Scott Feldman [Wed, 9 Nov 2005 07:18:52 +0000 (02:18 -0500)]
[netdrvr e100] experiment with doing RX in a similar manner to eepro100
I was going to say that eepro100's speedo_rx_link() does the same DMA
abuse as e100, but then I noticed one little detail: eepro100 sets both
EL (end of list) and S (suspend) bits in the RFD as it chains it to the
RFD list. e100 was only setting the EL bit. Hmmm, that's interesting.
That means that if HW reads a RFD with the S-bit set, it'll process
that RFD and then suspend the receive unit. The receive unit will
resume when SW clears the S-bit. There is no need for SW to restart
the receive unit. Which means a lot of the receive unit state tracking
code in the driver goes away.
So here's a patch against 2.6.14. (Sorry for inlining it; the mailer
I'm using now will mess with the word wrap). I can't test this on
XScale (unless someone has an e100 module for Gumstix :) . It should
be doing exactly what eepro100 does with RFDs. I don't believe this
change will introduce a performance hit because the S-bit and EL-bit go
hand-in-hand meaning if we're going to suspend because of the S- bit,
we're on the last resource anyway, so we'll have to wait for SW to
replenish.
(cherry picked from 29e79da9495261119e3b2e4e7c72507348e75976 commit)
Larry Finger [Tue, 17 Apr 2007 15:28:47 +0000 (10:28 -0500)]
[PATCH] ieee80211: add missing global needed by IEEE80211_DEBUG_XXXX
Using any of the IEEE80211_DEBUG_XXXX macros in any ieee80211_crypt
routine built as a module results in a missing global for
'ieee80211_debug_level'. The fix is to export the symbol in ieee80211.
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Michael Ellerman [Tue, 17 Apr 2007 03:12:55 +0000 (13:12 +1000)]
Fix sparse errors in drivers/net/ibmveth.c
drivers/net/ibmveth.c:96:46: error: marked inline, but without a definition
drivers/net/ibmveth.c:96: warning: 'ibmveth_rxq_harvest_buffer' declared inline after being called
drivers/net/ibmveth.c:96: warning: previous declaration of 'ibmveth_rxq_harvest_buffer' was here
Just let the compiler decide, as it happens gcc 4.~ inlines it anyway.
drivers/net/ibmveth.c:957:71: warning: Using plain integer as NULL pointer
drivers/net/ibmveth.c:964:85: warning: Using plain integer as NULL pointer
Split the long lines as well, ugly, but < 80 columns.
Signed-off-by: Michael Ellerman <michael@ellerman.id.au> Signed-off-by: Jeff Garzik <jeff@garzik.org>
Nicu Ioan Petru [Fri, 13 Apr 2007 06:26:29 +0000 (01:26 -0500)]
ucc_geth: fixes for ucc_geth_memclean
The ucc_geth_memclean function can be called before the Tx BD rings, Rx
BD rings and associated socket buffers are allocated (for example if
ucc_fast_init fails). The current code doesn't check if p_tx_bd_ring[i]
is null, generating a kernel panic when trying to free the associated
socket buffers.
The function can also fail when accessing the uninitialized list_head
structures ugeth->group_hash_q and ugeth->ind_hash_q. In the current
implementation the list heads are initialized only when
maxGroupAddrInHash and maxIndAddrInHash are positive values, although I
think it's better to always initialize them.
Signed-off-by: Ionut Nicu <ionut.nicu@freescale.com> Signed-off-by: Kim Phillips <kim.phillips@freescale.com> Signed-off-by: Jeff Garzik <jeff@garzik.org>
Michael Reiss [Fri, 13 Apr 2007 06:26:19 +0000 (01:26 -0500)]
ucc_geth: Implement Transmit on Demand support
Transmit on Demand: Fix spelling in config option, and make it actually enable TOD.
Signed-off-by: Michael Reiss <michael.f.reiss@freescale.com> Signed-off-by: Michael Barkowski <michael.barkowski@freescale.com> Signed-off-by: Kim Phillips <kim.phillips@freescale.com> Signed-off-by: Jeff Garzik <jeff@garzik.org>
ucc_geth: Fix interrupt coalescing size and alignment
The rx interrupt coalescing table alignment was "guessed" to be 4,
but should be 64. The size should be 8 * number of queues + 4.
