Commits · fc3d1bdd955f655daff3bd8c6ea468e6bf0a9444 · Projects / FMC TDC 1ns 5cha

07 Sep, 2018 4 commits

tools: keep last TS locally in the function · fc3d1bdd
Federico Vaga authored Sep 07, 2018
```
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>
```
fc3d1bdd
tools: always add debug symbols · 91411cac
Federico Vaga authored Sep 07, 2018
```
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>
```
91411cac

tools: introducing fmc-tdc-perftest · 6f2030b5

Tomasz Wlostowski authored Sep 03, 2018

a crude performance & validity testing tool for the DMA readout
Signed-off-by: Tomasz Włostowski <tomasz.wlostowski@cern.ch>

6f2030b5

drv: align with hardware changes · d5ec5b9f

Tomasz Wlostowski authored Aug 31, 2018

- FIFO readout
- offset is applied in hardware
- delta is calculated in hardware
- add RAW mode readout (no TS post-processing)
Signed-off-by: Tomasz Włostowski <tomasz.wlostowski@cern.ch>
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

d5ec5b9f

04 Sep, 2018 1 commit
- doc: ignore output files · 8c32848e
  Federico Vaga authored Sep 04, 2018
```
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>
```
  8c32848e
30 Aug, 2018 3 commits

drv: improve comments · 0b90ef44
Federico Vaga authored Aug 30, 2018
```
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>
```
0b90ef44
drv: bugfix check correct channels · 3be9df26
Federico Vaga authored Aug 30, 2018
```
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>
```
3be9df26

drv: ZIO linked list and DMA IRQ done · a822a358

Federico Vaga authored Aug 16, 2018

With this patch I try to improve performances by using interrupts
and the ZIO dma API. This allows me to program different blocks
form different channel at the same time
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

a822a358

29 Aug, 2018 4 commits
- drv: use SG by default · 82209b2e
  Federico Vaga authored Aug 29, 2018
```
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>
```
  82209b2e
- Merge branch 'tom-patches' into proposed_master · 0df5d97e
  Federico Vaga authored Aug 29, 2018
```
bugfixes
```
  0df5d97e
- drv: fixed the F***ING DMA issues (I mean they were fascinating). · 261b5ba6
  Tomasz Wlostowski authored Aug 28, 2018
```
- needs new bitstream (28 August onwards)
- fixed scatter list allocation (needs one more entry for the start item)
- fixed DMA direction attribute (should be ORed)
- fixed scatter list page calculation to point to different pages
```
  261b5ba6
- drv: return value is negative error or zero · 8b1c7927
  Tomasz Wlostowski authored Aug 28, 2018
  
  8b1c7927
24 Aug, 2018 2 commits
- drv: bugfix in test_dma_sg · 8451407c
  Federico Vaga authored Aug 24, 2018
```
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>
Signed-off-by: Tomasz Wlostowski <tomasz.wlostowski@cern.ch>
```
  8451407c
- drv: don't do anything if test data is not enabled · 0a416f20
  Federico Vaga authored Aug 24, 2018
```
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>
```
  0a416f20
23 Aug, 2018 6 commits
- drv: fix DMA wait helper · 8f269c14
  Federico Vaga authored Aug 23, 2018
  
  8f269c14
- drv: gennum status register has debug information about errors · 4fd46dd3
  Federico Vaga authored Aug 23, 2018
```
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>
```
  4fd46dd3
- drv: do not compile on warning · dc969cf9
  Federico Vaga authored Aug 23, 2018
```
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>
```
  dc969cf9
- drv: export testing DMA function to user-space · b3f907e8
  Federico Vaga authored Aug 23, 2018
```
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>
```
  b3f907e8
- Merge branch 'devfede-tst' into proposed_master · 954a5ad6
  Federico Vaga authored Aug 23, 2018
```
Mixed improvements
```
  954a5ad6
- drv: bugfix, register trigger type only once · 2d45c267
  Federico Vaga authored Aug 23, 2018
```
The bug is that on a second TDC instance, the driver will try to register
again the trigger type. The trigger type needs to be registered only once
at module_init
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>
```
  2d45c267
22 Aug, 2018 5 commits
- drv: remove print · c21b972a
  Federico Vaga authored Aug 22, 2018
```
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>
```
  c21b972a
- drv: debug print to inspect timestamps · ff36c141
  Federico Vaga authored Aug 22, 2018
```
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>
```
  ff36c141
- drv: reduce DMA transfer to maximum PAGE_SIZE · 3edd1f2b
  Federico Vaga authored Aug 22, 2018
```
Apparently the gn4124 does not handle properly bigger pages. I think I have
seen this before in the ADC and indeed thare we do PAGE_SIZE transfers.
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>
```
  3edd1f2b
- tools: add option to stop on error · b4eb3036
  Federico Vaga authored Aug 22, 2018
```
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>
```
  b4eb3036
- tools: handle seq overflow during test mode 1 · 2fe17105
  Federico Vaga authored Aug 22, 2018
```
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>
```
  2fe17105
21 Aug, 2018 7 commits
- drv: set default coalescing time to 10ms · 68d63acc
  Federico Vaga authored Aug 21, 2018
```
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>
```
  68d63acc
- drv: restore buffer burst size configuration · 398cd3ee
  Federico Vaga authored Aug 21, 2018
```
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>
```
  398cd3ee
- drv: add test mode on channel 0 · bf09403f
  Federico Vaga authored Aug 21, 2018
```
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>
```
  bf09403f
- lib: add comment · 007d1f05
  Federico Vaga authored Aug 21, 2018
```
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>
```
  007d1f05
- tools: set irq coalesing only when user ask for it · 62fe0b3a
  Federico Vaga authored Aug 21, 2018
```
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>
```
  62fe0b3a
- tools: improve printing code · f55e86f9
  Federico Vaga authored Aug 21, 2018
```
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>
```
  f55e86f9
- drv: improve debug message · 45003981
  Federico Vaga authored Aug 21, 2018
```
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>
```
  45003981
20 Aug, 2018 8 commits

tools:tstamp: improve testing mode output · c3f23abb
Federico Vaga authored Aug 20, 2018
```
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>
```
c3f23abb
lib: make the code simpler to read · 2ef88fe4
Federico Vaga authored Aug 20, 2018
```
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>
```
2ef88fe4
lib: fix to_ns convertion · 716b7a61
Federico Vaga authored Aug 17, 2018
```
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>
```
716b7a61
lib: convert all aquired timestamps in a dedicated func · a869ec6b
Federico Vaga authored Aug 17, 2018
```
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>
```
a869ec6b

*: set timeout coalescing dynamically · 013a37f9

Federico Vaga authored Aug 16, 2018

The driver exports a sysfs attrubte to change the coalescing timeout. A new
library funciton can be used to set/get this value; and the test program allows
the configuration before an acquisition.

Keep in mind that, for FIFOs the configuration is global (users are warned on
dmesg) while for DMA it is per-channel. I though it was better to have a common
interface for both. Users are warnd in the documentation as well. In the long
run, perhaps also the FIFO can have its own register.
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

013a37f9

drv: bugfix, do not touch burst size dynamically · 8f34868a

Federico Vaga authored Aug 17, 2018

It is not an IRQ coalescing size, but the size that triggers the dump
from the FIFO to the DDR. so, it is internal stuff that does not need
to get exposed to the user
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

8f34868a

drv: WARN if spurious IRQ · 19d22fd4
Federico Vaga authored Aug 20, 2018
```
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>
```
19d22fd4
drv: continue acquisition on ZIO failure · 4280de0d
Federico Vaga authored Aug 20, 2018
```
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>
```
4280de0d