* [PATCH 1/2 v4] Powerpc/eSDHC: Calculate the applicable mmc erase timeout value
@ 2012-11-29 4:00 Haijun Zhang
2012-11-29 4:20 ` Anton Vorontsov
2012-12-04 2:24 ` Powerpc : patches sent Zhang Haijun-B42677
0 siblings, 2 replies; 3+ messages in thread
From: Haijun Zhang @ 2012-11-29 4:00 UTC (permalink / raw)
To: linux-mmc; +Cc: Haijun Zhang, Jerry Huang, Anton Vorontsov
As large area erase needs long time usually a few minutes,
So we use max_discard_to to limit the size one request can
issue to avoid timeout. As our controller had the quirk
SDHCI_QUIRK_DATA_TIMEOUT_USES_SDCLK so we need to assign the
timeout_clk and max_discard_to according to the max_dtr which
can be read after the card had setup and decoded, As esdhc-imx
also use this func and this quirk so this patch will also work
on esdhci-imx, but this patch only test on powerpc platform, if
there are any negative impact on esdhc-imx please let me know.
Signed-off-by: Haijun Zhang <Haijun.Zhang@freescale.com>
Signed-off-by: Jerry Huang <Chang-Ming.Huang@freescale.com>
CC: Anton Vorontsov <cbouatmailru@gmail.com>
---
changes for v4:
- Apply the timeout clk after the card setup and reuse quirk
changes for v3:
- Add an empty line
drivers/mmc/host/sdhci-esdhc.h | 10 +++++++++-
1 files changed, 9 insertions(+), 1 deletions(-)
diff --git a/drivers/mmc/host/sdhci-esdhc.h b/drivers/mmc/host/sdhci-esdhc.h
index d25f9ab..49bc47c 100644
--- a/drivers/mmc/host/sdhci-esdhc.h
+++ b/drivers/mmc/host/sdhci-esdhc.h
@@ -46,7 +46,7 @@ static inline void esdhc_set_clock(struct sdhci_host *host, unsigned int clock)
{
int pre_div = 2;
int div = 1;
- u32 temp;
+ u32 temp, actual_clk;
if (clock == 0)
goto out;
@@ -65,6 +65,8 @@ static inline void esdhc_set_clock(struct sdhci_host *host, unsigned int clock)
dev_dbg(mmc_dev(host->mmc), "desired SD clock: %d, actual: %d\n",
clock, host->max_clk / pre_div / div);
+ actual_clk = host->max_clk / pre_div / div;
+
pre_div >>= 1;
div--;
@@ -74,6 +76,12 @@ static inline void esdhc_set_clock(struct sdhci_host *host, unsigned int clock)
| (pre_div << ESDHC_PREDIV_SHIFT));
sdhci_writel(host, temp, ESDHC_SYSTEM_CONTROL);
mdelay(1);
+
+ if (host->quirks & SDHCI_QUIRK_DATA_TIMEOUT_USES_SDCLK) {
+ host->timeout_clk = actual_clk / 1000;
+ host->mmc->max_discard_to = (1 << 27) / host->timeout_clk;
+ }
+
out:
host->clock = clock;
}
--
1.7.0.4
^ permalink raw reply related [flat|nested] 3+ messages in thread
* Re: [PATCH 1/2 v4] Powerpc/eSDHC: Calculate the applicable mmc erase timeout value
2012-11-29 4:00 [PATCH 1/2 v4] Powerpc/eSDHC: Calculate the applicable mmc erase timeout value Haijun Zhang
@ 2012-11-29 4:20 ` Anton Vorontsov
2012-12-04 2:24 ` Powerpc : patches sent Zhang Haijun-B42677
1 sibling, 0 replies; 3+ messages in thread
From: Anton Vorontsov @ 2012-11-29 4:20 UTC (permalink / raw)
To: Haijun Zhang; +Cc: linux-mmc, Jerry Huang
On Thu, Nov 29, 2012 at 12:00:16PM +0800, Haijun Zhang wrote:
> As large area erase needs long time usually a few minutes,
> So we use max_discard_to to limit the size one request can
> issue to avoid timeout. As our controller had the quirk
> SDHCI_QUIRK_DATA_TIMEOUT_USES_SDCLK so we need to assign the
> timeout_clk and max_discard_to according to the max_dtr which
> can be read after the card had setup and decoded, As esdhc-imx
> also use this func and this quirk so this patch will also work
> on esdhci-imx, but this patch only test on powerpc platform, if
> there are any negative impact on esdhc-imx please let me know.
