Re: [PATCH] add sched_getcpu

["nathan@nathan7.eu" <nathan@nathan7.eu>]

[-- Attachment #1: Type: text/plain, Size: 2405 bytes --]

On Mon, Feb 29, 2016 at 6:49 PM nathan@nathan7.eu <nathan@nathan7.eu> wrote:

> As far as I can tell, syscall() is supposed to set errno (using
> __sycall_ret), but maybe src/misc/syscall.c isn't what I think it is?
>
I indeed got the return value backwards, and I'll fix that, along with the
> #ifdef.
>
Actually, nope, that ternary is the right way round, and __sycall_ret
handles the -1 return.

>
>
> On Mon, Feb 29, 2016 at 6:23 PM Alexander Monakov <amonakov@ispras.ru>
> wrote:
>
>> On Mon, 29 Feb 2016, Nathan Zadoks wrote:
>>
>> > This is a GNU extension, but a fairly minor one, for a system call that
>> > otherwise has no libc wrapper.
>> >
>> > Adding it was discussed previously, without any objections:
>> > http://www.openwall.com/lists/musl/2015/05/08/24
>> > ---
>> >  include/sched.h          |  3 +++
>> >  src/sched/sched_getcpu.c | 10 ++++++++++
>> >  2 files changed, 13 insertions(+)
>> >  create mode 100644 src/sched/sched_getcpu.c
>> >
>> > diff --git a/include/sched.h b/include/sched.h
>> > index 3e34a72..17f5e06 100644
>> > --- a/include/sched.h
>> > +++ b/include/sched.h
>> > @@ -76,6 +76,9 @@ void free(void *);
>> >
>> >  typedef struct cpu_set_t { unsigned long __bits[128/sizeof(long)]; }
>> cpu_set_t;
>> >  int __sched_cpucount(size_t, const cpu_set_t *);
>> > +#ifdef _GNU_SOURCE
>>
>> This code is already under the same #ifdef.
>>
>> > +int sched_getcpu(void);
>> > +#endif
>> >  int sched_getaffinity(pid_t, size_t, cpu_set_t *);
>> >  int sched_setaffinity(pid_t, size_t, const cpu_set_t *);
>> >
>> > diff --git a/src/sched/sched_getcpu.c b/src/sched/sched_getcpu.c
>> > new file mode 100644
>> > index 0000000..070d6e7
>> > --- /dev/null
>> > +++ b/src/sched/sched_getcpu.c
>> > @@ -0,0 +1,10 @@
>> > +#define _GNU_SOURCE
>> > +#include <stdlib.h>
>>
>> Do you need this include?
>>
>> > +#include <sched.h>
>>
>> (this include could also be dropped; I think it's a matter of policy
>> whether
>> such includes are desirable or not, so please wait for comment from Rich)
>>
>> > +#include "syscall.h"
>> > +
>> > +int sched_getcpu(void) {
>> > +  int c, s;
>> > +  s = syscall(SYS_getcpu, &c, NULL, NULL);
>> > +  return (s == 0) ? c : s;
>>
>> This is wrong, as it doesn't set errno on error, and does not produce -1.
>> This
>> should be something like 'return s ? __syscall_ret(s) : c;' or maybe
>> 'return __syscall_ret(s ? s : c);'.
>>
>> Alexander
>>
>

[-- Attachment #2: Type: text/html, Size: 3690 bytes --]