| 72541ce | Mohamed Habib | 15 August 2011, 15:20:27 UTC | Xilinx: ARM: Updated xilinx_qemu_defconfig for SMP xilinx_qemu_defconfig has been updated to configure the kernel with processor local timers disabled since a problem was found with QEMU when trying to run the kernel with processor local timers enabled on SMP enabled QEMU. | 15 August 2011, 15:45:22 UTC |
| e0caf77 | John Linn | 09 August 2011, 19:47:35 UTC | Xilinx: ARM: SLCR: updated reset logic and unlock The driver was not allowing the resets in the FPGA reset register to be cleared or more than one at a time set. This is needed to allow bitstreams to be downloaded by holding the FPGA in reset, then clearing the resets after the bitstream is loaded. The driver was also not unlocking the SLCR so that the register could not be written. | 09 August 2011, 19:50:00 UTC |
| 0b48d74 | Srikanth Thokala | 27 July 2011, 05:19:46 UTC | Xilinx: ARM: net: PTP Hardware Tx Timestamp Support Added support for PTP Hardware Tx Timestamping Support. Tested on PEEP Board. Signed-off-by: SrikanthT <sthokal@xilinx.com> | 27 July 2011, 17:01:22 UTC |
| 662e092 | Srikanth Thokala | 27 July 2011, 05:14:04 UTC | Xilinx: ARM: net: Code Cleanup Applied checkpatch.pl and cleaned the code. Tested on PEEP board. Signed-off-by: SrikanthT <sthokal@xilinx.com> | 27 July 2011, 17:00:28 UTC |
| ddd2f57 | Mohamed Habib | 20 July 2011, 21:00:25 UTC | Xilinx: ARM: Updated xilinx_qemu_defconfig for device tree xilinx_qemu_defconfig has been updated with device tree enabled. It should be used to configure the kernel intended to be run on QEMU after updating QEMU to support device tree enabled kernel. Despite the fact that at this point not all device drivers have been upgraded to use devcie tree, it looks like by using this file to configure the kernek, there is no need anymore to change anything in devcies.c file before building the kernel that is intended to be run on QEMU. | 21 July 2011, 14:34:12 UTC |
| 5747dd9 | vidhum | 18 July 2011, 10:55:22 UTC | Xilinx: ARM: Device Config Driver: Fixed the jtag not working issue The driver was preventing any debug tools like XMD from attaching to the JTAG chain. This fixes that problem. Signed-off-by: Vidhumouli H <vidhum@xilinx.com> | 18 July 2011, 15:07:22 UTC |
| 9577e18 | John Linn | 05 July 2011, 16:10:17 UTC | Xilinx: ARM: SCU global timer: add 64 bit attributes in sysfs The counter and comparators are 64 bits even though the h/w is implemented as 2 32 bit registers. This makes it easier to use these attributes. | 05 July 2011, 16:10:17 UTC |
| de206f9 | John Linn | 01 July 2011, 16:31:15 UTC | Revert "Xilinx: ARM: SDHCI: removed ADMA support as it's not working" This reverts commit 2bc0c63dc02306745420235c4a57c1a0c32ac72d. | 01 July 2011, 16:31:15 UTC |
| c9ef8f3 | John Linn | 01 July 2011, 15:58:55 UTC | Xilinx: ARM: updated defconfigs to include unix sockets This change is needed to support booting an Ubuntu file system as upstart needs this feature. | 01 July 2011, 15:58:55 UTC |
| 34c7025 | John Linn | 30 June 2011, 21:02:52 UTC | Xilinx: ARM: added xdefcfg driver to defconfigs The device configuration driver is needed in the kernel to allow bitstreams to be downloaded. | 30 June 2011, 21:10:19 UTC |
| 23271b6 | John Linn | 30 June 2011, 16:12:59 UTC | Xilinx: ARM: BSP: Adding SCU global timer sysfs support This platform driver is a bit of an experiment to see if it really meets the needs. It only exports the timer registers to user space in sysfs so that they can be manipulated. They will appear in /sys/devices/platform/xscugtimer.0 directory. | 30 June 2011, 16:12:59 UTC |
| cf49b75 | John Linn | 28 June 2011, 20:57:57 UTC | Xilinx: ARM: I2C: Updated to support sub-devices in device tree This allows the i2c eeproms and rtc to be moved into the device tree so it is a more complete solution. | 28 June 2011, 20:59:26 UTC |
| 7942805 | John Linn | 24 June 2011, 21:46:44 UTC | Xilinx: ARM: update test defconfigs for emac driver name change The GEM driver changed from PSS to PS. | 24 June 2011, 21:46:44 UTC |
| 0edb721 | John Linn | 24 June 2011, 21:42:03 UTC | Xilinx: ARM: update defconfigs for emac driver name change The GEM driver changed from PSS to PS. | 24 June 2011, 21:42:03 UTC |
