RE: [PATCH v2] mmc: dw_mmc: remove UBSAN warning in dw_mci_setup_bus()

2016-06-19 Thread Seung-Woo Kim 
Hi,

> -Original Message-
> From: Seung-Woo Kim [mailto:sw0312@samsung.com]
> Sent: Monday, June 20, 2016 12:30 PM
> To: 'Jaehoon Chung'; ulf.hans...@linaro.org; linux-...@vger.kernel.org; 
> linux-kernel@vger.kernel.org
> Cc: sw0312@samsung.com
> Subject: RE: [PATCH v2] mmc: dw_mmc: remove UBSAN warning in 
> dw_mci_setup_bus()
> 
> Hello Jaehoon,
> 
> > -Original Message-
> > From: Jaehoon Chung [mailto:jh80.ch...@samsung.com]
> > Sent: Monday, June 20, 2016 11:34 AM
> > To: Seung-Woo Kim; ulf.hans...@linaro.org; linux-...@vger.kernel.org; 
> > linux-kernel@vger.kernel.org
> > Subject: Re: [PATCH v2] mmc: dw_mmc: remove UBSAN warning in 
> > dw_mci_setup_bus()
> >
> > Hi SeungWoo,
> >
> > On 06/17/2016 02:16 PM, Seung-Woo Kim wrote:
> > > This patch removes following UBSAN warnings in dw_mci_setup_bus().
> > >
> > >   UBSAN: Undefined behaviour in drivers/mmc/host/dw_mmc.c:1102:14
> > >   shift exponent 250 is too large for 32-bit type 'unsigned int'
> > >   Call trace:
> > >   [] dump_backtrace+0x0/0x380
> > >   [] show_stack+0x14/0x20
> > >   [] dump_stack+0xe0/0x120
> > >   [] ubsan_epilogue+0x18/0x68
> > >   [] __ubsan_handle_shift_out_of_bounds+0x18c/0x1bc
> > >   [] dw_mci_setup_bus+0x3a0/0x438
> > >   [...]
> > >
> > >   UBSAN: Undefined behaviour in drivers/mmc/host/dw_mmc.c:1132:27
> > >   shift exponent 250 is too large for 32-bit type 'unsigned int'
> > >   Call trace:
> > >   [] dump_backtrace+0x0/0x380
> > >   [] show_stack+0x14/0x20
> > >   [] dump_stack+0xe0/0x120
> > >   [] ubsan_epilogue+0x18/0x68
> > >   [] __ubsan_handle_shift_out_of_bounds+0x18c/0x1bc
> > >   [] dw_mci_setup_bus+0x384/0x438
> > >   [...]
> > >
> > > The warnings are caused because of shift with more than 31 on 32
> > > bit variable, so this patch fixes to keep both clock and divider
> > > instead of shift.
> > >
> > > Signed-off-by: Seung-Woo Kim <sw0312@samsung.com>
> > > ---
> > >  drivers/mmc/host/dw_mmc.c |8 +---
> > >  drivers/mmc/host/dw_mmc.h |8 +---
> > >  2 files changed, 10 insertions(+), 6 deletions(-)
> > >
> > > diff --git a/drivers/mmc/host/dw_mmc.c b/drivers/mmc/host/dw_mmc.c
> > > index 2cc6123..d05c8cc 100644
> > > --- a/drivers/mmc/host/dw_mmc.c
> > > +++ b/drivers/mmc/host/dw_mmc.c
> > > @@ -1099,7 +1099,8 @@ static void dw_mci_setup_bus(struct dw_mci_slot 
> > > *slot, bool force_clkinit)
> > >
> > >   div = (host->bus_hz != clock) ? DIV_ROUND_UP(div, 2) : 0;
> > >
> > > - if ((clock << div) != slot->__clk_old || force_clkinit)
> > > + if (clock != slot->__clk_old || div != slot->__div_old ||
> > > + force_clkinit)
> >
> > I have realized why we put this condition..after checking your below codes.
> > This was patch to avoid the spamming for CONFIG_MMC_CLKGATE..
> > but CONFIG_MMC_CLKAGET didn't use anymore.. I think we can remove this 
> > condition.
> >
> > How about?
> 
> There is no CONFIG_MMC_CLKAGET, but message is still printed. So without
> the condition check, there will be too much log. If you will just remove
> below print, then I agree about removing the condition and __clk_old, instead
> of my patch.

I tested again after removing the condition check, it prints same log and there
is not too much log. Let's just remove the condition and __clk_old.

