Re: [PATCH] Reduce locking in TX path of forcedth driver

2007-12-21 Thread Jeff Garzik
(Stephen, your mailer is snipping CC's, please fix) (Tom, you failed to CC the maintainer at NVIDIA, please fix...) Stephen Hemminger wrote: On Fri, 21 Dec 2007 17:41:34 -0800 (PST) [EMAIL PROTECTED] (Tom Herbert) wrote: Reduce the amount of locking in the TX path. Instead of using

[WEXT 0/12]: Un-crapify compat handling.

2007-12-21 Thread David Miller
This set of 12 patches pulls all of the existing compat handling for WEXT ioctls out of fs/compat_ioctl.c and allows implementations down in the driver layers to handle compat issues properly when necessary. In particular it fixes the bug where iw_point objects were not being size translated

[WEXT 1/12]: Remove inline from get_priv_size() and adjust_priv_size().

2007-12-21 Thread David Miller
[WEXT]: Remove inline from get_priv_size() and adjust_priv_size(). The compiler inlines when appropriate. Signed-off-by: David S. Miller [EMAIL PROTECTED] --- net/wireless/wext.c |5 ++--- 1 files changed, 2 insertions(+), 3 deletions(-) diff --git a/net/wireless/wext.c

[WEXT 2/12]: Make adjust_priv_size() take a struct iw_point *.

2007-12-21 Thread David Miller
[WEXT]: Make adjust_priv_size() take a struct iw_point *. Signed-off-by: David S. Miller [EMAIL PROTECTED] --- net/wireless/wext.c |6 +++--- 1 files changed, 3 insertions(+), 3 deletions(-) diff --git a/net/wireless/wext.c b/net/wireless/wext.c index 677be28..3ca9c3a 100644 ---

[WEXT 3/12]: Extract standard call iw_point handling into seperate function.

2007-12-21 Thread David Miller
[WEXT]: Extract standard call iw_point handling into seperate function. Signed-off-by: David S. Miller [EMAIL PROTECTED] --- net/wireless/wext.c | 258 ++ 1 files changed, 134 insertions(+), 124 deletions(-) diff --git a/net/wireless/wext.c

[WEXT 4/12]: Extract private call iw_point handling into seperate functions.

2007-12-21 Thread David Miller
[WEXT]: Extract private call iw_point handling into seperate functions. Signed-off-by: David S. Miller [EMAIL PROTECTED] --- net/wireless/wext.c | 141 +++ 1 files changed, 74 insertions(+), 67 deletions(-) diff --git a/net/wireless/wext.c

[WEXT 5/12]: Pull ioctl permission checking out into helper function.

2007-12-21 Thread David Miller
[WEXT]: Pull ioctl permission checking out into helper function. Signed-off-by: David S. Miller [EMAIL PROTECTED] --- net/wireless/wext.c | 22 +++--- 1 files changed, 15 insertions(+), 7 deletions(-) diff --git a/net/wireless/wext.c b/net/wireless/wext.c index

[WEXT 6/12]: Parameterize the standard/private handlers.

2007-12-21 Thread David Miller
[WEXT]: Parameterize the standard/private handlers. The WEXT standard and private handlers to use are now arguments to wireless_process_ioctl(). Signed-off-by: David S. Miller [EMAIL PROTECTED] --- net/wireless/wext.c | 24 1 files changed, 16 insertions(+), 8

[WEXT 7/12]: Pass iwreq pointer down into standard/private handlers.

2007-12-21 Thread David Miller
[WEXT]: Pass iwreq pointer down into standard/private handlers. They have no need to see the object as an ifreq. Signed-off-by: David S. Miller [EMAIL PROTECTED] --- net/wireless/wext.c | 17 - 1 files changed, 8 insertions(+), 9 deletions(-) diff --git a/net/wireless/wext.c

[WEXT 8/12]: Pull top-level ioctl dispatch logic into helper function.

2007-12-21 Thread David Miller
[WEXT]: Pull top-level ioctl dispatch logic into helper function. Signed-off-by: David S. Miller [EMAIL PROTECTED] --- net/wireless/wext.c | 26 -- 1 files changed, 20 insertions(+), 6 deletions(-) diff --git a/net/wireless/wext.c b/net/wireless/wext.c index

[WEXT 9/12]: Dispatch and handle compat ioctls entirely in net/wireless/wext.c

2007-12-21 Thread David Miller
[WEXT]: Dispatch and handle compat ioctls entirely in net/wireless/wext.c Next we can kill the hacks in fs/compat_ioctl.c and also dispatch compat ioctls down into the driver and 80211 protocol helper layers in order to handle iw_point objects embedded in stream replies which need to be

[WEXT 10/12]: Remove compat handling from fs/compat_ioctl.c

2007-12-21 Thread David Miller
[WEXT]: Remove compat handling from fs/compat_ioctl.c No longer used. Signed-off-by: David S. Miller [EMAIL PROTECTED] --- fs/compat_ioctl.c | 107 + 1 files changed, 1 insertions(+), 106 deletions(-) diff --git a/fs/compat_ioctl.c

[WEXT 11/12]: Create IW_REQUEST_FLAG_COMPAT and set it as needed.

2007-12-21 Thread David Miller
[WEXT]: Create IW_REQUEST_FLAG_COMPAT and set it as needed. Now low-level WEXT ioctl handlers can do compat handling when necessary. Signed-off-by: David S. Miller [EMAIL PROTECTED] --- include/net/iw_handler.h |2 +- net/wireless/wext.c | 73

[WEXT 12/12]: Emit event stream compat iw_point objects correctly.

2007-12-21 Thread David Miller
[WEXT]: Emit event stream compat iw_point objects correctly. Three major portions to this change: 1) Add IW_EV_COMPAT_POINT_LEN helper define. 2) Add iw_request_info argument to iwe_stream_add_point() and iwe_stream_check_add_point(), and use it to size the event and pointer lengths

hard_start_xmit struct sk_buff read-only?

2007-12-21 Thread Matti Linnanvuori
An old article about Linux's network drivers mentioned that the struct sk_buff whose pointer is passed to a driver's hard_start_xmit function is read-only to the function. Is that still so? If it is, there is no mention about it in the kernel tree as far as I know. Heute schon einen Blick

Re: [ETH]: Combine format_addr() with print_mac().

2007-12-21 Thread Joe Perches
On Fri, 2007-12-21 at 19:58 -0800, Michael Chan wrote: The dev_addr is declared as unsigned char* in struct net_device. To be consistent, can we change print_mac() and MAC_BUF to use unsigned char*? They are really the same. That's fine by me. I like consistency. I don't remember why it was

Re: [WEXT 3/12]: Extract standard call iw_point handling into seperate function.

2007-12-21 Thread Joe Perches
On Fri, 2007-12-21 at 20:54 -0800, David Miller wrote: [WEXT]: Extract standard call iw_point handling into seperate function. This potentially does copy_from_user twice, first to the stack, second to a kmalloc'ed area. Perhaps it's better to kmalloc first and copy once? -- To unsubscribe

<    1   2