Subject: [PATCH] mmc,sdio: Fix the panic due to devname NULL when calling pm_runtime_set_active()
Meet one panic as the below:
<1>[ 15.067350] BUG: unable to handle kernel NULL pointer dereference at (null)
<1>[ 15.074455] IP: [<c1496a42>] strlen+0x12/0x20
<4>[ 15.078803] *pde = 00000000
<0>[ 15.081674] Oops: 0000 [#1] PREEMPT SMP
<4>[ 15.101676] Pid: 5, comm: kworker/u:0 Tainted: G C 3.0.34-140729-g7f9d5c5 #1 Intel Corporation Medfield/BKB2
<4>[ 15.112282] EIP: 0060:[<c1496a42>] EFLAGS: 00010046 CPU: 0
<4>[ 15.117760] EIP is at strlen+0x12/0x20
<4>[ 15.121496] EAX: 00000000 EBX: f344cc04 ECX: ffffffff EDX: f344cc04
<4>[ 15.127754] ESI: c12bcee0 EDI: 00000000 EBP: f586fe74 ESP: f586fe70
<4>[ 15.134013] DS: 007b ES: 007b FS: 00d8 GS: 0000 SS: 0068
<0>[ 15.139406] Process kworker/u:0 (pid: 5, ti=f586e000 task=f585b440 task.ti=f586e000)
<0>[ 15.147140] Stack:
<4>[ 15.149141] f344cc04 f586feb0 c12bcf12 00000000 f586fe9c 00000000 00000007 00000082
<4>[ 15.156877] 00000092 00000002 c1b01ee4 f586feb8 c1635f31 f3b42330 c12bcee0 f344cc04
<4>[ 15.164616] f586fed0 c152f815 f3b42330 f3b42328 00000000 f344cc04 f589b804 00000000
<0>[ 15.172351] Call Trace:
<4>[ 15.174810] [<c12bcf12>] ftrace_raw_event_runtime_pm_status+0x32/0x140
<4>[ 15.181411] [<c1635f31>] ? sdio_enable_wide.part.8+0x61/0x80
<4>[ 15.187145] [<c12bcee0>] ? perf_trace_runtime_pm_usage+0x1a0/0x1a0
<4>[ 15.193407] [<c152f815>] __update_runtime_status+0x65/0x90
<4>[ 15.198968] [<c1531170>] __pm_runtime_set_status+0xe0/0x1b0
<4>[ 15.204621] [<c1637366>] mmc_attach_sdio+0x2f6/0x410
<4>[ 15.209666] [<c162f520>] mmc_rescan+0x240/0x2b0
<4>[ 15.214270] [<c12643ce>] process_one_work+0xfe/0x3f0
<4>[ 15.219311] [<c1242754>] ? wake_up_process+0x14/0x20
<4>[ 15.224357] [<c162f2e0>] ? mmc_detect_card_removed+0x80/0x80
<4>[ 15.230091] [<c12649c1>] worker_thread+0x121/0x2f0
<4>[ 15.234958] [<c12648a0>] ? rescuer_thread+0x1e0/0x1e0
<4>[ 15.240091] [<c12684cd>] kthread+0x6d/0x80
<4>[ 15.244264] [<c1268460>] ? __init_kthread_worker+0x30/0x30
<4>[ 15.245485] [<c186dc3a>] kernel_thread_helper+0x6/0x10
The reason is pm_runtime_set_active() is called before the device name
is set, and the dev name setting is done at mmc_add_card() laterly.
So when calling pm_runtime_set_active(), it will hit the strlen(devname==0)
which trigger the panic.
Here before calling pm_runtime_set_active(), set the dev name, although
it is duplicated with mmc_add_card(), but it do not break the original
design(commit 81968561b).
Signed-off-by: liu chuansheng <chuansheng.liu@xxxxxxxxx>
---
drivers/mmc/core/sdio.c | 9 +++++++++
1 files changed, 9 insertions(+), 0 deletions(-)
diff --git a/drivers/mmc/core/sdio.c b/drivers/mmc/core/sdio.c
index 2273ce6..73746af 100644
--- a/drivers/mmc/core/sdio.c
+++ b/drivers/mmc/core/sdio.c
@@ -1104,6 +1104,15 @@ int mmc_attach_sdio(struct mmc_host *host)
*/
if (host->caps & MMC_CAP_POWER_OFF_CARD) {
/*
+ * pm_runtime_set_active will use strlen(dev_name),
+ * we must set it in advance to avoid crash,
+ * although it is the duplication in mmc_add_card
+ * laterly.
+ */
+ dev_set_name(&card->dev, "%s:%04x", mmc_hostname(card->host),
+ card->rca);
+
+ /*
* Let runtime PM core know our card is active
*/
err = pm_runtime_set_active(&card->dev);
--
1.7.0.4
--
To unsubscribe from this list: send the line "unsubscribe linux-mmc" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