| 809437c | John Linn | 24 June 2011, 20:09:55 UTC | Xilinx: ARM: net: updated kconfig and make for new name The GEM driver changed from PSS to PS and the rest of the files needed to be changed also. | 24 June 2011, 20:26:17 UTC |
| 7879312 | Anirudha Sarangi | 24 June 2011, 13:56:50 UTC | Xilinx: EMAC PS (GEM) Driver: Addred device tree support Renamed xilinx_emacpss.c to xilinx_emacps.c. Made changes in driver file to rename all instances of emacpss to emacps. Added device tree support by adding entries in xilinx.dts and making other relevant changes. Signed-off-by: Anirudha Sarangi <anirudh@xilinx.com> | 24 June 2011, 19:25:57 UTC |
| 9ab6cd2 | John Linn | 21 June 2011, 22:32:39 UTC | Xilinx: ARM: WDT: fixed typo in driver name This error crept in during the device tree changes and kept the driver from running. | 21 June 2011, 22:32:39 UTC |
| 1ace8cb | Julie Zhu | 21 June 2011, 18:26:02 UTC | Xilinx: Dma: Driver update for loopback widget and test fix for receive buffer Driver updated for loopback widget v3.00.a. Test file, dmatest.c.cdma_dma, updated for receive buffer proper flush before test starts. Tested on peep10, with Linux version pele-beta1_1.1. | 21 June 2011, 18:26:02 UTC |
| 8564526 | John Linn | 21 June 2011, 16:27:34 UTC | Xilinx: ARM: Device Cfg: Fixed driver compile error A minor annoyance, the last patch was not correct. | 21 June 2011, 16:27:34 UTC |
| f53b509 | vidhum | 21 June 2011, 06:24:04 UTC | Xilinx: ARM: Device Config Driver: Completed device tree support The platform driver structure in driver file is updated with the .of_match_table parameter. The devices.c file is updated with the related device-tree changes. Signed-off-by: Vidhumouli H <vidhum@xilinx.com> | 21 June 2011, 16:17:19 UTC |
| 2413c44 | Mohamed Habib | 20 June 2011, 20:45:51 UTC | Xilinx: ARM: Adding QEMU kernel config file QEMU requires some special kernel configuration and this allows users to configure the kernel easier. | 20 June 2011, 21:04:08 UTC |
| 3600f8a | John Linn | 20 June 2011, 20:34:57 UTC | Xilinx: ARM: net: removed compiler warning from driver A call to platform_set_drvdata in mii_init() was removed as it appeared to be always wrong and now tighter typing in the kernel was exposing it. | 20 June 2011, 20:34:57 UTC |
| 8f3190d | John Linn | 17 June 2011, 20:48:45 UTC | Xilinx: Updated defconfig to add device tree partition for NOR Since we were using partitions for NOR on the command line the kernel config had to be changed to add the new device tree partition so that platform data matches the device tree operation. | 17 June 2011, 20:48:45 UTC |
| 466d533 | John Linn | 17 June 2011, 20:44:53 UTC | Xilinx: ARM: Add device tree partitions to NAND and QSPI Device trees will be stored in flash eventually and this is just getting ready for it before NAND and QSPI drivers are ready for device tree. This is only to make the partitions match between platform data and device tree operation. | 17 June 2011, 20:44:53 UTC |
| 59d7bba | John Linn | 16 June 2011, 21:20:43 UTC | Merge branch 'master-usb-dmazone' into master-of-sadanand-dma | 16 June 2011, 21:20:43 UTC |
| 9f18db8 | John Linn | 16 June 2011, 19:48:52 UTC | Xilinx: ARM: Update of test defconfigs for driver name changes The GPIO, I2C, and WDT drivers changes from PSS to PS. Also added other device tree support. | 16 June 2011, 19:52:56 UTC |
| 6c99b69 | John Linn | 16 June 2011, 18:43:51 UTC | Xilinx: ARM: Adding device tree defconfig Since there are getting to be a number of config items that must be turned on in the kernel config to support device tree it seems easier for development to have a device tree defconfig (xilinx_devtree_defconfig). | 16 June 2011, 18:43:51 UTC |
| 3f05b44 | John Linn | 16 June 2011, 18:35:06 UTC | Xilinx: ARM: Update defconfig for driver name changes The I2C, SPI, WDT, and GPIO drivers were updated to PS rather than PSS. The A9 WDT was changed to SCU WDT. | 16 June 2011, 18:35:06 UTC |
| 6b99a13 | John Linn | 16 June 2011, 17:44:07 UTC | Xilinx: ARM: updated BSP for SD driver in device tree The SD driver was being registered as a platform device when the device tree was being used. | 16 June 2011, 17:44:07 UTC |
