7 years agoharness: NFS share path for webserver test changed release2013-08-16
Stefan Kaestle [Thu, 15 Aug 2013 14:56:08 +0000]
harness: NFS share path for webserver test changed

7 years agoarm: Added missing capability for user-level drivers on armv5 and arm_gem5.
Stefan Kaestle [Tue, 13 Aug 2013 16:03:35 +0000]
arm: Added missing capability for user-level drivers on armv5 and arm_gem5.

These are currently dummy capabilities, as the only ARM platform for
which we provide user-level driver support is the Pandaboard.

7 years agoAdded missing cap_copy in arm elf_allocate().
Simon Gerber [Tue, 13 Aug 2013 14:11:06 +0000]
Added missing cap_copy in arm elf_allocate().

7 years agoMerge
Stefan Kaestle [Thu, 8 Aug 2013 11:23:02 +0000]

7 years agoarm_gem5: Need irq.c for user-level interrupt implementation
Stefan Kaestle [Thu, 8 Aug 2013 11:12:19 +0000]
arm_gem5: Need irq.c for user-level interrupt implementation

7 years agouser-level interrupts only supported on ARMv7
Stefan Kaestle [Wed, 7 Aug 2013 15:29:04 +0000]
user-level interrupts only supported on ARMv7

7 years agoAdded missing check for already mapped capabilities in omap44xx page-table manipulati...
Simon Gerber [Wed, 7 Aug 2013 13:27:19 +0000]
Added missing check for already mapped capabilities in omap44xx page-table manipulation code.

7 years agoFixed typo in SCC defines.
Simon Gerber [Tue, 6 Aug 2013 13:33:03 +0000]
Fixed typo in SCC defines.

7 years agoscalebench: Use Python 2.6 dictionary comprehension syntax (since this is what harnes...
Stefan Kaestle [Mon, 22 Jul 2013 13:39:35 +0000]
scalebench: Use Python 2.6 dictionary comprehension syntax (since this is what harness uses)

7 years agoFixed some errors in default_start_function passsing pci id.
Simon Gerber [Thu, 18 Jul 2013 12:29:06 +0000]
Fixed some errors in default_start_function passsing pci id.

7 years agoAdded passing of pci vendor and device id to drivers started by kaluga. Improved...
Simon Gerber [Fri, 5 Jul 2013 09:24:37 +0000]
Added passing of pci vendor and device id to drivers started by kaluga. Improved ahcid to use that information.

7 years agoAdded __x86__ preprocessor define on x86-based architectures.
Simon Gerber [Wed, 17 Jul 2013 13:07:27 +0000]
Added __x86__ preprocessor define on x86-based architectures.

7 years agoRemove some debug code.
Gerd Zellweger [Tue, 16 Jul 2013 11:24:13 +0000]
Remove some debug code.

7 years agoRename start_omap to default_start_function, changes driver start behaviour on pandab...
Gerd Zellweger [Tue, 16 Jul 2013 11:22:53 +0000]
Rename start_omap to default_start_function, changes driver start behaviour on pandaboard without manually overriding it.

7 years agoAdded debug prints instead of printf's.
Gerd Zellweger [Tue, 16 Jul 2013 10:07:19 +0000]
Added debug prints instead of printf's.

7 years agoProperly checking if slot exists during CNode walk. Cleaned up some debug printing.
Gerd Zellweger [Tue, 16 Jul 2013 10:06:31 +0000]
Properly checking if slot exists during CNode walk. Cleaned up some debug printing.

7 years agoUsing macros for offsets in Kaluga, added PRCM driver binary.
Gerd Zellweger [Tue, 16 Jul 2013 09:28:15 +0000]
Using macros for offsets in Kaluga, added PRCM driver binary.

7 years agoA set-up of the voltage that seems to work (i.e., the supply high bit is 1 and no...
Gerd Zellweger [Thu, 11 Jul 2013 08:24:40 +0000]
A set-up of the voltage that seems to work (i.e., the supply high bit is 1 and no vmode error at the end).

