On 4/14/21 9:26 AM, Shuah Khan wrote:
On 4/14/21 6:55 AM, Luis Chamberlain wrote:
Shuah, a question for you toward the end here.
On Wed, Apr 14, 2021 at 02:24:05PM +0530, Anirudh Rayabharam wrote:
This use-after-free happens when a fw_priv object has been freed but
hasn't been removed from the pending list (pending_fw_head). The next
time fw_load_sysfs_fallback tries to insert into the list, it ends up
accessing the pending_list member of the previoiusly freed fw_priv.
The root cause here is that all code paths that abort the fw load
don't delete it from the pending list. For example:
_request_firmware()
-> fw_abort_batch_reqs()
-> fw_state_aborted()
To fix this, delete the fw_priv from the list in __fw_set_state() if
the new state is DONE or ABORTED. This way, all aborts will remove
the fw_priv from the list. Accordingly, remove calls to list_del_init
that were being made before calling fw_state_(aborted|done).
Also, in fw_load_sysfs_fallback, don't add the fw_priv to the pending
list if it is already aborted. Instead, just jump out and return early.
Fixes: bcfbd3523f3c ("firmware: fix a double abort case with
fw_load_sysfs_fallback")
Reported-by: syzbot+de271708674e20930...@syzkaller.appspotmail.com
Tested-by: syzbot+de271708674e20930...@syzkaller.appspotmail.com
Signed-off-by: Anirudh Rayabharam <m...@anirudhrb.com>
---
Changes in v3:
Modified the patch to incorporate suggestions by Luis Chamberlain in
order to fix the root cause instead of applying a "band-aid" kind of
fix.
https://lore.kernel.org/lkml/20210403013143.gv4...@42.do-not-panic.com/
Changes in v2:
1. Fixed 1 error and 1 warning (in the commit message) reported by
checkpatch.pl. The error was regarding the format for referring to
another commit "commit <sha> ("oneline")". The warning was for line
longer than 75 chars.
---
drivers/base/firmware_loader/fallback.c | 8 ++++++--
drivers/base/firmware_loader/firmware.h | 6 +++++-
drivers/base/firmware_loader/main.c | 2 ++
3 files changed, 13 insertions(+), 3 deletions(-)
diff --git a/drivers/base/firmware_loader/fallback.c
b/drivers/base/firmware_loader/fallback.c
index 91899d185e31..73581b6998b4 100644
--- a/drivers/base/firmware_loader/fallback.c
+++ b/drivers/base/firmware_loader/fallback.c
@@ -94,7 +94,6 @@ static void __fw_load_abort(struct fw_priv *fw_priv)
if (fw_sysfs_done(fw_priv))
return;
- list_del_init(&fw_priv->pending_list);
fw_state_aborted(fw_priv);
}
@@ -280,7 +279,6 @@ static ssize_t firmware_loading_store(struct
device *dev,
* Same logic as fw_load_abort, only the DONE bit
* is ignored and we set ABORT only on failure.
*/
- list_del_init(&fw_priv->pending_list);
if (rc) {
fw_state_aborted(fw_priv);
written = rc;
@@ -513,6 +511,11 @@ static int fw_load_sysfs_fallback(struct
fw_sysfs *fw_sysfs, long timeout)
}
mutex_lock(&fw_lock);
+ if (fw_state_is_aborted(fw_priv)) {
+ mutex_unlock(&fw_lock);
+ retval = -EAGAIN;
+ goto out;
+ }
Thanks for the quick follow up!
This would regress commit 76098b36b5db1 ("firmware: send -EINTR on
signal abort on fallback mechanism") which I had mentioned in my follow
up email you posted a link to. It would regress it since the condition
is just being met earlier and you nullify the effort. So essentially
on Android you would make not being able to detect signal handlers
like the SIGCHLD signal sent to init, if init was the same process
dealing with the sysfs fallback firmware upload.
The way I dealt with this in my patch was I decided to return -EINTR
in the earlier case in the hunk you added, instead of -EAGAIN. In
addition to this, later on fw_load_sysfs_fallback() when
fw_sysfs_wait_timeout() is used that would also deal with checking
for error codes on wait, and only then check if it was a signal
that cancelled things (the check for -ERESTARTSYS). We therefore
only send to userspace -EAGAIN when the wait really did hit the
timeout.
But also note that my change added a check for
fw_state_is_aborted(fw_priv) inside fw_sysfs_wait_timeout(),
as that was a recently intended goal.
In either case I documented well *why* we do these error checks
before sending a code to userspace on fw_sysfs_wait_timeout() since
otherwise it would be easy to regress that code, so please also
document that as I did.
I'll re-iterate again also:
Shuah's commit 0542ad88fbdd81bb ("firmware loader: Fix
_request_firmware_load() return val for fw load abort") also
wanted to
distinguish the timeout vs -ENOMEM, but for some reason in the
timeout
case -EAGAIN was being sent back to userspace. I am no longer sure if
that is a good idea, but since we started doing that at some point I
guess we want to keep that behaviour.
Shuah, can you think of any reason to retain -EAGAIN other than you
introduced it here? If there's no real good reason I think it can
simplify the error handling here. But, we *would* change what we do
to userspace... and for that reason we may have to live with it.
As I recall the reason for this patch was to be able to differentiate
between timing out vs no memory case when driver was attempting to
load firmware. I wish I added why to the change log.
The code seems to have changed a lot since my commit. I will take a look
at the closely and let you know if this is still necessary late on
today.
Luis,
I did some digging and figured out why I added this timeout logic. Media
drivers expect request_firmware() timeout or fail.
It turns out I have notes saved on this problem.
When drivers attempt to load firmware while ext_fs is going through
recovery during boot or resume (after hibernate or suspend),
request_firmware() calls fail without this timeout handling leaving
drivers and devices that need firmware loaded in failed state.
I added this timeout handling so drivers can retry loading the firmware.
Several media drivers retry based on the timeoout return. This helps
differentiate timeout and other failures.
We still need this logic or at least we can't delete without verifying
that is indeed not needed.
thanks,
-- Shuah