| 0c0e211 | Sadanand M | 16 June 2011, 04:04:59 UTC | Xilinx: ARM: I2C Driver: Fix compilation errors The i2c ps driver was renamed from pss to ps, but the platform data structure wasn't renamed, which causes the compilation to fail when platform data is used. This patch fixes the error Signed-off-by: Sadanand M <sadanan@xilinx.com> | 16 June 2011, 15:40:16 UTC |
| 61b2235 | Sadanand M | 16 June 2011, 03:55:40 UTC | Xilinx: ARM: GPIO driver: Fix compilation errors All the functions, variables, etc. in the GPIO PS driver were renamed from *pss to *ps, but the constant XGPIOPSS_IRQBASE wasn't renamed, which causes a compilation failure. This patch fixes the error Signed-off-by: Sadanand M <sadanan@xilinx.com> | 16 June 2011, 15:40:15 UTC |
| 5760eeb | Sadanand M | 15 June 2011, 11:26:09 UTC | Xilinx: ARM: Add entries for sdhci in device-tree This patch adds nodes for sdhci devices to xilinx.dts. sdhci-of-core.c driver can be used with generic sdhci controller, when OF support is enabled. Signed-off-by: Sadanand M <sadanan@xilinx.com> | 15 June 2011, 19:52:57 UTC |
| 1b24f24 | Sadanand M | 15 June 2011, 11:18:01 UTC | Xilinx: ARM: sdhci OF driver: use dummy sdhci_ops sdhci-of-core.c driver doesn't setup any sdhci_ops when generic-sdhci device is used. Without any sdhci_ops, sdhci.c will access a NULL pointer, causing the kernel to crash. This patch hacks the OF driver to use dummy sdhci_ops. With dummy sdhci_ops, sdhci.c uses default readl/writel functions Signed-off-by: Sadanand M <sadanan@xilinx.com> | 15 June 2011, 19:52:47 UTC |
| 98e7a7e | Sadanand M | 15 June 2011, 10:41:22 UTC | Xilinx: ARM: Add scu wdt to device-tree Add a node for SCU WDT in mach-xilinx/devices.c Signed-off-by: Sadanand M <sadanan@xilinx.com> | 15 June 2011, 19:52:36 UTC |
| c3f02b2 | Sadanand M | 15 June 2011, 10:37:38 UTC | Xilinx: ARM: SCU WDT driver: Rename to SCU WDT and add OF support rename the driver from a9 wdt to scu wdt and add support for OF bus bindings Signed-off-by: Sadanand M <sadanan@xilinx.com> | 15 June 2011, 19:52:26 UTC |
| 7fcaed6 | Sadanand M | 15 June 2011, 10:24:09 UTC | Xilinx: ARM: Add entry for ps wdt in device-tree Add a node for ps wdt in mach-xilinx/xilinx.dts. ps wdt can be used with device tree Signed-off-by: Sadanand M <sadanan@xilinx.com> | 15 June 2011, 19:52:01 UTC |
| 90b2b63 | Sadanand M | 15 June 2011, 10:18:47 UTC | Xilinx: ARM: PS WDT driver: Rename to PS and add OF support Rename the driver, variables, constants, etc from PSS to PS and add support for OF bus bindings Signed-off-by: Sadanand M <sadanan@xilinx.com> | 15 June 2011, 19:51:45 UTC |
| 33ae4b2 | Sadanand M | 15 June 2011, 10:05:06 UTC | Xilinx: ARM: GPIO driver: Update Makefile and Kconfig Update the Makefile to use new driver filename in Makefile and change PSS to PS in Kconfig Signed-off-by: Sadanand M <sadanan@xilinx.com> | 15 June 2011, 19:51:28 UTC |
| 1f265a1 | Sadanand M | 15 June 2011, 09:59:37 UTC | Xilinx: ARM: I2C driver: Update Makefile and Kconfig Update Makefile to use new driver file name and change PSS to PS in Kconfig Signed-off-by: Sadanand M <sadanan@xilinx.com> | 15 June 2011, 19:51:14 UTC |
| ca4c7ad | Sadanand M | 15 June 2011, 09:43:00 UTC | Xilinx: ARM: BSP: Update SLCR registration code This patch registers SLCR platform device only when CONFIG_OF is not defined Signed-off-by: Sadanand M <sadanan@xilinx.com> | 15 June 2011, 19:50:56 UTC |
| b452198 | Sadanand M | 15 June 2011, 09:31:00 UTC | Xilinx: ARM: Add an entry for slcr in dts Add SLCR node to xilinx.dts in mach-xilinx. SLCR driver supports OF bus bindings Signed-off-by: Sadanand M <sadanan@xilinx.com> | 15 June 2011, 19:50:44 UTC |
| 742af2f | Sadanand M | 15 June 2011, 09:25:03 UTC | Xilinx: ARM: SLCR driver: Add device tree support Update the driver to support OF bus bindings and rename the driver name in plaform_driver structure to xslcr Signed-off-by: Sadanand M <sadanan@xilinx.com> | 15 June 2011, 19:50:29 UTC |
| ae9d6ca | Sadanand M | 15 June 2011, 08:13:28 UTC | Xilinx: ARM: Added gpio entry to xilinx.dts Add GPIO node to the dts file in mach-xilinx. GPIO PS driver can be used with device tree Signed-off-by: Sadanand M <sadanan@xilinx.com> | 15 June 2011, 19:50:20 UTC |
