diff mbox series

sched_attr: Do not define for glibc >= 2.41

Message ID 20241011054806.1014276-1-raj.khem@gmail.com (mailing list archive)
State New
Headers show
Series sched_attr: Do not define for glibc >= 2.41 | expand

Commit Message

Khem Raj Oct. 11, 2024, 5:48 a.m. UTC
glibc 2.41+ has added [1] definitions for sched_setattr and sched_getattr functions
and struct sched_attr. Therefore, it needs to be checked for here as well before
defining sched_attr

Fixes builds with glibc/trunk

[1] https://sourceware.org/git/?p=glibc.git;a=commitdiff;h=21571ca0d70302909cf72707b2a7736cf12190a0;hp=298bc488fdc047da37482f4003023cb9adef78f8

Signed-off-by: Khem Raj <raj.khem@gmail.com>
Cc: Laurent Vivier <laurent@vivier.eu> (m
---
 linux-user/syscall.c | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)

Comments

Paolo Bonzini Oct. 11, 2024, 9:08 a.m. UTC | #1
On 10/11/24 07:48, Khem Raj wrote:
> glibc 2.41+ has added [1] definitions for sched_setattr and sched_getattr functions
> and struct sched_attr. Therefore, it needs to be checked for here as well before
> defining sched_attr
> 
> Fixes builds with glibc/trunk
> 
> [1] https://sourceware.org/git/?p=glibc.git;a=commitdiff;h=21571ca0d70302909cf72707b2a7736cf12190a0;hp=298bc488fdc047da37482f4003023cb9adef78f8
> 
> Signed-off-by: Khem Raj <raj.khem@gmail.com>
> Cc: Laurent Vivier <laurent@vivier.eu> (m

Hi, I think it would be better to test in meson.build (with 
cc.has_type), as is already done in several other places in 
linux-user/syscall.c.

Thanks,

Paolo

> ---
>   linux-user/syscall.c | 10 +++++++++-
>   1 file changed, 9 insertions(+), 1 deletion(-)
> 
> diff --git a/linux-user/syscall.c b/linux-user/syscall.c
> index 1354e75694..9e6eebbf1a 100644
> --- a/linux-user/syscall.c
> +++ b/linux-user/syscall.c
> @@ -359,7 +359,13 @@ _syscall3(int, sys_sched_getaffinity, pid_t, pid, unsigned int, len,
>   #define __NR_sys_sched_setaffinity __NR_sched_setaffinity
>   _syscall3(int, sys_sched_setaffinity, pid_t, pid, unsigned int, len,
>             unsigned long *, user_mask_ptr);
> -/* sched_attr is not defined in glibc */
> +/* sched_attr is not defined in glibc < 2.41 */
> +#include <stdio.h>
> +
> +#if defined(__GLIBC__) && defined(__GLIBC_MINOR__)
> +# if (__GLIBC__ > 2) || (__GLIBC__ == 2 && __GLIBC_MINOR__ >= 41)
> +    /* do nothing */
> +# else
>   struct sched_attr {
>       uint32_t size;
>       uint32_t sched_policy;
> @@ -372,6 +378,8 @@ struct sched_attr {
>       uint32_t sched_util_min;
>       uint32_t sched_util_max;
>   };
> +# endif
> +#endif
>   #define __NR_sys_sched_getattr __NR_sched_getattr
>   _syscall4(int, sys_sched_getattr, pid_t, pid, struct sched_attr *, attr,
>             unsigned int, size, unsigned int, flags);
> 
>
Laurent Vivier Oct. 11, 2024, 9:15 a.m. UTC | #2
Le 11/10/2024 à 11:08, Paolo Bonzini a écrit :
> On 10/11/24 07:48, Khem Raj wrote:
>> glibc 2.41+ has added [1] definitions for sched_setattr and sched_getattr functions
>> and struct sched_attr. Therefore, it needs to be checked for here as well before
>> defining sched_attr
>>
>> Fixes builds with glibc/trunk
>>
>> [1] https://sourceware.org/git/? 
>> p=glibc.git;a=commitdiff;h=21571ca0d70302909cf72707b2a7736cf12190a0;hp=298bc488fdc047da37482f4003023cb9adef78f8
>>
>> Signed-off-by: Khem Raj <raj.khem@gmail.com>
>> Cc: Laurent Vivier <laurent@vivier.eu> (m
> 
> Hi, I think it would be better to test in meson.build (with cc.has_type), as is already done in 
> several other places in linux-user/syscall.c.

We can also test if SCHED_ATTR_SIZE_VER0 exists as it is defined when the structure is defined..

I don't know what is the best solution...

Thanks,
Laurent
Paolo Bonzini Oct. 11, 2024, 11:17 a.m. UTC | #3
On Fri, Oct 11, 2024 at 11:15 AM Laurent Vivier <laurent@vivier.eu> wrote:
>
> Le 11/10/2024 à 11:08, Paolo Bonzini a écrit :
> > On 10/11/24 07:48, Khem Raj wrote:
> >> glibc 2.41+ has added [1] definitions for sched_setattr and sched_getattr functions
> >> and struct sched_attr. Therefore, it needs to be checked for here as well before
> >> defining sched_attr
> >>
> >> Fixes builds with glibc/trunk
> >>
> >> [1] https://sourceware.org/git/?
> >> p=glibc.git;a=commitdiff;h=21571ca0d70302909cf72707b2a7736cf12190a0;hp=298bc488fdc047da37482f4003023cb9adef78f8
> >>
> >> Signed-off-by: Khem Raj <raj.khem@gmail.com>
> >> Cc: Laurent Vivier <laurent@vivier.eu> (m
> >
> > Hi, I think it would be better to test in meson.build (with cc.has_type), as is already done in
> > several other places in linux-user/syscall.c.
>
> We can also test if SCHED_ATTR_SIZE_VER0 exists as it is defined when the structure is defined..

Yes, that works too.

Paolo
diff mbox series

Patch

diff --git a/linux-user/syscall.c b/linux-user/syscall.c
index 1354e75694..9e6eebbf1a 100644
--- a/linux-user/syscall.c
+++ b/linux-user/syscall.c
@@ -359,7 +359,13 @@  _syscall3(int, sys_sched_getaffinity, pid_t, pid, unsigned int, len,
 #define __NR_sys_sched_setaffinity __NR_sched_setaffinity
 _syscall3(int, sys_sched_setaffinity, pid_t, pid, unsigned int, len,
           unsigned long *, user_mask_ptr);
-/* sched_attr is not defined in glibc */
+/* sched_attr is not defined in glibc < 2.41 */
+#include <stdio.h>
+
+#if defined(__GLIBC__) && defined(__GLIBC_MINOR__)
+# if (__GLIBC__ > 2) || (__GLIBC__ == 2 && __GLIBC_MINOR__ >= 41)
+    /* do nothing */
+# else
 struct sched_attr {
     uint32_t size;
     uint32_t sched_policy;
@@ -372,6 +378,8 @@  struct sched_attr {
     uint32_t sched_util_min;
     uint32_t sched_util_max;
 };
+# endif
+#endif
 #define __NR_sys_sched_getattr __NR_sched_getattr
 _syscall4(int, sys_sched_getattr, pid_t, pid, struct sched_attr *, attr,
           unsigned int, size, unsigned int, flags);