Re: [PATCH] binder: prevent transactions to context manager from its own process.

[Jann Horn]

On Mon, Jul 15, 2019 at 9:18 PM Hridya Valsaraju <hridya@xxxxxxxxxx> wrote:
> Currently, a transaction to context manager from its own process
> is prevented by checking if its binder_proc struct is the same as
> that of the sender. However, this would not catch cases where the
> process opens the binder device again and uses the new fd to send
> a transaction to the context manager.
>
> Reported-by: syzbot+8b3c354d33c4ac78bfad@xxxxxxxxxxxxxxxxxxxxxxxxx
> Signed-off-by: Hridya Valsaraju <hridya@xxxxxxxxxx>
> ---
>  drivers/android/binder.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/android/binder.c b/drivers/android/binder.c
> index e4d25ebec5be..89b9cedae088 100644
> --- a/drivers/android/binder.c
> +++ b/drivers/android/binder.c
> @@ -3138,7 +3138,7 @@ static void binder_transaction(struct binder_proc *proc,
>                         else
>                                 return_error = BR_DEAD_REPLY;
>                         mutex_unlock(&context->context_mgr_node_lock);
> -                       if (target_node && target_proc == proc) {
> +                       if (target_node && target_proc->pid == proc->pid) {
>                                 binder_user_error("%d:%d got transaction to context manager from process owning it\n",
>                                                   proc->pid, thread->pid);
>                                 return_error = BR_FAILED_REPLY;

This isn't a valid fix.

For context, the syzkaller report at
<https://lore.kernel.org/lkml/000000000000afe2c70589526668@xxxxxxxxxx/>
triggered this WARN_ON() in binder_transaction_buffer_release() in the
BINDER_TYPE_FD case, which Todd added in 44d8047f1d87 ("binder: use
standard functions to allocate fds"):

    case BINDER_TYPE_FD: {
        /*
         * No need to close the file here since user-space
         * closes it for for successfully delivered
         * transactions. For transactions that weren't
         * delivered, the new fd was never allocated so
         * there is no need to close and the fput on the
         * file is done when the transaction is torn
         * down.
         */
        WARN_ON(failed_at &&
            proc->tsk == current->group_leader);
    } break;

That check seems to be attempting to detect cases where
binder_transaction() fails and rolls back a partial transaction sent
by a process to itself. I think the intent there is probably to catch
cases that would cause the check in the BINDER_TYPE_FDA case below to
trip up?

About this fix: This prevents a task from sending binder transactions
to the context manager if they're running in the same process. (By the
way, I don't understand why that's a problem, conceptually.) But you
can still open a binder device twice (binder_proc instances A and B)
from a process that does not own the context manager instance, pass a
binder object from A to the context manager, let the context manager
pass it to B, and then A can transact with the same-process B. So this
merely looks fixed because syzkaller isn't able to construct such a
complicated testcase. (I think you could also let A receive a handle
to itself and then transact with itself, but I haven't tested that.)


I think this fix should probably be reverted (unless you actually want
to prevent intra-process transactions, which would probably require a
bunch of ugly extra checks), the WARN_ON() should be removed, and the
BINDER_TYPE_FDA case should be adjusted to make its decision based on
a flag passed from its parent instead of guessing based on what
`current` is. Since it looks like because of this bug, an aborted
intra-process transaction containing BINDER_TYPE_FDA (e.g. via the
err_translate_failed or err_dead_proc_or_thread cases) will cause file
descriptors to unexpectedly be released in the caller, leading to a
file-descriptor use-after-free in userspace, the fix should probably
also be stable-backported. (It's probably not a huge problem in
practice though, given that only hwbinder uses BINDER_TYPE_FDA and you
need to have an intra-process transaction at the same time as
something like a thread going away, or something like that? I don't
fully understand the failure conditions for binder transactions.)


Here's a reproducer for triggering the WARN_ON() on git master. The
helper files binder.c and binder.h are attached.

=================
#define _GNU_SOURCE
#include <unistd.h>
#include <stdio.h>
#include <stdint.h>
#include <err.h>
#include <stdlib.h>
#include <sys/signal.h>
#include <sys/prctl.h>
#include "binder.h"

#define BINDER_PATH "/dev/binder/binder"

static void do_exit(int dummy) {
  _exit(1);
}

static uint32_t ref_a_from_manager;

int my_handler(struct binder_state *bs, struct binder_transaction_data *txn,
               struct binder_io *msg, struct binder_io *reply) {
  if (txn->code == 1) {
    ref_a_from_manager = bio_get_ref(msg);
    if (ref_a_from_manager == 0)
      errx(1, "manager received bogus message 1");
    binder_acquire(bs, ref_a_from_manager);
    printf("manager received handle 0x%x from A\n", ref_a_from_manager);
    return 0;
  } else if (txn->code == 2) {
    if (ref_a_from_manager == 0)
      errx(1, "B asked too early");
    bio_put_ref(reply, ref_a_from_manager);
    printf("manager is sending handle to B\n");
    return 0;
  } else {
    errx(1, "manager got unexpected message");
  }
}

int main(void) {
  if (signal(SIGCHLD, do_exit))
    err(1, "signal");

  struct binder_state *bs_mgr = binder_open(BINDER_PATH, 0x400000);
  if (bs_mgr == NULL)
    err(1, "binder_open()");
  if (binder_become_context_manager(bs_mgr))
    err(1, "become mgr");

  pid_t child = fork();
  if (child == -1)
    err(1, "fork");
  if (child == 0) {
    prctl(PR_SET_PDEATHSIG, SIGKILL);
    if (getppid() == 1) exit(0);

