IRQ: ease out-of-tree migration to new irq_handler prototype
PATCH 1/1 IRQ: ease out-of-tree migration to new irq_handler prototype
From: [email protected]
Ease out-of-tree driver migration to new irq_handler prototype. Define empty 3rd argument macro for use in multi kernel version out-of-tree drivers going forward. Backportable drives can do:
(in a header)
#ifndef __PT_REGS
# define __PT_REGS , struct pt_regs *regs #endif
(in code body)
static irqreturn_t irq_handler(int irq, void *dev_id __PT_REGS)
---
Has this suggestion by Ingo Molnar been implemented? May I submit it? Is this how it's done?
include/linux/interrupt.h renamed from include/linux/interrupt.h.ORIG +8 -0
@@ -66,6 +66,14 @@typedef irqreturn_t (*irq_handler_t )(int ,void *); +/* +* Irq handler migration helper - empty 3rd argument +* #define __PT_REGS , struct pt_regs *regs +* for older kernel versions +*/ + +#define __PT_REGS +struct irqaction {irq_handler_t handler;unsigned long flags;
Bill
Signature · 7 lines
-------------------------------------------- William D Waddington Bainbridge Island, WA, USA [email protected] -------------------------------------------- "Even bugs...are unexpected signposts on the long road of creativity..." - Ken Burtch
Quoted 8 lines · William D Waddington
Ease out-of-tree driver migration to new
irq_handlerprototype. Define empty 3rd argument macro for use in multi kernel version out-of-tree drivers going forward. Backportable drives can do:(in a header)
#ifndef__PT_REGS# define __PT_REGS , struct pt_regs *regs #endif
Backportable drivers should check kernel version themselves and define __PT_REGS themselves.
Quoted 10 lines
(in code body)
staticirqreturn_tirq_handler(int irq, void *dev_id __PT_REGS)include/linux/interrupt.h renamed from include/linux/interrupt.h.ORIG +7
-0+/* +* Irq handler migration helper - empty 3rd argument +* #define __PT_REGS , struct pt_regs *regs +* for older kernel versions +*/ + +#define __PT_REGS
Quoted 6 lines · Alexey Dobriyan
Quoted 4 lines · William D Waddington
Ease out-of-tree driver migration to new
irq_handlerprototype. Define empty 3rd argument macro for use in multi kernel version out-of-tree drivers going forward. Backportable drives can do:
Quoted 8 lines
Quoted 4 lines
(in a header)
#ifndef__PT_REGS# define __PT_REGS , struct pt_regs *regs #endifBackportable drivers should check kernel version themselves and define
__PT_REGSthemselves.
I think I provided too much information :( It would be sufficiently helpful to just #define __PT_REGS <nothing> in interrupt.h to make things easier for low-life out-of-tree maintainers. There isn't any need to actualy detect version. Just detect __PT_REGS already defined.
The "in a header" above referred to the driver's header - #ifdefs in executable code really looks nasty IMHO.
The "#define __PT_REGS , ..." comment below was intended to be a "helpful" note to driver writers. Like I said, TMI.
Quoted 11 lines
include/linux/interrupt.h line 75
Quoted 11 lines
(in code body)
staticirqreturn_tirq_handler(int irq, void *dev_id __PT_REGS)include/linux/interrupt.h renamed from include/linux/interrupt.h.ORIG +7
-0+/* +* Irq handler migration helper - empty 3rd argument +* #define __PT_REGS , struct pt_regs *regs +* for older kernel versions +*/ + +#define __PT_REGS
How should I tidy this up - if it is acceptable at all? I'm a total noob struggling with a 1-line patch.
Bill
Signature · 7 lines
-------------------------------------------- William D Waddington Bainbridge Island, WA, USA [email protected] -------------------------------------------- "Even bugs...are unexpected signposts on the long road of creativity..." - Ken Burtch
Quoted 13 lines · William D Waddington
Quoted 12 lines · Alexey Dobriyan
Quoted 8 lines · William D Waddington
Ease out-of-tree driver migration to new
irq_handlerprototype. Define empty 3rd argument macro for use in multi kernel version out-of-tree drivers going forward. Backportable drives can do:(in a header)
#ifndef__PT_REGS# define __PT_REGS , struct pt_regs *regs #endif
Quoted 7 lines
Quoted 2 lines
Backportable drivers should check kernel version themselves and define
__PT_REGSthemselves.I think I provided too much information :( It would be sufficiently helpful to just #define
__PT_REGS<nothing> ininterrupt.hto make things easier for low-life out-of-tree maintainers. There isn't any need to actualy detect version. Just detect__PT_REGSalready defined.
