From: Benjamin Tissoires <benjamin.tissoires@redhat.com>
To: Corey Minyard <minyard@acm.org>
Cc: Jean Delvare <jdelvare@suse.com>,
linux-i2c@vger.kernel.org, linux-kernel@vger.kernel.org,
Corey Minyard <cminyard@mvista.com>
Subject: Re: [v2,02/10] i2c-i801: Move hostcfg set/reset to i801_access()
Date: Thu, 9 Jun 2016 11:39:45 +0200 [thread overview]
Message-ID: <20160609093945.GD24234@mail.corp.redhat.com> (raw)
In-Reply-To: <1464570544-975-3-git-send-email-minyard@acm.org>
On May 29 2016 or thereabouts, Corey Minyard wrote:
> From: Corey Minyard <cminyard@mvista.com>
>
> The HSTCFG register save/restore was done in i2c_block_transaction,
> but all the checks were already done in i801_access, so move it into
> those checks.
>
> This results in a small savings of code, and moves some special
> handing for I2C transactions into code that is already handling
> special things for I2C transactions.
>
> Signed-off-by: Corey Minyard <cminyard@mvista.com>
> ---
> drivers/i2c/busses/i2c-i801.c | 41 +++++++++++++++++++----------------------
> 1 file changed, 19 insertions(+), 22 deletions(-)
>
> diff --git a/drivers/i2c/busses/i2c-i801.c b/drivers/i2c/busses/i2c-i801.c
> index 818c0c8..205f9d0 100644
> --- a/drivers/i2c/busses/i2c-i801.c
> +++ b/drivers/i2c/busses/i2c-i801.c
> @@ -661,20 +661,6 @@ static int i801_block_transaction(struct i801_priv *priv,
> int command, int hwpec)
> {
> int result = 0;
> - unsigned char hostc;
> -
> - if (command == I2C_SMBUS_I2C_BLOCK_DATA) {
> - if (read_write == I2C_SMBUS_WRITE) {
> - /* set I2C_EN bit in configuration register */
> - pci_read_config_byte(priv->pci_dev, SMBHSTCFG, &hostc);
> - pci_write_config_byte(priv->pci_dev, SMBHSTCFG,
> - hostc | SMBHSTCFG_I2C_EN);
> - } else if (!(priv->features & FEATURE_I2C_BLOCK_READ)) {
> - dev_err(&priv->pci_dev->dev,
> - "I2C block read is unsupported!\n");
> - return -EOPNOTSUPP;
> - }
> - }
>
> if (read_write == I2C_SMBUS_WRITE
> || command == I2C_SMBUS_I2C_BLOCK_DATA) {
> @@ -699,11 +685,6 @@ static int i801_block_transaction(struct i801_priv *priv,
> read_write,
> command);
>
> - if (command == I2C_SMBUS_I2C_BLOCK_DATA
> - && read_write == I2C_SMBUS_WRITE) {
> - /* restore saved configuration register value */
> - pci_write_config_byte(priv->pci_dev, SMBHSTCFG, hostc);
> - }
> return result;
> }
>
> @@ -715,6 +696,7 @@ static s32 i801_access(struct i2c_adapter *adap, u16 addr,
> int hwpec;
> int block = 0;
> int ret = 0, xact = 0;
> + int hostc = -1;
> struct i801_priv *priv = i2c_get_adapdata(adap);
>
> pm_runtime_get_sync(&priv->pci_dev->dev);
> @@ -764,12 +746,24 @@ static s32 i801_access(struct i2c_adapter *adap, u16 addr,
> /* NB: page 240 of ICH5 datasheet shows that the R/#W
> * bit should be cleared here, even when reading */
> outb_p((addr & 0x7f) << 1, SMBHSTADD(priv));
> - if (read_write == I2C_SMBUS_READ) {
> + if (read_write == I2C_SMBUS_WRITE) {
> + unsigned char thostc;
> +
> + outb_p(command, SMBHSTCMD(priv));
> + /* set I2C_EN bit in configuration register */
> + pci_read_config_byte(priv->pci_dev, SMBHSTCFG, &thostc);
> + pci_write_config_byte(priv->pci_dev, SMBHSTCFG,
> + thostc | SMBHSTCFG_I2C_EN);
> + hostc = thostc;
> + } else if (!(priv->features & FEATURE_I2C_BLOCK_READ)) {
nitpick: I'd prefer seeing:
if (write) {
foo
} else {
if (!FEATURE_I2C_BLOCK_READ) {
goto out
}
bar
}
It's equivalent but I feel like it's easier to understand.
