linux-kernel - Re: [PATCH misc v2] misc: cardreader: fix overwritten return value in RTS5260 driver

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] [day] [month] [year] [list]

Message-ID: <2025081906-sandbar-shove-b79f@gregkh>
Date: Tue, 19 Aug 2025 12:39:23 +0200
From: Greg KH <gregkh@...uxfoundation.org>
To: Tian <27392025k@...il.com>
Cc: arnd@...db.de, linux-kernel@...r.kernel.org
Subject: Re: [PATCH misc v2] misc: cardreader: fix overwritten return value
 in RTS5260 driver

On Mon, Jul 28, 2025 at 06:35:10PM -0700, Tian wrote:
> From: Tian Liu <27392025k@...il.com>
> 
> In both rts5260.c and rtsx_pcr.c, a return value is assigned and then
> overwritten by a subsequent call, which causes the original result to be
> lost. This may result in silently ignoring errors from the first write.
> 
> This patch uses `err |=` to preserve earlier failure status.
> 
> This is an untested cleanup inspired by commit c55c7a85e02a ("um: ubd: Preserve earlier error value in ubd_user") which fixed similar error-handling logic. While this change has not been functionally tested on hardware, it is a mechanical fix and aligns with expected error-handling style.
> 
> Fixes: bede03a579b3 ("misc: rtsx: Enable OCP for rts522a rts524a rts525a rts5260")
> Fixes: c0e5f4e73a71 ("misc: rtsx: Add support for RTS5261")
> 
> Signed-off-by: Tian Liu <27392025k@...il.com>
> 
> Changes in v2:
> - Use full name in Signed-off-by and From
> - Add Fixes: tag
> - Clarify that patch is untested, but reasoned from upstream style
> - Cc: appropriate public mailing list
> 
> ---

The "changes..." stuff goes below the --- line.

Also, please wrap your changelog text at 72 columns.

As you have not tested this, I don't think this should be taken, do you?

>  drivers/misc/cardreader/rts5260.c  | 2 +-
>  drivers/misc/cardreader/rtsx_pcr.c | 2 +-
>  2 files changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/misc/cardreader/rts5260.c b/drivers/misc/cardreader/rts5260.c
> index d2d3a6ccb8f7..ed8adaab54a8 100644
> --- a/drivers/misc/cardreader/rts5260.c
> +++ b/drivers/misc/cardreader/rts5260.c
> @@ -269,7 +269,7 @@ static int rts5260_card_power_off(struct rtsx_pcr *pcr, int card)
>  	rts5260_card_before_power_off(pcr);
>  	err = rtsx_pci_write_register(pcr, LDO_VCC_CFG1,
>  			 LDO_POW_SDVDD1_MASK, LDO_POW_SDVDD1_OFF);
> -	err = rtsx_pci_write_register(pcr, LDO_CONFIG2,
> +	err |= rtsx_pci_write_register(pcr, LDO_CONFIG2,
>  			 DV331812_POWERON, DV331812_POWEROFF);

How can this write fail?  And just | all errors together really doesn't
do all that much, right?


>  	if (pcr->option.ocp_en)
>  		rtsx_pci_disable_ocp(pcr);
> diff --git a/drivers/misc/cardreader/rtsx_pcr.c b/drivers/misc/cardreader/rtsx_pcr.c
> index a7b066c48740..9fb22f2cedbd 100644
> --- a/drivers/misc/cardreader/rtsx_pcr.c
> +++ b/drivers/misc/cardreader/rtsx_pcr.c
> @@ -1196,7 +1196,7 @@ static int rtsx_pci_init_hw(struct rtsx_pcr *pcr)
>  		/* Gating real mcu clock */
>  		err = rtsx_pci_write_register(pcr, RTS5261_FW_CFG1,
>  			RTS5261_MCU_CLOCK_GATING, 0);
> -		err = rtsx_pci_write_register(pcr, RTS5261_REG_FPDCTL,
> +		err |= rtsx_pci_write_register(pcr, RTS5261_REG_FPDCTL,
>  			SSC_POWER_DOWN, 0);

Same here, it feels odd to do that.

thanks,

greg k-h