However, SDCard still doesn't work and we just ignore that we should wait for a PBIAS interrupt that never arrives (except
if you plugin/unplug a SDCard).

7 years agoAdded functions to power-on and off the vmmc power supply. Added them into the initia...
Gerd Zellweger [Wed, 10 Jul 2013 09:38:55 +0000]
Added functions to power-on and off the vmmc power supply. Added them into the initialization code.

This revision has supply_hi_out asserted in Step 7.

7 years agoAdded I2C_DEBUG macro to avoid excessive console prints.
Gerd Zellweger [Wed, 10 Jul 2013 09:38:12 +0000]
Added I2C_DEBUG macro to avoid excessive console prints.

7 years agoUse of sysctrl_padconf_core mackerel file instead of ctrlmod in ctrlmod.c.
Gerd Zellweger [Wed, 10 Jul 2013 09:35:08 +0000]
Use of sysctrl_padconf_core mackerel file instead of ctrlmod in ctrlmod.c.

7 years agoRemoved too verbose debugging during omap driver startup.
Gerd Zellweger [Wed, 10 Jul 2013 09:28:53 +0000]
Removed too verbose debugging during omap driver startup.

7 years agoCleaned up omap padconf core mackerel file and added it to hake.
Gerd Zellweger [Wed, 10 Jul 2013 09:27:34 +0000]
Cleaned up omap padconf core mackerel file and added it to hake.

7 years agoFixed indention in devices Hakefile.
Gerd Zellweger [Wed, 10 Jul 2013 09:26:19 +0000]
Fixed indention in devices Hakefile.

7 years agoFixed bug in OMAP ctrlmod mackerel file.
Gerd Zellweger [Wed, 10 Jul 2013 09:25:15 +0000]
Fixed bug in OMAP ctrlmod mackerel file.

Reserved bits added at the wrong end.

7 years agoChanged code for mmchs to use generated padconf core mackerel file.
Gerd Zellweger [Wed, 10 Jul 2013 07:24:48 +0000]
Changed code for mmchs to use generated padconf core mackerel file.

7 years agoCleanup headers and remove warnings.
Gerd Zellweger [Tue, 9 Jul 2013 16:16:32 +0000]
Cleanup headers and remove warnings.

7 years agoAdded user-level interrupts for mmchs driver.
Gerd Zellweger [Tue, 9 Jul 2013 15:49:03 +0000]
Added user-level interrupts for mmchs driver.

7 years agoDisable debugging code.
Gerd Zellweger [Tue, 9 Jul 2013 15:48:28 +0000]
Disable debugging code.

7 years agofdif: Command line argument for switching between poll-based and interrupt-based...
Stefan Kaestle [Tue, 9 Jul 2013 14:28:29 +0000]
fdif: Command line argument for switching between poll-based and interrupt-based mode. General cleanup

7 years agoTypo in comment
Stefan Kaestle [Tue, 9 Jul 2013 13:48:46 +0000]
Typo in comment

7 years agofdif: added support for interrupts
Stefan Kaestle [Tue, 9 Jul 2013 13:45:53 +0000]
fdif: added support for interrupts

7 years agointerrupts: Copied missing pieces from Reto Achermann's tree for user-level ARM inter...
Stefan Kaestle [Tue, 9 Jul 2013 13:44:54 +0000]
interrupts: Copied missing pieces from Reto Achermann's tree for user-level ARM interrupts

7 years agoAdded cap permissions for mmchs driver into kaluga.
Gerd Zellweger [Tue, 9 Jul 2013 13:36:31 +0000]
Added cap permissions for mmchs driver into kaluga.

7 years agoMake sure mmchs driver compiles again in user space & changed map_device_register...
Gerd Zellweger [Tue, 9 Jul 2013 13:34:35 +0000]
Make sure mmchs driver compiles again in user space & changed map_device_register API call slightly.

Interrupt acknowledge and enable methods are currently commented out and need to be reenabled once we have working interrupts.

