Why don't we have atomic_set_mask()?

Why don't we have atomic_set_mask()?

[KyongHo Cho]

Hi.

arch/arm/asm/atomic.h defines atomic_clear_mask() but its counterpart,
atomic_set_mask() is not defined.

Don't we need it?

Cho KyongHo.

Why don't we have atomic_set_mask()?

[Russell King - ARM Linux]

On Thu, Mar 17, 2011 at 08:36:15PM +0900, KyongHo Cho wrote:
> arch/arm/asm/atomic.h defines atomic_clear_mask() but its counterpart,
> atomic_set_mask() is not defined.

Nothing other than some s390 drivers seems to use it, so I'd say no.
I also feel that atomic_xxx() is over-used, many times in inappropriate
and buggy ways.

If you want to manipulate bitmasks or bitmaps, then imho you should be
using the bitops or bitmap support rather than atomic stuff.

Why don't we have atomic_set_mask()?

[Russell King - ARM Linux]

On Thu, Mar 17, 2011 at 08:36:15PM +0900, KyongHo Cho wrote:
> Hi.
> 
> arch/arm/asm/atomic.h defines atomic_clear_mask() but its counterpart,
> atomic_set_mask() is not defined.
> 
> Don't we need it?

I can't reply to you:

| A message that you sent could not be delivered to one or more of its
| recipients. This is a permanent error. The following address(es) failed:
| 
|   pullip.cho at samsung.com
|     SMTP error from remote mail server after end of data:
|     host mailin.samsung.com [203.254.224.12]: 550 esmtp:
|     protocol deviation

Which basically means that Samsung's MTA is currently broken.

Why don't we have atomic_set_mask()?

[KyongHo Cho]

Russell King - ARM Linux wrote:
> 
> On Thu, Mar 17, 2011 at 08:36:15PM +0900, KyongHo Cho wrote:
> > Hi.
> >
> > arch/arm/asm/atomic.h defines atomic_clear_mask() but its counterpart,
> > atomic_set_mask() is not defined.
> >
> > Don't we need it?
> 
> I can't reply to you:
> 
> | A message that you sent could not be delivered to one or more of its
> | recipients. This is a permanent error. The following address(es) failed:
> |
> |   pullip.cho at samsung.com
> |     SMTP error from remote mail server after end of data:
> |     host mailin.samsung.com [203.254.224.12]: 550 esmtp:
> |     protocol deviation
> 
> Which basically means that Samsung's MTA is currently broken.

I think it is currently ok because I just sent an email from gmail to
@samsung.com.

Thank you!

KyongHo.

Why don't we have atomic_set_mask()?

[Detlef Vollmann]

On 03/17/11 13:07, Russell King - ARM Linux wrote:
> On Thu, Mar 17, 2011 at 08:36:15PM +0900, KyongHo Cho wrote:
>> arch/arm/asm/atomic.h defines atomic_clear_mask() but its counterpart,
>> atomic_set_mask() is not defined.
>
> Nothing other than some s390 drivers seems to use it, so I'd say no.
> I also feel that atomic_xxx() is over-used, many times in inappropriate
> and buggy ways.
I fully agree.  Simply asking for a atomic_set_mask looks like you're
trying to do something wrong.
The convention is to use those bits as flags, and having the contention
on setting the flag, so you have test_and_set_bit(), and can then
deal with the case where the flag was already set.
Doing that with more than one flag at a time would get pretty complex.

   Detlef