From: Greg KH <gregkh@linuxfoundation.org>
To: Hayes Wang <hayeswang@realtek.com>
Cc: netdev@vger.kernel.org, linux-kernel@vger.kernel.org,
linux-usb@vger.kernel.org, nic_swsd@realtek.com
Subject: Re: [PATCH v3 1/3] net/usb/r815x: replace USB buffer from stack to DMA-able
Date: Tue, 30 Jul 2013 07:00:59 -0700 [thread overview]
Message-ID: <20130730140059.GE27962@kroah.com> (raw)
In-Reply-To: <1375172936-4145-1-git-send-email-hayeswang@realtek.com>
On Tue, Jul 30, 2013 at 04:28:54PM +0800, Hayes Wang wrote:
> Allocate the transfer buffer in probe(), and use the buffer for
> usb control transfer.
>
> Signed-off-by: Hayes Wang <hayeswang@realtek.com>
> ---
> drivers/net/usb/r815x.c | 117 +++++++++++++++++++++++++++++++++++++-----------
> 1 file changed, 90 insertions(+), 27 deletions(-)
>
> diff --git a/drivers/net/usb/r815x.c b/drivers/net/usb/r815x.c
> index 8523922..89ad63f 100644
> --- a/drivers/net/usb/r815x.c
> +++ b/drivers/net/usb/r815x.c
> @@ -21,37 +21,52 @@
> #define R815x_PHY_ID 32
> #define REALTEK_VENDOR_ID 0x0bda
>
> +struct r815x {
> + struct mutex transfer_mutex;
> + __le32 *transfer_buf;
> +};
Really? Why not just allocate it the times you need to send the data?
Is it really all that often that you need to do this?
> -static int pla_read_word(struct usb_device *udev, u16 index)
> +static int pla_read_word(struct usbnet *dev, u16 index)
> {
> - int data, ret;
> + struct r815x *tp = dev->driver_priv;
> + struct usb_device *udev = dev->udev;
> + int ret;
> u8 shift = index & 2;
> - __le32 ocp_data;
> + __le32 *tmp;
> +
> + mutex_lock(&tp->transfer_mutex);
> + tmp = tp->transfer_buf;
>
> index &= ~3;
>
> ret = usb_control_msg(udev, usb_rcvctrlpipe(udev, 0),
> RTL815x_REQ_GET_REGS, RTL815x_REQT_READ,
> - index, MCU_TYPE_PLA, &ocp_data, sizeof(ocp_data),
> - 500);
> + index, MCU_TYPE_PLA, tmp, sizeof(*tmp), 500);
This call is so slow, you can afford to make a call to kmalloc for the
data, as it sure just did for other structures it needed :)
So just do a kmalloc call for the data before this function, same for
your other patches in this series, no need for a lock for your "transfer
data" buffer, and a global one at all.
thanks,
greg k-h
next prev parent reply other threads:[~2013-07-30 14:00 UTC|newest]
Thread overview: 33+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-07-23 9:26 [PATCH 1/3] net/usb/r815x: replace USB buffer from stack to DMA-able hayeswang
2013-07-25 0:49 ` David Miller
2013-07-25 7:59 ` [PATCH v2 " Hayes Wang
2013-07-25 7:59 ` [PATCH v2 2/3] net/usb/r8152: make sure the USB buffer is DMA-able Hayes Wang
2013-07-25 7:59 ` [PATCH v2 3/3] net/usb/r8152: adjust relative ocp function Hayes Wang
[not found] ` <1374739144-732-1-git-send-email-hayeswang-Rasf1IRRPZFBDgjK7y7TUQ@public.gmane.org>
2013-07-28 3:21 ` [PATCH v2 1/3] net/usb/r815x: replace USB buffer from stack to DMA-able David Miller
2013-07-29 5:20 ` Oliver Neukum
2013-07-29 7:46 ` David Miller
2013-07-30 8:28 ` [PATCH v3 " Hayes Wang
2013-07-30 8:28 ` [PATCH v3 2/3] net/usb/r8152: " Hayes Wang
2013-07-30 14:01 ` Greg KH
2013-07-30 8:28 ` [PATCH v3 3/3] net/usb/r8152: adjust relative ocp function Hayes Wang
2013-07-30 13:58 ` [PATCH v3 1/3] net/usb/r815x: replace USB buffer from stack to DMA-able Greg KH
2013-07-30 15:17 ` [PATCH] MAINTAINERS: USB, remove F: drivers/net/usb/ pattern Joe Perches
2013-07-30 14:00 ` Greg KH [this message]
2013-07-30 18:33 ` [PATCH v3 1/3] net/usb/r815x: replace USB buffer from stack to DMA-able David Miller
2013-07-30 18:41 ` Joe Perches
2013-07-30 18:49 ` David Miller
2013-07-31 10:29 ` Ming Lei
2013-07-30 18:58 ` Greg KH
2013-07-30 14:29 ` Ming Lei
2013-07-31 9:21 ` [PATCH v4 1/5] " Hayes Wang
2013-07-31 9:21 ` [PATCH v4 2/5] net/usb/r815x: avoid to call mdio functions for runtime-suspended device Hayes Wang
2013-07-31 21:49 ` David Miller
2013-07-31 9:21 ` [PATCH v4 3/5] net/usb/r815x: change the return value for bind functions Hayes Wang
[not found] ` <1375262486-5373-3-git-send-email-hayeswang-Rasf1IRRPZFBDgjK7y7TUQ@public.gmane.org>
2013-07-31 21:49 ` David Miller
[not found] ` <1375262486-5373-1-git-send-email-hayeswang-Rasf1IRRPZFBDgjK7y7TUQ@public.gmane.org>
2013-07-31 9:21 ` [PATCH v4 4/5] net/usb/r8152: make sure the USB buffer is DMA-able Hayes Wang
2013-07-31 12:54 ` Greg KH
2013-07-31 21:50 ` David Miller
2013-07-31 9:21 ` [PATCH v4 5/5] net/usb/r8152: adjust relative ocp function Hayes Wang
2013-07-31 21:50 ` David Miller
2013-07-31 12:53 ` [PATCH v4 1/5] net/usb/r815x: replace USB buffer from stack to DMA-able Greg KH
2013-07-31 21:49 ` David Miller
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=20130730140059.GE27962@kroah.com \
--to=gregkh@linuxfoundation.org \
--cc=hayeswang@realtek.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-usb@vger.kernel.org \
--cc=netdev@vger.kernel.org \
--cc=nic_swsd@realtek.com \
/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).