[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: GSOC - valgrind-hurd queries

From: Subhashish
Subject: Re: GSOC - valgrind-hurd queries
Date: Sat, 31 May 2014 06:23:39 +0530
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:24.0) Gecko/20100101 Thunderbird/24.5.0

On Saturday 31 May 2014 04:29 AM, Samuel Thibault wrote:
Subhashish, le Sat 31 May 2014 04:24:27 +0530, a écrit :
The symbol VKI_PATH_MAX cannot be defined since symbol PATH_MAX is not
Can we have workarounds - based on realloc, or like xgethostname() as
Yes, but that'll be tedious (there are quite a few occurrences) and
uninteresting. I'd say for now just define it to some arbitrary value,
such as 4096.

I'm following vki-linux.h and vki-gnu.h to add type definitions to
vki-gnu.h - but I am in the dark as to what is to be added - I'm doing hit
and trial.
Well, everything that appears in vki-linux.h is probably used in
coregrind, so you'll have to define them too.

VKI_S_IRUSR, VKI_S_IWOTH are the few of many symbols to be
defined here.

The latter 2 are part of sys/stat.h in other unix systems - I suppose we
don't do stat.h
We do, see /usr/include/i386-gnu/sys/stat.h

Um, I have a doubt.

With this log[5], I need to define 3 types - rlimit, rlimit64 and rusage which I did.

Now since I've #include-ed "i386-gnu/bits/resource.h", do I need to #define the RLIMIT_* symbols with VKI_ prefixes?

vki-linux.h doesn't but darwin's does. Moreover, ours are inside eums, so maybe I think they will be automatically picked up by rlimit and friends. So for now I'm not doing so.


5 - https://gist.github.com/sprkv5/15b25ed9ee8ac5f20a47#file-makelog-7-txt

reply via email to

[Prev in Thread] Current Thread [Next in Thread]