| b5a4593 | Sadanand M | 15 June 2011, 08:07:34 UTC | Xilinx: ARM: BSP: Use new GPIO driver name Use new gpio driver name in the platform initialization code and register i2c and gpio devices as platfrom devices only when CONFIG_OF is not defined Signed-off-by: Sadanand M <sadanan@xilinx.com> | 15 June 2011, 19:50:06 UTC |
| d6f93e6 | Sadanand M | 15 June 2011, 07:56:02 UTC | Xilinx: ARM: GPIO PS driver: Rename driver and add device tree support Rename the driver from pss to ps and add support for OF bus bindings Signed-off-by: Sadanand M <sadanan@xilinx.com> | 15 June 2011, 19:49:16 UTC |
| e700d4f | Sadanand M | 15 June 2011, 07:31:05 UTC | Xilinx: ARM: Added I2C entries to device-tree Added I2 nodes to xilinx.dts in mach-xilinx. I2C driver can now be used with device tree Signed-off-by: Sadanand M <sadanan@xilinx.com> | 15 June 2011, 19:49:03 UTC |
| d3b4ebb | Sadanand M | 15 June 2011, 07:29:00 UTC | Xilinx: ARM: I2C PS driver: rename driver and add device tree suppport Rename the driver and variables, constants, functions, etc. from pss to ps and update to support OF bus bindings Signed-off-by: Sadanand M <sadanan@xilinx.com> | 15 June 2011, 19:48:50 UTC |
| b460003 | Sadanand M | 15 June 2011, 07:18:01 UTC | Xilinx: ARM: SPI PS driver: dont pull address from fdt _remove function pulls the address from the device-tree which is not required. this patch removes the code which gets the address from fdt and uses platfrom data. Signed-off-by: Sadanand M <sadanan@xilinx.com> | 15 June 2011, 19:48:38 UTC |
| 885dd15 | John Linn | 02 June 2011, 21:20:04 UTC | Xilinx: ARM: Adding DMA zone to avoid DMA into lower 512K The lower 512K of memory can not be DMAed into. A new memory zone is added with a hole in the lower memory to avoid this memory. We were previously able to see an issue with USB when copying a large file to the USB memory stick and some debug showed that some buffers were coming from lower memory. When the kernel releases init memory which is low this is possible. | 15 June 2011, 19:37:34 UTC |
| ba3514a | John Linn | 13 June 2011, 16:49:53 UTC | Revert "Xilinx: ARM: QSPI driver: Hack to transmit PP command & addr together" This reverts commit 39fe534878f220c3a66687aa308d2e4586071b47. Resolved Conflicts: drivers/spi/xilinx_qspipss.c | 13 June 2011, 16:49:53 UTC |
| 7c77452 | vidhum | 02 June 2011, 17:15:49 UTC | Xilinx: ARM: Device Config Driver: Updated with device tree changes The driver is updated to support device tree. The dts file is also updated with the device configuraton module. Signed-off-by: Vidhumouli H <vidhum@xilinx.com> | 02 June 2011, 17:31:03 UTC |
| 9ad84eb | John Linn | 01 June 2011, 13:46:15 UTC | Xilinx: ARM: update defconfig for new SPI driver The SPI driver changed from PSS to PS so that the defconfig needs to be updated. | 01 June 2011, 13:46:15 UTC |
| d05c92c | John Linn | 01 June 2011, 13:42:24 UTC | Xilinx: ARM: update test defconfigs to new SPI driver name The SPI driver was changed from PSS to PS such that the defconfigs need to be changed also. | 01 June 2011, 13:42:24 UTC |
| 2978dfa | John Linn | 31 May 2011, 22:30:24 UTC | Xilinx: ARM: updated test defconfigs to use NOR driver device tree This is needed to support NOR with device tree. | 31 May 2011, 22:30:24 UTC |
| 3ddcbfc | John Linn | 31 May 2011, 22:18:32 UTC | Xilinx: ARM: Added NOR flash to device tree boot NOR flash can now be used with the device tree. This also creates a new partition in flash after the Linux partition to hold the device tree. | 31 May 2011, 22:18:32 UTC |
| 870c412 | John Linn | 31 May 2011, 20:18:40 UTC | Xilinx: ARM: SPI: fixed non-device tree operation After the device tree changes, the platform bus was not working. This also has some general cleanup of the driver name to be more consistent and to make the device tree file easier to edit (tabs). This also removes device tree specific code for the base address of the device and the interrupt as it's handled automatically now. | 31 May 2011, 20:44:27 UTC |
| 9ab13cc | Sadanand M | 31 May 2011, 15:39:09 UTC | Xilinx: ARM: BSP: Dont register SPI as platform devices when device-tree is used Updated arch/arm/mach-xilinx/devices.c to not register SPI devices as platform devices when device tree is used. SPI PS driver has been updated to support device-tree Signed-off-by: Sadanand M <sadanan@xilinx.com> | 31 May 2011, 16:14:39 UTC |
| 5583850 | Sadanand M | 31 May 2011, 11:41:23 UTC | Xilinx: ARM: xilinx.dts: Added entries for SPI devices Updated the device tree to add nodes for the PS SPI devices. Signed-off-by: Sadanand M <sadanan@xilinx.com> | 31 May 2011, 16:14:28 UTC |
| 3ddb433 | Sadanand M | 31 May 2011, 11:34:11 UTC | Xilinx: ARM: PS SPI driver: device-tree support Updated the driver for device-tree and renamed the file from *pss.c to *ps.c. Also renamed the constants, variables, functions, etc. tested the changes with on-board EEPROM on PEEP. Signed-off-by: Sadanand M <sadanan@xilinx.com> | 31 May 2011, 16:13:00 UTC |
| a4b557e | John Linn | 31 May 2011, 14:36:53 UTC | Xilinx: ARM: change baud to 115K in mach-xilinx device tree The rest of the system changed from 9600 to 115K and this was forgot. Debug was also removed from the command line in the device tree. | 31 May 2011, 14:36:53 UTC |
| a93ff16 | John Linn | 30 May 2011, 04:54:11 UTC | Xilinx: ARM: BSP: fixed USB driver initialization hosed in AMP changes The conditional compilation was changed wrong in the previous AMP changes such that USB was not getting initialized when it should have been. | 30 May 2011, 04:55:42 UTC |
| cbf25dd | John Linn | 29 April 2011, 22:54:48 UTC | Xilinx: ARM: Adding system reset ability This change allows reboot -f in busybox to reset the system. I also changed SLCR to be part of the BSP all the time rather than configurable and based on SYSFS. | 27 May 2011, 20:05:43 UTC |
| 48b0a72 | John Linn | 27 May 2011, 16:55:13 UTC | Xilinx: ARM: UART: fixed UART0 lockup issue during resets The new UART driver that is going into 2.6.40 in the mainline seemed to add a new bug with the disable of irqs in the console write function. There is some interaction between the console polled code and the non-console interrupt code. Since the startup and shutdown functions were messing with interrupts also but not using the port lock the driver could get into a state such that interrupts were disabled for UART0 and UART1 was still working because the console polled code was only running on UART0. DT 612129 documented the issues seen. It seemed to get much worse on 2.6.39 but could be replicated with 2.6.38 and the new driver with enough time. 9600 baud also seemed to find the problem much sooner than 115k. This will need need to be patched in the mainline also. | 27 May 2011, 17:00:21 UTC |
| 817da81 | John Linn | 25 May 2011, 17:12:54 UTC | Xilinx: ARM: AMP mode operation GIC driver cleanup This change removes a lot of hard coded interrupt mangling that was added for AMP mode of operation. A new function was added to allow the BSP to move interrupts from CPU0 to CPU1 for the AMP mode. The BSP code should call the function for each interrupt that is to be moved to CPU1. | 25 May 2011, 20:20:29 UTC |
| 62151f6 | John Linn | 25 May 2011, 14:37:59 UTC | Xilinx: ARM: Minor updates for AMP prototype Updated the platform devices for CPU0 to match the non-AMP design. Also added conditional around the USB platfrom code so that it only runs on CPU0 for AMP. USB interrupts are not still not working and that will be addressed seperately. | 25 May 2011, 14:37:59 UTC |
| e1d292e | John Linn | 24 May 2011, 23:32:49 UTC | Xilinx: ARM: update of AMP defconfigs for 2.6.39 There have been a number of changes including the new UART driver and baud rate. | 25 May 2011, 14:28:40 UTC |
| 40dacc7 | John Linn | 24 May 2011, 21:42:18 UTC | Xilinx: ARM: Adding set_cpu function for AMP prototyping The current AMP prototype has hard coded interrupts moved over to the 2nd CPU. This function should allow the BSP to move the interrupts over rather than hard coding it in the gic driver. | 24 May 2011, 21:42:18 UTC |
| b960b76 | John Linn | 20 May 2011, 17:21:05 UTC | Xilinx: ARM: removed convert_vmlinux.sh calls from Makefile This image, vmlinux.bin, was a hack that we did a long time ago and needs to be removed. Users should be using the Image and zImage files in arch/arm/boot directory. | 20 May 2011, 22:36:47 UTC |