Verified in the MPC8323E manual.
Signed-off-by: Michael Barkowski <Michael.Barkowski@freescale.com> Signed-off-by: Kim Phillips <kim.phillips@freescale.com> Signed-off-by: Jeff Garzik <jeff@garzik.org>
Michael Reiss [Fri, 13 Apr 2007 06:26:11 +0000 (01:26 -0500)]
ucc_geth: NAPI-related bug fixes
Based partly on the gianfar driver, this patch fixes several
bugs which were causing NAPI to be completely unusable.
* An IRQ is still needed in NAPI, to kick off NAPI task,
and for Tx processing. Request the IRQ.
* If rx_work_limit = 0 we are not complete.
* While running Rx NAPI processing we must mask Rx events,
including Rx busy.
* ucc_geth_rx function does not need a lock.
Could lead to deadlock in NAPI case.
* There's no need to loop reading ucce multiple times in the ISR,
so while adding the call to schedule NAPI which was not there,
simplify the event processing into if-else format.
* Rx Busy now kicks off NAPI processing, while still
being counted as an error.
Signed-off-by: Michael Reiss <michael.f.reiss@freescale.com> Signed-off-by: Michael Barkowski <michael.barkowski@freescale.com> Signed-off-by: Kim Phillips <kim.phillips@freescale.com> Signed-off-by: Jeff Garzik <jeff@garzik.org>
Kim Phillips [Fri, 13 Apr 2007 06:26:03 +0000 (01:26 -0500)]
ucc_geth: migrate ucc_geth to phylib
migrate ucc_geth to use the common phylib code.
There are several side effects from doing this:
o deprecate 'interface' property specification present
in some old device tree source files in
favour of a split 'max-speed' and 'interface-type'
description to appropriately match definitions
in include/linux/phy.h. Note that 'interface' property
is still honoured if max-speed or interface-type
are not present (backward compatible).
o compile-time CONFIG_UGETH_HAS_GIGA is eliminated
in favour of probe time speed derivation logic.
o adjust_link streamlined to only operate on maccfg2
and upsmr.r10m, instead of reapplying static initial
values related to the interface-type.
o Addition of UEC MDIO of_platform driver requires
platform code add 'mdio' type to id list
prior to calling of_platform_bus_probe (separate patch).
o ucc_struct_init introduced to reduce ucc_geth_startup
complexity.
Signed-off-by: Li Yang <leoli@freescale.com> Signed-off-by: Kim Phillips <kim.phillips@freescale.com> Signed-off-by: Jeff Garzik <jeff@garzik.org>
Kim Phillips [Fri, 13 Apr 2007 06:25:57 +0000 (01:25 -0500)]
phylib: add RGMII-ID interface mode definition
The RGMII spec allows compliance for devices that implement an internal
delay on TXC or RXC inside the transmitter. This patch adds an RGMII_ID
definition to support RGMII-ID devices in the phylib.
Signed-off-by: Kim Phillips <kim.phillips@freescale.com> Signed-off-by: Jeff Garzik <jeff@garzik.org>
Larry Finger [Mon, 9 Apr 2007 16:24:41 +0000 (11:24 -0500)]
[PATCH] ieee80211-crypt: Make some TKIP and CCMP error logging conditional
There are messages arising from ieee80211_crypt that spam the logs
of casual users. These are changed to be logged only if the user
specifically requests the IEEE80211_DEBUG_DROP messages. In either
case, the error/drop count is incremented.
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Daniel Drake [Sat, 7 Apr 2007 15:00:15 +0000 (16:00 +0100)]
[PATCH] zd1211rw: rework band edge patching
This change allows RF drivers to provide their own 6M band edge patching
implementation, while providing a generic implementation shared by most
currently supported RF's.
The upcoming ZD1211B/AL7230B code will use this to define its own
patching function, which is different from the other RF configurations.
Signed-off-by: Daniel Drake <dsd@gentoo.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Gabriel Paubert [Fri, 23 Mar 2007 19:07:26 +0000 (12:07 -0700)]
mv643xx_eth: make eth_port_uc_addr_{get,set}() calls symmetric
There is no good reason for the asymmetry in the parameters of
eth_port_uc_addr_get() and eth_port_uc_addr_set(). Make them
symmetric. Remove some gratuitous block comments while we're here.