7 years agofdif: Code restructuring to allow switching between poll-based and interrupt-based...
Stefan Kaestle [Tue, 9 Jul 2013 11:41:03 +0000]
fdif: Code restructuring to allow switching between poll-based and interrupt-based face detection

7 years agoarmv7: Cleanup in init
Stefan Kaestle [Tue, 9 Jul 2013 11:44:46 +0000]
armv7: Cleanup in init

7 years agoUse Lazy THC on ARMv7
Stefan Kaestle [Tue, 9 Jul 2013 09:43:45 +0000]
Use Lazy THC on ARMv7

7 years agoMake sure we compile again on x86.
Gerd Zellweger [Tue, 9 Jul 2013 09:04:55 +0000]
Make sure we compile again on x86.

Added proper architecture for arm specific driver to Hakefiles.
Using proper printf specifier in some statements.

7 years agoAdded missing file.
Gerd Zellweger [Tue, 9 Jul 2013 06:56:40 +0000]
Added missing file.

7 years agoA more generic way of passing caps to user-space drivers.
Gerd Zellweger [Mon, 8 Jul 2013 09:17:02 +0000]
A more generic way of passing caps to user-space drivers.

Added driverkit client library. It should include functions that make
a driver developers life easier.

7 years agoFDIF functioning as a user level driver.
Gerd Zellweger [Thu, 4 Jul 2013 09:53:57 +0000]
FDIF functioning as a user level driver.

7 years agoMemory management functionality in driver.
Gerd Zellweger [Wed, 3 Jul 2013 06:26:24 +0000]
Memory management functionality in driver.

7 years agoTowards user-level sdcard driver. Still need to add interrupts to I2C driver and...
Gerd Zellweger [Tue, 2 Jul 2013 13:57:09 +0000]
Towards user-level sdcard driver. Still need to add interrupts to I2C driver and enable register mapping in user-level.

7 years agoGetting PBIAS Interrupt consistently with the following steps:
Gerd Zellweger [Tue, 2 Jul 2013 07:24:49 +0000]
Getting PBIAS Interrupt consistently with the following steps:
a) Unplug Power cord and SDCARD
b) Plug in power
c) Plug in SD Card
d) Boot
e) Unplug SD Card

7 years agoAdded debouncing register to the TWL 6030 mackerel file.
Gerd Zellweger [Tue, 2 Jul 2013 06:12:30 +0000]
Added debouncing register to the TWL 6030 mackerel file.

7 years agoCleaning up debug print function.
Gerd Zellweger [Fri, 28 Jun 2013 12:06:43 +0000]
Cleaning up debug print function.

7 years agoFixed an issue in i2c API that does not ACK register read ready event.
Gerd Zellweger [Fri, 28 Jun 2013 12:05:37 +0000]
Fixed an issue in i2c API that does not ACK register read ready event.

Also added some debug prints.

7 years agoChanges to twl6030 Mackerel file to allow send register manipulations over i2c bus...
Simon Gerber [Thu, 27 Jun 2013 09:17:55 +0000]
Changes to twl6030 Mackerel file to allow send register manipulations over i2c bus. TWL6030 driver to allow for voltage scaling.

7 years agoMMCHS driver cleanup
Gerd Zellweger [Tue, 25 Jun 2013 18:11:49 +0000]
MMCHS driver cleanup

Using generated mackerel files in CM2.

7 years agoSome cleanup in gic_enable_interrupt.
Gerd Zellweger [Tue, 25 Jun 2013 13:23:59 +0000]
Some cleanup in gic_enable_interrupt.

7 years agoMMCHS and I2C code imported into main tree right now as a kernel-space library.
Simon Gerber [Mon, 8 Jul 2013 09:30:04 +0000]
MMCHS and I2C code imported into main tree right now as a kernel-space library.

Needs to be changed to a user space driver.

7 years agoAdded implementation for oct_del() in skb_simple on ARM.
Gerd Zellweger [Mon, 24 Jun 2013 07:47:34 +0000]
Added implementation for oct_del() in skb_simple on ARM.

