From: Seungwon Jeon <tgih.jun@samsung.com>
To: 'Grant Grundler' <grundler@chromium.org>
Cc: linux-mmc@vger.kernel.org, 'Chris Ball' <cjb@laptop.org>,
'Jaehoon Chung' <jh80.chung@samsung.com>,
'Alim Akhtar' <alim.akhtar@samsung.com>
Subject: RE: [PATCH 02/14] mmc: dw_mmc: exynos: add variable delay tuning sequence
Date: Thu, 22 Aug 2013 11:25:41 +0900 [thread overview]
Message-ID: <000b01ce9ede$e5d939f0$b18badd0$%jun@samsung.com> (raw)
In-Reply-To: <CANEJEGseG+5S+S=er23qi4TgvwknQsJ0o0p9rMsOVH2g7SnXxA@mail.gmail.com>
On Wed, August 21, 2013, Grant Grundler
> On Wed, Aug 21, 2013 at 6:49 AM, Seungwon Jeon <tgih.jun@samsung.com> wrote:
> > Implements variable delay tuning. In this change,
> > exynos host can determine the correct sampling point
> > for the HS200 and SDR104 speed mode.
> >
> > Signed-off-by: Seungwon Jeon <tgih.jun@samsung.com>
> > ---
> > drivers/mmc/host/dw_mmc-exynos.c | 124 ++++++++++++++++++++++++++++++++++++++
> > 1 files changed, 124 insertions(+), 0 deletions(-)
> >
> > diff --git a/drivers/mmc/host/dw_mmc-exynos.c b/drivers/mmc/host/dw_mmc-exynos.c
> > index 866edef..90f9335 100644
> > --- a/drivers/mmc/host/dw_mmc-exynos.c
> > +++ b/drivers/mmc/host/dw_mmc-exynos.c
> > @@ -14,8 +14,10 @@
> > #include <linux/clk.h>
> > #include <linux/mmc/host.h>
> > #include <linux/mmc/dw_mmc.h>
> > +#include <linux/mmc/mmc.h>
> > #include <linux/of.h>
> > #include <linux/of_gpio.h>
> > +#include <linux/slab.h>
> >
> > #include "dw_mmc.h"
> > #include "dw_mmc-pltfm.h"
> > @@ -150,6 +152,127 @@ static int dw_mci_exynos_parse_dt(struct dw_mci *host)
> > return 0;
> > }
> >
> > +static inline u8 dw_mci_exynos_get_clksmpl(struct dw_mci *host)
> > +{
> > + return SDMMC_CLKSEL_CCLK_SAMPLE(mci_readl(host, CLKSEL));
> > +}
> > +
> > +static inline void dw_mci_exynos_set_clksmpl(struct dw_mci *host, u8 sample)
> > +{
> > + u32 clksel;
> > + clksel = mci_readl(host, CLKSEL);
> > + clksel = (clksel & ~0x7) | SDMMC_CLKSEL_CCLK_SAMPLE(sample);
> > + mci_writel(host, CLKSEL, clksel);
> > +}
> > +
> > +static inline u8 dw_mci_exynos_move_next_clksmpl(struct dw_mci *host)
>
> Hi Seungwon,
> I rewrote the dw_mmc HS200 tuning code for ChromeOS-3.4 kernel with
> this series of patches:
> https://gerrit.chromium.org/gerrit/#/c/57997 CHROMIUM: mmc:
> dw_mmc: move struct dw_mci_slot to public header
> https://gerrit.chromium.org/gerrit/#/c/58269 CHROMIUM: mmc:
> dw_mmc: move dw_mci_set_timeout() to dw_mmc.h
> https://gerrit.chromium.org/gerrit/#/c/57996/ CHROMIUM: mmc:
> dw_mmc: add execute_tuning hook
> https://gerrit.chromium.org/gerrit/#/c/57861 CHROMIUM: mmc:
> dw_mmc: add HS200 tuning feature
>
> This code was reviewed and approved by Alim Akhtar.
As mentioned previously, this change comes from below:
@ mmc: dw_mmc: add support of tuning feature
https://android.googlesource.com/kernel/exynos/+/016ebd43ab27cf8b0258d1e7d3e3288d7f7aab6a%5E%21/#F0
Here, origin is split into basic feature and error handling part.
>
> Most of the differences I don't care enough about to quibble. But the
> handling of "clksmpl" was difficult to read and understand. I added a
> lot of comments so the next person without access to Samsung's
> confidential and incomplete Exynos manuals could have a small chance
> of debugging this code.
>
> In particular look at dw_mci_exynos_execute_tuning() and
> find_median_of_5bits() in this patch:
> https://gerrit.chromium.org/gerrit/#/c/57861/9/drivers/mmc/host/dw_mmc-exynos.c
>
> Is there some reason you would NOT want to use find_median_of_5bits()
> and related code?
I didn't check 'find_median_of_5bits function'.
But let me explain my case.
First, try to find the continuous 5-bit of tuning success.
If not, then try to find 3-bit for the second best.