Signed-off-by: Gabriel Paubert <paubert@iram.es> Signed-off-by: Dale Farnsworth <dale@farnsworth.org> Signed-off-by: Jeff Garzik <jeff@garzik.org>
Ron Mercer [Mon, 26 Mar 2007 19:43:52 +0000 (12:43 -0700)]
qla3xxx: Adding support for the Agere PHY (ET1011C)
This PHY support patch was written by Benjamin Li. Signed-off-by: Benjamin Li <benjamin.li@qlogic.com> Signed-off-by: Ron Mercer <ron.mercer@qlogic.com> Signed-off-by: Jeff Garzik <jeff@garzik.org>
Signed-off-by: Robert P. J. Day <rpjday@mindspring.com> Cc: Jeff Garzik <jeff@garzik.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Jeff Garzik <jeff@garzik.org>
baycom_ser_fdx: also allow i/o ports >= 0x1000 and enhanced failure logging
The baycom_ser_fdx driver did not allow i/o ports >= 0x1000. Now that
there are pci cards (with rs232 ports) which use for example 0xb800 this
limit should not exists.
Also, for non kernel coders find the cause of problems was challenging so I
added extra logging.
Signed-off-by: Folkert van Heusden <folkert@vanheusden.com> Cc: Jeff Garzik <jeff@garzik.org> Cc: Alan Cox <alan@lxorguk.ukuu.org.uk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Jeff Garzik <jeff@garzik.org>
Jean Delvare [Tue, 27 Mar 2007 05:47:22 +0000 (21:47 -0800)]
strlcpy is smart enough
strlcpy already accounts for the trailing zero in its length
computation, so there is no need to substract one to the buffer size.
Signed-off-by: Jean Delvare <khali@linux-fr.org> Cc: Jan-Bernd Themann <themann@de.ibm.com> Cc: Thomas Klein <osstklei@de.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Jeff Garzik <jeff@garzik.org>
Jouni Malinen [Sun, 25 Mar 2007 00:15:30 +0000 (17:15 -0700)]
[PATCH] Update my email address from jkmaline@cc.hut.fi to j@w1.fi
After 13 years of use, it looks like my email address is finally going
to disappear. While this is likely to drop the amount of incoming spam
greatly ;-), it may also affect more appropriate messages, so let's
update my email address in various places. In addition, Host AP mailing
list is subscribers-only and linux-wireless can also be used for
discussing issues related to this driver which is now shown in
MAINTAINERS.
Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Pavel Roskin [Sun, 11 Mar 2007 03:54:22 +0000 (22:54 -0500)]
[PATCH] sparse-annotate radiotap header
Document that all fields must be little endian. Use annotated types
even in the comments. Consistently use shorter type names (u8, s8).
Realign the comments.
Signed-off-by: Pavel Roskin <proski@gnu.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Michal Schmidt [Fri, 16 Mar 2007 11:44:40 +0000 (12:44 +0100)]
[PATCH] airo: Make /sys/bus/pci/drivers/airo/{,un}bind work
The way airo.c keeps track of all its devices is complicated and buggy
as well (del_airo_dev forgets to free the memory add_airo_dev allocates).
It's cleaner to use the standard list primitives.
While we're at it, it's not necessary to put PCI cards in the list, because
the kernel already keeps track of them. We can take advantage of it and
use the .remove callback as it was meant to be.
This makes /sys/bus/pci/drivers/airo/{,un}bind work.
Signed-off-by: Michal Schmidt <mschmidt@redhat.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Larry Finger [Wed, 14 Mar 2007 20:06:22 +0000 (15:06 -0500)]
[PATCH] bcm43xx:Eliminate some 'G Mode Enable' magic numbers
In code manipulating the TM State Low register of 802.11 cores, two
different magic numbers are used to reference the 'G Mode Enable' bit.
One of these, 0x20000000, is clear, but the other, (0x800 << 18), is not.
This patch replaces both types with a defined constant. In addition, two
bits in the TM State High registers are given definitions to help in
following the code.