| d9b7317 | John Linn | 20 May 2011, 20:01:06 UTC | ARM: Xilinx: update of infrastructure for SMP Only minor updates of Kconfig and Makefile to support SMP. Signed-off-by: John Linn <john.linn@xilinx.com> | 20 May 2011, 22:25:24 UTC |
| 821b0c4 | John Linn | 20 May 2011, 19:51:41 UTC | ARM: Xilinx: add SMP specific support files These files are the core processing for SMP which are similar to most other platforms SMP. The biggest difference is the way the 2nd CPU is started. Since the versatile platform code is reused to minimize code duplication, it must also have weak functions for the secondary CPU functions. Signed-off-by: John Linn <john.linn@xilinx.com> | 20 May 2011, 22:17:14 UTC |
| e10ef20 | John Linn | 20 May 2011, 22:12:23 UTC | ARM: Xilinx: Adding local timer to support SMP Each Cortex A9 CPU has a local timer in it and it can be used during SMP mode. This is using the local timer code from the versatile platform. Signed-off-by: John Linn <john.linn@xilinx.com> | 20 May 2011, 22:12:23 UTC |
| 7f9a8d8 | John Linn | 20 May 2011, 17:07:57 UTC | ARM: Xilinx: Add on chip memory for SMP support On chip memory (OCM) is always present and is used for communication between the CPUs when in SMP mode. Signed-off-by: John Linn <john.linn@xilinx.com> | 20 May 2011, 17:07:57 UTC |
| 956486d | John Linn | 20 May 2011, 16:46:53 UTC | Xilinx: ARM: updated ep107 device tree for 115K and 2nd UART It's much easier to use our ramdisk if the device tree has both UARTs in it. We also changed to 115K so the device tree should use it. | 20 May 2011, 16:46:53 UTC |
| b0ff2b0 | John Linn | 19 May 2011, 22:50:28 UTC | Xilinx: ARM: updated early printk uart to 115K baud This code doesn't really run when u-boot is used, but could be used when loading the kernel using a jtag pod on the real board (faster). | 19 May 2011, 22:50:28 UTC |
| 53cc170 | John Linn | 19 May 2011, 21:08:07 UTC | Xilinx: ARM: update test defconfigs for 115200 baud rate 115200 is now the normal baud so test with it also. | 19 May 2011, 21:08:07 UTC |
| f6a0e75 | John Linn | 19 May 2011, 20:59:53 UTC | Xilinx: ARM: update defconfig to use 115200 baud rate Not much more to say, 9600 is old school. | 19 May 2011, 20:59:53 UTC |
| d5f50fc | John Linn | 19 May 2011, 20:15:02 UTC | Xilinx: ARM: Removed old UART driver as new driver is being used Since we're now using the new UART driver that will hopefully be in 2.6.40, remove the old one. | 19 May 2011, 20:15:02 UTC |
| 875deae | John Linn | 19 May 2011, 20:11:38 UTC | Xilinx: ARM: update test defconfigs for new UART driver The driver now uses ps rather than pss and required devtmpfs for the device nodes. | 19 May 2011, 20:11:38 UTC |
| e3857c8 | John Linn | 19 May 2011, 19:36:06 UTC | Xilinx: ARM: Update xilinx device tree for new UART The UART is now ps rather than pss in the name and it doesn't use port-number any more. | 19 May 2011, 19:36:06 UTC |
| 6f260b3 | John Linn | 19 May 2011, 18:11:33 UTC | Xilinx: ARM: update of defconfig for new UART driver The new UART driver that should be going into 2.6.40 uses PS rather than DF for the tty name. Devtmpfs is also needed for the new driver. The SMP defconfig was also removed as the normal xilinx_defconfig is SMP by default now. | 19 May 2011, 18:11:33 UTC |
| 7a27d1b | John Linn | 19 May 2011, 17:58:25 UTC | Merge branch 'master' of /home/linnj/linux_arm/devel/linux-2.6 into master-2.6-39 | 19 May 2011, 17:58:25 UTC |
| 4d99281 | John Linn | 19 May 2011, 17:39:41 UTC | Xilinx: ARM: GPIO: updated driver to build with 2.6.39 Due to interrupt changes in the kernel, there were a number of changes to this driver. | 19 May 2011, 17:56:01 UTC |
| 83e16d1 | John Linn | 19 May 2011, 17:54:19 UTC | Xilinx: ARM: Updated platform data to use new UART driver The new driver uses ps rather than pss so the platform data must reflect that minor change. Note that devtmpfs must also be used now with the new driver as it's using dynamic device nodes as required by the open source community. | 19 May 2011, 17:54:19 UTC |