Out-of-tree maintainer will have to change his code ANYWAY. And while he is doing that, he can spend 10 seconds to add 5-line version check.
More, if you've followed pt_regs removal patches, you'd noticed that some of them were not trivial. In this case version check is least of his worries.
Quoted 10 lines
The "in a header" above referred to the driver's header - #ifdefs in executable code really looks nasty IMHO.
The "#define
__PT_REGS, ..." comment below was intended to be a "helpful" note to driver writers. Like I said, TMI.Quoted 4 lines
Quoted 2 lines
(in code body)
staticirqreturn_tirq_handler(int irq, void *dev_id __PT_REGS)
Quoted 9 lines
Quoted 7 lines
include/linux/interrupt.h line 75
Quoted 7 lines
include/linux/interrupt.h renamed from include/linux/interrupt.h.ORIG +7
-0+/* +* Irq handler migration helper - empty 3rd argument +* #define __PT_REGS , struct pt_regs *regs +* for older kernel versions +*/ + +#define __PT_REGSHow should I tidy this up - if it is acceptable at all?
No, this is not acceptable.
Quoted 20 lines · Alexey Dobriyan
Quoted 18 lines · William D Waddington
Quoted 15 lines · Alexey Dobriyan
Quoted 8 lines · William D Waddington
Ease out-of-tree driver migration to new
irq_handlerprototype. Define empty 3rd argument macro for use in multi kernel version out-of-tree drivers going forward. Backportable drives can do:(in a header)
#ifndef__PT_REGS# define __PT_REGS , struct pt_regs *regs #endifBackportable drivers should check kernel version themselves and define
__PT_REGSthemselves.
Quoted 8 lines
Quoted 4 lines
I think I provided too much information :( It would be sufficiently helpful to just #define
__PT_REGS<nothing> ininterrupt.hto make things easier for low-life out-of-tree maintainers. There isn't any need to actualy detect version. Just detect__PT_REGSalready defined.Out-of-tree maintainer will have to change his code ANYWAY. And while he is doing that, he can spend 10 seconds to add 5-line version check.
I'm a little out of my depth here (obviously) but why is it "better" to require a version check rather than the kernel simply flagging the function change? Rather like HAVE_COMPAT_IOCTL.
With the proposed 1-line kernel patch my drivers build against kernels back to 2.6.9 (the earliest I have around here) without any version checking.
Quoted 3 lines
More, if you've followed
pt_regsremoval patches, you'd noticed that some of them were not trivial. In this case version check is least of his worries.
Ah. I don't use the *regs arg. The one line patch just keeps the compiler happy. And provides a "tidy" way to detect the interface change if necessary.
Quoted 22 lines
Quoted 11 lines
The "in a header" above referred to the driver's header - #ifdefs in executable code really looks nasty IMHO.
The "#define
__PT_REGS, ..." comment below was intended to be a "helpful" note to driver writers. Like I said, TMI.Quoted 4 lines
Quoted 2 lines
(in code body)
staticirqreturn_tirq_handler(int irq, void *dev_id __PT_REGS)Quoted 9 lines
Quoted 7 lines
include/linux/interrupt.h line 75
Quoted 7 lines
include/linux/interrupt.h renamed from include/linux/interrupt.h.ORIG +7
-0+/* +* Irq handler migration helper - empty 3rd argument +* #define __PT_REGS , struct pt_regs *regs +* for older kernel versions +*/ + +#define __PT_REGSHow should I tidy this up - if it is acceptable at all?
Quoted 1 line
No, this is not acceptable.
Oh well.
Thanks for your time,
Bill
Signature · 7 lines
-------------------------------------------- William D Waddington Bainbridge Island, WA, USA [email protected] -------------------------------------------- "Even bugs...are unexpected signposts on the long road of creativity..." - Ken Burtch