Regards,
- Seung-Woo Kim

> 
> Thanks,
> - Seung-Woo Kim
> 
> >
> > Best Regards,
> > Jaehoon Chung
> >
> > >   dev_info(>mmc->class_dev,
> > >"Bus speed (slot %d) = %dHz (slot req %dHz, 
> > > actual %dHZ div = %d)\n",
> > >slot->id, host->bus_hz, clock,
> > > @@ -1128,8 +1129,9 @@ static void dw_mci_setup_bus(struct dw_mci_slot 
> > > *slot, bool force_clkinit)
> > >   /* inform CIU */
> > >   mci_send_cmd(slot, sdmmc_cmd_bits, 0);
> > >
> > > - /* keep the clock with reflecting clock dividor */
> > > - slot->__clk_old = clock << div;
> > > + /* keep the clock and clock divider */
> > > + slot->__clk_old = clock;
> > > + slot->__div_old = div;
> > >   }
> > >
> > >   host->current_speed = clock;
> > > diff --git a/d

RE: [PATCH v2] mmc: dw_mmc: remove UBSAN warning in dw_mci_setup_bus()

2016-06-19 Thread Seung-Woo Kim 
Hi,

> -Original Message-
> From: Seung-Woo Kim [mailto:sw0312@samsung.com]
> Sent: Monday, June 20, 2016 12:30 PM
> To: 'Jaehoon Chung'; ulf.hans...@linaro.org; linux-...@vger.kernel.org; 
> linux-kernel@vger.kernel.org
> Cc: sw0312@samsung.com
> Subject: RE: [PATCH v2] mmc: dw_mmc: remove UBSAN warning in 
> dw_mci_setup_bus()
> 
> Hello Jaehoon,
> 
> > -Original Message-
> > From: Jaehoon Chung [mailto:jh80.ch...@samsung.com]
> > Sent: Monday, June 20, 2016 11:34 AM
> > To: Seung-Woo Kim; ulf.hans...@linaro.org; linux-...@vger.kernel.org; 
> > linux-kernel@vger.kernel.org
> > Subject: Re: [PATCH v2] mmc: dw_mmc: remove UBSAN warning in 
> > dw_mci_setup_bus()
> >
> > Hi SeungWoo,
> >
> > On 06/17/2016 02:16 PM, Seung-Woo Kim wrote:
> > > This patch removes following UBSAN warnings in dw_mci_setup_bus().
> > >
> > >   UBSAN: Undefined behaviour in drivers/mmc/host/dw_mmc.c:1102:14
> > >   shift exponent 250 is too large for 32-bit type 'unsigned int'
> > >   Call trace:
> > >   [] dump_backtrace+0x0/0x380
> > >   [] show_stack+0x14/0x20
> > >   [] dump_stack+0xe0/0x120
> > >   [] ubsan_epilogue+0x18/0x68
> > >   [] __ubsan_handle_shift_out_of_bounds+0x18c/0x1bc
> > >   [] dw_mci_setup_bus+0x3a0/0x438
> > >   [...]
> > >
> > >   UBSAN: Undefined behaviour in drivers/mmc/host/dw_mmc.c:1132:27
> > >   shift exponent 250 is too large for 32-bit type 'unsigned int'
> > >   Call trace:
> > >   [] dump_backtrace+0x0/0x380
> > >   [] show_stack+0x14/0x20
> > >   [] dump_stack+0xe0/0x120
> > >   [] ubsan_epilogue+0x18/0x68
> > >   [] __ubsan_handle_shift_out_of_bounds+0x18c/0x1bc
> > >   [] dw_mci_setup_bus+0x384/0x438
> > >   [...]
> > >
> > > The warnings are caused because of shift with more than 31 on 32
> > > bit variable, so this patch fixes to keep both clock and divider
> > > instead of shift.
> > >
> > > Signed-off-by: Seung-Woo Kim 
> > > ---
> > >  drivers/mmc/host/dw_mmc.c |8 +---
> > >  drivers/mmc/host/dw_mmc.h |8 +---
> > >  2 files changed, 10 insertions(+), 6 deletions(-)
> > >
> > > diff --git a/drivers/mmc/host/dw_mmc.c b/drivers/mmc/host/dw_mmc.c
> > > index 2cc6123..d05c8cc 100644
> > > --- a/drivers/mmc/host/dw_mmc.c
> > > +++ b/drivers/mmc/host/dw_mmc.c
> > > @@ -1099,7 +1099,8 @@ static void dw_mci_setup_bus(struct dw_mci_slot 
> > > *slot, bool force_clkinit)
> > >
> > >   div = (host->bus_hz != clock) ? DIV_ROUND_UP(div, 2) : 0;
> > >
> > > - if ((clock << div) != slot->__clk_old || force_clkinit)
> > > + if (clock != slot->__clk_old || div != slot->__div_old ||
> > > + force_clkinit)
> >
> > I have realized why we put this condition..after checking your below codes.
> > This was patch to avoid the spamming for CONFIG_MMC_CLKGATE..
> > but CONFIG_MMC_CLKAGET didn't use anymore.. I think we can remove this 
> > condition.
> >
> > How about?
> 
> There is no CONFIG_MMC_CLKAGET, but message is still printed. So without
> the condition check, there will be too much log. If you will just remove
> below print, then I agree about removing the condition and __clk_old, instead
> of my patch.

I tested again after removing the condition check, it prints same log and there
is not too much log. Let's just remove the condition and __clk_old.