    /* create endpoint A and send message with handle to manager */
    {
      struct binder_state *bs_a = binder_open(BINDER_PATH, 0x400000);
      if (bs_a == NULL) err(1, "binder_open()");

        struct binder_io msg;
        struct binder_io reply;
        char data[0x1000];
        bio_init(&msg, data, sizeof(data), 4);
        bio_put_obj(&msg, (void*)1);
        if (binder_call(bs_a, &msg, &reply, 0, 1/*code*/))
          errx(1, "binder_call");
        binder_done(bs_a, &msg, &reply);
    }

    /* create endpoint B and retrieve handle from manager */
    struct binder_state *bs_b;
    uint32_t ref_a_from_b;
    {
      bs_b = binder_open(BINDER_PATH, 0x400000);
      if (bs_b == NULL) err(1, "binder_open()");

        struct binder_io msg;
        struct binder_io reply;
        char data[0x1000];
        bio_init(&msg, data, sizeof(data), 4);
        if (binder_call(bs_b, &msg, &reply, 0, 2/*code*/))
          errx(1, "binder_call");
        ref_a_from_b = bio_get_ref(&reply);
        if (ref_a_from_b == 0)
          errx(1, "B received bogus reply");
        binder_acquire(bs_b, ref_a_from_b);
        printf("B received handle 0x%x from manager\n", ref_a_from_b);
        binder_done(bs_b, &msg, &reply);
    }

    /* let B send a message with a valid FD and an invalid FD to A */
    {
      struct binder_io msg;
      struct binder_io reply;
      char data[0x1000];
      bio_init(&msg, data, sizeof(data), 4);
      bio_put_fd(&msg, 0); /*valid*/
      bio_put_fd(&msg, -1); /*invalid*/
      if (binder_call(bs_b, &msg, &reply, ref_a_from_b, 3/*code*/))
        errx(1, "binder_call");
    }

    exit(0);
  }

  binder_loop(bs_mgr, my_handler);
}
=================
/* Copyright 2008 The Android Open Source Project
 */
#ifndef _BINDER_H_
#define _BINDER_H_

#include <sys/ioctl.h>
#include </h/aosp-walleye/bionic/libc/kernel/uapi/linux/android/binder.h>

struct binder_state
{
    int fd;
    void *mapped;
    size_t mapsize;
};
struct binder_io
{
    char *data;            /* pointer to read/write from */
    binder_size_t *offs;   /* array of offsets */
    size_t data_avail;     /* bytes available in data buffer */
    size_t offs_avail;     /* entries available in offsets array */
    char *data0;           /* start of data buffer */
    binder_size_t *offs0;  /* start of offsets buffer */
    uint32_t flags;
    uint32_t unused;
    uint64_t buffers_size;
};
struct binder_death {
    void (*func)(struct binder_state *bs, void *ptr);
    void *ptr;
};
/* the one magic handle */
#define BINDER_SERVICE_MANAGER  0U
#define SVC_MGR_NAME "android.os.IServiceManager"
enum {
    /* Must match definitions in IBinder.h and IServiceManager.h */
    PING_TRANSACTION  = B_PACK_CHARS('_','P','N','G'),
    SVC_MGR_GET_SERVICE = 1,
    SVC_MGR_CHECK_SERVICE,
    SVC_MGR_ADD_SERVICE,
    SVC_MGR_LIST_SERVICES,
};
typedef int (*binder_handler)(struct binder_state *bs,
                              struct binder_transaction_data *txn,
                              struct binder_io *msg,
                              struct binder_io *reply);
struct binder_state *binder_open(char *device, size_t mapsize);
void binder_close(struct binder_state *bs);
/* initiate a blocking binder call
 * - returns zero on success
 */
int binder_call(struct binder_state *bs,
                struct binder_io *msg, struct binder_io *reply,
                uint32_t target, uint32_t code);
int binder_call_async(struct binder_state *bs,
                struct binder_io *msg,
                uint32_t target, uint32_t code);
int binder_read_reply(struct binder_state* bs,
                struct binder_io* reply);
int binder_read_reply_handler(struct binder_state* bs,
                struct binder_io* reply, binder_handler func);
/* release any state associate with the binder_io
 * - call once any necessary data has been extracted from the
 *   binder_io after binder_call() returns
 * - can safely be called even if binder_call() fails
 */
void binder_done(struct binder_state *bs,
                 struct binder_io *msg, struct binder_io *reply);

