[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <DD3F340E-4D88-497D-9DCC-3A4C93F03E5B@dilger.ca>
Date: Tue, 27 Jun 2017 12:43:42 -0600
From: Andreas Dilger <adilger@...ger.ca>
To: Tahsin Erdogan <tahsin@...gle.com>
Cc: "Darrick J . Wong" <darrick.wong@...cle.com>,
Theodore Ts'o <tytso@....edu>,
Ext4 Developers List <linux-ext4@...r.kernel.org>
Subject: Re: [PATCH 07/12] e2fsck: track ea_inode references
> On Jun 26, 2017, at 7:25 PM, Tahsin Erdogan <tahsin@...gle.com> wrote:
>
>>> +#ifdef EXT2_NO_64_TYPE
>>> + fprintf(f, "%*u", width, ctx->num2);
>>> +#else
>>> + fprintf(f, "%*llu", width, (long long)ctx->num2);
>>> #endif
>>
>> Rather than a series of "#ifdef EXT2_NO_64_TYPE" checks, it would be cleaner
>> to have a single #define to set the printf type, like:
>>
>> #ifdef EXT2_NO_64_TYPE
>> #define EXT2_64U "%*u"
>> #define EXT2_64D "%*d"
>> #define EXT2_64X "%*x"
>> #else
>> #define EXT2_64U "%*llu"
>> #define EXT2_64D "%*lld"
>> #define EXT2_64X "%*llx"
>> #endif
>
> I am trying to figure out the purpose of #ifdef EXT2_NO_64_TYPE checks
> now. Who defines it and what problem does it solve?
Sorry, I can't answer that. Git history shows that it has been around since
2000 but is not set by the configure script. I just thought it made sense to
clean up the code, but it might be even better to get rid of it completely.
Cheers, Andreas
Download attachment "signature.asc" of type "application/pgp-signature" (196 bytes)
Powered by blists - more mailing lists