Commit
4cc1def5378c ("ath10k: pci: use mutex for diagnostic window CE
polling") introduced a regression where we try to sleep (grab a mutex)
in an atomic context:
[ 233.602619] BUG: sleeping function called from invalid context at kernel/locking/mutex.c:254
[ 233.602626] in_atomic(): 1, irqs_disabled(): 0, pid: 0, name: swapper/0
[ 233.602636] CPU: 0 PID: 0 Comm: swapper/0 Tainted: G W 5.1.0-rc2 #4
[ 233.602642] Hardware name: Google Scarlet (DT)
[ 233.602647] Call trace:
[ 233.602663] dump_backtrace+0x0/0x11c
[ 233.602672] show_stack+0x20/0x28
[ 233.602681] dump_stack+0x98/0xbc
[ 233.602690] ___might_sleep+0x154/0x16c
[ 233.602696] __might_sleep+0x78/0x88
[ 233.602704] mutex_lock+0x2c/0x5c
[ 233.602717] ath10k_pci_diag_read_mem+0x68/0x21c [ath10k_pci]
[ 233.602725] ath10k_pci_diag_read32+0x48/0x74 [ath10k_pci]
[ 233.602733] ath10k_pci_dump_registers+0x5c/0x16c [ath10k_pci]
[ 233.602741] ath10k_pci_fw_crashed_dump+0xb8/0x548 [ath10k_pci]
[ 233.602749] ath10k_pci_napi_poll+0x60/0x128 [ath10k_pci]
[ 233.602757] net_rx_action+0x140/0x388
[ 233.602766] __do_softirq+0x1b0/0x35c
[...]
ath10k_pci_fw_crashed_dump() is called from NAPI contexts, and firmware
memory dumps are retrieved using the diag memory interface.
A simple reproduction case is to run this on QCA6174A /
WLAN.RM.4.4.1-00132-QCARMSWP-1, which happens to be a way to b0rk the
firmware:
dd if=/sys/kernel/debug/ieee80211/phy0/ath10k/mem_value bs=4K count=1
of=/dev/null
(NB: simulated firmware crashes, via debugfs, don't trigger firmware
dumps.)
The fix is to move the crash-dump into a workqueue context, and avoid
relying on 'data_lock' for most mutual exclusion. We only keep using it
here for protecting 'fw_crash_counter', while the rest of the coredump
buffers are protected by a new 'dump_mutex'.
I've tested the above with simulated firmware crashes (debugfs 'reset'
file), real firmware crashes (the 'dd' command above), and a variety of
reboot and suspend/resume configurations on QCA6174A.
Reported here:
http://lkml.kernel.org/linux-wireless/
20190325202706.GA68720@google.com
Fixes: 4cc1def5378c ("ath10k: pci: use mutex for diagnostic window CE polling")
Signed-off-by: Brian Norris <briannorris@chromium.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
struct ath10k_ce_crash_data ce_data;
u32 addr, id;
- lockdep_assert_held(&ar->data_lock);
+ lockdep_assert_held(&ar->dump_mutex);
ath10k_err(ar, "Copy Engine register dump:\n");
goto err_free_wq;
mutex_init(&ar->conf_mutex);
+ mutex_init(&ar->dump_mutex);
spin_lock_init(&ar->data_lock);
INIT_LIST_HEAD(&ar->peers);
/* prevents concurrent FW reconfiguration */
struct mutex conf_mutex;
+ /* protects coredump data */
+ struct mutex dump_mutex;
+
/* protects shared structure data */
spinlock_t data_lock;
{
struct ath10k_fw_crash_data *crash_data = ar->coredump.fw_crash_data;
- lockdep_assert_held(&ar->data_lock);
+ lockdep_assert_held(&ar->dump_mutex);
if (ath10k_coredump_mask == 0)
/* coredump disabled */
if (!buf)
return NULL;
- spin_lock_bh(&ar->data_lock);
+ mutex_lock(&ar->dump_mutex);
dump_data = (struct ath10k_dump_file_data *)(buf);
strlcpy(dump_data->df_magic, "ATH10K-FW-DUMP",
sofar += sizeof(*dump_tlv) + crash_data->ramdump_buf_len;
}
- spin_unlock_bh(&ar->data_lock);
+ mutex_unlock(&ar->dump_mutex);
return dump_data;
}
__le32 reg_dump_values[REG_DUMP_COUNT_QCA988X] = {};
int i, ret;
- lockdep_assert_held(&ar->data_lock);
+ lockdep_assert_held(&ar->dump_mutex);
ret = ath10k_pci_diag_read_hi(ar, ®_dump_values[0],
hi_failure_state,
int ret, i;
u8 *buf;
- lockdep_assert_held(&ar->data_lock);
+ lockdep_assert_held(&ar->dump_mutex);
if (!crash_data)
return;
}
}
-static void ath10k_pci_fw_crashed_dump(struct ath10k *ar)
+static void ath10k_pci_fw_dump_work(struct work_struct *work)
{
+ struct ath10k_pci *ar_pci = container_of(work, struct ath10k_pci,
+ dump_work);
struct ath10k_fw_crash_data *crash_data;
+ struct ath10k *ar = ar_pci->ar;
char guid[UUID_STRING_LEN + 1];
- spin_lock_bh(&ar->data_lock);
+ mutex_lock(&ar->dump_mutex);
+ spin_lock_bh(&ar->data_lock);
ar->stats.fw_crash_counter++;
+ spin_unlock_bh(&ar->data_lock);
crash_data = ath10k_coredump_new(ar);
ath10k_ce_dump_registers(ar, crash_data);
ath10k_pci_dump_memory(ar, crash_data);
- spin_unlock_bh(&ar->data_lock);
+ mutex_unlock(&ar->dump_mutex);
queue_work(ar->workqueue, &ar->restart_work);
}
+static void ath10k_pci_fw_crashed_dump(struct ath10k *ar)
+{
+ struct ath10k_pci *ar_pci = ath10k_pci_priv(ar);
+
+ queue_work(ar->workqueue, &ar_pci->dump_work);
+}
+
void ath10k_pci_hif_send_complete_check(struct ath10k *ar, u8 pipe,
int force)
{
spin_lock_init(&ar_pci->ps_lock);
mutex_init(&ar_pci->ce_diag_mutex);
+ INIT_WORK(&ar_pci->dump_work, ath10k_pci_fw_dump_work);
+
timer_setup(&ar_pci->rx_post_retry, ath10k_pci_rx_replenish_retry, 0);
if (QCA_REV_6174(ar) || QCA_REV_9377(ar))
/* For protecting ce_diag */
struct mutex ce_diag_mutex;
+ struct work_struct dump_work;
+
struct ath10k_ce ce;
struct timer_list rx_post_retry;