linux-kernel - Re: [PATCH 2/2] mtd: rawnand: use bounce buffer when vmalloced data buf detected

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]

Date:   Mon, 30 Sep 2019 12:33:06 -0400
From:   Kamal Dasu <kamal.dasu@...adcom.com>
To:     Boris Brezillon <boris.brezillon@...labora.com>
Cc:     Kamal Dasu <kdasu.kdev@...il.com>,
        Brian Norris <computersforpeace@...il.com>,
        Miquel Raynal <miquel.raynal@...tlin.com>,
        Richard Weinberger <richard@....at>,
        David Woodhouse <dwmw2@...radead.org>,
        Marek Vasut <marek.vasut@...il.com>,
        Vignesh Raghavendra <vigneshr@...com>,
        Boris Brezillon <bbrezillon@...nel.org>,
        Frieder Schrempf <frieder.schrempf@...tron.de>,
        MTD Maling List <linux-mtd@...ts.infradead.org>,
        bcm-kernel-feedback-list@...adcom.com,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH 2/2] mtd: rawnand: use bounce buffer when vmalloced data
 buf detected

On Mon, Sep 30, 2019 at 12:25 PM Boris Brezillon
<boris.brezillon@...labora.com> wrote:
>
> On Mon, 30 Sep 2019 12:01:28 -0400
> Kamal Dasu <kdasu.kdev@...il.com> wrote:
>
> > Does anyone have any comments on this patch ?.
> >
> > Kamal
> >
> > On Fri, Sep 6, 2019 at 3:49 PM Kamal Dasu <kdasu.kdev@...il.com> wrote:
> > >
> > > For controller drivers that use DMA and set NAND_USE_BOUNCE_BUFFER
> > > option use data buffers that are not vmalloced, aligned and have
> > > valid virtual address to be able to do DMA transfers. This change
> > > adds additional check and makes use of data buffer allocated
> > > in nand_base driver when it is passed a vmalloced data buffer for
> > > DMA transfers.
> > >
> > > Signed-off-by: Kamal Dasu <kdasu.kdev@...il.com>
> > > ---
> > >  drivers/mtd/nand/raw/nand_base.c | 14 ++++++++------
> > >  1 file changed, 8 insertions(+), 6 deletions(-)
> > >
> > > diff --git a/drivers/mtd/nand/raw/nand_base.c b/drivers/mtd/nand/raw/nand_base.c
> > > index 91f046d4d452..46f6965a896a 100644
> > > --- a/drivers/mtd/nand/raw/nand_base.c
> > > +++ b/drivers/mtd/nand/raw/nand_base.c
> > > @@ -45,6 +45,12 @@
> > >
> > >  #include "internals.h"
> > >
> > > +static int nand_need_bounce_buf(const void *buf, struct nand_chip *chip)
> > > +{
> > > +       return !virt_addr_valid(buf) || is_vmalloc_addr(buf) ||
>
> I thought virt_addr_valid() was implying !is_vmalloc_addr(), are you
> sure you need that test, and can you tell me on which arch(es) this is
> needed.
>

This has been observed on MIPS4K and MIPS5K architectures. There is a
check on the controller driver to use pio in such cases.

> > > +               !IS_ALIGNED((unsigned long)buf, chip->buf_align);
> > > +}
> > > +
> > >  /* Define default oob placement schemes for large and small page devices */
> > >  static int nand_ooblayout_ecc_sp(struct mtd_info *mtd, int section,
> > >                                  struct mtd_oob_region *oobregion)
> > > @@ -3183,9 +3189,7 @@ static int nand_do_read_ops(struct nand_chip *chip, loff_t from,
> > >                 if (!aligned)
> > >                         use_bufpoi = 1;
> > >                 else if (chip->options & NAND_USE_BOUNCE_BUFFER)
> > > -                       use_bufpoi = !virt_addr_valid(buf) ||
> > > -                                    !IS_ALIGNED((unsigned long)buf,
> > > -                                                chip->buf_align);
> > > +                       use_bufpoi = nand_need_bounce_buf(buf, chip);
> > >                 else
> > >                         use_bufpoi = 0;
> > >
> > > @@ -4009,9 +4013,7 @@ static int nand_do_write_ops(struct nand_chip *chip, loff_t to,
> > >                 if (part_pagewr)
> > >                         use_bufpoi = 1;
> > >                 else if (chip->options & NAND_USE_BOUNCE_BUFFER)
> > > -                       use_bufpoi = !virt_addr_valid(buf) ||
> > > -                                    !IS_ALIGNED((unsigned long)buf,
> > > -                                                chip->buf_align);
> > > +                       use_bufpoi = nand_need_bounce_buf(buf, chip);
> > >                 else
> > >                         use_bufpoi = 0;
> > >
> > > --
> > > 2.17.1
> > >
>