Linux Test Project
 help / color / mirror / Atom feed
From: Cyril Hrubis <chrubis@suse.cz>
To: ltp@lists.linux.it
Subject: [LTP] [RFC PATCH 1/8] lib: Fix linking error multiple TCID definitions with -fno-common
Date: Wed, 15 Apr 2020 12:49:07 +0200	[thread overview]
Message-ID: <20200415104907.GC12705@rei.lan> (raw)
In-Reply-To: <20200415094932.GA29975@dell5510>

Hi!
> > > +extern const char *TCID;
> 
> > Do we really need this extern or can we remove it?
> Some code is using TCID, not sure if it could be replaced by something else or
> whether is worth to add API function which returns TCID.
> 
> In file included from memfd_create01.c:17:
> memfd_create01.c: In function ???verify_memfd_create???:
> memfd_create01.c:244:21: error: ???TCID??? undeclared (first use in this function)
>   244 |  fd = CHECK_MFD_NEW(TCID, MFD_DEF_SIZE, tc->flags);
>       |                     ^~~~
> ---

This one is used as a unique test identifier, we can hardcode something
as "ltp_memfd_create01" there instead.

> 
> In file included from testcases/kernel/mem/include/mem.h:4,
>                  from overcommit_memory.c:72:
> overcommit_memory.c: In function ???setup???:
> overcommit_memory.c:105:33: error: ???TCID??? undeclared (first use in this function)
>   105 |     "can't support to test %s", TCID);
>       |                                 ^~~~
> ---
> 
> In file included from ima_mmap.c:9:
> ima_mmap.c: In function ???run???:
> ima_mmap.c:36:43: error: ???TCID??? undeclared (first use in this function)
>    36 |   tst_brk(TBROK, "Usage: %s -f filename", TCID);
>       |                                           ^~~~

Hmm these two are using the TCID as a test binary name, which kind of
works by accident. Because:

1. The library uses argv[0] to initialize TCID
2. Then TCID is used as a prefix for the temporary directory
   the test has created and also for filename on tmpfs

There is no guarantee that TCID is equal to argv[0], it's only supposed
to be unique identifier for a test.

So I would rather be for removing TCID from these tests. The question is
if we want to replace it with something that is guaranteed to contain
the test binary name or remove it completely.

-- 
chrubis@suse.cz

  reply	other threads:[~2020-04-15 10:49 UTC|newest]

Thread overview: 14+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-04-15  9:28 [LTP] [RFC PATCH 0/8] Fix compilation with -fno-common (gcc-10) Petr Vorel
2020-04-15  9:28 ` [LTP] [RFC PATCH 1/8] lib: Fix linking error multiple TCID definitions with -fno-common Petr Vorel
2020-04-15  9:30   ` Cyril Hrubis
2020-04-15  9:49     ` Petr Vorel
2020-04-15 10:49       ` Cyril Hrubis [this message]
2020-04-15 11:08         ` Petr Vorel
2020-04-15  9:28 ` [LTP] [RFC PATCH 2/8] controllers: Simplify fullpath definition Petr Vorel
2020-04-15  9:28 ` [LTP] [RFC PATCH 3/8] controllers: Fix linking with -fno-common Petr Vorel
2020-04-15  9:28 ` [LTP] [RFC PATCH 4/8] kernel/mem: Fix linking error " Petr Vorel
2020-04-15  9:28 ` [LTP] [RFC PATCH 5/8] hugepage: " Petr Vorel
2020-04-15  9:28 ` [LTP] [RFC PATCH 6/8] hugetlb: " Petr Vorel
2020-04-15  9:28 ` [LTP] [RFC PATCH 7/8] realtime: " Petr Vorel
2020-04-15  9:28 ` [LTP] [RFC PATCH 8/8] travis: Build " Petr Vorel
2020-04-15  9:39 ` [LTP] [RFC PATCH 0/8] Fix compilation with -fno-common (gcc-10) Cyril Hrubis

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20200415104907.GC12705@rei.lan \
    --to=chrubis@suse.cz \
    --cc=ltp@lists.linux.it \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox