summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2010-08-02tg3: Improve small packet performanceMatt Carlson
smp_mb() inside tg3_tx_avail() is used twice in the normal tg3_start_xmit() path (see illustration below). The full memory barrier is only necessary during race conditions with tx completion. We can speed up the tx path by replacing smp_mb() in tg3_tx_avail() with a compiler barrier. The compiler barrier is to force the compiler to fetch the tx_prod and tx_cons from memory. In the race condition between tg3_start_xmit() and tg3_tx(), we have the following situation: tg3_start_xmit() tg3_tx() if (!tg3_tx_avail()) BUG(); ... if (!tg3_tx_avail()) netif_tx_stop_queue(); update_tx_index(); smp_mb(); smp_mb(); if (tg3_tx_avail()) if (netif_tx_queue_stopped() && netif_tx_wake_queue(); tg3_tx_avail()) With smp_mb() removed from tg3_tx_avail(), we need to add smp_mb() to tg3_start_xmit() as shown above to properly order netif_tx_stop_queue() and tg3_tx_avail() to check the ring index. If it is not strictly ordered, the tx queue can be stopped forever. This improves performance by about 3% with 2 ports running bi-directional 64-byte packets. Reviewed-by: Benjamin Li <benli@broadcom.com> Signed-off-by: Michael Chan <mchan@broadcom.com> Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-08-02tg3: Remove 5720, 5750, and 5750MMatt Carlson
These devices were never released to the public. Reviewed-by: Benjamin Li <benli@broadcom.com> Reviewed-by: Michael Chan <mchan@broadcom.com> Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-08-02tg3: Detect APE firmware typesMatt Carlson
This patch adds code to determine the APE firmware type and report this along with the firmware version. Reviewed-by: Benjamin Li <benli@broadcom.com> Reviewed-by: Michael Chan <mchan@broadcom.com> Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-08-02tg3: Restrict ASPM workaround devlistMatt Carlson
The ASPM workaround setting obtained from NVRAM only works with devices older than 5717. This patch enforces the restriction. Reviewed-by: Benjamin Li <benli@broadcom.com> Reviewed-by: Michael Chan <mchan@broadcom.com> Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-08-02tg3: Manage gphy power for CPMU-less devs onlyMatt Carlson
This patch changes the code to only manage the PCIe gphy power for CPMU-less devices only. The CPMU takes over management for newer chips. Reviewed-by: Benjamin Li <benli@broadcom.com> Reviewed-by: Michael Chan <mchan@broadcom.com> Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-08-02tg3: Don't access phy test ctrl reg for 5717+Matt Carlson
The phy test register location has been repurposed for 5717+ devices. This patch changes the code to avoid this location for these devices. Reviewed-by: Benjamin Li <benli@broadcom.com> Reviewed-by: Michael Chan <mchan@broadcom.com> Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-08-02tg3: Create TG3_FLG3_5717_PLUS flagMatt Carlson
This patch creates a TG3_FLG3_5717_PLUS flag to collectively describe the set of changes in the ASIC that will apply to all future chip revisions. Signed-off-by: Michael Chan <mchan@broadcom.com> Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-08-02tg3: Disable TSS also during tg3_close()Matt Carlson
The TSS flag needs to be turned off during tg3_close(). If the device fails to allocate more than one MSI-X vector the next time the device is brought up, transmits will fail. Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-08-02tg3: Add 5784 ASIC rev to earlier PCIe MPS fixMatt Carlson
tg3 commit e7126997342560533317d8467e8516119ebcbd21 entitled "tg3: Preserve PCIe MPS setting for new devs" attempted to ensure the PCIe link negotiated Maximum Payload Size (MPS) setting was 128 bytes for all devices that didn't support higher speeds. The 5784 device was mistakenly added to this list when it shouldn't have. This patch removes the 5784 ASIC rev devices from that list. Reviewed-by: Benjamin Li <benli@broadcom.com> Reviewed-by: Michael Chan <mchan@broadcom.com> Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-08-02Merge branch 'master' of ↵David S. Miller
git://git.kernel.org/pub/scm/linux/kernel/git/kaber/nf-next-2.6
2010-08-02V4L/DVB: sms: Convert IR support to use the Remote Controller coreMauro Carvalho Chehab
Rewrites the siano IR implementation. The previous implementation were non-standard. As such, it has issues if more than one device registers IR, as there used to have some static constants used during protocol decoding phase. Also, it used to implement its on RAW decoder, and only for RC5. The new code uses RC core subsystem for handling IR. This brings several new features to the driver, including: - Allow to dynamically replace the IR keycodes; - Supports all existing raw decoders (JVC, NEC, RC-5, RC-6, SONY); - Supports lirc dev; - Doesn't have race conditions when more than one sms IR is registered; - The code size for the IR implementation is very small; - it exports the IR features via /sys/class/rc. Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-08-02V4L/DVB: sms: properly initialize IR phys and IR nameMauro Carvalho Chehab
sms were using a non-compliant nomenclature for the USB devices. Fix it. Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-08-02V4L/DVB: standardize names at rc-dib0700 tablesMauro Carvalho Chehab
Use a more standard way to name those tables, as they're currently used by the script that coverts those tables to be loaded via userspace. Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-08-02V4L/DVB: smsusb: enable IR port for Hauppauge WinTV MiniStickMauro Carvalho Chehab
Add the proper gpio port for WinTV MiniStick, with the information provided by Michael. Thanks-to: Michael Krufky <mkrufky@kernellabs.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-08-02V4L/DVB: dib0700: Fix RC protocol logic to properly handle NEC/NECx and RC-5Mauro Carvalho Chehab
Simplifies the logic for handling firmware 1.20 RC messages, fixing the logic. While here, I tried to use a RC-6 remote controller from my TV set, but it didn't work with dib0700. Not sure why, but maybe this never worked. Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-08-02V4L/DVB: dib0700: properly implement IR change_protocolMauro Carvalho Chehab
This patch implements change_protocol callback. With this change, there's no need for an extra modprobe parameter to specify the protocol. When a table is loaded (either from in-kernel rc-map tables or via ir-keytable program), the driver will automatically change the protocol, in order to work with the given table. Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-08-02V4L/DVB: dib0700: break keytable into NEC and RC-5 variantsMauro Carvalho Chehab
Instead of having one big keytable with 2 protocols inside, break it into two separate tables, being one for NEC and another for RC-5 variants, and properly identify what variant should be used at the boards entries. Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-08-02V4L/DVB: dib0700: avoid bad repeatMauro Carvalho Chehab
a 250ms delay is too low for this device. It ends by producing false repeat events. Increase the delay time to 500 ms to avoid troubles. Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-08-02V4L/DVB: Port dib0700 to rc-coreMauro Carvalho Chehab
Use the new rc-core handler at dvb-usb-remote for dib0700 driver. Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-08-02V4L/DVB: Add a keymap file with dib0700 tableMauro Carvalho Chehab
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-08-02V4L/DVB: dvb-usb: add support for rc-core modeMauro Carvalho Chehab
Allows dvb-usb drivers to use rc-core, instead of the legacy implementation. No driver were ported yet to rc-core, so, some small adjustments may be needed, when starting to migrate the drivers. Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-08-02V4L/DVB: dvb-usb: prepare drivers for using rc-coreMauro Carvalho Chehab