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Daniel Drake [Sun, 11 Mar 2007 19:54:39 +0000 (19:54 +0000)]
[PATCH] zd1211rw: More device IDs
ASUS A9Rp
Tested by Serge
zd1211b chip 0b05:171b v4802 high 00-17-31 AL2230_RF pa0 g--
ZyXEL G-202
Tested by Marcus D. Hanwell
zd1211b chip 0586:3410 v4810 high 00-13-49 AL2230_RF pa0 g---
US Robotics USR805423
Tested by Pascal S. de Kloe
FCC ID: RAXWN4501H
zd1211b chip 0baf:0121 v4810 high 00-14-c1 AL2230_RF pa0 g--N
Julien Pinon reports this also comes in AL2230S form
Signed-off-by: Daniel Drake <dsd@gentoo.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Don Fry [Tue, 6 Mar 2007 18:55:00 +0000 (10:55 -0800)]
pcnet32: change to use netdev_priv
use netdev_priv() instead of dev->priv
Signed-off-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de> Signed-off-by: Don Fry <pcnet32@verizon.net> Signed-off-by: Jeff Garzik <jeff@garzik.org>
Shan Lu [Tue, 6 Mar 2007 10:42:03 +0000 (02:42 -0800)]
network: add the missing phy_device speed information to phy_mii_ioctl
Function `phy_mii_ioctl' returns physical device's information based on
user requests. When requested to return the basic mode control register
information (BMCR), the original implementation only returns the physical
device's duplex information and forgets to return speed information, which
should not be because BMCR register is used to hold both duplex and speed
information.
The patch checks the BMCR value against speed-related flags and fills the
return structure's speed field accordingly.
Signed-off-by: Shan <shanlu@cs.uiuc.edu> Cc: Jeff Garzik <jeff@garzik.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Jeff Garzik <jeff@garzik.org>
Randy Dunlap [Tue, 6 Mar 2007 10:41:48 +0000 (02:41 -0800)]
phy layer: add kernel-doc + DocBook
Convert function documentation in drivers/net/phy/ to kernel-doc
and add it to DocBook.
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Jeff Garzik <jeff@garzik.org>
These timers all were of the "about once a second" or "about once every X
seconds" variety and several showed up in the "what wakes the cpu up" profiles
that the tickless patches provide. Some timers are highly dynamic based on
network load; but even on low activity systems they still show up so the
rounding is done only in cases of low activity, allowing higher frequency
timers in the high activity case.
The various hardware watchdogs are an obvious case; they run every 2 seconds
but aren't otherwise specific of exactly when they need to run.
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com> Acked-by: Auke Kok <auke-jan.h.kok@intel.com> Cc: Jeff Garzik <jeff@garzik.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Jeff Garzik <jeff@garzik.org>
Atsushi Nemoto [Sat, 3 Mar 2007 14:54:59 +0000 (23:54 +0900)]
tc35815 driver update (take 2)
Current tc35815 driver is very obsolete and less maintained for a long
time. Replace it with a new driver based on one from CELF patch
archive.
Major advantages of CELF version (version 1.23, for kernel 2.6.10) are:
* Independent of JMR3927.
(Actually independent of MIPS, but AFAIK the chip is used only on
MIPS platforms)
* TX4938 support.
* 64-bit proof.
* Asynchronous and on-demand auto negotiation.
* High performance on non-coherent architecture.
* ethtool support.
* Many bugfixes and cleanups.
And improvoments since version 1.23 are:
* TX4939 support.
* NETPOLL support.
* NAPI support. (disabled by default)
* Reduce memcpy on receiving.
* PM support.
* Many cleanups and bugfixes.
Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp> Signed-off-by: Jeff Garzik <jeff@garzik.org>
This patch introduces functionality to dynamically add / remove
ehea ports via an userspace DLPAR tool. It creates a subnode for
each logical port in the sysfs.
Signed-off-by: Jan-Bernd Themann <themann@de.ibm.com> Signed-off-by: Jeff Garzik <jeff@garzik.org>
Some code for Chelsio 1G boards was put in the driver
based on the vendor version (minus TOE). Well some of those board
versions are only supported with TOE on the vendor driver, so additional
dead code was added.
Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org> Signed-off-by: Jeff Garzik <jeff@garzik.org>
Sam Ravnborg [Sun, 4 Mar 2007 05:55:08 +0000 (23:55 -0600)]
[PATCH] bcm43xx: do not rebuild when kernel version changes
In bcm43xx_ethtool, UTS_RELEASE is used. Replacing this with utsname()->release
avoids rebuilding this module each time the kernel version changes.