Regards,
- Seung-Woo Kim

> 
> Thanks,
> - Seung-Woo Kim
> 
> >
> > Best Regards,
> > Jaehoon Chung
> >
> > >   dev_info(>mmc->class_dev,
> > >"Bus speed (slot %d) = %dHz (slot req %dHz, 
> > > actual %dHZ div = %d)\n",
> > >slot->id, host->bus_hz, clock,
> > > @@ -1128,8 +1129,9 @@ static void dw_mci_setup_bus(struct dw_mci_slot 
> > > *slot, bool force_clkinit)
> > >   /* inform CIU */
> > >   mci_send_cmd(slot, sdmmc_cmd_bits, 0);
> > >
> > > - /* keep the clock with reflecting clock dividor */
> > > - slot->__clk_old = clock << div;
> > > + /* keep the clock and clock divider */
> > > + slot->__clk_old = clock;
> > > + slot->__div_old = div;
> > >   }
> > >
> > >   host->current_speed = clock;
> > > diff --git a/drivers/mmc/host/dw_mmc.h b/dri

RE: [PATCH v2] mmc: dw_mmc: remove UBSAN warning in dw_mci_setup_bus()

2016-06-19 Thread Seung-Woo Kim 
Hello Jaehoon,

> -Original Message-
> From: Jaehoon Chung [mailto:jh80.ch...@samsung.com]
> Sent: Monday, June 20, 2016 11:34 AM
> To: Seung-Woo Kim; ulf.hans...@linaro.org; linux-...@vger.kernel.org; 
> linux-kernel@vger.kernel.org
> Subject: Re: [PATCH v2] mmc: dw_mmc: remove UBSAN warning in 
> dw_mci_setup_bus()
> 
> Hi SeungWoo,
> 
> On 06/17/2016 02:16 PM, Seung-Woo Kim wrote:
> > This patch removes following UBSAN warnings in dw_mci_setup_bus().
> >
> >   UBSAN: Undefined behaviour in drivers/mmc/host/dw_mmc.c:1102:14
> >   shift exponent 250 is too large for 32-bit type 'unsigned int'
> >   Call trace:
> >   [] dump_backtrace+0x0/0x380
> >   [] show_stack+0x14/0x20
> >   [] dump_stack+0xe0/0x120
> >   [] ubsan_epilogue+0x18/0x68
> >   [] __ubsan_handle_shift_out_of_bounds+0x18c/0x1bc
> >   [] dw_mci_setup_bus+0x3a0/0x438
> >   [...]
> >
> >   UBSAN: Undefined behaviour in drivers/mmc/host/dw_mmc.c:1132:27
> >   shift exponent 250 is too large for 32-bit type 'unsigned int'
> >   Call trace:
> >   [] dump_backtrace+0x0/0x380
> >   [] show_stack+0x14/0x20
> >   [] dump_stack+0xe0/0x120
> >   [] ubsan_epilogue+0x18/0x68
> >   [] __ubsan_handle_shift_out_of_bounds+0x18c/0x1bc
> >   [] dw_mci_setup_bus+0x384/0x438
> >   [...]
> >
> > The warnings are caused because of shift with more than 31 on 32
> > bit variable, so this patch fixes to keep both clock and divider
> > instead of shift.
> >
> > Signed-off-by: Seung-Woo Kim <sw0312@samsung.com>
> > ---
> >  drivers/mmc/host/dw_mmc.c |8 +---
> >  drivers/mmc/host/dw_mmc.h |8 +---
> >  2 files changed, 10 insertions(+), 6 deletions(-)
> >
> > diff --git a/drivers/mmc/host/dw_mmc.c b/drivers/mmc/host/dw_mmc.c
> > index 2cc6123..d05c8cc 100644
> > --- a/drivers/mmc/host/dw_mmc.c
> > +++ b/drivers/mmc/host/dw_mmc.c
> > @@ -1099,7 +1099,8 @@ static void dw_mci_setup_bus(struct dw_mci_slot 
> > *slot, bool force_clkinit)
> >
> > div = (host->bus_hz != clock) ? DIV_ROUND_UP(div, 2) : 0;
> >
> > -   if ((clock << div) != slot->__clk_old || force_clkinit)
> > +   if (clock != slot->__clk_old || div != slot->__div_old ||
> > +   force_clkinit)
> 
> I have realized why we put this condition..after checking your below codes.
> This was patch to avoid the spamming for CONFIG_MMC_CLKGATE..
> but CONFIG_MMC_CLKAGET didn't use anymore.. I think we can remove this 
> condition.
> 
> How about?

There is no CONFIG_MMC_CLKAGET, but message is still printed. So without
the condition check, there will be too much log. If you will just remove
below print, then I agree about removing the condition and __clk_old, instead
of my patch.

