From: Carlos Llamas <cmllamas@google.com>
To: Hui Peng <benquike@gmail.com>
Cc: gregkh@linuxfoundation.org, arve@android.com, tkjos@android.com,
brauner@kernel.org, aliceryhl@google.com,
linux-kernel@vger.kernel.org
Subject: Re: [PATCH] binder: fix buffer/node leak and premature fd_install on read -EFAULT
Date: Wed, 23 Sep 2026 22:54:22 +0000 [thread overview]
Message-ID: <arRYnh6kMWPIKiAn@google.com> (raw)
In-Reply-To: <20260919213650.3316812-1-benquike@gmail.com>
On Sat, Sep 19, 2026 at 09:36:49PM +0000, Hui Peng wrote:
> In `binder_thread_read()`, `binder_apply_fd_fixups()` installs
> translated file descriptors into `current->files` via `fd_install()`
> BEFORE `put_user()` and `copy_to_user()` copy the `BR_TRANSACTION` /
> `BR_REPLY` command and `binder_transaction_data` payload to userspace.
>
> If `put_user()` or `copy_to_user()` returns `-EFAULT`:
>
> 1. Any file descriptors in `t->fd_fixups` have already been installed
> into the recipient process's file descriptor table and removed from
> `t->fd_fixups`, even though the transaction failed and the sender
> receives `BR_FAILED_REPLY`.
> 2. Unlike the `binder_apply_fd_fixups()` failure path immediately above
> it, the `put_user()` and `copy_to_user()` failure paths call
> `binder_cleanup_transaction()` without setting
> `t->buffer->transaction = NULL` or calling `binder_free_buf(proc,
> thread, buffer, true)`. Because `allow_user_free` is still `0`, the
> `binder_buffer` and all translated `binder_node`/`binder_ref`
> references inside it are permanently leaked, and for `TF_ONE_WAY`
> transactions `node->has_async_transaction` remains `true` forever,
> wedging all subsequent async transactions to that node.
>
> Split `fd_install()` out of `binder_apply_fd_fixups()` into
> `binder_fd_fixups_install(t)` called only after `put_user()` and
> `copy_to_user()` succeed, and free `t->buffer` via
> `binder_free_buf(proc, thread, buffer, true)` on `-EFAULT`.
>
> Fixes: 44d8047f1d87 ("binder: use standard functions to allocate fds")
> Assisted-by: LLM
> Signed-off-by: Hui Peng <benquike@gmail.com>
>
> ---
> drivers/android/binder.c | 30 +++++++++++++++++++++++-------
> 1 file changed, 23 insertions(+), 7 deletions(-)
>
> diff --git a/drivers/android/binder.c b/drivers/android/binder.c
> index fb9ff072bd6c..efff59853752 100644
> --- a/drivers/android/binder.c
> +++ b/drivers/android/binder.c
> @@ -4705,7 +4705,7 @@ static int binder_wait_for_work(struct binder_thread *thread,
> static int binder_apply_fd_fixups(struct binder_proc *proc,
> struct binder_transaction *t)
> {
> - struct binder_txn_fd_fixup *fixup, *tmp;
> + struct binder_txn_fd_fixup *fixup;
> int ret = 0;
>
> list_for_each_entry(fixup, &t->fd_fixups, fixup_entry) {
> @@ -4730,19 +4730,25 @@ static int binder_apply_fd_fixups(struct binder_proc *proc,
> goto err;
> }
> }
> - list_for_each_entry_safe(fixup, tmp, &t->fd_fixups, fixup_entry) {
> - fd_install(fixup->target_fd, fixup->file);
> - list_del(&fixup->fixup_entry);
> - kfree(fixup);
> - }
>
> - return ret;
> + return 0;
>
> err:
> binder_free_txn_fixups(t);
> return ret;
> }
>
> +static void binder_fd_fixups_install(struct binder_transaction *t)
> +{
> + struct binder_txn_fd_fixup *fixup, *tmp;
> +
> + list_for_each_entry_safe(fixup, tmp, &t->fd_fixups, fixup_entry) {
> + fd_install(fixup->target_fd, fixup->file);
> + list_del(&fixup->fixup_entry);
> + kfree(fixup);
> + }
> +}
> +
> static int binder_thread_read(struct binder_proc *proc,
> struct binder_thread *thread,
> binder_uintptr_t binder_buffer, size_t size,
> @@ -5121,26 +5127,36 @@ static int binder_thread_read(struct binder_proc *proc,
> trsize = sizeof(tr);
> }
> if (put_user(cmd, (uint32_t __user *)ptr)) {
> + struct binder_buffer *buffer = t->buffer;
> +
> if (t_from)
> binder_thread_dec_tmpref(t_from);
>
> + buffer->transaction = NULL;
> binder_cleanup_transaction(t, "put_user failed",
> BR_FAILED_REPLY);
> + binder_free_buf(proc, thread, buffer, true);
>
> return -EFAULT;
> }
> ptr += sizeof(uint32_t);
> if (copy_to_user(ptr, &tr, trsize)) {
> + struct binder_buffer *buffer = t->buffer;
> +
> if (t_from)
> binder_thread_dec_tmpref(t_from);
>
> + buffer->transaction = NULL;
> binder_cleanup_transaction(t, "copy_to_user failed",
> BR_FAILED_REPLY);
> + binder_free_buf(proc, thread, buffer, true);
>
> return -EFAULT;
> }
> ptr += trsize;
>
> + binder_fd_fixups_install(t);
> +
> trace_binder_transaction_received(t);
> binder_stat_br(proc, thread, cmd);
> binder_debug(BINDER_DEBUG_TRANSACTION,
> --
> 2.55.0.1082.g2b9226bbc0-goog
>
This seems correct. Although I assume you'll be sending a new patchset
addressing Greg's feedback right? e.g.
https://lore.kernel.org/all/2026092022-wad-vertigo-3361@gregkh/
next prev parent reply other threads:[~2026-09-23 22:54 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-09-19 21:36 Hui Peng
2026-09-23 22:54 ` Carlos Llamas [this message]
2026-09-24 10:00 ` [PATCH v2 0/2] binder: fix premature fd_install() and buffer leak " Hui Peng
2026-09-24 10:00 ` [PATCH v2 1/2] binder: defer fd_install() until after copy_to_user() in binder_thread_read() Hui Peng
2026-09-24 10:00 ` [PATCH v2 2/2] binder: free transaction buffer via binder_free_buf() on read -EFAULT Hui Peng
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=arRYnh6kMWPIKiAn@google.com \
--to=cmllamas@google.com \
--cc=aliceryhl@google.com \
--cc=arve@android.com \
--cc=benquike@gmail.com \
--cc=brauner@kernel.org \
--cc=gregkh@linuxfoundation.org \
--cc=linux-kernel@vger.kernel.org \
--cc=tkjos@android.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®