This is a big patch, yet trivial. It just move the RC properties to a separate struct, in order to prepare the dvb-usb drivers to use rc-core. There's no change on the behavior of the drivers. With this change, it is possible to have both legacy and rc-core based code inside the dvb-usb-remote, allowing a gradual migration to rc-core, driver per driver. Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-08-02V4L/DVB: dvb-usb: get rid of struct dvb_usb_rc_keyMauro Carvalho Chehab
dvb-usb has its own IR handle code. Now that we have a Remote Controller subsystem, we should start using it. So, remove this struct, in favor of the similar struct defined at the RC subsystem. This is a big, but trivial patch. It is a 3 line delect, plus lots of rename on several dvb-usb files. Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-08-02V4L/DVB: rj54n1cb0c: fix a comment in the driverGuennadi Liakhovetski
RJ54N1CB0C is a Sharp camera sensor. Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-08-02V4L/DVB: V4L2: sh_vou: VOU does support the full PAL resolution tooGuennadi Liakhovetski
SH7724 datasheet specifies 480 pixels as the VOU maximum vertical resolution. This is a bug in the datasheet, VOU also supports the full PAL resolution: 576 lines. Adjust the driver accordingly. Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-08-02V4L/DVB: V4L2: sh_mobile_camera_ceu: add support for CSI2Guennadi Liakhovetski
Using CEU with CSI2 on SH-Mobile requires some special configuration of the former. We also have to switch from calling only one subdev .s_mbus_fmt and .try_mbus_fmt to calling all subdevices. Take care to increment CSI2 driver use count to prevent it from unloading, while in use. Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-08-02V4L/DVB: V4L2: soc-camera: add a MIPI CSI-2 driver for SH-Mobile platformsGuennadi Liakhovetski
Some SH-Mobile SoCs implement a MIPI CSI-2 controller, that can interface to several video clients and send data to the CEU or to the Image Signal Processor. This patch implements a v4l2-subdevice driver for CSI-2 to be used within the soc-camera framework, implementing the second subdevice in addition to the actual video clients. Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-08-02V4L/DVB: V4L2: soc-camera: export soc-camera bus type for notificationsGuennadi Liakhovetski
Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-08-02V4L/DVB: V4L2: mediabus: add 12-bit Bayer and YUV420 pixel formatsGuennadi Liakhovetski
These formats belong to the standard format set, defined by the MIPI CSI-2 specification. Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-08-02V4L/DVB: mediabus: fix ambiguous pixel code namesGuennadi Liakhovetski
Endianness notation is meaningless for 8 bit YUYV codes. Switch pixel code names to explicitly state the order of colour components in the data stream. Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-08-02V4L/DVB: staging/lirc: CodingStyle cleanupsMauro Carvalho Chehab
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-08-02V4L/DVB: staging/lirc: wire up Kconfig and Makefile bitsJarod Wilson
Make the bits under staging/lirc/ buildable, and add a TODO note. Signed-off-by: Jarod Wilson <jarod@redhat.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-08-02V4L/DVB: staging/lirc: add lirc_zilog driverJarod Wilson
Commonly found on several Hauppauge video capture devices. Signed-off-by: Jarod Wilson <jarod@redhat.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-08-02V4L/DVB: staging/lirc: add lirc_ttusbir driverJarod Wilson
Signed-off-by: Jarod Wilson <jarod@redhat.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-08-02V4L/DVB: staging/lirc: add lirc_streamzap driverJarod Wilson
Signed-off-by: Jarod Wilson <jarod@redhat.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-08-02V4L/DVB: staging/lirc: add lirc_sir driverJarod Wilson
Signed-off-by: Jarod Wilson <jarod@redhat.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-08-02V4L/DVB: staging/lirc: add lirc_serial driverJarod Wilson
Signed-off-by: Jarod Wilson <jarod@redhat.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-08-02V4L/DVB: staging/lirc: add lirc_sasem driverJarod Wilson
Signed-off-by: Jarod Wilson <jarod@redhat.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-08-02V4L/DVB: staging/lirc: add lirc_parallel driverJarod Wilson
Signed-off-by: Jarod Wilson <jarod@redhat.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-08-02V4L/DVB: staging/lirc: add lirc_it87 driverJarod Wilson
Signed-off-by: Jarod Wilson <jarod@redhat.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-08-02V4L/DVB: staging/lirc: add lirc_ite8709 driverJarod Wilson
Signed-off-by: Jarod Wilson <jarod@redhat.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-08-02V4L/DVB: staging/lirc: add lirc_imon driverJarod Wilson
Nb: this is a very trimmed-down lirc_imon, which only supports the oldest generation of imon devices that don't do onboard signal decoding like the most recent generation imon devices -- those are now supported by the ir-core imon driver. Signed-off-by: Jarod Wilson <jarod@redhat.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-08-02V4L/DVB: staging/lirc: add lirc_igorplugusb driverJarod Wilson
Signed-off-by: Jarod Wilson <jarod@redhat.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-08-02V4L/DVB: staging: Add an specific TODO note for lirc_i2cMauro Carvalho Chehab
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-08-02V4L/DVB: staging/lirc: add lirc_i2c driverJarod Wilson
Signed-off-by: Jarod Wilson <jarod@redhat.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-08-02V4L/DVB: staging/lirc: add lirc_ene0100 driverJarod Wilson
[mchehab@redhat.com: fixed a few WARNING: please, no space before tabs] Signed-off-by: Jarod Wilson <jarod@redhat.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-08-02V4L/DVB: staging/lirc: add lirc_bt829 driverJarod Wilson
Signed-off-by: Jarod Wilson <jarod@redhat.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-08-02V4L/DVB: af9005: use generic_bulk_ctrl_endpoint_responseMichael Krufky
Signed-off-by: Michael Krufky <mkrufky@kernellabs.com> Acked-by: Luca Olivetti <luca@ventoso.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-08-02V4L/DVB: dvb: add support for kworld 340u and ub435-q to em28xx-dvbJarod Wilson
This adds support for the KWorld PlusTV 340U and KWorld UB345-Q ATSC sticks, which are really the same device. The sticks have an eMPIA em2870 usb bridge chipset, an LG Electronics LGDT3304 ATSC/QAM demodulator and an NXP TDA18271HD tuner -- early versions of the 340U have a a TDA18271HD/C1, later models and the UB435-Q have a C2. The stick has been tested succesfully with both VSB_8 and QAM_256 signals. Its using lgdt3304 support added to the lgdt3305 driver by a prior patch, rather than the current lgdt3304 driver, as its severely lacking in functionality by comparison (see said patch for details). Signed-off-by: Jarod Wilson <jarod@redhat.com> Signed-off-by: Michael Krufky <mkrufky@kernellabs.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-08-02V4L/DVB: DVB: fix dvr node refcountingJiri Slaby
In dvb_dvr_release, there is a test dvbdev->users==-1, but users are never negative. This error results in hung tasks: task PC stack pid father bash D ffffffffa000c948 0 3264 3170 0x00000000 ffff88003aec5ce8 0000000000000086 0000000000011f80 0000000000011f80 ffff88003aec5fd8 ffff88003aec5fd8 ffff88003b848670 0000000000011f80 ffff88003aec5fd8 0000000000011f80 ffff88003e02a030 ffff88003b848670 Call Trace: [<ffffffff813dd4a5>] dvb_dmxdev_release+0xc5/0x130 [<ffffffff8107b750>] ? autoremove_wake_function+0x0/0x40 [<ffffffffa00013a2>] dvb_usb_adapter_dvb_exit+0x42/0x70 [dvb_usb] [<ffffffffa0000525>] dvb_usb_exit+0x55/0xd0 [dvb_usb] [<ffffffffa00005ee>] dvb_usb_device_exit+0x4e/0x70 [dvb_usb] [<ffffffffa000a065>] af9015_usb_device_exit+0x55/0x60 [dvb_usb_af9015] [<ffffffff813a3f05>] usb_unbind_interface+0x55/0x1a0 [<ffffffff81316000>] __device_release_driver+0x70/0xe0 ... So check against 1 there instead. BTW why's the TODO there? Adding TODOs to the code without descriptions is like adding nothing. Signed-off-by: Jiri Slaby <jslaby@suse.cz> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
generated by cgit (git 2.34.1) at 2025-09-12 16:56:35 +0000