usb_udc_vbus_handler() can be invoked from interrupt context by irq
handlers of the gadget drivers, however, usb_udc_connect_control() has
to run in non-atomic context due to the following:
a. Some of the gadget driver implementations expect the ->pullup
callback to be invoked in non-atomic context.
b. usb_gadget_disconnect() acquires udc_lock which is a mutex.
Hence offload invocation of usb_udc_connect_control()
to workqueue.
UDC should not be pulled up unless gadget driver is bound. The new flag
"allow_connect" is now set by gadget_bind_driver() and cleared by
gadget_unbind_driver(). This prevents work item to pull up the gadget
even if queued when the gadget driver is already unbound.
Cc: stable(a)vger.kernel.org
Fixes: 1016fc0c096c ("USB: gadget: Fix obscure lockdep violation for udc_mutex")
Signed-off-by: Badhri Jagan Sridharan <badhri(a)google.com>
---
Changes since v1:
- Address Alan Stern's comment on usb_udc_vbus_handler invocation from
atomic context:
* vbus_events_lock is now a spinlock and allocations in
* usb_udc_vbus_handler are atomic now.
Changes since v2:
- Addressing Alan Stern's comments:
** connect_lock is now held by callers of
* usb_gadget_pullup_update_locked() and gadget_(un)bind_driver() does
* notdirectly hold the lock.
** Both usb_gadget_(dis)connect() and usb_udc_vbus_handler() would
* set/clear udc->vbus and invoke usb_gadget_pullup_update_locked.
** Add "unbinding" to prevent new connections after the gadget is being
* unbound.
Changes since v3:
** Made a minor cleanup which I missed to do in v3 in
* usb_udc_vbus_handler().
Changes since v4:
- Addressing Alan Stern's comments:
** usb_udc_vbus_handler() now offloads invocation of usb_udc_connect_control()
* from workqueue.
** Dropped vbus_events list as this was redundant. Updating to the
* latest value is suffice
Changes since v5:
- Addressing Alan Stern's comments:
** Squashed allow_connect logic to this patch.
** Fixed comment length to wrap at 76
** Cancelling vbus_work in del_gadget()
---
drivers/usb/gadget/udc/core.c | 29 +++++++++++++++++++++++++++--
1 file changed, 27 insertions(+), 2 deletions(-)
diff --git a/drivers/usb/gadget/udc/core.c b/drivers/usb/gadget/udc/core.c
index 52e6d2e84e35..d2e4f78c53e3 100644
--- a/drivers/usb/gadget/udc/core.c
+++ b/drivers/usb/gadget/udc/core.c
@@ -37,6 +37,9 @@ static const struct bus_type gadget_bus_type;
* @vbus: for udcs who care about vbus status, this value is real vbus status;
* for udcs who do not care about vbus status, this value is always true
* @started: the UDC's started state. True if the UDC had started.
+ * @allow_connect: Indicates whether UDC is allowed to be pulled up.
+ * Set/cleared by gadget_(un)bind_driver() after gadget driver is bound or
+ * unbound.
*
* This represents the internal data structure which is used by the UDC-class
* to hold information about udc driver and gadget together.
@@ -48,6 +51,8 @@ struct usb_udc {
struct list_head list;
bool vbus;
bool started;
+ bool allow_connect;
+ struct work_struct vbus_work;
};
static struct class *udc_class;
@@ -706,7 +711,7 @@ int usb_gadget_connect(struct usb_gadget *gadget)
goto out;
}
- if (gadget->deactivated) {
+ if (gadget->deactivated || !gadget->udc->allow_connect) {
/*
* If gadget is deactivated we only save new state.
* Gadget will be connected automatically after activation.
