;
+    uint16_t sectors_per_fat_16;
+    uint16_t sectors_per_track;
+    uint16_t head_count;
+    uint32_t hidden_sector_count;
+    uint32_t logical_sector_count;
+    uint32_t sectors_per_fat_32;
+    uint16_t extended_flags;
+    uint16_t version;
+    uint32_t root_cluster;
+    uint16_t system_information;
+    uint16_t backup_boot_sector;
+    uint8_t reserved[12];
+    uint8_t drive_num;
+    uint8_t reserved2;
+    uint8_t extended_signature;
+    uint32_t volume_id;
+    uint8_t partition_label[11];
+    uint8_t filesystem_type[8];
+} full_bpb_t;
+
+typedef struct {
+    uint32_t sig;
+    uint8_t reserved[480];
+    uint32_t sig2;
+    uint32_t free_data_clusters;
+    uint32_t last_allocated_data_cluster;
+    uint32_t reserved2;
+    uint32_t sig3;
+} fs_info_sector_t;
+
+typedef struct {
+    uint8_t sequence_number;
+    uint16_t filename0[5];
+    uint8_t attributes;
+    uint8_t type;
+    uint8_t checksum;
+    uint16_t filename1[6];
+    uint16_t reserved;
+    uint16_t filename2[2];
+} vfat_dentry_t;
+
+typedef struct {
+    uint8_t filename[8];
+    uint8_t extension[3];
+    uint8_t attributes;
+    uint8_t reserved;
+    uint8_t create_time_10ms;
+    uint32_t create_date;
+    uint16_t access_date;
+    uint16_t first_cluster_h;
+    uint32_t modify_cluster;
+    uint16_t first_cluster_l;
+    uint32_t file_size;
+} dos_dentry_t;
+
+void fat_init();
+
+uint16_t fat_parse_path_to_cluster(char* filename);
+void fat_read_cluster(uint16_t cluster, uint8_t* buf);
+uint32_t fat_get_chain_value(uint16_t cluster);
+
+#endif
diff --git a/sw/bootloader/filesystem/o65.c b/sw/bootloader/filesystem/o65.c
new file mode 100644
index 0000000..9e76dc0
--- /dev/null
+++ b/sw/bootloader/filesystem/o65.c
@@ -0,0 +1,28 @@
+#include <conio.h>
+
+#include "o65.h"
+
+void o65_print_option(o65_opt_t* opt) {
+    int i;
+
+    cprintf("Option Length: %d\n", opt->olen);
+    cprintf("Option Type: %x ", opt->type);
+
+    switch (opt->type) {
+        case O65_OPT_FILENAME: cprintf("Filename\n"); break;
+        case O65_OPT_OS: cprintf("OS\n"); break;
+        case O65_OPT_ASSEMBLER: cprintf("Assembler\n"); break;
+        case O65_OPT_AUTHOR: cprintf("Author\n"); break;
+        case O65_OPT_DATE: cprintf("Creation Date\n"); break;
+        default: cprintf("Invalid\n"); break;
+    }
+
+    if (opt->type != O65_OPT_OS) {
+        for (i = 0; i < opt->olen - 2; i++) {
+            cprintf("%c", opt->data[i]);
+        }
+    } else {
+        cprintf("%x", opt->data[0]);
+    }
+    cprintf("\n\n");
+}
diff --git a/sw/bootloader/filesystem/o65.h b/sw/bootloader/filesystem/o65.h
new file mode 100644
index 0000000..719305c
--- /dev/null
+++ b/sw/bootloader/filesystem/o65.h
@@ -0,0 +1,65 @@
+#ifndef _O65_H
+#define _O65_H
+
+#include <stdint.h>
+
+#define O65_NON_C64 0x0001
+#define O65_MAGIC_0 0x6f
+#define O65_MAGIC_1 0x36
+#define O65_MAGIC_2 0x35
+
+#define O65_OPT_FILENAME 0
+#define O65_OPT_OS 1
+#define O65_OPT_ASSEMBLER 2
+#define O65_OPT_AUTHOR 3
+#define O65_OPT_DATE 4
+
+#define O65_OS_OSA65 1
+#define O65_OS_LUNIX 2
+#define O65_OS_CC65 3
+#define O65_OS_OPENCBM 4
+#define O65_OS_SUPER6502 5
+
+typedef union {
+    struct {
+        int cpu : 1;
+        int reloc : 1;
+        int size : 1;
+        int obj : 1;
+        int simple : 1;
+        int chain : 1;
+        int bsszero : 1;
+        int cpu2 : 4;
+        int align : 2;
+    };
+    uint16_t _mode;
+} o65_mode_t;
+
+typedef struct {
+    uint16_t c64_marker;
+    uint8_t magic[3];
+    uint8_t version;
+
+    o65_mode_t mode;
+
+    uint16_t tbase;
+    uint16_t tlen;
+    uint16_t dbase;
+    uint16_t dlen;
+    uint16_t bbase;