Signed-off-by: Sam Ravnborg <sam@ravnborg.org> Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Jouni Malinen [Wed, 28 Feb 2007 03:46:46 +0000 (19:46 -0800)]
[PATCH] hostap: Add D-Link DWL-650 Rev. P1 product id
It looks like some of the PC Card manfid/product strings were lost
when Host AP driver was converted to use PCMCIA_DEVICE_* helpers. This
patch adds back D-Link DWL-650 Rev. P1 using the same product ID
string match as the pcmcia-cs/cardmgr configuration used before.
Signed-off-by: Jouni Malinen <jkmaline@cc.hut.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Pavel Roskin [Mon, 19 Feb 2007 01:44:06 +0000 (20:44 -0500)]
[PATCH] hostap: use offsetof() instead of own equivalent
The original macros result in gcc 4.2 warning about "cast from pointer
to integer of different size" on 64-bit systems.
Use of offsetof() on fields in substructures is widespread throughout
the kernel code and should work whether offsetof() is defined using
__compiler_offsetof() or a cast.
Signed-off-by: Pavel Roskin <proski@gnu.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Ulrich Kunitz [Sun, 18 Feb 2007 20:28:23 +0000 (20:28 +0000)]
[PATCH] zd1211rw: changed GFP_NOFS to GFP_KERNEL
Michael Buesch commented that GFP_NOFS should not be used in a
network driver.
Signed-off-by: Ulrich Kunitz <kune@deine-taler.de> Signed-off-by: Daniel Drake <dsd@gentoo.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Neil Horman [Thu, 26 Apr 2007 17:47:36 +0000 (13:47 -0400)]
sis900: Allocate rx replacement buffer before rx operation
Just found a hole in my last patch. It was reported to me that shortly after we
integrated this patch. The report was of an oops that took place inside of
netif_rx when using the sis900 driver. Looking at my origional patch I noted
that there was a spot between the new skb_alloc and the refill_rx_ring label
where skb got reassigned to the pointer currently held in the rx_ring for the
purposes of receiveing the frame. The result of this is however that the buffer
that gets passed to netif_rx (if it is called), then gets placed right back into
the rx_ring. So if you receive frames fast enough the skb being processed by
the network stack can get corrupted. The reporter is testing out the fix I've
written for this below (I'm not near my hardware at the moment to test myself),
but I wanted to post it for review ASAP. I'll post test results when I hear
them, but I think this is a pretty straightforward fix. It just uses a separate
pointer to do the rx operation, so that we don't improperly reassign the pointer
that we use to refill the rx ring.
Signed-off-by: Neil Horman <nhorman@tuxdriver.com> Signed-off-by: Jeff Garzik <jeff@garzik.org>
* git://git.infradead.org/mtd-2.6: (46 commits)
[MTD] [MAPS] drivers/mtd/maps/ck804xrom.c: convert pci_module_init()
[MTD] [NAND] CM-x270 MTD driver
[MTD] [NAND] Wrong calculation of page number in nand_block_bad()
[MTD] [MAPS] fix plat-ram printk format
[JFFS2] Fix compr_rubin.c build after include file elimination.
[JFFS2] Handle inodes with only a single metadata node with non-zero isize
[JFFS2] Tidy up licensing/copyright boilerplate.
[MTD] [OneNAND] Exit loop only when column start with 0
[MTD] [OneNAND] Fix access the past of the real oobfree array
[MTD] [OneNAND] Update Samsung OneNAND official URL
[JFFS2] Better fix for all-zero node headers
[JFFS2] Improve read_inode memory usage, v2.
[JFFS2] Improve failure mode if inode checking leaves unchecked space.
[JFFS2] Fix cross-endian build.
[MTD] Finish conversion mtd_blkdevs to use the kthread API
[JFFS2] Obsolete dirent nodes immediately on unlink, where possible.
Use menuconfig objects: MTD
[MTD] mtd_blkdevs: Convert to use the kthread API
[MTD] Fix fwh_lock locking
[JFFS2] Speed up mount for directly-mapped NOR flash
...
The simplification made in that change works with the assumption that
the 'offset' parameter to these functions is always positive or zero,
which is not true. It can be and often is negative in order to access
SKB header values in front of skb->data.
Signed-off-by: David S. Miller <davem@davemloft.net>
projects / linux-2.6-omap-h63xx.git / log