USB implementation needs to delete records in case devices are detached.

7 years agoMake GCC binary for pandaboard_image depend on ARM_GCC variable in Makefile.
Gerd Zellweger [Tue, 25 Jun 2013 07:51:35 +0000]
Make GCC binary for pandaboard_image depend on ARM_GCC variable in Makefile.

7 years ago(library) driver for face detection that can run inside the kernel right now.
Gerd Zellweger [Mon, 10 Jun 2013 11:51:45 +0000]
(library) driver for face detection that can run inside the kernel right now.

7 years agoFixed binary selection in
Gerd Zellweger [Mon, 10 Jun 2013 07:18:49 +0000]
Fixed binary selection in

Merge this with build fixes commit.

7 years agoComplete set of Mackerel Files for the OMAP4460 platform added.
Gerd Zellweger [Wed, 5 Jun 2013 13:13:52 +0000]
Complete set of Mackerel Files for the OMAP4460 platform added.

These files have been auto-generated based on XML files found in
the TI RDT v1.0.0.4p tool (download at
The files will most likely compile. However, because their XML format
is less expressive than Mackerel some renaming might be helpful before use.
Especially to save some typing (i.e., remove duplicated namings, constant names etc.).
Also, some constants may be duplicated if their descriptions was
slightly different.

7 years agoBuild fixes for ARMv7.
Gerd Zellweger [Thu, 6 Jun 2013 15:12:44 +0000]
Build fixes for ARMv7.

Make choice about GCC binary. Fixed typos in usbconf mackerel file.
Remove MMU from kernel Hakefile because it is not needed.
Don't do stack protection (leads to compiler errors on newer versions).
Don't add build-id when linking on ARMv7. This screwed up ELF loading in our
kernel when it tried to load a section at 0x8000 -- which is not a valid
virtual address -- for the init binary. For some reason
stripping the section manually did not help.

7 years agoPassing IO Capability to the monitor.
Achermann Reto [Mon, 8 Jul 2013 09:39:26 +0000]
Passing IO Capability to the monitor.

7 years agoAdded arm_irq_handle message to monitor interface.
Achermann Reto [Tue, 2 Jul 2013 11:35:52 +0000]
Added arm_irq_handle message to monitor interface.

7 years agoAllow driver start-up on ARM by relying on Kaluga.
Achermann Reto [Tue, 2 Jul 2013 11:29:32 +0000]
Allow driver start-up on ARM by relying on Kaluga.

7 years agoEnable user-level interrupt forwarding on ARM.
Achermann Reto [Tue, 2 Jul 2013 11:21:03 +0000]
Enable user-level interrupt forwarding on ARM.

7 years agoAdded a section on connecting the tracing framework to Aquarium 2 directly in the...
Simon Gerber [Mon, 1 Jul 2013 11:17:20 +0000]
Added a section on connecting the tracing framework to Aquarium 2 directly in the tracing tech-note.

7 years agoAdded brie machines to scalebench
Stefan Kaestle [Mon, 24 Jun 2013 13:12:03 +0000]
Added brie machines to scalebench

7 years agoMerge
Stefan Kaestle [Mon, 24 Jun 2013 12:06:09 +0000]

7 years agoAdded basic usage information for Tracing framework
Stefan Kaestle [Mon, 24 Jun 2013 11:56:20 +0000]
Added basic usage information for Tracing framework

7 years agoChanged default bfscope port to 6666 to make everyone's lifes easier.
Simon Gerber [Mon, 24 Jun 2013 11:43:29 +0000]
Changed default bfscope port to 6666 to make everyone's lifes easier.

