2006-08-01 18:57:42

by Dave Jones

[permalink] [raw]
Subject: wire up missing syscalls on x86-64

Signed-off-by: Dave Jones <[email protected]>

diff --git a/arch/x86_64/ia32/ia32entry.S b/arch/x86_64/ia32/ia32entry.S
index 5a92fed..07399fe 100644
--- a/arch/x86_64/ia32/ia32entry.S
+++ b/arch/x86_64/ia32/ia32entry.S
@@ -687,8 +687,8 @@ #endif
.quad sys_readlinkat /* 305 */
.quad sys_fchmodat
.quad sys_faccessat
- .quad quiet_ni_syscall /* pselect6 for now */
- .quad quiet_ni_syscall /* ppoll for now */
+ .quad compat_sys_pselect6
+ .quad compat_sys_ppoll
.quad sys_unshare /* 310 */
.quad compat_sys_set_robust_list
.quad compat_sys_get_robust_list
diff --git a/include/asm-x86_64/unistd.h b/include/asm-x86_64/unistd.h
index feb77cb..0a08bbf 100644
--- a/include/asm-x86_64/unistd.h
+++ b/include/asm-x86_64/unistd.h
@@ -600,9 +600,9 @@ __SYSCALL(__NR_fchmodat, sys_fchmodat)
#define __NR_faccessat 269
__SYSCALL(__NR_faccessat, sys_faccessat)
#define __NR_pselect6 270
-__SYSCALL(__NR_pselect6, sys_ni_syscall) /* for now */
+__SYSCALL(__NR_pselect6, sys_pselect6)
#define __NR_ppoll 271
-__SYSCALL(__NR_ppoll, sys_ni_syscall) /* for now */
+__SYSCALL(__NR_ppoll, sys_ppoll)
#define __NR_unshare 272
__SYSCALL(__NR_unshare, sys_unshare)
#define __NR_set_robust_list 273
--
http://www.codemonkey.org.uk


2006-08-01 19:56:41

by Andi Kleen

[permalink] [raw]
Subject: Re: wire up missing syscalls on x86-64

On Tuesday 01 August 2006 20:57, Dave Jones wrote:
> Signed-off-by: Dave Jones <[email protected]>

No, the reason they're not wired up yet is that some infrastructure
for them is still missing. It's currently queued up for .19

-Andi

2006-08-01 19:59:24

by Dave Jones

[permalink] [raw]
Subject: Re: wire up missing syscalls on x86-64

On Tue, Aug 01, 2006 at 09:55:34PM +0200, Andi Kleen wrote:
> On Tuesday 01 August 2006 20:57, Dave Jones wrote:
> > Signed-off-by: Dave Jones <[email protected]>
>
> No, the reason they're not wired up yet is that some infrastructure
> for them is still missing. It's currently queued up for .19

Ah, I missed the fact that we're also carrying another diff
in the Fedora kernel for this. Ok, as long as you're aware
of it, all is good.

Dave

--
http://www.codemonkey.org.uk