@@ -1086,6 +1091,13 @@ static void usb_udc_connect_control(struct usb_udc *udc)
usb_gadget_disconnect(udc->gadget);
}
+static void vbus_event_work(struct work_struct *work)
+{
+ struct usb_udc *udc = container_of(work, struct usb_udc, vbus_work);
+
+ usb_udc_connect_control(udc);
+}
+
/**
* usb_udc_vbus_handler - updates the udc core vbus status, and try to
* connect or disconnect gadget
@@ -1094,6 +1106,14 @@ static void usb_udc_connect_control(struct usb_udc *udc)
*
* The udc driver calls it when it wants to connect or disconnect gadget
* according to vbus status.
+ *
+ * This function can be invoked from interrupt context by irq handlers of
+ * the gadget drivers, however, usb_udc_connect_control() has to run in
+ * non-atomic context due to the following:
+ * a. Some of the gadget driver implementations expect the ->pullup
+ * callback to be invoked in non-atomic context.
+ * b. usb_gadget_disconnect() acquires udc_lock which is a mutex.
+ * Hence offload invocation of usb_udc_connect_control() to workqueue.
*/
void usb_udc_vbus_handler(struct usb_gadget *gadget, bool status)
{
@@ -1101,7 +1121,7 @@ void usb_udc_vbus_handler(struct usb_gadget *gadget, bool status)
if (udc) {
udc->vbus = status;
- usb_udc_connect_control(udc);
+ schedule_work(&udc->vbus_work);
}
}
EXPORT_SYMBOL_GPL(usb_udc_vbus_handler);
@@ -1328,6 +1348,7 @@ int usb_add_gadget(struct usb_gadget *gadget)
mutex_lock(&udc_lock);
list_add_tail(&udc->list, &udc_list);
mutex_unlock(&udc_lock);
+ INIT_WORK(&udc->vbus_work, vbus_event_work);
ret = device_add(&udc->dev);
if (ret)
@@ -1459,6 +1480,7 @@ void usb_del_gadget(struct usb_gadget *gadget)
flush_work(&gadget->work);
device_del(&gadget->dev);
ida_free(&gadget_id_numbers, gadget->id_number);
+ cancel_work_sync(&udc->vbus_work);
device_unregister(&udc->dev);
}
EXPORT_SYMBOL_GPL(usb_del_gadget);
@@ -1527,6 +1549,7 @@ static int gadget_bind_driver(struct device *dev)
if (ret)
goto err_start;
usb_gadget_enable_async_callbacks(udc);
+ udc->allow_connect = true;
usb_udc_connect_control(udc);
kobject_uevent(&udc->dev.kobj, KOBJ_CHANGE);
@@ -1558,6 +1581,8 @@ static void gadget_unbind_driver(struct device *dev)
kobject_uevent(&udc->dev.kobj, KOBJ_CHANGE);
+ udc->allow_connect = false;
+ cancel_work_sync(&udc->vbus_work);
usb_gadget_disconnect(gadget);
usb_gadget_disable_async_callbacks(udc);
if (gadget->irq)
base-commit: 922c0cb578ac9104a22c11a093cc1e0575c35a39
--
2.41.0.rc0.172.g3f132b7071-goog
When commit 19343b5bdd16 ("mm/page-writeback: introduce tracepoint for
wait_on_page_writeback()") repurposed the writeback_dirty_page trace event
as a template to create its new wait_on_page_writeback trace event, it
ended up opening a window to NULL pointer dereference crashes due to
the (infrequent) occurrence of a race where an access to a page in the
swap-cache happens concurrently with the moment this page is being
written to disk and the tracepoint is enabled:
BUG: kernel NULL pointer dereference, address: 0000000000000040
#PF: supervisor read access in kernel mode
#PF: error_code(0x0000) - not-present page
PGD 800000010ec0a067 P4D 800000010ec0a067 PUD 102353067 PMD 0
Oops: 0000 [#1] PREEMPT SMP PTI
CPU: 1 PID: 1320 Comm: shmem-worker Kdump: loaded Not tainted 6.4.0-rc5+ #13
