Linux MultiMedia Card development
 help / color / mirror / Atom feed
From: Giulio Benetti <giulio.benetti@benettiengineering.com>
To: Avri Altman <Avri.Altman@wdc.com>,
	"linux-mmc@vger.kernel.org" <linux-mmc@vger.kernel.org>
Subject: Re: [PATCH] mmc-utils: mmc_cmds: fix type-punned warning on &ext_csd[] casting
Date: Tue, 5 Mar 2024 19:46:49 +0100	[thread overview]
Message-ID: <9de33c0d-1fe5-4114-8e11-306db56576ed@benettiengineering.com> (raw)
In-Reply-To: <DM6PR04MB657504D3DD4B66F9873E3094FC222@DM6PR04MB6575.namprd04.prod.outlook.com>

On 05/03/24 16:01, Avri Altman wrote:
>>> When building with -Werror=strict-aliasing error is thrown:
>>>
>>> mmc_cmds.c: In function 'do_ffu':
>>> mmc_cmds.c:2972:2: error: dereferencing type-punned pointer will break
>>> strict- aliasing rules [-Werror=strict-aliasing]
>>>    sect_done = htole32(*((__u32
>>> *)&ext_csd[EXT_CSD_NUM_OF_FW_SEC_PROG_0]));
>>>    ^
>>> cc1: all warnings being treated as errors
>>>
>>> Let's fix type-punned breaking strict-aliasing by memcpy() the variable.
>> If any, I would just revert commit a1b594ca735e.
>>
>> Thanks,
>> Avri
> Or, how about this untested option.

This is a good solution, just re-spin and renamed to
per_byte_htole32(__u8 *). Hope it sounds good the name to you.

Thank you
Best regards
-- 
Giulio Benetti
CEO&CTO@Benetti Engineering sas

> Thanks,
> Avri
> 
> 
> index ae7b876..10f7532 100644
> --- a/mmc_cmds.c
> +++ b/mmc_cmds.c
> @@ -58,6 +58,11 @@
>   #define WPTYPE_PWRON 2
>   #define WPTYPE_PERM 3
> 
> +static inline __u32 strict_htole32(__u8 *arr)
> +{
> +       return arr[0] | arr[1] << 8 | arr[2] << 16 | arr[3] << 24;
> +}
> +
>   int read_extcsd(int fd, __u8 *ext_csd)
>   {
>          int ret = 0;
> @@ -2807,7 +2812,7 @@ static void set_ffu_single_cmd(struct mmc_ioc_multi_cmd *multi_cmd,
>                                 __u8 *ext_csd, unsigned int bytes, __u8 *buf,
>                                 off_t offset)
>   {
> -       __u32 arg = htole32(*((__u32 *)&ext_csd[EXT_CSD_FFU_ARG_0]));
> +       __u32 arg = safe_htole32(&ext_csd[EXT_CSD_FFU_ARG_0]);
> 
>          /* send block count */
>          set_single_cmd(&multi_cmd->cmds[1], MMC_SET_BLOCK_COUNT, 0, 0,
> @@ -2827,7 +2832,7 @@ static void set_ffu_single_cmd(struct mmc_ioc_multi_cmd *multi_cmd,
>   int do_ffu(int nargs, char **argv)
>   {
>          int dev_fd, img_fd;
> -       int sect_done = 0, retry = 3, ret = -EINVAL;
> +       int retry = 3, ret = -EINVAL;
>          unsigned int sect_size;
>          __u8 ext_csd[512];
>          __u8 *buf = NULL;
> @@ -2835,6 +2840,7 @@ int do_ffu(int nargs, char **argv)
>          char *device;
>          struct mmc_ioc_multi_cmd *multi_cmd = NULL;
>          unsigned int default_chunk = MMC_IOC_MAX_BYTES;
> +       __u32 sect_done = 0;
> 
>          assert (nargs == 3 || nargs == 4);
> 
> @@ -2968,7 +2974,7 @@ do_retry:
>          }
> 
>          /* Test if we need to restart the download */
> -       sect_done = htole32(*((__u32 *)&ext_csd[EXT_CSD_NUM_OF_FW_SEC_PROG_0]));
> +       sect_done = strict_htole32(&ext_csd[EXT_CSD_NUM_OF_FW_SEC_PROG_0]);
>          /* By spec, host should re-start download from the first sector if sect_done is 0 */
>          if (sect_done == 0) {
>                  if (retry--) {


      parent reply	other threads:[~2024-03-05 18:49 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-03-05 10:59 [PATCH] mmc-utils: mmc_cmds: fix type-punned warning on &ext_csd[] casting Giulio Benetti
2024-03-05 14:52 ` Avri Altman
2024-03-05 15:01   ` Avri Altman
2024-03-05 18:45     ` [PATCH v2] " Giulio Benetti
2024-03-05 19:48       ` Avri Altman
2024-03-07 11:02       ` Ulf Hansson
2024-03-05 18:46     ` Giulio Benetti [this message]

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=9de33c0d-1fe5-4114-8e11-306db56576ed@benettiengineering.com \
    --to=giulio.benetti@benettiengineering.com \
    --cc=Avri.Altman@wdc.com \
    --cc=linux-mmc@vger.kernel.org \
    /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