linux-kernel - Re: [PATCH] mtd: rawnand: marvell: Use correct logic for nand-keep-config

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: <20221004120212.6389b96a@xps-13>
Date:   Tue, 4 Oct 2022 12:02:12 +0200
From:   Miquel Raynal <miquel.raynal@...tlin.com>
To:     Chris Packham <Chris.Packham@...iedtelesis.co.nz>
Cc:     "richard@....at" <richard@....at>,
        "vigneshr@...com" <vigneshr@...com>,
        "bbrezillon@...nel.org" <bbrezillon@...nel.org>,
        Tony O'Brien <Tony.OBrien@...iedtelesis.co.nz>,
        "linux-mtd@...ts.infradead.org" <linux-mtd@...ts.infradead.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] mtd: rawnand: marvell: Use correct logic for
 nand-keep-config

Hi Chris,

Chris.Packham@...iedtelesis.co.nz wrote on Tue, 27 Sep 2022 02:54:40
+0000:

> On 27/09/22 15:47, Chris Packham wrote:
> > From: Tony O'Brien <tony.obrien@...iedtelesis.co.nz>
> >
> > Originally the absence of the marvell,nand-keep-config property caused
> > the setup_data_interface function to be provided. However when
> > setup_data_interface was moved into nand_controller_ops the logic was
> > unintentionally inverted. Update the logic so that only if the
> > marvell,nand-keep-config property is present the bootloader NAND config
> > kept.
> >
> > Fixes: 7a08dbaedd36 ("mtd: rawnand: Move ->setup_data_interface() to nand_controller_ops")
> > Signed-off-by: Tony O'Brien <tony.obrien@...iedtelesis.co.nz>
> > Signed-off-by: Chris Packham <chris.packham@...iedtelesis.co.nz>
> > ---
> >
> > Notes:
> >      I think this is a bug that's been lurking for 4 years or so. I'm not
> >      sure that's particularly long in the life of an embedded device but it
> >      does make me wonder if there have been other bug reports about it.
> >      
> >      We noticed this because we had a bootloader that used maxed out NAND
> >      timings which made the time it took the kernel to do anything on the
> >      file system longer than we expected.  
> 
> I think there might be a similar logic inversion bug in 
> drivers/mtd/nand/raw/denali.c but I lack the ability to test for that 
> platform.

Agreed, the denali driver has the same issue. Could you please send a
patch?
 
> 
> >   drivers/mtd/nand/raw/marvell_nand.c | 2 +-
> >   1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --git a/drivers/mtd/nand/raw/marvell_nand.c b/drivers/mtd/nand/raw/marvell_nand.c
> > index 2455a581fd70..b248c5f657d5 100644
> > --- a/drivers/mtd/nand/raw/marvell_nand.c
> > +++ b/drivers/mtd/nand/raw/marvell_nand.c
> > @@ -2672,7 +2672,7 @@ static int marvell_nand_chip_init(struct device *dev, struct marvell_nfc *nfc,
> >   	chip->controller = &nfc->controller;
> >   	nand_set_flash_node(chip, np);
> >   
> > -	if (!of_property_read_bool(np, "marvell,nand-keep-config"))
> > +	if (of_property_read_bool(np, "marvell,nand-keep-config"))
> >   		chip->options |= NAND_KEEP_TIMINGS;
> >   
> >   	mtd = nand_to_mtd(chip)  


Thanks,
Miquèl