[LTP] [PATCH] bugfix for ssetmask01.c

[LTP] [PATCH] bugfix for ssetmask01.c

[zhanglianjie]

mips architecture syscall(69) __NR_ssetmask not supported on your arch.

Signed-off-by: zhanglianjie <zhanglianjie@uniontech.com>
---
 include/lapi/syscalls/mips_o32.in | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/include/lapi/syscalls/mips_o32.in b/include/lapi/syscalls/mips_o32.in
index c71a62a35..c2d43b984 100644
--- a/include/lapi/syscalls/mips_o32.in
+++ b/include/lapi/syscalls/mips_o32.in
@@ -66,8 +66,8 @@ getppid 64
 getpgrp 65
 setsid 66
 sigaction 67
-sgetmask 68
-ssetmask 69
+sgetmask 4068
+ssetmask 4069
 setreuid 70
 setregid 71
 sigsuspend 72
--
2.20.1

[LTP] [PATCH] bugfix for ssetmask01.c

[Petr Vorel]

Hi,

> mips architecture syscall(69) __NR_ssetmask not supported on your arch.

nit: "mips_o32: Fix __NR_s{g,s}etmask" for bug title would be more descriptive
than "bugfix for ssetmask01.c". And then in text mention ssetmask01.c.

> +++ b/include/lapi/syscalls/mips_o32.in
> @@ -66,8 +66,8 @@ getppid 64
>  getpgrp 65
>  setsid 66
>  sigaction 67
> -sgetmask 68
> -ssetmask 69
> +sgetmask 4068
> +ssetmask 4069

But more important: I guess you're right because we don't add any offset to the
numbers. But IMHO all numbers are wrong, all mips files should contain offset
(01e4dc222 was wrong). mips_o32.in should be offset 4000, mips_n32.in 6000 and
mips_n64.in 5000 (comparing scall32-o32.S, scall64-64.S, scall64-n32.S in
arch/mips/kernel in kernel and mips{,n32,64}/bits/syscall.h.in in musl).

Kind regards,
Petr

[LTP] [PATCH] bugfix for ssetmask01.c

[He Zhe]

On 4/1/21 2:12 PM, Petr Vorel wrote:
> Hi,
>
>> mips architecture syscall(69) __NR_ssetmask not supported on your arch.
> nit: "mips_o32: Fix __NR_s{g,s}etmask" for bug title would be more descriptive
> than "bugfix for ssetmask01.c". And then in text mention ssetmask01.c.
>
>> +++ b/include/lapi/syscalls/mips_o32.in
>> @@ -66,8 +66,8 @@ getppid 64
>>  getpgrp 65
>>  setsid 66
>>  sigaction 67
>> -sgetmask 68
>> -ssetmask 69
>> +sgetmask 4068
>> +ssetmask 4069
> But more important: I guess you're right because we don't add any offset to the
> numbers. But IMHO all numbers are wrong, all mips files should contain offset
> (01e4dc222 was wrong). mips_o32.in should be offset 4000, mips_n32.in 6000 and
> mips_n64.in 5000 (comparing scall32-o32.S, scall64-64.S, scall64-n32.S in
> arch/mips/kernel in kernel and mips{,n32,64}/bits/syscall.h.in in musl).

Petr is correct. I probably missed interpreting the offsets for each sub-arch.
Sorry for the inconvenience. I'll send a patch to correct this later.

Thanks,
Zhe

>
> Kind regards,
> Petr