/* manipulate strong references */
void binder_acquire(struct binder_state *bs, uint32_t target);
void binder_release(struct binder_state *bs, uint32_t target);
void binder_increfs(struct binder_state *bs, uint32_t target);
void binder_decrefs(struct binder_state *bs, uint32_t target);
void binder_link_to_death(struct binder_state *bs, uint32_t target, struct binder_death *death);
void binder_loop(struct binder_state *bs, binder_handler func);
int binder_become_context_manager(struct binder_state *bs);
/* allocate a binder_io, providing a stack-allocated working
 * buffer, size of the working buffer, and how many object
 * offset entries to reserve from the buffer
 */
void bio_init(struct binder_io *bio, void *data,
           size_t maxdata, size_t maxobjects);
void bio_put_obj(struct binder_io *bio, void *ptr);
void bio_put_fd(struct binder_io *bio, int fd);
void bio_put_ref(struct binder_io *bio, uint32_t handle);
void bio_put_uint32(struct binder_io *bio, uint32_t n);
void bio_put_string16(struct binder_io *bio, const uint16_t *str);
void bio_put_string16_x(struct binder_io *bio, const char *_str);
void bio_put_string8_x(struct binder_io *bio, const char *_str);
void bio_put_cstring(struct binder_io *bio, const char *_str);
uint32_t bio_get_uint32(struct binder_io *bio);
uint16_t *bio_get_string16(struct binder_io *bio, size_t *sz);
char *bio_get_string8(struct binder_io *bio, size_t *sz);
uint32_t bio_get_ref(struct binder_io *bio);
uint32_t bio_get_ref_cookie(struct binder_io *bio, uint64_t* cookie);
void bio_put_buf(struct binder_io *bio, void *data, size_t len, int *buf_id);
void bio_put_sub_buf(struct binder_io *bio, int parent_id, int parent_offset, void *data, size_t len, int *buf_id);

