[LTP] [PATCH V2] Use a union of sigval instead of sigval_t

[LTP] [PATCH V2] Use a union of sigval instead of sigval_t

[Khem Raj]

sigval_t is glibc only construct, we use a union of sigval
which pretty much is same effect as sigval_t

Signed-off-by: Khem Raj <raj.khem@gmail.com>
---
 lib/tlibio.c | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/lib/tlibio.c b/lib/tlibio.c
index 6b6103f..7b6bb33 100644
--- a/lib/tlibio.c
+++ b/lib/tlibio.c
@@ -130,9 +130,6 @@
 #endif
 
 static void lio_async_signal_handler();
-#ifdef sgi
-static void lio_async_callback_handler();
-#endif
 
 /*
  * Define the structure as used in lio_parse_arg1 and lio_help1
@@ -441,7 +438,7 @@ static void lio_async_signal_handler(int sig)
  * If the handler is called, it will increment the Received_callback
  * global variable.
  ***********************************************************************/
-static void lio_async_callback_handler(sigval_t sigval)
+static void lio_async_callback_handler(union sigval sigval)
 {
 	if (Debug_level)
 		printf
-- 
2.7.0

[LTP] [PATCH V2] Use a union of sigval instead of sigval_t

[Jan Stancek]

----- Original Message -----
> From: "Khem Raj" <raj.khem@gmail.com>
> To: ltp@lists.linux.it
> Sent: Thursday, 7 January, 2016 1:25:16 PM
> Subject: [LTP] [PATCH V2] Use a union of sigval instead of sigval_t
> 
> sigval_t is glibc only construct, we use a union of sigval
> which pretty much is same effect as sigval_t
> 
> Signed-off-by: Khem Raj <raj.khem@gmail.com>

Reviewed-by: Jan Stancek <jstancek@redhat.com>

Looks good to me, I think this should be safe to apply before release.
(I tested it across RHEL5.6/6.0/7.0)

Cyril, do you agree?

Regards,
Jan

> ---
>  lib/tlibio.c | 5 +----
>  1 file changed, 1 insertion(+), 4 deletions(-)
> 
> diff --git a/lib/tlibio.c b/lib/tlibio.c
> index 6b6103f..7b6bb33 100644
> --- a/lib/tlibio.c
> +++ b/lib/tlibio.c
> @@ -130,9 +130,6 @@
>  #endif
>  
>  static void lio_async_signal_handler();
> -#ifdef sgi
> -static void lio_async_callback_handler();
> -#endif
>  
>  /*
>   * Define the structure as used in lio_parse_arg1 and lio_help1
> @@ -441,7 +438,7 @@ static void lio_async_signal_handler(int sig)
>   * If the handler is called, it will increment the Received_callback
>   * global variable.
>   ***********************************************************************/
> -static void lio_async_callback_handler(sigval_t sigval)
> +static void lio_async_callback_handler(union sigval sigval)
>  {
>  	if (Debug_level)
>  		printf
> --
> 2.7.0
> 
> 
> --
> Mailing list info: http://lists.linux.it/listinfo/ltp
> 

[LTP] [PATCH V2] Use a union of sigval instead of sigval_t

[Cyril Hrubis]

Hi!
> > sigval_t is glibc only construct, we use a union of sigval
> > which pretty much is same effect as sigval_t
> > 
> > Signed-off-by: Khem Raj <raj.khem@gmail.com>
> 
> Reviewed-by: Jan Stancek <jstancek@redhat.com>
> 
> Looks good to me, I think this should be safe to apply before release.
> (I tested it across RHEL5.6/6.0/7.0)
> 
> Cyril, do you agree?

Looking into POSIX it explicitly says that the callback takes union
sigval paramter and not sigval_t.
(http://pubs.opengroup.org/onlinepubs/9699919799/basedefs/signal.h.html)

So this change should not break anything. Acked to go in before release
by me as well.

-- 
Cyril Hrubis
chrubis@suse.cz

[LTP] [PATCH V2] Use a union of sigval instead of sigval_t

[Jan Stancek]

----- Original Message -----
> From: "Cyril Hrubis" <chrubis@suse.cz>
> To: "Jan Stancek" <jstancek@redhat.com>
> Cc: "Khem Raj" <raj.khem@gmail.com>, ltp@lists.linux.it
> Sent: Thursday, 7 January, 2016 3:08:51 PM
> Subject: Re: [LTP] [PATCH V2] Use a union of sigval instead of sigval_t
> 
> Hi!
> > > sigval_t is glibc only construct, we use a union of sigval
> > > which pretty much is same effect as sigval_t
> > > 
> > > Signed-off-by: Khem Raj <raj.khem@gmail.com>
> > 
> > Reviewed-by: Jan Stancek <jstancek@redhat.com>
> > 
> > Looks good to me, I think this should be safe to apply before release.
> > (I tested it across RHEL5.6/6.0/7.0)
> > 
> > Cyril, do you agree?
> 
> Looking into POSIX it explicitly says that the callback takes union
> sigval paramter and not sigval_t.
> (http://pubs.opengroup.org/onlinepubs/9699919799/basedefs/signal.h.html)
> 
> So this change should not break anything. Acked to go in before release
> by me as well.

Pushed with 2nd fwd declaration removed too and added POSIX reference
to commit message.

Regards,
Jan

> 
> --
> Cyril Hrubis
> chrubis@suse.cz
>