[PATCH v7] Implement -e trace=%statfs option to trace and group statfs, statfs64, statvfs syscalls

[Dmitry V. Levin]

On Sat, Apr 01, 2017 at 01:02:33AM +0530, Abhishek Tiwari wrote:
> while grouping I found that osf_statfs has following entries
> 
>   linux/alpha/syscallent.h:188:[160] = { 3,   TF|TSF,
> SEN(osf_statfs),        "osf_statfs"        },
>   linux/alpha/syscallent.h:233:[227] = { 3,   TF|TSF,
> SEN(osf_statfs),        "osf_statfs64"      },
> 
> Should the second entry be
>  linux/alpha/syscallent.h:233:[227] = { 3,   TF|TSF,
> SEN(osf_statfs64),        "osf_statfs64"      },
> 
> and not
> linux/alpha/syscallent.h:233:[227] = { 3,   TF|TSF,
> SEN(osf_statfs),        "osf_statfs64"      },

No, I don't think so.  There is a single parser for both syscalls:

$ grep osf_statfs alpha.c 
SYS_FUNC(osf_statfs)

This is a dumb parser, it doesn't decode neither struct osf_statfs
nor struct osf_statfs64.

> and then grouped under statfs64 ?
> 
> If entry is correct what group should I put for this file statfs or statfs64 ?

Why do you want to separate statfs and statfs64?


-- 
ldv
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 801 bytes
Desc: not available
URL: <http://lists.strace.io/pipermail/strace-devel/attachments/20170331/dc6b4388/attachment.bin>