int binder_write(struct binder_state *bs, void *data, size_t len);
#endif/* Copyright 2008 The Android Open Source Project
 */
#include <inttypes.h>
#include <stdbool.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <errno.h>
#include <unistd.h>
#include <fcntl.h>
#include <err.h>
#include <sys/mman.h>
#include <sys/wait.h>
#include "binder.h"
#define MAX_BIO_SIZE (1 << 30)
#define TRACE 0
#define LOG_TAG "Binder"
#define ALOGE(...)
#define ALOGW(...)
#define ALOGD(...)
void bio_init_from_txn(struct binder_io *io, struct binder_transaction_data *txn);
#if TRACE
void hexdump(void *_data, size_t len)
{
    unsigned char *data = _data;
    size_t count;
    for (count = 0; count < len; count++) {
        if ((count & 15) == 0)
            fprintf(stderr,"%04zu:", count);
        fprintf(stderr," %02x %c", *data,
                (*data < 32) || (*data > 126) ? '.' : *data);
        data++;
        if ((count & 15) == 15)
            fprintf(stderr,"\n");
    }
    if ((count & 15) != 0)
        fprintf(stderr,"\n");
}
void binder_dump_txn(struct binder_transaction_data *txn)
{
    struct flat_binder_object *obj;
    binder_size_t *offs = (binder_size_t *)(uintptr_t)txn->data.ptr.offsets;
    size_t count = txn->offsets_size / sizeof(binder_size_t);
    fprintf(stderr,"  target %016"PRIx64"  cookie %016"PRIx64"  code %08x  flags %08x\n",
            (uint64_t)txn->target.ptr, (uint64_t)txn->cookie, txn->code, txn->flags);
    fprintf(stderr,"  pid %8d  uid %8d  data %"PRIu64"  offs %"PRIu64"\n",
            txn->sender_pid, txn->sender_euid, (uint64_t)txn->data_size, (uint64_t)txn->offsets_size);
    hexdump((void *)(uintptr_t)txn->data.ptr.buffer, txn->data_size);
    while (count--) {
        obj = (struct flat_binder_object *) (((char*)(uintptr_t)txn->data.ptr.buffer) + *offs++);
        fprintf(stderr,"  - type %08x  flags %08x  ptr %016"PRIx64"  cookie %016"PRIx64"\n",
                obj->hdr.type, obj->flags, (uint64_t)obj->binder, (uint64_t)obj->cookie);
    }
}
#define NAME(n) case n: return #n
const char *cmd_name(uint32_t cmd)
{
    switch(cmd) {
        NAME(BR_NOOP);
        NAME(BR_TRANSACTION_COMPLETE);
        NAME(BR_INCREFS);
        NAME(BR_ACQUIRE);
        NAME(BR_RELEASE);
        NAME(BR_DECREFS);
        NAME(BR_TRANSACTION);
        NAME(BR_REPLY);
        NAME(BR_FAILED_REPLY);
        NAME(BR_DEAD_REPLY);
        NAME(BR_DEAD_BINDER);
    default: return "???";
    }
}
#else
#define hexdump(a,b) do{} while (0)
#define binder_dump_txn(txn)  do{} while (0)
#endif
#define BIO_F_SHARED    0x01  /* needs to be buffer freed */
#define BIO_F_OVERFLOW  0x02  /* ran out of space */
#define BIO_F_IOERROR   0x04
#define BIO_F_MALLOCED  0x08  /* needs to be free()'d */
struct binder_state *binder_open(char *device, size_t mapsize)
{
    struct binder_state *bs;
    struct binder_version vers;
    bs = malloc(sizeof(*bs));
    if (!bs) {
        errno = ENOMEM;
        return NULL;
    }
    bs->fd = open(device, O_RDWR | O_CLOEXEC);
    if (bs->fd < 0) {
        fprintf(stderr,"binder: cannot open device (%s)\n",
                strerror(errno));
        goto fail_open;
    }
    if ((ioctl(bs->fd, BINDER_VERSION, &vers) == -1) ||
        (vers.protocol_version != BINDER_CURRENT_PROTOCOL_VERSION)) {
        fprintf(stderr,
                "binder: kernel driver version (%d) differs from user space version (%d)\n",
                vers.protocol_version, BINDER_CURRENT_PROTOCOL_VERSION);
        goto fail_open;
    }
    bs->mapsize = mapsize;
    bs->mapped = mmap(NULL, mapsize, PROT_READ, MAP_PRIVATE, bs->fd, 0);
    if (bs->mapped == MAP_FAILED) {
        fprintf(stderr,"binder: cannot map device (%s)\n",
                strerror(errno));
        goto fail_map;
    }
    if (madvise(bs->mapped, mapsize, MADV_DOFORK)) err(1, "MADV_DOFORK");
    return bs;
fail_map:
    close(bs->fd);
fail_open:
    free(bs);
    return NULL;
}
void binder_close(struct binder_state *bs)
{
    munmap(bs->mapped, bs->mapsize);
    close(bs->fd);
    free(bs);
}
int binder_become_context_manager(struct binder_state *bs)
{
    return ioctl(bs->fd, BINDER_SET_CONTEXT_MGR, 0);
}
int binder_write(struct binder_state *bs, void *data, size_t len)
{
    struct binder_write_read bwr;
    int res;
    bwr.write_size = len;
    bwr.write_consumed = 0;
    bwr.write_buffer = (uintptr_t) data;
    bwr.read_size = 0;
    bwr.read_consumed = 0;
    bwr.read_buffer = 0;
    res = ioctl(bs->fd, BINDER_WRITE_READ, &bwr);
    if (res < 0) {
        fprintf(stderr,"binder_write: ioctl failed (%s)\n",
                strerror(errno));
    }
    return res;
}
void binder_send_reply(struct binder_state *bs,
                       struct binder_io *reply,
                       binder_uintptr_t buffer_to_free,
                       int status)
{
    printf("binder_send_reply(status=%d)\n", status);
    struct {
        uint32_t cmd_free;
        binder_uintptr_t buffer;
        uint32_t cmd_reply;
        struct binder_transaction_data_sg txn_sg;
    } __attribute__((packed)) data;
    data.cmd_free = BC_FREE_BUFFER;
    data.buffer = buffer_to_free;
    data.cmd_reply = BC_REPLY_SG;
    data.txn_sg.buffers_size = reply->buffers_size;
    data.txn_sg.transaction_data.target.ptr = 0;
    data.txn_sg.transaction_data.cookie = 0;
    data.txn_sg.transaction_data.code = 0;
    if (status) {
        data.txn_sg.transaction_data.flags = TF_STATUS_CODE;
        data.txn_sg.transaction_data.data_size = sizeof(int);
        data.txn_sg.transaction_data.offsets_size = 0;
        data.txn_sg.transaction_data.data.ptr.buffer = (uintptr_t)&status;
        data.txn_sg.transaction_data.data.ptr.offsets = 0;
    } else {
        data.txn_sg.transaction_data.flags = 0;