+    uint16_t blen;
+    uint16_t zbase;
+    uint16_t zlen;
+    uint16_t stack;
+
+} o65_header_t;
+
+typedef struct {
+    uint8_t olen;
+    uint8_t type;
+    uint8_t data[1];    //This is actually variable length
+} o65_opt_t;
+
+void o65_print_option(o65_opt_t* opt);
+
+#endif
diff --git a/sw/bootloader/irq.c b/sw/bootloader/irq.c
new file mode 100644
index 0000000..4d2c0d8
--- /dev/null
+++ b/sw/bootloader/irq.c
@@ -0,0 +1,24 @@
+
+#include <stdint.h>
+#include <conio.h>
+
+#include "devices/interrupt.h"
+#include "devices/uart.h"
+
+char lastchar;
+
+
+void handle_irq() {
+    uint8_t status;
+
+    status = irq_get_status();
+
+    if (status & BUTTON) {
+        cputs("Button Interrupt!\n");
+        irq_set_status(status & ~BUTTON);
+    }
+    if (status & UART) {
+        lastchar = uart_rxb();
+        irq_set_status(status & ~UART);
+    }
+}
\ No newline at end of file
diff --git a/sw/bootloader/link.ld b/sw/bootloader/link.ld
new file mode 100644
index 0000000..46ee428
--- /dev/null
+++ b/sw/bootloader/link.ld
@@ -0,0 +1,35 @@
+MEMORY
+{
+  ZP:  start = $0,    size = $100,  type = rw, define = yes;
+  SDRAM: start = $200, size = $7cf0, type = rw, define = yes;
+  ROM: start = $8000, size = $8000, file = %O;
+}
+
+SEGMENTS {
+    ZEROPAGE: load = ZP,  type = zp,  define   = yes;
+    DATA:     load = ROM, type = rw,  define   = yes, run = SDRAM;
+    BSS:      load = SDRAM, type = bss, define   = yes;
+    HEAP:     load = SDRAM, type = bss, optional = yes;
+    STARTUP:  load = ROM, type = ro;
+    ONCE:     load = ROM, type = ro,  optional = yes;
+    CODE:     load = ROM, type = ro;
+    RODATA:   load = ROM, type = ro;
+    VECTORS:  load = ROM, type = ro,  start    = $FFFA;
+}
+
+FEATURES {
+    CONDES:    segment = STARTUP,
+               type    = constructor,
+               label   = __CONSTRUCTOR_TABLE__,
+               count   = __CONSTRUCTOR_COUNT__;
+    CONDES:    segment = STARTUP,
+               type    = destructor,
+               label   = __DESTRUCTOR_TABLE__,
+               count   = __DESTRUCTOR_COUNT__;
+}
+
+SYMBOLS {
+    # Define the stack size for the application
+    __STACKSIZE__:  value = $0200, type = weak;
+    __STACKSTART__: type = weak, value = $0800; # 2k stack
+}
diff --git a/sw/bootloader/main.c b/sw/bootloader/main.c
new file mode 100644
index 0000000..e6e8407
--- /dev/null
+++ b/sw/bootloader/main.c
@@ -0,0 +1,62 @@
+#include <stdint.h>
+#include <conio.h>
+#include <string.h>
+#include <stdlib.h>
+
+#include "devices/board_io.h"
+#include "devices/uart.h"
+#include "devices/mapper.h"
+#include "devices/sd_card.h"
+#include "filesystem/fat.h"
+
+#define KERNEL_LOAD_ADDR 0xD000
+
+uint8_t buf[512];
+
+int main() {
+	int i;
+	uint16_t rca;
+	char* filename;
+	uint32_t cluster;
+	uint8_t* kernel_load;
+
+	clrscr();
+	cprintf("boot\n");
+
+	for (i = 0; i < 16; i++){
+		//cprintf("Mapping %1xxxx to %2xxxx\n", i, i);
+		mapper_write(i, i);
+	}
+
+	cprintf("Enabling Mapper\n");
+	mapper_enable(1);
+
+	mapper_write(0x10, 0xd);
+	mapper_write(0x11, 0xe);
+	mapper_write(0x12, 0xf);
+
+	sd_init();
+
+	rca = sd_get_rca();
+	cprintf("rca: %x\n", rca);
+
+	sd_select_card(rca);
+
+	fat_init();
+
+	filename = (char*)malloc(FAT_MAX_FILE_NAME);
+
+	cluster = fat_parse_path_to_cluster("/kernel.bin");
+	for (kernel_load = (uint8_t*)KERNEL_LOAD_ADDR; cluster < FAT_CLUSTERMASK; kernel_load+=(8*512)) {
+		cprintf("cluster: %lx\n", cluster);
+		cprintf("Writing to %p\n", kernel_load);
+		fat_read_cluster(cluster, kernel_load);
+		cluster = fat_get_chain_value(cluster);
+	}
+
+	cprintf("Done!\n");
+
+	cprintf("Reset vector: %x\n", *((uint16_t*)0xfffc));
+