Thanks,
- Seung-Woo Kim

> 
> Best Regards,
> Jaehoon Chung
> 
> > dev_info(>mmc->class_dev,
> >  "Bus speed (slot %d) = %dHz (slot req %dHz, 
> > actual %dHZ div = %d)\n",
> >  slot->id, host->bus_hz, clock,
> > @@ -1128,8 +1129,9 @@ static void dw_mci_setup_bus(struct dw_mci_slot 
> > *slot, bool force_clkinit)
> > /* inform CIU */
> > mci_send_cmd(slot, sdmmc_cmd_bits, 0);
> >
> > -   /* keep the clock with reflecting clock dividor */
> > -   slot->__clk_old = clock << div;
> > +   /* keep the clock and clock divider */
> > +   slot->__clk_old = clock;
> > +   slot->__div_old = div;
> > }
> >
> > host->current_speed = clock;
> > diff --git a/drivers/mmc/host/dw_mmc.h b/drivers/mmc/host/dw_mmc.h
> > index 1e8d838..fdfc3f5 100644
> > --- a/drivers/mmc/host/dw_mmc.h
> > +++ b/drivers/mmc/host/dw_mmc.h
> > @@ -245,9 +245,10 @@ extern int dw_mci_resume(struct dw_mci *host);
> >   * @queue_node: List node for placing this node in the @queue list of
> >   *  dw_mci.
> >   * @clock: Clock rate configured by set_ios(). Protected by host->lock.
> > - * @__clk_old: The last updated clock with reflecting clock divider.
> > - * Keeping track of this helps us to avoid spamming the console
> > - * with CONFIG_MMC_CLKGATE.
> > + * @__clk_old: The last updated clock.
> > + * @__div_old: The last updated clock divider.
> > + * Keeping track of clock and clock divider helps us to avoid spamming
> > + * the console with CONFIG_MMC_CLKGATE.
> >   * @flags: Random state bits associated with the slot.
> >   * @id: Number of this slot.
> >   * @sdio_id: Number of this slot in the SDIO interrupt registers.
> > @@ -263,6 +264,7 @@ struct dw_mci_slot {
> >
> > unsigned intclock;
> > unsigned int__clk_old;
> > +   unsigned int__div_old;
> >
> > unsigned long   flags;
> >  #define DW_MMC_CARD_PRESENT0
> >

RE: [PATCH v2] mmc: dw_mmc: remove UBSAN warning in dw_mci_setup_bus()

2016-06-19 Thread Seung-Woo Kim 
Hello Jaehoon,

> -Original Message-
> From: Jaehoon Chung [mailto:jh80.ch...@samsung.com]
> Sent: Monday, June 20, 2016 11:34 AM
> To: Seung-Woo Kim; ulf.hans...@linaro.org; linux-...@vger.kernel.org; 
> linux-kernel@vger.kernel.org
> Subject: Re: [PATCH v2] mmc: dw_mmc: remove UBSAN warning in 
> dw_mci_setup_bus()
> 
> Hi SeungWoo,
> 
> On 06/17/2016 02:16 PM, Seung-Woo Kim wrote:
> > This patch removes following UBSAN warnings in dw_mci_setup_bus().
> >
> >   UBSAN: Undefined behaviour in drivers/mmc/host/dw_mmc.c:1102:14
> >   shift exponent 250 is too large for 32-bit type 'unsigned int'
> >   Call trace:
> >   [] dump_backtrace+0x0/0x380
> >   [] show_stack+0x14/0x20
> >   [] dump_stack+0xe0/0x120
> >   [] ubsan_epilogue+0x18/0x68
> >   [] __ubsan_handle_shift_out_of_bounds+0x18c/0x1bc
> >   [] dw_mci_setup_bus+0x3a0/0x438
> >   [...]
> >
> >   UBSAN: Undefined behaviour in drivers/mmc/host/dw_mmc.c:1132:27
> >   shift exponent 250 is too large for 32-bit type 'unsigned int'
> >   Call trace:
> >   [] dump_backtrace+0x0/0x380
> >   [] show_stack+0x14/0x20
> >   [] dump_stack+0xe0/0x120
> >   [] ubsan_epilogue+0x18/0x68
> >   [] __ubsan_handle_shift_out_of_bounds+0x18c/0x1bc
> >   [] dw_mci_setup_bus+0x384/0x438
> >   [...]
> >
> > The warnings are caused because of shift with more than 31 on 32
> > bit variable, so this patch fixes to keep both clock and divider
> > instead of shift.
> >
> > Signed-off-by: Seung-Woo Kim 
> > ---
> >  drivers/mmc/host/dw_mmc.c |8 +---
> >  drivers/mmc/host/dw_mmc.h |8 +---
> >  2 files changed, 10 insertions(+), 6 deletions(-)
> >
> > diff --git a/drivers/mmc/host/dw_mmc.c b/drivers/mmc/host/dw_mmc.c
> > index 2cc6123..d05c8cc 100644
> > --- a/drivers/mmc/host/dw_mmc.c
> > +++ b/drivers/mmc/host/dw_mmc.c
> > @@ -1099,7 +1099,8 @@ static void dw_mci_setup_bus(struct dw_mci_slot 
> > *slot, bool force_clkinit)
> >
> > div = (host->bus_hz != clock) ? DIV_ROUND_UP(div, 2) : 0;
> >
> > -   if ((clock << div) != slot->__clk_old || force_clkinit)
> > +   if (clock != slot->__clk_old || div != slot->__div_old ||
> > +   force_clkinit)
> 
> I have realized why we put this condition..after checking your below codes.
> This was patch to avoid the spamming for CONFIG_MMC_CLKGATE..
> but CONFIG_MMC_CLKAGET didn't use anymore.. I think we can remove this 
> condition.
> 
> How about?

There is no CONFIG_MMC_CLKAGET, but message is still printed. So without
the condition check, there will be too much log. If you will just remove
below print, then I agree about removing the condition and __clk_old, instead
of my patch.