        data.txn_sg.transaction_data.data_size = reply->data - reply->data0;
        data.txn_sg.transaction_data.offsets_size = ((char*) reply->offs) - ((char*) reply->offs0);
        data.txn_sg.transaction_data.data.ptr.buffer = (uintptr_t)reply->data0;
        data.txn_sg.transaction_data.data.ptr.offsets = (uintptr_t)reply->offs0;
        if (data.txn_sg.transaction_data.offsets_size) {
            *(volatile unsigned long *)data.txn_sg.transaction_data.data.ptr.offsets;
        }
        printf("offsets=%p, offsets_size=%lu\n",
            reply->offs0,
            (unsigned long)data.txn_sg.transaction_data.offsets_size);
    }
    binder_write(bs, &data, sizeof(data));
}
int binder_parse(struct binder_state *bs, struct binder_io *bio,
                 uintptr_t ptr, size_t size, binder_handler func)
{
    int r = 1;
    uintptr_t end = ptr + (uintptr_t) size;
    while (ptr < end) {
        uint32_t cmd = *(uint32_t *) ptr;
        ptr += sizeof(uint32_t);
#if TRACE
        fprintf(stderr,"%s:\n", cmd_name(cmd));
#endif
        switch(cmd) {
        case BR_NOOP:
            break;
        case BR_TRANSACTION_COMPLETE:
            break;
        case BR_INCREFS:
        case BR_ACQUIRE:
        case BR_RELEASE:
        case BR_DECREFS:
#if TRACE
            fprintf(stderr,"  %p, %p\n", (void *)ptr, (void *)(ptr + sizeof(void *)));
#endif
            ptr += sizeof(struct binder_ptr_cookie);
            break;
        case BR_TRANSACTION: {
            struct binder_transaction_data *txn = (struct binder_transaction_data *) ptr;
            if ((end - ptr) < sizeof(*txn)) {
                ALOGE("parse: txn too small!\n");
                return -1;
            }
            binder_dump_txn(txn);
            if (func) {
                unsigned rdata[256/4];
                struct binder_io msg;
                struct binder_io reply;
                int res;
                bio_init(&reply, rdata, sizeof(rdata), 4);
                bio_init_from_txn(&msg, txn);
                res = func(bs, txn, &msg, &reply);
                if ((txn->flags & 1) == 0) {
                    binder_send_reply(bs, &reply, txn->data.ptr.buffer, res);
                }
            }
            ptr += sizeof(*txn);
            break;
        }
        case BR_REPLY: {
            struct binder_transaction_data *txn = (struct binder_transaction_data *) ptr;
            if ((end - ptr) < sizeof(*txn)) {
                ALOGE("parse: reply too small!\n");
                return -1;
            }
            binder_dump_txn(txn);
            if (bio) {
                bio_init_from_txn(bio, txn);
                bio = 0;
            } else {
                /* todo FREE BUFFER */
            }
            ptr += sizeof(*txn);
            r = 0;
            break;
        }
        case BR_DEAD_BINDER: {
            struct binder_death *death = (struct binder_death *)(uintptr_t) *(binder_uintptr_t *)ptr;
            ptr += sizeof(binder_uintptr_t);
            death->func(bs, death->ptr);
            break;
        }
        case BR_FAILED_REPLY:
            r = -1;
            break;
        case BR_DEAD_REPLY:
            r = -1;
            break;
        default:
            ALOGE("parse: OOPS %d\n", cmd);
            return -1;
        }
    }
    return r;
}
void binder_acquire(struct binder_state *bs, uint32_t target)
{
    uint32_t cmd[2];
    cmd[0] = BC_ACQUIRE;
    cmd[1] = target;
    binder_write(bs, cmd, sizeof(cmd));
}
void binder_release(struct binder_state *bs, uint32_t target)
{
    uint32_t cmd[2];
    cmd[0] = BC_RELEASE;
    cmd[1] = target;
    binder_write(bs, cmd, sizeof(cmd));
}
void binder_increfs(struct binder_state *bs, uint32_t target)
{
    uint32_t cmd[2];
    cmd[0] = BC_INCREFS;
    cmd[1] = target;
    binder_write(bs, cmd, sizeof(cmd));
}
void binder_decrefs(struct binder_state *bs, uint32_t target)
{
    uint32_t cmd[2];
    cmd[0] = BC_DECREFS;
    cmd[1] = target;
    binder_write(bs, cmd, sizeof(cmd));
}
void binder_link_to_death(struct binder_state *bs, uint32_t target, struct binder_death *death)
{
    struct {
        uint32_t cmd;
        struct binder_handle_cookie payload;
    } __attribute__((packed)) data;
    data.cmd = BC_REQUEST_DEATH_NOTIFICATION;
    data.payload.handle = target;
    data.payload.cookie = (uintptr_t) death;
    binder_write(bs, &data, sizeof(data));
}
int binder_call(struct binder_state *bs,
                struct binder_io *msg, struct binder_io *reply,
                uint32_t target, uint32_t code)
{
    int res;
    struct binder_write_read bwr;
    struct {
        uint32_t cmd;
        struct binder_transaction_data_sg txn_sg;
    } __attribute__((packed)) writebuf;
    unsigned readbuf[32];
    if (msg->flags & BIO_F_OVERFLOW) {
        fprintf(stderr,"binder: txn buffer overflow\n");
        goto fail;
    }
    writebuf.cmd = BC_TRANSACTION_SG;
    writebuf.txn_sg.buffers_size = msg->buffers_size;
    writebuf.txn_sg.transaction_data.target.handle = target;
    writebuf.txn_sg.transaction_data.code = code;
    writebuf.txn_sg.transaction_data.flags = TF_ACCEPT_FDS;
    writebuf.txn_sg.transaction_data.data_size = msg->data - msg->data0;
    writebuf.txn_sg.transaction_data.offsets_size = ((char*) msg->offs) - ((char*) msg->offs0);
    writebuf.txn_sg.transaction_data.data.ptr.buffer = (uintptr_t)msg->data0;
    writebuf.txn_sg.transaction_data.data.ptr.offsets = (uintptr_t)msg->offs0;
    bwr.write_size = sizeof(writebuf);
    bwr.write_consumed = 0;
    bwr.write_buffer = (uintptr_t) &writebuf;
    hexdump(msg->data0, msg->data - msg->data0);
    bool first_iter = false;
    for (;;) {
        bwr.read_size = sizeof(readbuf);
        bwr.read_consumed = 0;
        bwr.read_buffer = (uintptr_t) readbuf;

