linux-kernel - Re: [PATCH] mtd: spi-nor: Add support for S3AN spi-nor devices

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: <20160921102007.47181440@bbrezillon>
Date:   Wed, 21 Sep 2016 10:20:07 +0200
From:   Boris Brezillon <boris.brezillon@...e-electrons.com>
To:     Ricardo Ribalda Delgado <ricardo.ribalda@...il.com>
Cc:     Cyrille Pitchen <cyrille.pitchen@...el.com>,
        David Woodhouse <dwmw2@...radead.org>,
        Brian Norris <computersforpeace@...il.com>,
        Javier Martinez Canillas <javier@....samsung.com>,
        Stephen Warren <swarren@...dia.com>,
        Jagan Teki <jteki@...nedev.com>, Vignesh R <vigneshr@...com>,
        Marek Vasut <marex@...x.de>,
        Ezequiel García 
        <ezequiel@...guardiasur.com.ar>,
        Rafał Miłecki <zajec5@...il.com>,
        Furquan Shaikh <furquan@...gle.com>,
        "linux-mtd@...ts.infradead.org" <linux-mtd@...ts.infradead.org>,
        LKML <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] mtd: spi-nor: Add support for S3AN spi-nor devices

On Wed, 21 Sep 2016 10:14:20 +0200
Boris Brezillon <boris.brezillon@...e-electrons.com> wrote:

> On Wed, 21 Sep 2016 09:57:23 +0200
> Ricardo Ribalda Delgado <ricardo.ribalda@...il.com> wrote:
> 
> > Hi Boris
> > 
> > On Wed, Sep 21, 2016 at 9:07 AM, Boris Brezillon
> > <boris.brezillon@...e-electrons.com> wrote:  
> > > Hi Ricardo,
> > >
> > > Please try to pass the version in your subject prefix (pass
> > > --subject-prefix="PATCH vX" to git format-patch).
> > >    
> > 
> > Sorry about that.
> > 
> >   
> > > On Tue, 20 Sep 2016 17:45:51 +0200
> > > Ricardo Ribalda Delgado <ricardo.ribalda@...il.com> wrote:    
> >   
> > >>  /*
> > >> + * This code converts an address to the Default Address Mode, that has non
> > >> + * power of two page sizes. We must support this mode because it is the default
> > >> + * mode supported by Xilinx tools, it can access the whole flash area and
> > >> + * changing over to the Power-of-two mode is irreversible and corrupts the
> > >> + * original data.
> > >> + */
> > >> +static loff_t spi_nor_s3an_addr_convert(struct spi_nor *nor, unsigned int addr)
> > >> +{
> > >> +     unsigned int offset;
> > >> +
> > >> +     offset = (nor->page_size == 264) ? (addr % 264) : (addr % 528);    
> > >
> > > Can you send a new version with
> > >
> > >         offset = addr % nor->page_size;
> > >
> > > to see if kbuild test robot still complains.
> > >    
> > 
> > This code:
> > 
> > static loff_t spi_nor_s3an_addr_convert(struct spi_nor *nor, loff_t addr)
> > {
> > unsigned int offset;
> > 
> > offset = addr % nor->page_size;
> > 
> > return ((addr - offset) << 1) | offset;
> > }
> > 
> > 
> > When built like:
> > 
> >    wget https://git.kernel.org/cgit/linux/kernel/git/wfg/lkp-tests.git/plain/sbin/make.cross
> > -O ~/bin/make.cross
> >         chmod +x ~/bin/make.cross
> >         # save the attached .config to linux build tree
> >         make.cross ARCH=blackfin
> > 
> > Produces this error:
> > 
> > 
> > drivers/built-in.o: In function `spi_nor_write':
> > drivers/mtd/spi-nor/spi-nor.c:(.text+0xfeb8): undefined reference to `__moddi3'
> > drivers/built-in.o: In function `spi_nor_read':
> > drivers/mtd/spi-nor/spi-nor.c:(.text+0x10248): undefined reference to `__moddi3'
> > drivers/built-in.o: In function `spi_nor_erase':
> > drivers/mtd/spi-nor/spi-nor.c:(.text+0x1034e): undefined reference to `__moddi3'
> > Makefile:956: recipe for target 'vmlinux' failed
> > 
> > 
> > But I think I found the right combination:
> > 
> > 
> > static loff_t spi_nor_s3an_addr_convert(struct spi_nor *nor, loff_t addr)
> > {
> > unsigned int offset = addr;
> > 
> > offset %= nor->page_size;
> > 
> > return ((addr - offset) << 1) | offset;
> > }
> > 
> > 
> > This one works fine on x86 and blackfin
> > 
> > Sending v7  
> 
> Wait. If you really want to manipulate an loff_t variable, you can do
> 
> 	offset = do_div(addr, nor->page_size);
> 
> 

Actually, you should have an intermediate u64 var, or just turn addr
into an u64 (loff_t is a long long not an unsigned long long).