Thanks,
- Seung-Woo Kim

> 
> Best Regards,
> Jaehoon Chung
> 
> > dev_info(>mmc->class_dev,
> >  "Bus speed (slot %d) = %dHz (slot req %dHz, 
> > actual %dHZ div = %d)\n",
> >  slot->id, host->bus_hz, clock,
> > @@ -1128,8 +1129,9 @@ static void dw_mci_setup_bus(struct dw_mci_slot 
> > *slot, bool force_clkinit)
> > /* inform CIU */
> > mci_send_cmd(slot, sdmmc_cmd_bits, 0);
> >
> > -   /* keep the clock with reflecting clock dividor */
> > -   slot->__clk_old = clock << div;
> > +   /* keep the clock and clock divider */
> > +   slot->__clk_old = clock;
> > +   slot->__div_old = div;
> > }
> >
> > host->current_speed = clock;
> > diff --git a/drivers/mmc/host/dw_mmc.h b/drivers/mmc/host/dw_mmc.h
> > index 1e8d838..fdfc3f5 100644
> > --- a/drivers/mmc/host/dw_mmc.h
> > +++ b/drivers/mmc/host/dw_mmc.h
> > @@ -245,9 +245,10 @@ extern int dw_mci_resume(struct dw_mci *host);
> >   * @queue_node: List node for placing this node in the @queue list of
> >   *  dw_mci.
> >   * @clock: Clock rate configured by set_ios(). Protected by host->lock.
> > - * @__clk_old: The last updated clock with reflecting clock divider.
> > - * Keeping track of this helps us to avoid spamming the console
> > - * with CONFIG_MMC_CLKGATE.
> > + * @__clk_old: The last updated clock.
> > + * @__div_old: The last updated clock divider.
> > + * Keeping track of clock and clock divider helps us to avoid spamming
> > + * the console with CONFIG_MMC_CLKGATE.
> >   * @flags: Random state bits associated with the slot.
> >   * @id: Number of this slot.
> >   * @sdio_id: Number of this slot in the SDIO interrupt registers.
> > @@ -263,6 +264,7 @@ struct dw_mci_slot {
> >
> > unsigned intclock;
> > unsigned int__clk_old;
> > +   unsigned int__div_old;
> >
> > unsigned long   flags;
> >  #define DW_MMC_CARD_PRESENT0
> >

Re: [PATCH v2] mmc: dw_mmc: remove UBSAN warning in dw_mci_setup_bus()

2016-06-19 Thread Jaehoon Chung 
Hi SeungWoo,

On 06/17/2016 02:16 PM, Seung-Woo Kim wrote:
> This patch removes following UBSAN warnings in dw_mci_setup_bus().
> 
>   UBSAN: Undefined behaviour in drivers/mmc/host/dw_mmc.c:1102:14
>   shift exponent 250 is too large for 32-bit type 'unsigned int'
>   Call trace:
>   [] dump_backtrace+0x0/0x380
>   [] show_stack+0x14/0x20
>   [] dump_stack+0xe0/0x120
>   [] ubsan_epilogue+0x18/0x68
>   [] __ubsan_handle_shift_out_of_bounds+0x18c/0x1bc
>   [] dw_mci_setup_bus+0x3a0/0x438
>   [...]
> 
>   UBSAN: Undefined behaviour in drivers/mmc/host/dw_mmc.c:1132:27
>   shift exponent 250 is too large for 32-bit type 'unsigned int'
>   Call trace:
>   [] dump_backtrace+0x0/0x380
>   [] show_stack+0x14/0x20
>   [] dump_stack+0xe0/0x120
>   [] ubsan_epilogue+0x18/0x68
>   [] __ubsan_handle_shift_out_of_bounds+0x18c/0x1bc
>   [] dw_mci_setup_bus+0x384/0x438
>   [...]
> 
> The warnings are caused because of shift with more than 31 on 32
> bit variable, so this patch fixes to keep both clock and divider
> instead of shift.
> 
> Signed-off-by: Seung-Woo Kim 
> ---
>  drivers/mmc/host/dw_mmc.c |8 +---
>  drivers/mmc/host/dw_mmc.h |8 +---
>  2 files changed, 10 insertions(+), 6 deletions(-)
> 
> diff --git a/drivers/mmc/host/dw_mmc.c b/drivers/mmc/host/dw_mmc.c
> index 2cc6123..d05c8cc 100644
> --- a/drivers/mmc/host/dw_mmc.c
> +++ b/drivers/mmc/host/dw_mmc.c
> @@ -1099,7 +1099,8 @@ static void dw_mci_setup_bus(struct dw_mci_slot *slot, 
> bool force_clkinit)
>  
>   div = (host->bus_hz != clock) ? DIV_ROUND_UP(div, 2) : 0;
>  
> - if ((clock << div) != slot->__clk_old || force_clkinit)
> + if (clock != slot->__clk_old || div != slot->__div_old ||
> + force_clkinit)

I have realized why we put this condition..after checking your below codes.
This was patch to avoid the spamming for CONFIG_MMC_CLKGATE..
but CONFIG_MMC_CLKAGET didn't use anymore.. I think we can remove this 
condition.

How about?