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS edk2-20230301gitf80f052277c8-1.fc37 03/01/2023
RIP: 0010:trace_event_raw_event_writeback_folio_template+0x76/0xf0
Code: 4d 85 e4 74 5c 49 8b 3c 24 e8 06 98 ee ff 48 89 c7 e8 9e 8b ee ff ba 20 00 00 00 48 89 ef 48 89 c6 e8 fe d4 1a 00 49 8b 04 24 <48> 8b 40 40 48 89 43 28 49 8b 45 20 48 89 e7 48 89 43 30 e8 a2 4d
RSP: 0000:ffffaad580b6fb60 EFLAGS: 00010246
RAX: 0000000000000000 RBX: ffff90e38035c01c RCX: 0000000000000000
RDX: 0000000000000000 RSI: 0000000000000000 RDI: ffff90e38035c044
RBP: ffff90e38035c024 R08: 0000000000000002 R09: 0000000000000006
R10: ffff90e38035c02e R11: 0000000000000020 R12: ffff90e380bac000
R13: ffffe3a7456d9200 R14: 0000000000001b81 R15: ffffe3a7456d9200
FS: 00007f2e4e8a15c0(0000) GS:ffff90e3fbc80000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000000000000040 CR3: 00000001150c6003 CR4: 0000000000170ee0
Call Trace:
<TASK>
? __die+0x20/0x70
? page_fault_oops+0x76/0x170
? kernelmode_fixup_or_oops+0x84/0x110
? exc_page_fault+0x65/0x150
? asm_exc_page_fault+0x22/0x30
? trace_event_raw_event_writeback_folio_template+0x76/0xf0
folio_wait_writeback+0x6b/0x80
shmem_swapin_folio+0x24a/0x500
? filemap_get_entry+0xe3/0x140
shmem_get_folio_gfp+0x36e/0x7c0
? find_busiest_group+0x43/0x1a0
shmem_fault+0x76/0x2a0
? __update_load_avg_cfs_rq+0x281/0x2f0
__do_fault+0x33/0x130
do_read_fault+0x118/0x160
do_pte_missing+0x1ed/0x2a0
__handle_mm_fault+0x566/0x630
handle_mm_fault+0x91/0x210
do_user_addr_fault+0x22c/0x740
exc_page_fault+0x65/0x150
asm_exc_page_fault+0x22/0x30
This problem arises from the fact that the repurposed writeback_dirty_page
trace event code was written assuming that every pointer to mapping
(struct address_space) would come from a file-mapped page-cache object,
thus mapping->host would always be populated, and that was a valid case
before commit 19343b5bdd16. The swap-cache address space (swapper_spaces),
however, doesn't populate its ->host (struct inode) pointer, thus leading
to the crashes in the corner-case aforementioned.
commit 19343b5bdd16 ended up breaking the assignment of __entry->name and
__entry->ino for the wait_on_page_writeback tracepoint -- both dependent
on mapping->host carrying a pointer to a valid inode. The assignment of
__entry->name was fixed by commit 68f23b89067f ("memcg: fix a crash in
wb_workfn when a device disappears"), and this commit fixes the remaining
case, for __entry->ino.
Fixes: 19343b5bdd16 ("mm/page-writeback: introduce tracepoint for wait_on_page_writeback()")
Signed-off-by: Rafael Aquini <aquini(a)redhat.com>
---
include/trace/events/writeback.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/include/trace/events/writeback.h b/include/trace/events/writeback.h
index 86b2a82da546..54e353c9f919 100644
--- a/include/trace/events/writeback.h
+++ b/include/trace/events/writeback.h
@@ -68,7 +68,7 @@ DECLARE_EVENT_CLASS(writeback_folio_template,
strscpy_pad(__entry->name,
bdi_dev_name(mapping ? inode_to_bdi(mapping->host) :
NULL), 32);
- __entry->ino = mapping ? mapping->host->i_ino : 0;
+ __entry->ino = (mapping && mapping->host) ? mapping->host->i_ino : 0;
__entry->index = folio->index;
),
--
2.39.2