- 04 Mar, 2010 40 commits
-
-
Alan Cox authored
Signed-off-by:
Alan Cox <alan@linux.intel.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@suse.de>
-
Alan Cox authored
Signed-off-by:
-
Alan Cox authored
More we don't need Signed-off-by:
-
Alan Cox authored
This completes the structures within the txmac block so we can now propogate a name change and type removal up a layer and clean up TXMAC as well Signed-off-by:
-
Alan Cox authored
This is another one we don't really need to do much to get rid of Signed-off-by:
-
Alan Cox authored
Signed-off-by:
-
Alan Cox authored
Only used for one trivial thing so turn that into something trivial instead Signed-off-by:
-
Alan Cox authored
We have lots of tiny files right now that could be one Signed-off-by:
-
Alan Cox authored
Signed-off-by:
-
Alan Cox authored
Signed-off-by:
-
Alan Cox authored
This completes the typedef clean up of the rx specific structures, although there is plenty do on field names and the like Signed-off-by:
-
Alan Cox authored
Signed-off-by:
-
Alan Cox authored
All the subtypes are sane so just turn it into something struct and linux like Signed-off-by:
-
Alan Cox authored
Use the proper pointer types for the higher level pointers to the rx_status object and kill casts Signed-off-by:
-
Alan Cox authored
So kill off the top level type and turn it into a struct Signed-off-by:
-
Alan Cox authored
Another verbose enum we don't need Signed-off-by:
-
Alan Cox authored
Signed-off-by:
-
Alan Cox authored
All the contents of this type are now clean, so kill the top level type as well Signed-off-by:
-
Alan Cox authored
The contents of MMC_t are clean so kill off the MMC_t typedef Signed-off-by:
-
Alan Cox authored
All the contents are clean so kill off the top level typedefs Signed-off-by:
-
Alan Cox authored
Signed-off-by:
-
Alan Cox authored
This is another unused type for the bin Signed-off-by:
-
Alan Cox authored
We have a clean struct of this now so turn the top level typedefs into a struct Signed-off-by:
-
Alan Cox authored
This is another set of flags as typedef that can be cleaned up. Signed-off-by:
-
Alan Cox authored
Clean up the typdef for fbr_desc itself so we know it is done Signed-off-by:
-
Alan Cox authored
This is only used meaningfully as a definition, we never mask and fetch the bits apart Signed-off-by:
-
Alan Cox authored
This is another 10 bit value with the high bits clear, and where the type doesn't get used anywhere properly anyway Signed-off-by:
-
Dave Airlie authored
Add necessary include to fix build on PowerPC Signed-off-by:
Dave Airlie <airlied@redhat.com> Signed-off-by:
Jarod Wilson <jarod@redhat.com> Signed-off-by:
-
Dan Carpenter authored
I removed a misplace semicolon. It is clear from the indentation that TxPwrTracking87SE() was only supposed to be called if CheckTxPwrTracking() returned true. Signed-off-by:
Dan Carpenter <error27@gmail.com> Signed-off-by:
-
Ben Hutchings authored
When originally introduced into staging, these drivers had custom firmware-loading code which checked a version number and CRC at the end of each blob. This reintroduces those checks, using crc-ccitt instead of custom code. The removed firmware will be added to the linux-firmware.git repository. Based on work by Darren Salt <linux@youmustbejoking.demon.co.uk>. Signed-off-by:
Ben Hutchings <ben@decadent.org.uk> Tested-By:
Darren Salt <linux@youmustbejoking.demon.co.uk> Signed-off-by:
-
Ben Hutchings authored
These functions do not modify the data they are passed. Signed-off-by:
-
Roel Kluin authored
Regardless of the condition, the branches executed the same code Signed-off-by:
Roel Kluin <roel.kluin@gmail.com> Signed-off-by:
-
Roel Kluin authored
Regardless of the condition, the branches execute the same code Signed-off-by:
-
Ian Abbott authored
For AI command testing, the driver does not need to range-check the length of the channel list as the comedi core has already checked it. Signed-off-by:
Ian Abbott <abbotti@mev.co.uk> Signed-off-by:
-
Ian Abbott authored
Check the channel list is valid in step 5 of the AI command test. Split function check_and_setup_channel_list() in two. Also, remove unnecessary chanlist_len tests in step 3 of the AI command test as the comedi core has already checked it. Signed-off-by:
-
Ian Abbott authored
For multi-channel AI commands, the interrupt handler is missing code to switch to the next channel. Add some (untested) code to handle this. Signed-off-by:
-
Ian Abbott authored
For AI command testing, the driver does not need to range-check the length of the channel list as the comedi core has already checked it. Signed-off-by:
-
Ian Abbott authored
For AI commands, the internal channel list used to check incoming data was only set up partially if the channel list contained repeats. Set it up fully. Also the current scan position was not updated properly when moving a block of data from the FIFO unless a whole number of scans was moved. Signed-off-by:
-
Ian Abbott authored
For AI commands, the scan counter should be updated after every scan. It was being updated after every sample except for DMA mode where it was being updated after every repeated segment of the channel list. Also AI commands with multiple channels were being terminated with an error prematurely except in DMA mode. This was because the driver was comparing channel numbers received from the hardware (combined with the sample value) with the expected channel numbers to check for a "channel dropout". This test was failing incorrectly because the driver was not keeping the current position within the (repeated segment of the) channel list up to date. Signed-off-by:
-
Ian Abbott authored
For AI commands, the scan counter should be updated after every scan. In DMA mode, it was being updated after every repeated segment of the channel list. In non-DMA mode, it was being updated after every sample. Signed-off-by:
-
