| 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
| ||
|
Message-ID: <01010164e92bcc18-d4fd119e-62e5-4aba-a6e3-9d5787f31a1b-000000@us-west-2.amazonses.com>
Date: Mon, 30 Jul 2018 03:11:59 +0000
From: Matthew Leon <matthewleon@...ux.com>
To: Jisheng Zhang <Jisheng.Zhang@...aptics.com>
Cc: Adrian Hunter <adrian.hunter@...el.com>,
Ulf Hansson <ulf.hansson@...aro.org>,
linux-mmc@...r.kernel.org, linux-kernel@...r.kernel.org,
linux-arm-kernel@...ts.infradead.org
Subject: Re: [PATCH mmc-next v3 3/3] mmc: sdhci-of-dwcmshc: solve 128MB DMA
boundary limitation
>> Hey Jisheng,
>Hi,
>>
>In LKML, we'd better not top post.
Noted. My apologies.
>> Shouldn't we be splitting until all DMA blocks are less than 128M
boundary?
>> I am a noob, but I think we should be prepared for boundaries that when
>> split in two, will still be greater than 128M. Feel free to disagree but
>> please explain why I may be wrong. Thank-you.
>the limitation is "DMA addr can't span 128MB boundary" rather than "must be
>less than 128MB", they are different.
>And the max transfer size of one DMA desc is 64KB.
>thanks
I have misspoken. What if the DMA transfer size is 1024M? If we split in
two, then we have 2 transfers, each of which span 512M. So wouldn't we need
to split again to have 4 transfers, each of which span 128M?
Sincerely,
Matthew Leon
On Sun, Jul 29, 2018 at 10:59 PM, Jisheng Zhang <Jisheng.Zhang@...aptics.com
> wrote:
> On Mon, 30 Jul 2018 02:56:20 +0000 Matthew Leon <matthewleon@...ux.com>
> wrote:
>
> > Hey Jisheng,
>
> Hi,
>
> >
>
> In LKML, we'd better not top post.
>
> > Shouldn't we be splitting until all DMA blocks are less than 128M
> boundary?
> > I am a noob, but I think we should be prepared for boundaries that when
> > split in two, will still be greater than 128M. Feel free to disagree but
> > please explain why I may be wrong. Thank-you.
>
> the limitation is "DMA addr can't span 128MB boundary" rather than "must be
> less than 128MB", they are different.
>
> And the max transfer size of one DMA desc is 64KB.
>
> thanks
>
> >
> > Sincerely,
> > Matthew Leon
> >
> > On Sun, Jul 29, 2018 at 10:46 PM, Jisheng Zhang <
> Jisheng.Zhang@...aptics.com
> > > wrote:
> >
> > > When using DMA, if the DMA addr spans 128MB boundary, we have to split
> > > the DMA transfer into two so that each one doesn't exceed the boundary.
> > >
> > > Signed-off-by: Jisheng Zhang <Jisheng.Zhang@...aptics.com>
> > > ---
> > > drivers/mmc/host/sdhci-of-dwcmshc.c | 43
> +++++++++++++++++++++++++++++
> > > 1 file changed, 43 insertions(+)
> > >
> > > diff --git a/drivers/mmc/host/sdhci-of-dwcmshc.c
> > > b/drivers/mmc/host/sdhci-of-dwcmshc.c
> > > index 1b7cd144fb01..e890fc8f5284 100644
> > > --- a/drivers/mmc/host/sdhci-of-dwcmshc.c
> > > +++ b/drivers/mmc/host/sdhci-of-dwcmshc.c
> > > @@ -8,21 +8,52 @@
> > > */
> > >
> > > #include <linux/clk.h>
> > > +#include <linux/mm.h>
> > > #include <linux/module.h>
> > > #include <linux/of.h>
> > > +#include <linux/sizes.h>
> > >
> > > #include "sdhci-pltfm.h"
> > >
> > > +#define BOUNDARY_OK(addr, len) \
> > > + ((addr | (SZ_128M - 1)) == ((addr + len - 1) | (SZ_128M - 1)))
> > > +
> > > struct dwcmshc_priv {
> > > struct clk *bus_clk;
> > > };
> > >
> > > +/*
> > > + * if DMA addr spans 128MB boundary, we split the DMA transfer into
> two
> > > + * so that the DMA transfer doesn't exceed the boundary.
> > > + */
> > > +static unsigned int dwcmshc_adma_write_desc(struct sdhci_host *host,
> > > + void *desc, dma_addr_t
> addr,
> > > + int len, unsigned int cmd)
> > > +{
> > > + int tmplen, offset;
> > > +
> > > + if (likely(!len || BOUNDARY_OK(addr, len)))
> > > + return _sdhci_adma_write_desc(host, desc, addr, len,
> cmd);
> > > +
> > > + offset = addr & (SZ_128M - 1);
> > > + tmplen = SZ_128M - offset;
> > > + _sdhci_adma_write_desc(host, desc, addr, tmplen, cmd);
> > > +
> > > + addr += tmplen;
> > > + len -= tmplen;
> > > + desc += host->desc_sz;
> > > + _sdhci_adma_write_desc(host, desc, addr, len, cmd);
> > > +
> > > + return host->desc_sz * 2;
> > > +}
> > > +
> > > static const struct sdhci_ops sdhci_dwcmshc_ops = {
> > > .set_clock = sdhci_set_clock,
> > > .set_bus_width = sdhci_set_bus_width,
> > > .set_uhs_signaling = sdhci_set_uhs_signaling,
> > > .get_max_clock = sdhci_pltfm_clk_get_max_clock,
> > > .reset = sdhci_reset,
> > > + .adma_write_desc = dwcmshc_adma_write_desc,
> > > };
> > >
> > > static const struct sdhci_pltfm_data sdhci_dwcmshc_pdata = {
> > > @@ -36,12 +67,24 @@ static int dwcmshc_probe(struct platform_device
> *pdev)
> > > struct sdhci_host *host;
> > > struct dwcmshc_priv *priv;
> > > int err;
> > > + u32 extra;
> > >
> > > host = sdhci_pltfm_init(pdev, &sdhci_dwcmshc_pdata,
> > > sizeof(struct dwcmshc_priv));
> > > if (IS_ERR(host))
> > > return PTR_ERR(host);
> > >
> > > + /*
> > > + * The DMA descriptor table number is calculated as the maximum
> > > + * number of segments times 2, to allow for an alignment
> > > + * descriptor for each segment, plus 1 for a nop end
> descriptor,
> > > + * plus extra number for cross 128M boundary handling.
> > > + */
> > > + extra = DIV_ROUND_UP(totalram_pages, SZ_128M / PAGE_SIZE);
> > > + if (extra > SDHCI_MAX_SEGS)
> > > + extra = SDHCI_MAX_SEGS;
> > > + host->adma_table_num = SDHCI_MAX_SEGS * 2 + 1 + extra;
> > > +
> > > pltfm_host = sdhci_priv(host);
> > > priv = sdhci_pltfm_priv(pltfm_host);
> > >
> > > --
> > > 2.18.0
> > >
> > >
>
>
Content of type "text/html" skipped
Powered by blists - more mailing lists