Best Regards,
Jaehoon Chung

>   dev_info(>mmc->class_dev,
>"Bus speed (slot %d) = %dHz (slot req %dHz, 
> actual %dHZ div = %d)\n",
>slot->id, host->bus_hz, clock,
> @@ -1128,8 +1129,9 @@ static void dw_mci_setup_bus(struct dw_mci_slot *slot, 
> bool force_clkinit)
>   /* inform CIU */
>   mci_send_cmd(slot, sdmmc_cmd_bits, 0);
>  
> - /* keep the clock with reflecting clock dividor */
> - slot->__clk_old = clock << div;
> + /* keep the clock and clock divider */
> + slot->__clk_old = clock;
> + slot->__div_old = div;
>   }
>  
>   host->current_speed = clock;
> diff --git a/drivers/mmc/host/dw_mmc.h b/drivers/mmc/host/dw_mmc.h
> index 1e8d838..fdfc3f5 100644
> --- a/drivers/mmc/host/dw_mmc.h
> +++ b/drivers/mmc/host/dw_mmc.h
> @@ -245,9 +245,10 @@ extern int dw_mci_resume(struct dw_mci *host);
>   * @queue_node: List node for placing this node in the @queue list of
>   *dw_mci.
>   * @clock: Clock rate configured by set_ios(). Protected by host->lock.
> - * @__clk_old: The last updated clock with reflecting clock divider.
> - *   Keeping track of this helps us to avoid spamming the console
> - *   with CONFIG_MMC_CLKGATE.
> + * @__clk_old: The last updated clock.
> + * @__div_old: The last updated clock divider.
> + *   Keeping track of clock and clock divider helps us to avoid spamming
> + *   the console with CONFIG_MMC_CLKGATE.
>   * @flags: Random state bits associated with the slot.
>   * @id: Number of this slot.
>   * @sdio_id: Number of this slot in the SDIO interrupt registers.
> @@ -263,6 +264,7 @@ struct dw_mci_slot {
>  
>   unsigned intclock;
>   unsigned int__clk_old;
> + unsigned int__div_old;
>  
>   unsigned long   flags;
>  #define DW_MMC_CARD_PRESENT  0
>

Re: [PATCH v2] mmc: dw_mmc: remove UBSAN warning in dw_mci_setup_bus()

2016-06-19 Thread Jaehoon Chung 
Hi SeungWoo,

On 06/17/2016 02:16 PM, Seung-Woo Kim wrote:
> This patch removes following UBSAN warnings in dw_mci_setup_bus().
> 
>   UBSAN: Undefined behaviour in drivers/mmc/host/dw_mmc.c:1102:14
>   shift exponent 250 is too large for 32-bit type 'unsigned int'
>   Call trace:
>   [] dump_backtrace+0x0/0x380
>   [] show_stack+0x14/0x20
>   [] dump_stack+0xe0/0x120
>   [] ubsan_epilogue+0x18/0x68
>   [] __ubsan_handle_shift_out_of_bounds+0x18c/0x1bc
>   [] dw_mci_setup_bus+0x3a0/0x438
>   [...]
> 
>   UBSAN: Undefined behaviour in drivers/mmc/host/dw_mmc.c:1132:27
>   shift exponent 250 is too large for 32-bit type 'unsigned int'
>   Call trace:
>   [] dump_backtrace+0x0/0x380
>   [] show_stack+0x14/0x20
>   [] dump_stack+0xe0/0x120
>   [] ubsan_epilogue+0x18/0x68
>   [] __ubsan_handle_shift_out_of_bounds+0x18c/0x1bc
>   [] dw_mci_setup_bus+0x384/0x438
>   [...]
> 
> The warnings are caused because of shift with more than 31 on 32
> bit variable, so this patch fixes to keep both clock and divider
> instead of shift.
> 
> Signed-off-by: Seung-Woo Kim 
> ---
>  drivers/mmc/host/dw_mmc.c |8 +---
>  drivers/mmc/host/dw_mmc.h |8 +---
>  2 files changed, 10 insertions(+), 6 deletions(-)
> 
> diff --git a/drivers/mmc/host/dw_mmc.c b/drivers/mmc/host/dw_mmc.c
> index 2cc6123..d05c8cc 100644
> --- a/drivers/mmc/host/dw_mmc.c
> +++ b/drivers/mmc/host/dw_mmc.c
> @@ -1099,7 +1099,8 @@ static void dw_mci_setup_bus(struct dw_mci_slot *slot, 
> bool force_clkinit)
>  
>   div = (host->bus_hz != clock) ? DIV_ROUND_UP(div, 2) : 0;
>  
> - if ((clock << div) != slot->__clk_old || force_clkinit)
> + if (clock != slot->__clk_old || div != slot->__div_old ||
> + force_clkinit)

I have realized why we put this condition..after checking your below codes.
This was patch to avoid the spamming for CONFIG_MMC_CLKGATE..
but CONFIG_MMC_CLKAGET didn't use anymore.. I think we can remove this 
condition.

How about?