>
> Signed-off-by: Haijun Zhang <Haijun.Zhang@freescale.com>
> Signed-off-by: Jerry Huang <Chang-Ming.Huang@freescale.com>
> CC: Anton Vorontsov <cbouatmailru@gmail.com>
> ---
> changes for v4:
> - Apply the timeout clk after the card setup and reuse quirk
> changes for v3:
> - Add an empty line
>
> drivers/mmc/host/sdhci-esdhc.h | 10 +++++++++-
> 1 files changed, 9 insertions(+), 1 deletions(-)
>
> diff --git a/drivers/mmc/host/sdhci-esdhc.h b/drivers/mmc/host/sdhci-esdhc.h
> index d25f9ab..49bc47c 100644
> --- a/drivers/mmc/host/sdhci-esdhc.h
> +++ b/drivers/mmc/host/sdhci-esdhc.h
> @@ -46,7 +46,7 @@ static inline void esdhc_set_clock(struct sdhci_host *host, unsigned int clock)
> {
> int pre_div = 2;
> int div = 1;
> - u32 temp;
> + u32 temp, actual_clk;
These two variables unrelated, please put them on their own lines.
Thanks,
>
> if (clock == 0)
> goto out;
> @@ -65,6 +65,8 @@ static inline void esdhc_set_clock(struct sdhci_host *host, unsigned int clock)
> dev_dbg(mmc_dev(host->mmc), "desired SD clock: %d, actual: %d\n",
> clock, host->max_clk / pre_div / div);
>
> + actual_clk = host->max_clk / pre_div / div;
> +
> pre_div >>= 1;
> div--;
>
> @@ -74,6 +76,12 @@ static inline void esdhc_set_clock(struct sdhci_host *host, unsigned int clock)
> | (pre_div << ESDHC_PREDIV_SHIFT));
> sdhci_writel(host, temp, ESDHC_SYSTEM_CONTROL);
> mdelay(1);
> +
> + if (host->quirks & SDHCI_QUIRK_DATA_TIMEOUT_USES_SDCLK) {
> + host->timeout_clk = actual_clk / 1000;
> + host->mmc->max_discard_to = (1 << 27) / host->timeout_clk;
> + }
> +
> out:
> host->clock = clock;
> }
> --
> 1.7.0.4
^ permalink raw reply [flat|nested] 3+ messages in thread
* Powerpc : patches sent
2012-11-29 4:00 [PATCH 1/2 v4] Powerpc/eSDHC: Calculate the applicable mmc erase timeout value Haijun Zhang
2012-11-29 4:20 ` Anton Vorontsov
@ 2012-12-04 2:24 ` Zhang Haijun-B42677
1 sibling, 0 replies; 3+ messages in thread
From: Zhang Haijun-B42677 @ 2012-12-04 2:24 UTC (permalink / raw)
To: Chris Ball
Cc: Huang Changming-R66093, Anton Vorontsov,
linux-mmc@vger.kernel.org
Hi Chris
I had sent six patches list below, they all Acked-by or Reviewed-by Anton.
Could you please help review these patches again? If resending these patches is needed Pls let me know.
They are sorted by date.
Thanks in advance.
1) Subject: [PATCH V5] Powerpc eSDHC Recover from the ADMA error
2) Subject: [PATCH] Powerpc eSDHC: Use u64 to calculate the timeout value to avoid overflow
3) Subject: [PATCH 2/2 v3] mmc:core: Add limit to data and erase timeout value calculation
4) Subject: [PATCH 1/2 v5] Powerpc/eSDHC: Calculate the applicable mmc erase timeout value
5)
6) Subject: [PATCH 1/2 V2] mmc:block : Convert the cmd_timeou_ms to u64 to avoid overflow
7) Subject: [PATCH 2/2 V2] mmc:host: Use u64 to calculate the timeout value to avoid overflow
Regards
Haijun.
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2012-12-04 2:24 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2012-11-29 4:00 [PATCH 1/2 v4] Powerpc/eSDHC: Calculate the applicable mmc erase timeout value Haijun Zhang
2012-11-29 4:20 ` Anton Vorontsov
2012-12-04 2:24 ` Powerpc : patches sent Zhang Haijun-B42677
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).