        if (first_iter) {
            bwr.read_size = 0;
            printf("%d forking...\n", getpid());
            pid_t child = fork();
            if (child == -1) err(1, "fork");
            if (child == 0) {

                printf("entering child: %d\n", getpid());
                res = ioctl(bs->fd, BINDER_WRITE_READ, &bwr);
                if (res < 0) {
                    fprintf(stderr,"binder: ioctl failed (%s)\n", strerror(errno));
                }
                if (bwr.write_consumed != bwr.write_size) {
                    errx(1, "write_consumed != write_size");
                }

                printf("child exiting\n");
                exit(0);
            }
            int status;
            if (wait(&status) != child) err(1, "wait for child");
            bwr.write_consumed = bwr.write_size;
            printf("child is dead\n");
        } else {
            res = ioctl(bs->fd, BINDER_WRITE_READ, &bwr);
            if (res < 0) {
                fprintf(stderr,"binder: ioctl failed (%s)\n", strerror(errno));
                goto fail;
            }
            res = binder_parse(bs, reply, (uintptr_t) readbuf, bwr.read_consumed, 0);
            if (res == 0) return 0;
            if (res < 0) goto fail;
        }

        first_iter = false;
    }
fail:
    memset(reply, 0, sizeof(*reply));
    reply->flags |= BIO_F_IOERROR;
    return -1;
}
int binder_call_async(struct binder_state *bs,
                struct binder_io *msg, uint32_t target, uint32_t code)
{
    int res;
    struct binder_write_read bwr;
    struct {
        uint32_t cmd;
        struct binder_transaction_data txn;
    } __attribute__((packed)) writebuf;
    if (msg->flags & BIO_F_OVERFLOW) {
        fprintf(stderr,"binder: txn buffer overflow\n");
        goto fail;
    }
    writebuf.cmd = BC_TRANSACTION;
    writebuf.txn.target.handle = target;
    writebuf.txn.code = code;
    writebuf.txn.flags = TF_ACCEPT_FDS;
    writebuf.txn.data_size = msg->data - msg->data0;
    writebuf.txn.offsets_size = ((char*) msg->offs) - ((char*) msg->offs0);
    writebuf.txn.data.ptr.buffer = (uintptr_t)msg->data0;
    writebuf.txn.data.ptr.offsets = (uintptr_t)msg->offs0;
    bwr.write_size = sizeof(writebuf);
    bwr.write_consumed = 0;
    bwr.write_buffer = (uintptr_t) &writebuf;
    bwr.read_size = 0;
    bwr.read_buffer = 0;
    bwr.read_consumed = 0;
    hexdump(msg->data0, msg->data - msg->data0);
    res = ioctl(bs->fd, BINDER_WRITE_READ, &bwr);
    if (res < 0) {
        fprintf(stderr,"binder: ioctl failed (%s)\n", strerror(errno));
        goto fail;
    }
    return 0;
fail:
    return -1;
}


int binder_read_reply(struct binder_state* bs,
                struct binder_io* reply)
{
    int res;
    struct binder_write_read bwr;
    unsigned readbuf[32];
    bwr.write_size = 0;
    bwr.write_consumed = 0;
    bwr.write_buffer = 0;
    for (;;) {
        bwr.read_size = sizeof(readbuf);
        bwr.read_consumed = 0;
        bwr.read_buffer = (uintptr_t) readbuf;
        res = ioctl(bs->fd, BINDER_WRITE_READ, &bwr);
        if (res < 0) {
            fprintf(stderr,"binder: ioctl failed (%s)\n", strerror(errno));
            goto fail;
        }
        res = binder_parse(bs, reply, (uintptr_t) readbuf, bwr.read_consumed, 0);
        if (res == 0) return 0;
        if (res < 0) goto fail;
    }
fail:
    memset(reply, 0, sizeof(*reply));
    reply->flags |= BIO_F_IOERROR;
    return -1;
}

int binder_read_reply_handler(struct binder_state* bs,
                struct binder_io* reply, binder_handler func)
{
    int res;
    struct binder_write_read bwr;
    unsigned readbuf[32];
    bwr.write_size = 0;
    bwr.write_consumed = 0;
    bwr.write_buffer = 0;
    for (;;) {
        bwr.read_size = sizeof(readbuf);
        bwr.read_consumed = 0;
        bwr.read_buffer = (uintptr_t) readbuf;
        res = ioctl(bs->fd, BINDER_WRITE_READ, &bwr);
        if (res < 0) {
            fprintf(stderr,"binder: ioctl failed (%s)\n", strerror(errno));
            goto fail;
        }
        res = binder_parse(bs, reply, (uintptr_t) readbuf, bwr.read_consumed, func);
        if (res == 0) return 0;
        if (res < 0) goto fail;
    }
fail:
    memset(reply, 0, sizeof(*reply));
    reply->flags |= BIO_F_IOERROR;
    return -1;
}