Best Regards,
Jaehoon Chung

>   dev_info(>mmc->class_dev,
>"Bus speed (slot %d) = %dHz (slot req %dHz, 
> actual %dHZ div = %d)\n",
>slot->id, host->bus_hz, clock,
> @@ -1128,8 +1129,9 @@ static void dw_mci_setup_bus(struct dw_mci_slot *slot, 
> bool force_clkinit)
>   /* inform CIU */
>   mci_send_cmd(slot, sdmmc_cmd_bits, 0);
>  
> - /* keep the clock with reflecting clock dividor */
> - slot->__clk_old = clock << div;
> + /* keep the clock and clock divider */
> + slot->__clk_old = clock;
> + slot->__div_old = div;
>   }
>  
>   host->current_speed = clock;
> diff --git a/drivers/mmc/host/dw_mmc.h b/drivers/mmc/host/dw_mmc.h
> index 1e8d838..fdfc3f5 100644
> --- a/drivers/mmc/host/dw_mmc.h
> +++ b/drivers/mmc/host/dw_mmc.h
> @@ -245,9 +245,10 @@ extern int dw_mci_resume(struct dw_mci *host);
>   * @queue_node: List node for placing this node in the @queue list of
>   *dw_mci.
>   * @clock: Clock rate configured by set_ios(). Protected by host->lock.
> - * @__clk_old: The last updated clock with reflecting clock divider.
> - *   Keeping track of this helps us to avoid spamming the console
> - *   with CONFIG_MMC_CLKGATE.
> + * @__clk_old: The last updated clock.
> + * @__div_old: The last updated clock divider.
> + *   Keeping track of clock and clock divider helps us to avoid spamming
> + *   the console with CONFIG_MMC_CLKGATE.
>   * @flags: Random state bits associated with the slot.
>   * @id: Number of this slot.
>   * @sdio_id: Number of this slot in the SDIO interrupt registers.
> @@ -263,6 +264,7 @@ struct dw_mci_slot {
>  
>   unsigned intclock;
>   unsigned int__clk_old;
> + unsigned int__div_old;
>  
>   unsigned long   flags;
>  #define DW_MMC_CARD_PRESENT  0
>

[PATCH v2] mmc: dw_mmc: remove UBSAN warning in dw_mci_setup_bus()

2016-06-16 Thread Seung-Woo Kim 
This patch removes following UBSAN warnings in dw_mci_setup_bus().

  UBSAN: Undefined behaviour in drivers/mmc/host/dw_mmc.c:1102:14
  shift exponent 250 is too large for 32-bit type 'unsigned int'
  Call trace:
  [] dump_backtrace+0x0/0x380
  [] show_stack+0x14/0x20
  [] dump_stack+0xe0/0x120
  [] ubsan_epilogue+0x18/0x68
  [] __ubsan_handle_shift_out_of_bounds+0x18c/0x1bc
  [] dw_mci_setup_bus+0x3a0/0x438
  [...]

  UBSAN: Undefined behaviour in drivers/mmc/host/dw_mmc.c:1132:27
  shift exponent 250 is too large for 32-bit type 'unsigned int'
  Call trace:
  [] dump_backtrace+0x0/0x380
  [] show_stack+0x14/0x20
  [] dump_stack+0xe0/0x120
  [] ubsan_epilogue+0x18/0x68
  [] __ubsan_handle_shift_out_of_bounds+0x18c/0x1bc
  [] dw_mci_setup_bus+0x384/0x438
  [...]

The warnings are caused because of shift with more than 31 on 32
bit variable, so this patch fixes to keep both clock and divider
instead of shift.

Signed-off-by: Seung-Woo Kim 
---
 drivers/mmc/host/dw_mmc.c |8 +---
 drivers/mmc/host/dw_mmc.h |8 +---
 2 files changed, 10 insertions(+), 6 deletions(-)

diff --git a/drivers/mmc/host/dw_mmc.c b/drivers/mmc/host/dw_mmc.c
index 2cc6123..d05c8cc 100644
--- a/drivers/mmc/host/dw_mmc.c
+++ b/drivers/mmc/host/dw_mmc.c
@@ -1099,7 +1099,8 @@ static void dw_mci_setup_bus(struct dw_mci_slot *slot, 
bool force_clkinit)
 
div = (host->bus_hz != clock) ? DIV_ROUND_UP(div, 2) : 0;
 
-   if ((clock << div) != slot->__clk_old || force_clkinit)
+   if (clock != slot->__clk_old || div != slot->__div_old ||
+   force_clkinit)
dev_info(>mmc->class_dev,
 "Bus speed (slot %d) = %dHz (slot req %dHz, 
actual %dHZ div = %d)\n",
 slot->id, host->bus_hz, clock,
@@ -1128,8 +1129,9 @@ static void dw_mci_setup_bus(struct dw_mci_slot *slot, 
bool force_clkinit)
/* inform CIU */
mci_send_cmd(slot, sdmmc_cmd_bits, 0);
 