7 years agoadding a tag for release 2013-06-07
Stefan Kaestle [Fri, 7 Jun 2013 05:02:51 +0000]
adding a tag for release 2013-06-07

7 years agomerge release2013-06-07
Kornilios Kourtis [Thu, 6 Jun 2013 06:13:14 +0000]

7 years mkdir first when generating romfs_size.h
Kornilios Kourtis [Thu, 6 Jun 2013 06:12:33 +0000] mkdir first when generating romfs_size.h

There seems to be a race in parallel builds:
echo "size_t romfs_cpio_archive_size = 0; //should not see this" > armv5/include/romfs_size.h
/bin/sh: cannot create armv5/include/romfs_size.h: Directory nonexistent

7 years agoRemoved
Simon Peter [Wed, 5 Jun 2013 02:31:54 +0000]

7 years agoNew debug scripts.
Simon Peter [Wed, 5 Jun 2013 02:31:37 +0000]
New debug scripts.

* Removed a lot of old, unused cruft
* This does not depend on a pre-processed anymore
* Added example
* Uses user-specific TCP port for GDB->QEMU connection

7 years agoe1000: add octopus flounder defs/bindings
Kornilios Kourtis [Tue, 4 Jun 2013 13:02:31 +0000]
e1000: add octopus flounder defs/bindings

7 years agomerge
Kornilios Kourtis [Mon, 3 Jun 2013 14:45:23 +0000]

7 years agoCheck if child processes exist before terminating them
Stefan Kaestle [Mon, 3 Jun 2013 10:19:55 +0000]
Check if child processes exist before terminating them

7 years agoAdded omap44xx_mmu again to fix dependency error when building arm_gem5_image.
Stefan Kaestle [Mon, 3 Jun 2013 10:18:35 +0000]
Added omap44xx_mmu again to fix dependency error when building arm_gem5_image.
Improved code formatting.

7 years agoDummy commit
Stefan Kaestle [Fri, 31 May 2013 13:46:51 +0000]
Dummy commit

7 years agoarmv7: disable vfp instructions for the cpu driver
Wang Nan [Fri, 31 May 2013 12:39:44 +0000]
armv7: disable vfp instructions for the cpu driver

Some compiler generates vfp instructions such as vpush, vldr for normal code in
cpu driver, which introduces complexity (related to fpu state).  I propose a
patch which disables vfp instructions in cpu driver code.

7 years agoarmv7: align exception stack with 8 bytes
Wang Nan [Fri, 31 May 2013 12:11:44 +0000]
armv7: align exception stack with 8 bytes

Here I propose a small patch which adjust arm exception stacks.

You can try 'printf("0x%016llx", 0x12345678ULL)' in undef handler. ARM abi
require stack be 8 bytes aligned, but original code setups 4   bytes gaps
between 2 stacks, makes irq_stack and undef_stack misaligned.


7 years agoAuto append the current year to the results path for a test.
Gerd Zellweger [Thu, 30 May 2013 09:35:03 +0000]
Auto append the current year to the results path for a test.

This should avoid the "OSError: [Errno 31] Too many links" error in future tests.
We ran into this because we hit the ext3 limit in maximum subdirectories (32000)
in our test result directory. For now, lets hope we don't have more than 32000
test runs per year.

7 years agoPEP-8 Autoformatting.
Gerd Zellweger [Thu, 30 May 2013 09:34:39 +0000]
PEP-8 Autoformatting.

7 years agomerge
Kornilios Kourtis [Wed, 29 May 2013 08:25:52 +0000]

7 years agoUsing platform independent size specifier 'z' in printf() for size_t.
Gerd Zellweger [Wed, 29 May 2013 07:48:22 +0000]
Using platform independent size specifier 'z' in printf() for size_t.

Makes sure we can compile for 32bit.

7 years agoDon't write txce register in PHY mode.
Gerd Zellweger [Fri, 24 May 2013 13:38:54 +0000]
Don't write txce register in PHY mode.
Don't write asde register for 8257X variants.

7 years agoUse 82575 auto-negotiation for 82576 cards as well -- even though it already worked...
Gerd Zellweger [Fri, 24 May 2013 08:51:20 +0000]
Use 82575 auto-negotiation for 82576 cards as well -- even though it already worked for these cards with the old code.

