blob: 07b1ab8e9e8905586e93337a9df9e8909d976ced (
plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
|
From 58c679541448da375ccc262585f3172e913811f3 Mon Sep 17 00:00:00 2001
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: Fri, 8 Nov 2013 17:34:54 +0100
Subject: [PATCH 035/347] usb: do no disable interrupts in giveback
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.19/older/patches-4.19.246-rt110.tar.xz
Since commit 94dfd7ed ("USB: HCD: support giveback of URB in tasklet
context") the USB code disables interrupts before invoking the complete
callback.
This should not be required the HCD completes the URBs either in hard-irq
context or in BH context. Lockdep may report false positives if one has two
HCDs (one completes in IRQ and the other in BH context) and is using the same
USB driver (device) with both HCDs. This is safe since the same URBs are never
mixed with those two HCDs.
Longeterm we should force all HCDs to complete in the same context.
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
---
drivers/usb/core/hcd.c | 3 ---
1 file changed, 3 deletions(-)
diff --git a/drivers/usb/core/hcd.c b/drivers/usb/core/hcd.c
index df661460e9f9..1a1487d312d0 100644
--- a/drivers/usb/core/hcd.c
+++ b/drivers/usb/core/hcd.c
@@ -1752,7 +1752,6 @@ static void __usb_hcd_giveback_urb(struct urb *urb)
struct usb_hcd *hcd = bus_to_hcd(urb->dev->bus);
struct usb_anchor *anchor = urb->anchor;
int status = urb->unlinked;
- unsigned long flags;
urb->hcpriv = NULL;
if (unlikely((urb->transfer_flags & URB_SHORT_NOT_OK) &&
@@ -1780,9 +1779,7 @@ static void __usb_hcd_giveback_urb(struct urb *urb)
* and no one may trigger the above deadlock situation when
* running complete() in tasklet.
*/
- local_irq_save(flags);
urb->complete(urb);
- local_irq_restore(flags);
usb_anchor_resume_wakeups(anchor);
atomic_dec(&urb->use_count);
--
2.36.1
|