-   /* keep the clock with reflecting clock dividor */
-   slot->__clk_old = clock << div;
+   /* keep the clock and clock divider */
+   slot->__clk_old = clock;
+   slot->__div_old = div;
}
 
host->current_speed = clock;
diff --git a/drivers/mmc/host/dw_mmc.h b/drivers/mmc/host/dw_mmc.h
index 1e8d838..fdfc3f5 100644
--- a/drivers/mmc/host/dw_mmc.h
+++ b/drivers/mmc/host/dw_mmc.h
@@ -245,9 +245,10 @@ extern int dw_mci_resume(struct dw_mci *host);
  * @queue_node: List node for placing this node in the @queue list of
  *  dw_mci.
  * @clock: Clock rate configured by set_ios(). Protected by host->lock.
- * @__clk_old: The last updated clock with reflecting clock divider.
- * Keeping track of this helps us to avoid spamming the console
- * with CONFIG_MMC_CLKGATE.
+ * @__clk_old: The last updated clock.
+ * @__div_old: The last updated clock divider.
+ * Keeping track of clock and clock divider helps us to avoid spamming
+ * the console with CONFIG_MMC_CLKGATE.
  * @flags: Random state bits associated with the slot.
  * @id: Number of this slot.
  * @sdio_id: Number of this slot in the SDIO interrupt registers.
@@ -263,6 +264,7 @@ struct dw_mci_slot {
 
unsigned intclock;
unsigned int__clk_old;
+   unsigned int__div_old;
 
unsigned long   flags;
 #define DW_MMC_CARD_PRESENT0
-- 
1.7.4.1

[PATCH v2] mmc: dw_mmc: remove UBSAN warning in dw_mci_setup_bus()

2016-06-16 Thread Seung-Woo Kim 
This patch removes following UBSAN warnings in dw_mci_setup_bus().

  UBSAN: Undefined behaviour in drivers/mmc/host/dw_mmc.c:1102:14
  shift exponent 250 is too large for 32-bit type 'unsigned int'
  Call trace:
  [] dump_backtrace+0x0/0x380
  [] show_stack+0x14/0x20
  [] dump_stack+0xe0/0x120
  [] ubsan_epilogue+0x18/0x68
  [] __ubsan_handle_shift_out_of_bounds+0x18c/0x1bc
  [] dw_mci_setup_bus+0x3a0/0x438
  [...]

  UBSAN: Undefined behaviour in drivers/mmc/host/dw_mmc.c:1132:27
  shift exponent 250 is too large for 32-bit type 'unsigned int'
  Call trace:
  [] dump_backtrace+0x0/0x380
  [] show_stack+0x14/0x20
  [] dump_stack+0xe0/0x120
  [] ubsan_epilogue+0x18/0x68
  [] __ubsan_handle_shift_out_of_bounds+0x18c/0x1bc
  [] dw_mci_setup_bus+0x384/0x438
  [...]

The warnings are caused because of shift with more than 31 on 32
bit variable, so this patch fixes to keep both clock and divider
instead of shift.

Signed-off-by: Seung-Woo Kim 
---
 drivers/mmc/host/dw_mmc.c |8 +---
 drivers/mmc/host/dw_mmc.h |8 +---
 2 files changed, 10 insertions(+), 6 deletions(-)

diff --git a/drivers/mmc/host/dw_mmc.c b/drivers/mmc/host/dw_mmc.c
index 2cc6123..d05c8cc 100644
--- a/drivers/mmc/host/dw_mmc.c
+++ b/drivers/mmc/host/dw_mmc.c
@@ -1099,7 +1099,8 @@ static void dw_mci_setup_bus(struct dw_mci_slot *slot, 
bool force_clkinit)
 
div = (host->bus_hz != clock) ? DIV_ROUND_UP(div, 2) : 0;
 
-   if ((clock << div) != slot->__clk_old || force_clkinit)
+   if (clock != slot->__clk_old || div != slot->__div_old ||
+   force_clkinit)
dev_info(>mmc->class_dev,
 "Bus speed (slot %d) = %dHz (slot req %dHz, 
actual %dHZ div = %d)\n",
 slot->id, host->bus_hz, clock,
@@ -1128,8 +1129,9 @@ static void dw_mci_setup_bus(struct dw_mci_slot *slot, 
bool force_clkinit)
/* inform CIU */
mci_send_cmd(slot, sdmmc_cmd_bits, 0);
 
-   /* keep the clock with reflecting clock dividor */
-   slot->__clk_old = clock << div;
+   /* keep the clock and clock divider */
+   slot->__clk_old = clock;
+   slot->__div_old = div;
}
 
host->current_speed = clock;
diff --git a/drivers/mmc/host/dw_mmc.h b/drivers/mmc/host/dw_mmc.h
index 1e8d838..fdfc3f5 100644
--- a/drivers/mmc/host/dw_mmc.h
+++ b/drivers/mmc/host/dw_mmc.h
@@ -245,9 +245,10 @@ extern int dw_mci_resume(struct dw_mci *host);
  * @queue_node: List node for placing this node in the @queue list of
  *  dw_mci.
  * @clock: Clock rate configured by set_ios(). Protected by host->lock.
- * @__clk_old: The last updated clock with reflecting clock divider.
- * Keeping track of this helps us to avoid spamming the console
- * with CONFIG_MMC_CLKGATE.
+ * @__clk_old: The last updated clock.
+ * @__div_old: The last updated clock divider.
+ * Keeping track of clock and clock divider helps us to avoid spamming
+ * the console with CONFIG_MMC_CLKGATE.
  * @flags: Random state bits associated with the slot.
  * @id: Number of this slot.
  * @sdio_id: Number of this slot in the SDIO interrupt registers.
@@ -263,6 +264,7 @@ struct dw_mci_slot {
 
unsigned intclock;
unsigned int__clk_old;
+   unsigned int__div_old;
 
unsigned long   flags;
 #define DW_MMC_CARD_PRESENT0
-- 
1.7.4.1