Other than that, no strong opinion on the patch itself, but it's correct
(the code path stays the same):
Reviewed-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Cheers,
Benjamin
> + dev_err(&priv->pci_dev->dev,
> + "I2C block read is unsupported!\n");
> + ret = -EOPNOTSUPP;
> + goto out;
> + } else
> /* NB: page 240 of ICH5 datasheet also shows
> * that DATA1 is the cmd field when reading */
> outb_p(command, SMBHSTDAT1(priv));
> - } else
> - outb_p(command, SMBHSTCMD(priv));
> block = 1;
> break;
> default:
> @@ -798,6 +792,9 @@ static s32 i801_access(struct i2c_adapter *adap, u16 addr,
> outb_p(inb_p(SMBAUXCTL(priv)) &
> ~(SMBAUXCTL_CRC | SMBAUXCTL_E32B), SMBAUXCTL(priv));
>
> + if (hostc >= 0)
> + pci_write_config_byte(priv->pci_dev, SMBHSTCFG, hostc);
> +
> if (block)
> goto out;
> if (ret)
next prev parent reply other threads:[~2016-06-09 9:39 UTC|newest]
Thread overview: 24+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-05-30 1:08 [PATCH v2 00/10] i2c-i801: Various cleanups minyard
2016-05-30 1:08 ` [PATCH v2 01/10] i2c-i801: Remove hwpec from block byte-by-byte function minyard
2016-06-09 9:36 ` [v2,01/10] " Benjamin Tissoires
2016-05-30 1:08 ` [PATCH v2 02/10] i2c-i801: Move hostcfg set/reset to i801_access() minyard
2016-06-09 9:39 ` Benjamin Tissoires [this message]
2016-06-10 10:43 ` [v2,02/10] " Corey Minyard
2016-05-30 1:08 ` [PATCH v2 03/10] i2c-i801: Move hwpec handling into block transaction minyard
2016-06-09 9:42 ` [v2,03/10] " Benjamin Tissoires
2016-05-30 1:08 ` [PATCH v2 04/10] i2c-i801: Consolidate calls to i801_check_pre() minyard
2016-06-09 9:44 ` [v2,04/10] " Benjamin Tissoires
2016-06-10 10:52 ` Corey Minyard
2016-05-30 1:08 ` [PATCH v2 05/10] i2c-i801: Consolidate calls to i801_check_post minyard
2016-06-09 10:03 ` [v2,05/10] " Benjamin Tissoires
2016-06-10 11:09 ` Corey Minyard
2016-05-30 1:09 ` [PATCH v2 06/10] i2c-i801: Pass around a boolean read/write variable minyard
2016-06-09 10:05 ` [v2,06/10] " Benjamin Tissoires
2016-05-30 1:09 ` [PATCH v2 07/10] i2c-i801: Fix some inconsistent variable names minyard
2016-06-09 14:01 ` [v2,07/10] " Benjamin Tissoires
2016-06-10 11:12 ` Corey Minyard
2016-05-30 1:09 ` [PATCH v2 08/10] i2c-i801: Handle a protocol error in byte-by-byte isr minyard
2016-06-09 14:07 ` [v2,08/10] " Benjamin Tissoires
2016-05-30 1:09 ` [PATCH v2 09/10] i2c-i801: Null isr data buffer when done with it minyard
2016-06-09 14:14 ` [v2,09/10] " Benjamin Tissoires
2016-05-30 1:09 ` [PATCH v2 10/10] i2c-i801: Only write the host control reg when necessary minyard
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=20160609093945.GD24234@mail.corp.redhat.com \
--to=benjamin.tissoires@redhat.com \
--cc=cminyard@mvista.com \
--cc=jdelvare@suse.com \
--cc=linux-i2c@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=minyard@acm.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).