+	return 0;
+}
diff --git a/sw/bootloader/tests/test_main.c b/sw/bootloader/tests/test_main.c
new file mode 100644
index 0000000..99e3dd8
--- /dev/null
+++ b/sw/bootloader/tests/test_main.c
@@ -0,0 +1,89 @@
+#include <stdio.h>
+
+#include "devices/board_io.h"
+#include "devices/uart.h"
+#include "devices/interrupt.h"
+
+int main(void)
+{
+    int retval = 0;
+
+    int i;
+
+    printf("\nStarting tests...\n\n");
+
+    printf("Testing hex_set_8...\n");
+    for (i = 0; i < 3; i++) {
+        if (hex_set_8(i+1, i)) {
+            printf("Failed to write to idx %d!\n", i);
+            retval++;
+        }
+        if (*(uint8_t*)0x7ff0+i != i+1) {
+            printf("Incorrect value at idx %d!\n", i);
+            retval++;
+        }
+    }
+
+    if (!hex_set_8(0xab, 3)) {
+        printf("Writing to idx 3 should fail!\n");
+        retval++;
+    }
+    printf("Done!\n\n");
+
+    printf("Testing hex_set_16...\n");
+    if (hex_set_16(0xabcd)){
+        printf("Failed to write!\n");
+    }
+    if (*(uint16_t*)0x7ff0 != 0xabcd) {
+        printf("Incorrect value!\n", i);
+        retval++;
+    }
+    printf("Done!\n\n");
+
+    printf("Testing hex_set_24...\n");
+    if (hex_set_24(0xabcdef)){
+        printf("Failed to write!\n");
+    }
+    if (*(uint16_t*)0x7ff0 != 0xcdef && *(uint8_t*)0x7ff2 != 0xab) {
+        printf("Incorrect value!\n", i);
+        retval++;
+    }
+    printf("Done!\n\n");
+
+    printf("Testing hex_enable...\n");
+    hex_enable(0xa5);
+    if (*(uint8_t*)0x7ff3 != 0xa5) {
+        printf("Incorrect value!\n", i);
+        retval++;
+    }
+    printf("Done!\n\n");
+
+    printf("Testing uart_txb_block...\n");
+    *(uint8_t*)0x7ff5 = 0;
+    uart_txb_block(0xa5);
+    if (*(uint8_t*)0x7ff4 != 0xa5) {
+        printf("Incorrect value!\n", i);
+        retval++;
+    }
+    printf("Done!\n\n");
+
+    printf("Testing uart_status...\n");
+    *(uint8_t*)0x7ff5 = 0xa5;
+    if (uart_status() != 0xa5) {
+        printf("Incorrect value!\n", i);
+        retval++;
+    }
+    printf("Done!\n\n");
+
+
+    printf("Testing irq_get_status...\n");
+    *(uint8_t*)0x7fff = 0xa5;
+    if (irq_get_status() != 0xa5) {
+        printf("Incorrect value!\n", i);
+        retval++;
+    }
+    printf("Done!\n\n");
+
+
+    return retval != 0;
+}
\ No newline at end of file
diff --git a/sw/bootloader/vectors.s b/sw/bootloader/vectors.s
new file mode 100644
index 0000000..81ae6e0
--- /dev/null
+++ b/sw/bootloader/vectors.s
@@ -0,0 +1,14 @@
+; ---------------------------------------------------------------------------
+; vectors.s
+; ---------------------------------------------------------------------------
+;
+; Defines the interrupt vector table.
+
+.import    _init
+.import    _nmi_int, _irq_int
+
+.segment  "VECTORS"
+
+.addr      _nmi_int    ; NMI vector
+.addr      _init       ; Reset vector
+.addr      _irq_int    ; IRQ/BRK vector
\ No newline at end of file
diff --git a/sw/cc65 b/sw/cc65
new file mode 160000
index 0000000..23a984f
--- /dev/null
+++ b/sw/cc65
@@ -0,0 +1 @@
+Subproject commit 23a984f0ddd00712bb29bc0568e2e14cca637ed8
diff --git a/sw/fsdir/hello.txt b/sw/fsdir/hello.txt
new file mode 100644
index 0000000..270c611
--- /dev/null
+++ b/sw/fsdir/hello.txt
@@ -0,0 +1 @@
+hello, world!
diff --git a/sw/kernel/Makefile b/sw/kernel/Makefile
index a84801d..ce9ac23 100644
--- a/sw/kernel/Makefile
+++ b/sw/kernel/Makefile
@@ -1,11 +1,13 @@
-CC=cl65
-CFLAGS=-T -t none -I. --cpu "65C02"
+CC=../cc65/bin/cl65
+CFLAGS=-T -t super6502 -I. --cpu "65C02"
 test: CFLAGS=-T -t sim65c02 -I.
 LDFLAGS=-C link.ld -m $(NAME).map
-SIM=sim65
+SIM=../cc65/bin/sim65
 SIMARGS=-v -c -x 1000000
 