Thanks,
Seungwon Jeon
>
> thanks,
> grant
>
> > +{
> > + u32 clksel;
> > + u8 sample;
> > +
> > + clksel = mci_readl(host, CLKSEL);
> > + sample = (clksel + 1) & 0x7;
> > + clksel = (clksel & ~0x7) | sample;
> > + mci_writel(host, CLKSEL, clksel);
> > + return sample;
> > +}
> > +
> > +static s8 dw_mci_exynos_get_best_clksmpl(u8 candiates)
> > +{
> > + const u8 iter = 8;
> > + u8 __c;
> > + s8 i, loc = -1;
> > +
> > + for (i = 0; i < iter; i++) {
> > + __c = ror8(candiates, i);
> > + if ((__c & 0xc7) == 0xc7) {
> > + loc = i;
> > + goto out;
> > + }
> > + }
> > +
> > + for (i = 0; i < iter; i++) {
> > + __c = ror8(candiates, i);
> > + if ((__c & 0x83) == 0x83) {
> > + loc = i;
> > + goto out;
> > + }
> > + }
> > +
> > +out:
> > + return loc;
> > +}
> > +
> > +static int dw_mci_exynos_execute_tuning(struct dw_mci_slot *slot, u32 opcode,
> > + struct dw_mci_tuning_data *tuning_data)
> > +{
> > + struct dw_mci *host = slot->host;
> > + struct mmc_host *mmc = slot->mmc;
> > + const u8 *blk_pattern = tuning_data->blk_pattern;
> > + u8 *blk_test;
> > + unsigned int blksz = tuning_data->blksz;
> > + u8 start_smpl, smpl, candiates = 0;
> > + s8 found = -1;
> > + int ret = 0;
> > +
> > + blk_test = kmalloc(blksz, GFP_KERNEL);
> > + if (!blk_test)
> > + return -ENOMEM;
> > +
> > + start_smpl = dw_mci_exynos_get_clksmpl(host);
> > +
> > + do {
> > + struct mmc_request mrq = {NULL};
> > + struct mmc_command cmd = {0};
> > + struct mmc_command stop = {0};
> > + struct mmc_data data = {0};
> > + struct scatterlist sg;
> > +
> > + cmd.opcode = opcode;
> > + cmd.arg = 0;
> > + cmd.flags = MMC_RSP_R1 | MMC_CMD_ADTC;
> > +
> > + stop.opcode = MMC_STOP_TRANSMISSION;
> > + stop.arg = 0;
> > + stop.flags = MMC_RSP_R1B | MMC_CMD_AC;
> > +
> > + data.blksz = blksz;
> > + data.blocks = 1;
> > + data.flags = MMC_DATA_READ;
> > + data.sg = &sg;
> > + data.sg_len = 1;
> > +
> > + sg_init_one(&sg, blk_test, blksz);
> > + mrq.cmd = &cmd;
> > + mrq.stop = &stop;
> > + mrq.data = &data;
> > + host->mrq = &mrq;
> > +
> > + mci_writel(host, TMOUT, ~0);
> > + smpl = dw_mci_exynos_move_next_clksmpl(host);
> > +
> > + mmc_wait_for_req(mmc, &mrq);
> > +
> > + if (!cmd.error && !data.error) {
> > + if (!memcmp(blk_pattern, blk_test, blksz))
> > + candiates |= (1 << smpl);
> > + } else {
> > + dev_dbg(host->dev,
> > + "Tuning error: cmd.error:%d, data.error:%d\n",
> > + cmd.error, data.error);
> > + }
> > + } while (start_smpl != smpl);
> > +
> > + found = dw_mci_exynos_get_best_clksmpl(candiates);
> > + if (found >= 0)
> > + dw_mci_exynos_set_clksmpl(host, found);
> > + else
> > + ret = -EIO;
> > +
> > + kfree(blk_test);
> > + return ret;
> > +}
> > +
> > /* Common capabilities of Exynos4/Exynos5 SoC */
> > static unsigned long exynos_dwmmc_caps[4] = {
> > MMC_CAP_UHS_DDR50 | MMC_CAP_1_8V_DDR |
> > @@ -166,6 +289,7 @@ static const struct dw_mci_drv_data exynos_drv_data = {
> > .prepare_command = dw_mci_exynos_prepare_command,
> > .set_ios = dw_mci_exynos_set_ios,
> > .parse_dt = dw_mci_exynos_parse_dt,
> > + .execute_tuning = dw_mci_exynos_execute_tuning,
> > };
> >
> > static const struct of_device_id dw_mci_exynos_match[] = {
> > --
> > 1.7.0.4
> >
> >
> > --
> > 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
prev parent reply other threads:[~2013-08-22 2:25 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-08-21 13:49 [PATCH 02/14] mmc: dw_mmc: exynos: add variable delay tuning sequence Seungwon Jeon
2013-08-21 14:58 ` Grant Grundler
2013-08-22 2:25 ` Seungwon Jeon [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to='000b01ce9ede$e5d939f0$b18badd0$%jun@samsung.com' \
--to=tgih.jun@samsung.com \
--cc=alim.akhtar@samsung.com \
--cc=cjb@laptop.org \
--cc=grundler@chromium.org \
--cc=jh80.chung@samsung.com \
--cc=linux-mmc@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
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).