Messages in this thread | | | From | Andy Shevchenko <> | Date | Wed, 21 Feb 2018 21:16:26 +0200 | Subject | Re: [PATCH] mmc: dw_mmc: Fix the DTO timeout overflow calculation for 32-bit systems |
| |
On Wed, Feb 21, 2018 at 7:57 PM, Evgeniy Didin <Evgeniy.Didin@synopsys.com> wrote: > In commit 9d9491a7da2a ("mmc: dw_mmc: Fix the DTO timeout calculation") have been made > changes which can cause multiply overflow for 32-bit systems. It was caught on arc/hsdk board > when "drto_ms" should be 671, but it was 70 and that caused a flow of stack-traces just copying > file from mmc. > > Lets cast this multiply to long long int which prevents overflow. > > Signed-off-by: Evgeniy Didin <Evgeniy.Didin@synopsys.com> > CC: Alexey Brodkin <abrodkin@synopsys.com> > CC: Douglas Anderson <dianders@chromium.org> > CC: Ulf Hansson <ulf.hansson@linaro.org> > CC: linux-kernel@vger.kernel.org > CC: linux-snps-arc@lists.infradead.org
> --- a/drivers/mmc/host/dw_mmc.c > +++ b/drivers/mmc/host/dw_mmc.c > @@ -1944,7 +1944,7 @@ static void dw_mci_set_drto(struct dw_mci *host) > drto_div = (mci_readl(host, CLKDIV) & 0xff) * 2; > if (drto_div == 0) > drto_div = 1; > - drto_ms = DIV_ROUND_UP(MSEC_PER_SEC * drto_clks * drto_div, > + drto_ms = DIV_ROUND_UP((uint64_t)MSEC_PER_SEC * drto_clks * drto_div, > host->bus_hz);
Does the driver use uint64_t elsewhere? Or simple u64?
-- With Best Regards, Andy Shevchenko
| |