| 61c4f2c | Linus Torvalds | 19 May 2011, 04:06:34 UTC | Linux 2.6.39 | 19 May 2011, 04:06:34 UTC |
| 3f80fbf | Linus Torvalds | 18 May 2011, 23:50:28 UTC | Merge branch 'fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jlbec/ocfs2 * 'fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jlbec/ocfs2: configfs: Fix race between configfs_readdir() and configfs_d_iput() configfs: Don't try to d_delete() negative dentries. ocfs2/dlm: Target node death during resource migration leads to thread spin ocfs2: Skip mount recovery for hard-ro mounts ocfs2/cluster: Heartbeat mismatch message improved ocfs2/cluster: Increase the live threshold for global heartbeat ocfs2/dlm: Use negotiated o2dlm protocol version ocfs2: skip existing hole when removing the last extent_rec in punching-hole codes. ocfs2: Initialize data_ac (might be used uninitialized) | 18 May 2011, 23:50:28 UTC |
| 856b712 | John Linn | 15 April 2011, 23:14:15 UTC | tty/serial: add support for Xilinx PS UART The Xilinx PS Uart is used on the new ARM based SoC. This UART is not compatible with others such that a seperate driver is required. Signed-off-by: John Linn <john.linn@xilinx.com> | 18 May 2011, 21:23:27 UTC |
| b5345f6 | John Linn | 18 May 2011, 21:22:15 UTC | Xilinx: ARM: removed out of tree UART change This should allow the UART driver patch that has been accepted into the mainline to apply. | 18 May 2011, 21:22:15 UTC |
| 3c9c8d4 | John Linn | 18 May 2011, 20:03:13 UTC | Merge branch 'grant-mainline-v6' of /home/linnj/linux_arm/devel/linux-2.6 into master-2.6.39-rc7 Resolved Conflicts: Documentation/devicetree/booting-without-of.txt arch/arm/Kconfig arch/arm/Makefile arch/arm/include/asm/setup.h arch/arm/kernel/devtree.c arch/arm/kernel/head.S arch/arm/kernel/setup.c arch/arm/mach-tegra/board-harmony.c arch/arm/mm/Kconfig arch/arm/mm/init.c drivers/dma/Kconfig drivers/usb/Kconfig drivers/usb/core/hub.c drivers/usb/gadget/Kconfig drivers/usb/gadget/Makefile | 18 May 2011, 21:00:15 UTC |
| fce5195 | Linus Torvalds | 18 May 2011, 20:25:57 UTC | Merge branch 'devicetree/merge' of git://git.secretlab.ca/git/linux-2.6 * 'devicetree/merge' of git://git.secretlab.ca/git/linux-2.6: drivercore: revert addition of of_match to struct device of: fix race when matching drivers | 18 May 2011, 20:25:57 UTC |
| 7103dbe | Linus Torvalds | 18 May 2011, 20:21:43 UTC | Merge branch 'upstream' of git://git.linux-mips.org/pub/scm/upstream-linus * 'upstream' of git://git.linux-mips.org/pub/scm/upstream-linus: MIPS: Kludge IP27 build for 2.6.39. MIPS: AR7: Fix GPIO register size for Titan variant. MIPS: Fix duplicate invocation of notify_die. MIPS: RB532: Fix iomap resource size miscalculation. | 18 May 2011, 20:21:43 UTC |
| b1608d6 | Grant Likely | 18 May 2011, 17:19:24 UTC | drivercore: revert addition of of_match to struct device Commit b826291c, "drivercore/dt: add a match table pointer to struct device" added an of_match pointer to struct device to cache the of_match_table entry discovered at driver match time. This was unsafe because matching is not an atomic operation with probing a driver. If two or more drivers are attempted to be matched to a driver at the same time, then the cached matching entry pointer could get overwritten. This patch reverts the of_match cache pointer and reworks all users to call of_match_device() directly instead. Signed-off-by: Grant Likely <grant.likely@secretlab.ca> | 18 May 2011, 18:32:23 UTC |
| 01294d8 | Milton Miller | 18 May 2011, 15:27:39 UTC | of: fix race when matching drivers If two drivers are probing devices at the same time, both will write their match table result to the dev->of_match cache at the same time. Only write the result if the device matches. In a thread titled "SBus devices sometimes detected, sometimes not", Meelis reported his SBus hme was not detected about 50% of the time. From the debug suggested by Grant it was obvious another driver matched some devices between the call to match the hme and the hme discovery failling. Reported-by: Meelis Roos <mroos@linux.ee> Signed-off-by: Milton Miller <miltonm@bga.com> [grant.likely: modified to only call of_match_device() once] Signed-off-by: Grant Likely <grant.likely@secretlab.ca> | 18 May 2011, 16:19:36 UTC |