-NAME=bootrom
+FSDIR=../fsdir
+
+NAME=kernel
 
 TEST_BIN=test.bin
 BIN=$(NAME).bin
@@ -24,7 +26,8 @@ TEST_OBJS+=$(patsubst %.s,%.o,$(filter %s,$(TEST_SRCS)))
 TEST_OBJS+=$(patsubst %.c,%.o,$(filter %c,$(TEST_SRCS)))
 TEST_OBJS+=$(filter-out boot.o,$(filter-out main.o,$(filter-out vectors.o,$(OBJS))))
 
-all: $(HEX)
+all: $(BIN)
+	cp $(BIN) $(FSDIR)
 
 test: $(TEST_BIN)
 	$(SIM) $(SIMARGS) $(TEST_BIN)
@@ -53,4 +56,5 @@ $(LISTS):
 clean:
 	rm -rf $(OBJS) $(BIN) $(HEX) $(LISTS) $(NAME).map
 	rm -rf $(TEST_OBJS) $(TEST_BIN)
+	rm -rf $(FSDIR)/$(BIN)
 
diff --git a/sw/kernel/boot.s b/sw/kernel/boot.s
index 8cb6a79..a30963b 100644
--- a/sw/kernel/boot.s
+++ b/sw/kernel/boot.s
@@ -29,9 +29,10 @@ _init:    LDX     #$FF                 ; Initialize stack pointer to $01FF
 ; ---------------------------------------------------------------------------
 ; Set cc65 argument stack pointer
 
-          LDA     #<(__SDRAM_START__ + __SDRAM_SIZE__)
+          ;LDA     #<(__SDRAM_START__ + __SDRAM_SIZE__)
+          lda     #<($200 + $7cf0)
           STA     sp
-          LDA     #>(__SDRAM_START__ + __SDRAM_SIZE__)
+          LDA     #>($200 + $7cf0)
           STA     sp+1
 
 ; ---------------------------------------------------------------------------
diff --git a/sw/kernel/exec.c b/sw/kernel/exec.c
index 0824eb6..cbf1ccb 100644
--- a/sw/kernel/exec.c
+++ b/sw/kernel/exec.c
@@ -52,7 +52,7 @@ void exec(char* filename) {
 
 		ret = 0;
 
-		ret = (*exec)();
+		//ret = (*exec)();
 
 		cprintf("ret: %x\n", ret);
 
diff --git a/sw/kernel/filesystem/fat.c b/sw/kernel/filesystem/fat.c
index e5187f3..cdcbdfd 100644
--- a/sw/kernel/filesystem/fat.c
+++ b/sw/kernel/filesystem/fat.c
@@ -45,7 +45,7 @@ void fat_print_pbp_info(full_bpb_t* bpb){
 }
 
 void fat_init(){
-	int i;
+	//int i;
 
     sd_readblock(0, fat_buf);
 
diff --git a/sw/kernel/link.ld b/sw/kernel/link.ld
index 172f0a5..96adc89 100644
--- a/sw/kernel/link.ld
+++ b/sw/kernel/link.ld
@@ -2,12 +2,16 @@ MEMORY
 {
   ZP:  start = $0,    size = $100,  type = rw, define = yes;
   SDRAM: start = $200, size = $7cf0, type = rw, define = yes;
-  ROM: start = $8000, size = $8000, fill = yes, fillval = $ff, file = %O;
+  ROM: start = $D000, size = $3000, type = rw, define = yes;
+}
+
+FILES {
+    %O: format = bin;
 }
 