7 years agoLink-up routine for 82575 cards.
Gerd Zellweger [Fri, 24 May 2013 08:32:57 +0000]
Link-up routine for 82575 cards.

7 years agoSet bsize to 0 for 82575 cards (i.e., use 2048 bytes receive buffers).
Gerd Zellweger [Fri, 24 May 2013 06:51:35 +0000]
Set bsize to 0 for 82575 cards (i.e., use 2048 bytes receive buffers).

This avoids splitted packets in the receive handler.

7 years agoWrite Interrupt interval in extended register for 82575 cards.
Gerd Zellweger [Wed, 22 May 2013 14:44:21 +0000]
Write Interrupt interval in extended register for 82575 cards.

7 years agobsex bit does no longer exist in 82575 cards.
Gerd Zellweger [Fri, 24 May 2013 06:49:21 +0000]
bsex bit does no longer exist in 82575 cards.

7 years agoUse different rxdctl register for 82575 cards.
Gerd Zellweger [Fri, 24 May 2013 06:44:29 +0000]
Use different rxdctl register for 82575 cards.

7 years agoSet e1000_glci as default case for all cards that are not special cases.
Gerd Zellweger [Fri, 24 May 2013 07:17:32 +0000]
Set e1000_glci as default case for all cards that are not special cases.

7 years agoDon't write int_tca register for 82575 variants.
Gerd Zellweger [Fri, 24 May 2013 06:39:06 +0000]
Don't write int_tca register for 82575 variants.

7 years agoSet IPG Register for 82575 card variants.
Gerd Zellweger [Fri, 24 May 2013 06:37:25 +0000]
Set IPG Register for 82575 card variants.

7 years agoAdded another version for the txdctl register for the 82575 cards.
Gerd Zellweger [Thu, 23 May 2013 16:09:45 +0000]
Added another version for the txdctl register for the 82575 cards.

This is especially important because the new cards have the enable bit inside this register to enable/disable queues.
The code for the older cards set this bit to 0, therefore disabling the transmit queue.

7 years agoFixed false register name in e1000 mackerel file (was bst, now cold) and adjusted...
Gerd Zellweger [Wed, 22 May 2013 11:03:36 +0000]
Fixed false register name in e1000 mackerel file (was bst, now cold) and adjusted the function call in the C code.

7 years agoFix Flounder messaging (TX_BUSY) for iref allocation inside the monitor.
Simon Peter [Thu, 16 May 2013 06:51:12 +0000]
Fix Flounder messaging (TX_BUSY) for iref allocation inside the monitor.

7 years agoMoved mmap region resizing to general mmap formatting code.
Simon Gerber [Tue, 21 May 2013 14:11:10 +0000]
Moved mmap region resizing to general mmap formatting code.

7 years agoMake IO APIC work on machines where its memory map entry is smaller than a page.
Stefan Kaestle [Thu, 16 May 2013 13:51:50 +0000]
Make IO APIC work on machines where its memory map entry is smaller than a page.

On some machines (brie1) the IOAPIC region is only 1kB. Currently we are not able to map regions that are less than the page size so we increase the size of the region for such cases to one page.

7 years agoAdded licensing and formatted according to Barrelfish coding guidelines.
Gerd Zellweger [Tue, 14 May 2013 15:48:18 +0000]
Added licensing and formatted according to Barrelfish coding guidelines.

7 years agoAdded e1000 support for the follwing e1000 cards:
Gerd Zellweger [Tue, 14 May 2013 14:11:39 +0000]
Added e1000 support for the follwing e1000 cards:
 - 82574L
 - 82575EB
 - 82576EG
The 82575EB, 82576EG are somewhat different and do not transmit
anything at the moment. However since they are in our rack we
might want to work on those at some point.

Fixed some issues encountered on our test machines:
- In case auto-negotiation fails (in our machines with 82572 cards),
  unset bit for auto-negotiation and force the link manually.
- 82574L does not seem to have ee_pres bit set
  However, if we don't read from the EEPROM we cannot
  get a mac address for our card, therefore we force the
  read for this device right now.