| a2b9c1f | Linus Torvalds | 18 May 2011, 13:49:02 UTC | Merge branch 'for-linus' of git://git.kernel.dk/linux-2.6-block * 'for-linus' of git://git.kernel.dk/linux-2.6-block: block: don't delay blk_run_queue_async scsi: remove performance regression due to async queue run blk-throttle: Use task_subsys_state() to determine a task's blkio_cgroup block: rescan partitions on invalidated devices on -ENOMEDIA too cdrom: always check_disk_change() on open block: unexport DISK_EVENT_MEDIA_CHANGE for legacy/fringe drivers | 18 May 2011, 13:49:02 UTC |
| a5602a3 | Ralf Baechle | 18 May 2011, 12:14:36 UTC | MIPS: Kludge IP27 build for 2.6.39. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> | 18 May 2011, 13:18:27 UTC |
| 3e9957b | Florian Fainelli | 13 May 2011, 15:41:21 UTC | MIPS: AR7: Fix GPIO register size for Titan variant. The 'size' variable contains the correct register size for both AR7 and Titan, but we never used it to ioremap the correct register size. This problem only shows up on Titan. [ralf@linux-mips.org: Fixed the fix. The original patch as in patchwork recognizes the problem correctly then fails to fix it ...] Reported-by: Alexander Clouter <alex@digriz.org.uk> Signed-off-by: Florian Fainelli <florian@openwrt.org> Patchwork: https://patchwork.linux-mips.org/patch/2380/ Signed-off-by: Ralf Baechle <ralf@linux-mips.org> | 18 May 2011, 13:18:27 UTC |
| 10423c9 | Ralf Baechle | 13 May 2011, 09:33:28 UTC | MIPS: Fix duplicate invocation of notify_die. Initial patch by Yury Polyanskiy <ypolyans@princeton.edu>. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> Patchwork: https://patchwork.linux-mips.org/patch/2373/ | 18 May 2011, 13:18:26 UTC |
| 3436830 | Ralf Baechle | 12 May 2011, 12:55:48 UTC | MIPS: RB532: Fix iomap resource size miscalculation. This is the MIPS portion of Joe Perches <joe@perches.com>'s https://patchwork.linux-mips.org/patch/2172/ which seems to have been lost in time and space. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> | 18 May 2011, 13:18:26 UTC |
| 24307aa | Joel Becker | 18 May 2011, 11:08:16 UTC | configfs: Fix race between configfs_readdir() and configfs_d_iput() configfs_readdir() will use the existing inode numbers of inodes in the dcache, but it makes them up for attribute files that aren't currently instantiated. There is a race where a closing attribute file can be tearing down at the same time as configfs_readdir() is trying to get its inode number. We want to get the inode number of open attribute files, because they should match while instantiated. We can't lock down the transition where dentry->d_inode is set to NULL, so we just check for NULL there. We can, however, ensure that an inode we find isn't iput() in configfs_d_iput() until after we've accessed it. Signed-off-by: Joel Becker <jlbec@evilplan.org> | 18 May 2011, 11:08:16 UTC |
| df7f996 | Joel Becker | 22 February 2011, 09:09:49 UTC | configfs: Don't try to d_delete() negative dentries. When configfs is faking mkdir() on its subsystem or default group objects, it starts by adding a negative dentry. It then tries to instantiate the group. If that should fail, it must clean up after itself. I was using d_delete() here, but configfs_attach_group() promises to return an empty dentry on error. d_delete() explodes with the entry dentry. Let's try d_drop() instead. The unhashing is what we want for our dentry. Signed-off-by: Joel Becker <jlbec@evilplan.org> | 18 May 2011, 10:30:58 UTC |
| 3ec717b | Shaohua Li | 18 May 2011, 09:22:43 UTC | block: don't delay blk_run_queue_async Let's check a scenario: 1. blk_delay_queue(q, SCSI_QUEUE_DELAY); 2. blk_run_queue_async(); the second one will became a noop, because q->delay_work already has WORK_STRUCT_PENDING_BIT set, so the delayed work will still run after SCSI_QUEUE_DELAY. But blk_run_queue_async actually hopes the delayed work runs immediately. Fix this by doing a cancel on potentially pending delayed work before queuing an immediate run of the workqueue. Signed-off-by: Shaohua Li <shaohua.li@intel.com> Signed-off-by: Jens Axboe <jaxboe@fusionio.com> | 18 May 2011, 10:24:03 UTC |