void binder_loop(struct binder_state *bs, binder_handler func)
{
    int res;
    struct binder_write_read bwr;
    uint32_t readbuf[32];
    bwr.write_size = 0;
    bwr.write_consumed = 0;
    bwr.write_buffer = 0;
    readbuf[0] = BC_ENTER_LOOPER;
    binder_write(bs, readbuf, sizeof(uint32_t));
    for (;;) {
        bwr.read_size = sizeof(readbuf);
        bwr.read_consumed = 0;
        bwr.read_buffer = (uintptr_t) readbuf;
        res = ioctl(bs->fd, BINDER_WRITE_READ, &bwr);
        if (res < 0) {
            fprintf(stderr,"binder_loop: ioctl failed (%s)\n", strerror(errno));
            break;
        }
        res = binder_parse(bs, 0, (uintptr_t) readbuf, bwr.read_consumed, func);
        if (res == 0) {
            fprintf(stderr,"binder_loop: unexpected reply?!\n");
            break;
        }
        if (res < 0) {
            fprintf(stderr,"binder_loop: io error %d %s\n", res, strerror(errno));
            break;
        }
    }
}
void bio_init_from_txn(struct binder_io *bio, struct binder_transaction_data *txn)
{
    bio->data = bio->data0 = (char *)(intptr_t)txn->data.ptr.buffer;
    bio->offs = bio->offs0 = (binder_size_t *)(intptr_t)txn->data.ptr.offsets;
    bio->data_avail = txn->data_size;
    bio->offs_avail = txn->offsets_size / sizeof(size_t);
    bio->flags = BIO_F_SHARED;
}
void bio_init(struct binder_io *bio, void *data,
              size_t maxdata, size_t maxoffs)
{
    size_t n = maxoffs * sizeof(size_t);
    if (n > maxdata) {
        bio->flags = BIO_F_OVERFLOW;
        bio->data_avail = 0;
        bio->offs_avail = 0;
        return;
    }
    bio->data = bio->data0 = (char *) data + n;
    bio->offs = bio->offs0 = data;
    bio->data_avail = maxdata - n;
    bio->offs_avail = maxoffs;
    bio->flags = 0;
    bio->buffers_size = 0;
}
static void *bio_alloc(struct binder_io *bio, size_t size)
{
    size = (size + 3) & (~3);
    if (size > bio->data_avail) {
        bio->flags |= BIO_F_OVERFLOW;
        return NULL;
    } else {
        void *ptr = bio->data;
        bio->data += size;
        bio->data_avail -= size;
        return ptr;
    }
}
void binder_done(struct binder_state *bs,
                 struct binder_io *msg,
                 struct binder_io *reply)
{
    struct {
        uint32_t cmd;
        uintptr_t buffer;
    } __attribute__((packed)) data;
    if (reply->flags & BIO_F_SHARED) {
        printf("binder_done: freeing buffer\n");
        data.cmd = BC_FREE_BUFFER;
        data.buffer = (uintptr_t) reply->data0;
        binder_write(bs, &data, sizeof(data));
        reply->flags = 0;
        printf("binder_done: free done\n");
    }
}
static struct flat_binder_object *bio_alloc_obj(struct binder_io *bio)
{
    struct flat_binder_object *obj;
    obj = bio_alloc(bio, sizeof(*obj));
    if (obj && bio->offs_avail) {
        bio->offs_avail--;
        *bio->offs++ = ((char*) obj) - ((char*) bio->data0);
        return obj;
    }
    bio->flags |= BIO_F_OVERFLOW;
    return NULL;
}
static struct binder_fd_array_object *bio_alloc_fda(struct binder_io *bio)
{
    struct binder_fd_array_object *obj;
    obj = bio_alloc(bio, sizeof(*obj));
    if (obj && bio->offs_avail) {
        bio->offs_avail--;
        *bio->offs++ = ((char*) obj) - ((char*) bio->data0);
        return obj;
    }
    bio->flags |= BIO_F_OVERFLOW;
    return NULL;
}
static struct binder_buffer_object *bio_alloc_buf(struct binder_io *bio, int *buf_id)
{
    struct binder_buffer_object *obj;
    obj = bio_alloc(bio, sizeof(*obj));
    if (obj && bio->offs_avail) {
        bio->offs_avail--;
        if (buf_id) *buf_id = bio->offs - bio->offs0;
        *bio->offs++ = ((char*) obj) - ((char*) bio->data0);
        return obj;
    }
    bio->flags |= BIO_F_OVERFLOW;
    return NULL;
}
void bio_put_uint32(struct binder_io *bio, uint32_t n)
{
    uint32_t *ptr = bio_alloc(bio, sizeof(n));
    if (ptr)
        *ptr = n;
}
void bio_put_obj(struct binder_io *bio, void *ptr)
{
    struct flat_binder_object *obj;
    obj = bio_alloc_obj(bio);
    if (!obj)
        return;
    obj->flags = 0x7f | FLAT_BINDER_FLAG_ACCEPTS_FDS;
    obj->hdr.type = BINDER_TYPE_BINDER;
    obj->binder = (uintptr_t)ptr;
    obj->cookie = 0;
}
void bio_put_fd(struct binder_io *bio, int fd)
{
    struct flat_binder_object *obj;
    obj = bio_alloc_obj(bio);
    if (!obj)
        return;
    obj->flags = 0x7f | FLAT_BINDER_FLAG_ACCEPTS_FDS;
    obj->hdr.type = BINDER_TYPE_FD;
    obj->binder = (uintptr_t)fd;
    obj->cookie = 0;
}

void bio_put_buf(struct binder_io *bio, void *data, size_t len, int *buf_id) {
    struct binder_buffer_object *obj;
    obj = bio_alloc_buf(bio, buf_id);
    if (!obj)
        return;
    obj->hdr.type = BINDER_TYPE_PTR;
    obj->flags = 0;
    obj->buffer = (unsigned long)data;
    obj->length = len;
    obj->parent = 0; // unused
    obj->parent_offset = 0; // unused
    bio->buffers_size += (len+7)&~7UL; // TODO rounding blargh
}

void bio_put_sub_buf(struct binder_io *bio, int parent_id, int parent_offset, void *data, size_t len, int *buf_id) {
    struct binder_buffer_object *obj;
    obj = bio_alloc_buf(bio, buf_id);
    if (!obj)
        return;
    obj->hdr.type = BINDER_TYPE_PTR;
    obj->flags = BINDER_BUFFER_FLAG_HAS_PARENT;
    obj->buffer = (unsigned long)data;
    obj->length = len;
    obj->parent = parent_id;
    obj->parent_offset = parent_offset;
    bio->buffers_size += (len+7)&~7UL; // TODO rounding blargh
}

void bio_put_fda(struct binder_io *bio, int *fds, int fd_count) {
    int buf_id = -1;
    bio_put_buf(bio, fds, sizeof(int)*fd_count, &buf_id);
    if (buf_id == -1) errx(1, "bio_put_buf fail");
    struct binder_fd_array_object *obj;
    obj = bio_alloc_fda(bio);
    if (!obj)
        return;
    obj->hdr.type = BINDER_TYPE_FDA;
    obj->num_fds = fd_count;
    printf("fda->parent = %d\n", buf_id);
    obj->parent = buf_id;
    obj->parent_offset = 0;
}

