sdhci-stm performance issues

sdhci-stm performance issues

[Peppe CAVALLARO]

Hello,
testing the sdhci-stm.c driver, I sent some weeks ago to the mailing list,
I'm facing some performance issues.
For example, when copy a file (e.g. ~10MiB) to the card, I've noticed that,
in the sdhci_prepare_data, the maximum value for data->blocks is 248.

The sdhci-stm driver uses the ADMA and:
  max_seg_size = 65536, 
  max_hw_segs = 128,
  max_phys_segs = 128,
  max_req_size = 524288, 
  max_blk_size = 2048,
  max_blk_count = 65535.

Using my arasan.c code, I've also sent to this mailing list in July,
I could see that the data->blocks was bigger than 248 and the performances were better.
Note that this driver used the same mmc_host parameters (max_blk_size etc.) reported
above for the sdhci-stm.

What am I missing?

Welcome advice.

Best Regards,
Giuseppe

Re: sdhci-stm performance issues

[Peppe CAVALLARO]

[-- Attachment #1: Type: text/plain, Size: 1316 bytes --]

 Hello
The problem was due to the missing both 8BIT and MMC_SPEED supports in
my sdhci driver (comes from the 2.6.32 Kernel series).
I've also added the patch in attachment.

Best Regards
Peppe

On 8/24/2010 5:11 PM, Peppe CAVALLARO wrote:
>
> Hello,
> testing the sdhci-stm.c driver, I sent some weeks ago to the mailing list,
> I'm facing some performance issues.
> For example, when copy a file (e.g. ~10MiB) to the card, I've noticed
> that,
> in the sdhci_prepare_data, the maximum value for data->blocks is 248.
>
> The sdhci-stm driver uses the ADMA and:
>   max_seg_size = 65536,
>   max_hw_segs = 128,
>   max_phys_segs = 128,
>   max_req_size = 524288,
>   max_blk_size = 2048,
>   max_blk_count = 65535.
>
> Using my arasan.c code, I've also sent to this mailing list in July,
> I could see that the data->blocks was bigger than 248 and the
> performances were better.
> Note that this driver used the same mmc_host parameters (max_blk_size
> etc.) reported
> above for the sdhci-stm.
>
> What am I missing?
>
> Welcome advice.
>
> Best Regards,
> Giuseppe
> --
> To unsubscribe from this list: send the line "unsubscribe linux-mmc" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>


[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: 0001-sdhci-add-MMC_CAP_8_BIT_DATA-in-the-Host-capabiliti.patch --]
[-- Type: text/x-patch; name="0001-sdhci-add-MMC_CAP_8_BIT_DATA-in-the-Host-capabiliti.patch", Size: 912 bytes --]

From b3427a938ed04febdaa729ca500f266274e06bd2 Mon Sep 17 00:00:00 2001
From: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Date: Tue, 31 Aug 2010 05:43:45 +0200
Subject: [PATCH (sh-2.6.32.y)] sdhci: add MMC_CAP_8_BIT_DATA in the Host capabilities

Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
---
 drivers/mmc/host/sdhci.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/drivers/mmc/host/sdhci.c b/drivers/mmc/host/sdhci.c
index 955cad9..1c49326 100644
--- a/drivers/mmc/host/sdhci.c
+++ b/drivers/mmc/host/sdhci.c
@@ -1799,7 +1799,7 @@ int sdhci_add_host(struct sdhci_host *host)
 	mmc->caps = MMC_CAP_SDIO_IRQ;
 
 	if (!(host->quirks & SDHCI_QUIRK_FORCE_1_BIT_DATA))
-		mmc->caps |= MMC_CAP_4_BIT_DATA;
+		mmc->caps |= MMC_CAP_4_BIT_DATA | MMC_CAP_8_BIT_DATA;
 
 	if (caps & SDHCI_CAN_DO_HISPD)
 		mmc->caps |= MMC_CAP_SD_HIGHSPEED;
-- 
1.5.5.6

[PATCH RESEND] sdhci: add MMC_CAP_8_BIT_DATA in the Host capabilities

[Chris Ball]

From: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Date: Tue, 31 Aug 2010 05:43:45 +0200
Subject: [PATCH (sh-2.6.32.y)] sdhci: add MMC_CAP_8_BIT_DATA in the Host capabilities

Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
---
 drivers/mmc/host/sdhci.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/drivers/mmc/host/sdhci.c b/drivers/mmc/host/sdhci.c
index 955cad9..1c49326 100644
--- a/drivers/mmc/host/sdhci.c
+++ b/drivers/mmc/host/sdhci.c
@@ -1799,7 +1799,7 @@ int sdhci_add_host(struct sdhci_host *host)
 	mmc->caps = MMC_CAP_SDIO_IRQ;
 
 	if (!(host->quirks & SDHCI_QUIRK_FORCE_1_BIT_DATA))
-		mmc->caps |= MMC_CAP_4_BIT_DATA;
+		mmc->caps |= MMC_CAP_4_BIT_DATA | MMC_CAP_8_BIT_DATA;
 
 	if (caps & SDHCI_CAN_DO_HISPD)
 		mmc->caps |= MMC_CAP_SD_HIGHSPEED;
-- 
Chris Ball   <cjb@laptop.org>   <http://printf.net/>
One Laptop Per Child

Re: [PATCH RESEND] sdhci: add MMC_CAP_8_BIT_DATA in the Host capabilities

[Peppe CAVALLARO]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Thanks Chris to have resent the patch again.

I'm going to resend it just today ;-).

I'm using 8BIT_DATA mode on my HC without any problems and with good
performances.

Regards
Peppe

On 09/18/2010 12:19 AM, Chris Ball wrote:
> From: Giuseppe Cavallaro <peppe.cavallaro@st.com>
> Date: Tue, 31 Aug 2010 05:43:45 +0200
> Subject: [PATCH (sh-2.6.32.y)] sdhci: add MMC_CAP_8_BIT_DATA in the Host 
> capabilities
> 
> Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
> ---
> drivers/mmc/host/sdhci.c | 2 +-
> 1 files changed, 1 insertions(+), 1 deletions(-)
> 
> diff --git a/drivers/mmc/host/sdhci.c b/drivers/mmc/host/sdhci.c
> index 955cad9..1c49326 100644
> --- a/drivers/mmc/host/sdhci.c
> +++ b/drivers/mmc/host/sdhci.c
> @@ -1799,7 +1799,7 @@ int sdhci_add_host(struct sdhci_host *host)
> mmc->caps = MMC_CAP_SDIO_IRQ;
> 
> if (!(host->quirks & SDHCI_QUIRK_FORCE_1_BIT_DATA))
> - mmc->caps |= MMC_CAP_4_BIT_DATA;
> + mmc->caps |= MMC_CAP_4_BIT_DATA | MMC_CAP_8_BIT_DATA;
> 
> if (caps & SDHCI_CAN_DO_HISPD)
> mmc->caps |= MMC_CAP_SD_HIGHSPEED;
> --
> Chris Ball <cjb@laptop.org> <http://printf.net/>
> One Laptop Per Child
> 

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iQEcBAEBAgAGBQJMlwSGAAoJEAm9vY9TshdvbJAH+wZXUrPEOIGr/4i/uAGKW+jN
6yYukjyyCHerXLmI9vmmV8NekDU/DOi5jMHlUaFHkCUZcSVFfK+68xfOuH4Sz2Cu
7L3wTPXrQabIeM6cV31DOu15wLpgmTZhXECBxlZOAjMnnE2UyzszMbep4v5ZMZM4
Tj8lbCkdnINjBZc/lNUUwgIdOGvoIdxt5KIMy0k84BZY01ggsurTzMwtnXQEK6J4
YkOT4Ut57TIR/9ZxEzii/QbJjVNvm4hrZOwbooTu8cYcy2di66dJ7TwrOa5rKcev
z5QkmH7UZqVja7WAhgK1f9TlBSMiTmhRnJeOoQvELpq8lHV0tOajIBsK7EmJiwI=
=W0a6
-----END PGP SIGNATURE-----

Re: [PATCH RESEND] sdhci: add MMC_CAP_8_BIT_DATA in the Host capabilities

[Chris Ball]

Hi Peppe,

On Mon, Sep 20, 2010 at 08:51:50AM +0200, Peppe CAVALLARO wrote:
> Thanks Chris to have resent the patch again.
> 
> I'm going to resend it just today ;-).
> 
> I'm using 8BIT_DATA mode on my HC without any problems and with good
> performances.

Thanks, I've applied this to mmc-next now.

-- 
Chris Ball   <cjb@laptop.org>   <http://printf.net/>
One Laptop Per Child