linux-kernel - Re: [PATCH 5/6] udf: Simplify calls to udf_disk_stamp_to

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20180514135759.r4jxnw7mp3h6ak36@quack2.suse.cz>
Date:   Mon, 14 May 2018 15:57:59 +0200
From:   Jan Kara <jack@...e.cz>
To:     Deepa Dinamani <deepa.kernel@...il.com>
Cc:     viro@...iv.linux.org.uk, tglx@...utronix.de, arnd@...db.de,
        linux-kernel@...r.kernel.org, linux-fsdevel@...r.kernel.org,
        y2038@...ts.linaro.org, jack@...e.com
Subject: Re: [PATCH 5/6] udf: Simplify calls to udf_disk_stamp_to_time

Hello,

On Fri 11-05-18 21:59:20, Deepa Dinamani wrote:
> Subsequent patches in the series convert inode timestamps
> to use struct timespec64 instead of struct timespec as
> part of solving the y2038 problem.
> 
> commit fd3cfad374d4 ("udf: Convert udf_disk_stamp_to_time() to use mktime64()")
> eliminated the NULL return condition from udf_disk_stamp_to_time().
> udf_time_to_disk_time() is always called with a valid dest pointer and
> the return value is ignored.
> Further, caller can as well check the dest pointer being passed in rather
> than return argument.
> Make both the functions return void.
> 
> This will make the inode timestamp conversion simpler.
> 
> Signed-off-by: Deepa Dinamani <deepa.kernel@...il.com>
> Cc: jack@...e.com

I'm fine with this. Just one bug below:

> diff --git a/fs/udf/super.c b/fs/udf/super.c
> index 0d27d41f5c6e..bd0ae64bc31c 100644
> --- a/fs/udf/super.c
> +++ b/fs/udf/super.c
> @@ -862,6 +862,9 @@ static int udf_load_pvoldesc(struct super_block *sb, sector_t block)
>  	struct buffer_head *bh;
>  	uint16_t ident;
>  	int ret = -ENOMEM;
> +#ifdef UDFFS_DEBUG
> +	struct timestamp *ts;
> +#endif
>  
>  	outstr = kmalloc(128, GFP_NOFS);
>  	if (!outstr)
> @@ -880,15 +883,14 @@ static int udf_load_pvoldesc(struct super_block *sb, sector_t block)
>  
>  	pvoldesc = (struct primaryVolDesc *)bh->b_data;
>  
> -	if (udf_disk_stamp_to_time(&UDF_SB(sb)->s_record_time,
> -			      pvoldesc->recordingDateAndTime)) {
> +	udf_disk_stamp_to_time(&UDF_SB(sb)->s_record_time,
> +			      pvoldesc->recordingDateAndTime);
>  #ifdef UDFFS_DEBUG
> -		struct timestamp *ts = &pvoldesc->recordingDateAndTime;
> -		udf_debug("recording time %04u/%02u/%02u %02u:%02u (%x)\n",
> -			  le16_to_cpu(ts->year), ts->month, ts->day, ts->hour,
> -			  ts->minute, le16_to_cpu(ts->typeAndTimezone));
> +	*ts = &pvoldesc->recordingDateAndTime;

This should be 'ts ='.

Do you prefer me taking this patch through my tree or will you carry it as
a part of your series?

								Honza
-- 
Jan Kara <jack@...e.com>
SUSE Labs, CR