void bio_put_ref(struct binder_io *bio, uint32_t handle)
{
    struct flat_binder_object *obj;
    if (handle)
        obj = bio_alloc_obj(bio);
    else
        obj = bio_alloc(bio, sizeof(*obj));
    if (!obj)
        return;
    obj->flags = 0x7f | FLAT_BINDER_FLAG_ACCEPTS_FDS;
    obj->hdr.type = BINDER_TYPE_HANDLE;
    obj->handle = handle;
    obj->cookie = 0;
}
void bio_put_string16(struct binder_io *bio, const uint16_t *str)
{
    size_t len;
    uint16_t *ptr;
    if (!str) {
        bio_put_uint32(bio, 0xffffffff);
        return;
    }
    len = 0;
    while (str[len]) len++;
    if (len >= (MAX_BIO_SIZE / sizeof(uint16_t))) {
        bio_put_uint32(bio, 0xffffffff);
        return;
    }
    /* Note: The payload will carry 32bit size instead of size_t */
    bio_put_uint32(bio, (uint32_t) len);
    len = (len + 1) * sizeof(uint16_t);
    ptr = bio_alloc(bio, len);
    if (ptr)
        memcpy(ptr, str, len);
}

void bio_put_cstring(struct binder_io *bio, const char *str)
{
    char* ptr = NULL;
    size_t len = 0;

    len = strlen(str) + 1;
    ptr = bio_alloc(bio, len);
    if (ptr)
        memcpy(ptr, str, len);
}

void bio_put_string16_x(struct binder_io *bio, const char *_str)
{
    unsigned char *str = (unsigned char*) _str;
    size_t len;
    uint16_t *ptr;
    if (!str) {
        bio_put_uint32(bio, 0xffffffff);
        return;
    }
    len = strlen(_str);
    if (len >= (MAX_BIO_SIZE / sizeof(uint16_t))) {
        bio_put_uint32(bio, 0xffffffff);
        return;
    }
    /* Note: The payload will carry 32bit size instead of size_t */
    bio_put_uint32(bio, len);
    ptr = bio_alloc(bio, (len + 1) * sizeof(uint16_t));
    if (!ptr)
        return;
    while (*str)
        *ptr++ = *str++;
    *ptr++ = 0;
}

void bio_put_string8_x(struct binder_io *bio, const char *_str)
{
    unsigned char *str = (unsigned char*) _str;
    size_t len;
    uint8_t *ptr;
    if (!str) {
        bio_put_uint32(bio, 0xffffffff);
        return;
    }
    len = strlen(_str);
    if (len >= (MAX_BIO_SIZE / sizeof(uint8_t))) {
        bio_put_uint32(bio, 0xffffffff);
        return;
    }
    /* Note: The payload will carry 32bit size instead of size_t */
    bio_put_uint32(bio, len);
    ptr = bio_alloc(bio, (len + 1) * sizeof(uint8_t));
    if (!ptr)
        return;
    while (*str)
        *ptr++ = *str++;
    *ptr++ = 0;
}
static void *bio_get(struct binder_io *bio, size_t size)
{
    size = (size + 3) & (~3);
    if (bio->data_avail < size){
        bio->data_avail = 0;
        bio->flags |= BIO_F_OVERFLOW;
        return NULL;
    }  else {
        void *ptr = bio->data;
        bio->data += size;
        bio->data_avail -= size;
        return ptr;
    }
}
uint32_t bio_get_uint32(struct binder_io *bio)
{
    uint32_t *ptr = bio_get(bio, sizeof(*ptr));
    return ptr ? *ptr : 0;
}
uint16_t *bio_get_string16(struct binder_io *bio, size_t *sz)
{
    size_t len;
    /* Note: The payload will carry 32bit size instead of size_t */
    len = (size_t) bio_get_uint32(bio);
    if (sz)
        *sz = len;
    return bio_get(bio, (len + 1) * sizeof(uint16_t));
}
char *bio_get_string8(struct binder_io *bio, size_t *sz) {
    size_t len;
    /* Note: The payload will carry 32bit size instead of size_t */
    len = (size_t) bio_get_uint32(bio);
    if (sz)
        *sz = len;
    return bio_get(bio, len + 1);  
}
static struct flat_binder_object *_bio_get_obj(struct binder_io *bio)
{
    size_t n;
    size_t off = bio->data - bio->data0;
    /* TODO: be smarter about this? */
    for (n = 0; n < bio->offs_avail; n++) {
        if (bio->offs[n] == off)
            return bio_get(bio, sizeof(struct flat_binder_object));
    }
    bio->data_avail = 0;
    bio->flags |= BIO_F_OVERFLOW;
    return NULL;
}
uint32_t bio_get_ref(struct binder_io *bio)
{
    struct flat_binder_object *obj;
    obj = _bio_get_obj(bio);
    if (!obj)
        return 0;
    if (obj->hdr.type == BINDER_TYPE_HANDLE)
        return obj->handle;
    return 0;
}


uint32_t bio_get_ref_cookie(struct binder_io *bio, uint64_t* cookie) {
    struct flat_binder_object *obj;
    obj = _bio_get_obj(bio);
    if (!obj)
        return 0;
    if (obj->hdr.type == BINDER_TYPE_HANDLE || obj->hdr.type == BINDER_TYPE_WEAK_HANDLE) {
        *cookie = obj->cookie;
        return obj->handle;
    }
    return 0;
}