 SEGMENTS {
     ZEROPAGE: load = ZP,  type = zp,  define   = yes;
-    DATA:     load = ROM, type = rw,  define   = yes, run = SDRAM;
+    DATA:     load = SDRAM, type = rw,  define   = yes;
     BSS:      load = SDRAM, type = bss, define   = yes;
     HEAP:     load = SDRAM, type = bss, optional = yes;
     STARTUP:  load = ROM, type = ro;
diff --git a/sw/kernel/main.c b/sw/kernel/main.c
index 21850fd..1039619 100644
--- a/sw/kernel/main.c
+++ b/sw/kernel/main.c
@@ -13,20 +13,11 @@
 uint8_t buf[512];
 
 int main() {
-	int i;
 	uint16_t rca;
 	char* filename;
 
 	clrscr();
-	cprintf("Hello, world!\n");
-
-	for (i = 0; i < 16; i++){
-		//cprintf("Mapping %1xxxx to %2xxxx\n", i, i);
-		mapper_write(i, i);
-	}
-
-	cprintf("Enabling Mapper\n");
-	mapper_enable(1);
+	cprintf("Hello, world! Modified\n");
 
 	sd_init();
 
diff --git a/sw/script/copy_files.sh b/sw/script/copy_files.sh
new file mode 100644
index 0000000..17bc144
--- /dev/null
+++ b/sw/script/copy_files.sh
@@ -0,0 +1,19 @@
+DEVICE=/dev/mmcblk0
+TMPMOUNT=/tmp/sd
+FSDIR=fsdir
+
+V=
+
+echo "$(tput bold setaf 11)Mounting Device$(tput sgr 0)"
+mkdir $V -p $TMPMOUNT
+sudo mount $V $DEVICE $TMPMOUNT
+echo
+
+echo "$(tput bold setaf 11)Copying Files$(tput sgr 0)"
+sudo cp $V -r $FSDIR/* $TMPMOUNT
+echo
+
+echo "$(tput bold setaf 11)Unmounting Device$(tput sgr 0)"
+sudo umount $V $DEVICE
+rmdir $V $TMPMOUNT
+echo
\ No newline at end of file
diff --git a/sw/script/format_disk.sh b/sw/script/format_disk.sh
new file mode 100644
index 0000000..99b82a6
--- /dev/null
+++ b/sw/script/format_disk.sh
@@ -0,0 +1,23 @@
+#!/bin/bash
+
+BOOTLOADER=bootloader/bootloader.bin
+DEVICE=/dev/mmcblk0
+TMPBOOTSECT=/tmp/bootsect
+TMPMOUNT=/tmp/sd
+
+V=
+STATUS="status=none"
+
+echo "$(tput bold setaf 11)Creating Filesystem$(tput sgr 0)"
+sudo mkfs.vfat -F32 $DEVICE -n SUPER6502 $V
+echo
+
+echo "$(tput bold setaf 11)Modifying Boot Sector$(tput sgr 0)"
+sudo dd if=$DEVICE of=$TMPBOOTSECT bs=512 count=1 $STATUS
+sudo dd conv=notrunc if=$BOOTLOADER of=$DEVICE bs=512 skip=0 count=1 $STATUS
+sudo dd conv=notrunc if=$TMPBOOTSECT of=$DEVICE bs=1 skip=0 count=90 iflag=skip_bytes,count_bytes $STATUS
+sudo dd conv=notrunc if=$BOOTLOADER of=$DEVICE bs=1 skip=0 count=3 iflag=skip_bytes,count_bytes $STATUS
+echo
+
+echo "$(tput bold setaf 10)Done!$(tput sgr 0)"
+
diff --git a/sw/test_exec/Makefile b/sw/test_exec/Makefile
index cfcfe6d..a02cbfe 100644
--- a/sw/test_exec/Makefile
+++ b/sw/test_exec/Makefile
@@ -1,5 +1,5 @@
 CC=cl65
-CFLAGS=-T -t none -I. --cpu "65C02"
+CFLAGS=-T -t super6502 -I. --cpu "65C02"
 test: CFLAGS=-T -t sim65c02 -I.
 LDFLAGS=-C link.ld -m $(NAME).map
 SIM=sim65