Commits · 40e311798bf39da993d08e4a5b30a7e3ba6c93a0 · Projects / Simple PCIe FMC carrier SPEC

03 Mar, 2016 1 commit

Revert "kernel: fix memory leakage" · 40e31179

Federico Vaga authored Mar 03, 2016

This reverts commit 0598de40.

This was not a memory leakage because the fmc-bus release that
pointer for us. Is it good?

We allocate and it is someone else us free, it lacks of simmetry.
Probably the register function should be called devm_ like all
the others in the kernel who relase for your the memory. In this
way it is clear the behaviour.

40e31179

10 Feb, 2016 7 commits

kernel: swap irq ack between VIC and GN4124 · d56cd47c

Federico Vaga authored Nov 20, 2015

We ack the GN4124 **before** the VIC. The other way around makes the
interrupt handling timing dependent. For instance. When the VIC
VIC_CTL_EMU_LEN_W is too short it may happens that the VIC asserts
the interrupt **before** the GN4124 is acked; then, so do not receive
interrupts anymore from the GN4124 because it misses the VIC edge.
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>
Signed-off-by: Michel Arruat <michel.arruat@cern.ch>

d56cd47c

kernel: use the vic ack function to ack the VIC · 86b5196f
Federico Vaga authored Nov 19, 2015
```
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>
```
86b5196f
Add support for SPEC 100T version · 182e30e7
Federico Vaga authored Feb 10, 2016

182e30e7
kernel: load a different golden bitstream according to SPEC type · 83d3fa57
Federico Vaga authored Feb 08, 2016
```
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>
```
83d3fa57
lib: detect 100T spec variant · 3afdfcba
Federico Vaga authored Feb 09, 2016
```
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>
```
3afdfcba
kernel: detect 100T spec variant · 92f3a560
Federico Vaga authored Feb 08, 2016
```
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>
```
92f3a560
kernel: fix memory leakage · 0598de40
Federico Vaga authored Jul 31, 2014
```
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>
```
0598de40

01 Dec, 2015 1 commit
- uplift of submodules · 08b08c1b
  Adam Wujek authored Dec 01, 2015
```
Signed-off-by: Adam Wujek <adam.wujek@cern.ch>
```
  08b08c1b
30 Nov, 2015 1 commit
- tools: add printing version of the repo · aab33787
  Adam Wujek authored Nov 30, 2015
```
with -V switch
Signed-off-by: Adam Wujek <adam.wujek@cern.ch>
```
  aab33787
20 Nov, 2015 3 commits
- kernel: update .gitignore · bc0e12b9
  Adam Wujek authored Nov 20, 2015
```
Signed-off-by: Adam Wujek <adam.wujek@cern.ch>
```
  bc0e12b9
- fmc-bus: uplift fmc-bus · 54dbd11d
  Adam Wujek authored Nov 10, 2015
```
Signed-off-by: Adam Wujek <adam.wujek@cern.ch>
```
  54dbd11d
- Merge branch 'adam-coht' · 0cf43726
  Adam Wujek authored Nov 20, 2015
```
Changes in the buildsystem.
Signed-off-by: Adam Wujek <adam.wujek@cern.ch>
```
  0cf43726
13 Nov, 2015 13 commits

kernel: add versions of used submodules to the kernel module · 3a731d20
Adam Wujek authored Oct 27, 2015
```
Signed-off-by: Adam Wujek <adam.wujek@cern.ch>
```
3a731d20
tools: use static version of libspec · 952d547d
Adam Wujek authored Oct 27, 2015
```
Signed-off-by: Adam Wujek <adam.wujek@cern.ch>
```
952d547d

use absolute paths in Makefiles · 9fcf3c70

Adam Wujek authored Oct 27, 2015

Use _ABS in Makefiles wherever possible.
Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

9fcf3c70

initialize fmc-bus submodule if required · f30e2daa
Adam Wujek authored Oct 19, 2015
```
Signed-off-by: Adam Wujek <adam.wujek@cern.ch>
```
f30e2daa

kernel: always use absolute path of fmc-bus · 3bce9901

Adam Wujek authored Oct 19, 2015

FMC_BUS_ABS is passed to Kbuild, which uses different current directory.
It is not trivial to handle FMC_BUS to be absolute or relative path.
Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

3bce9901

kernel: get rid of FMC_DRV from Makefiles · ed6f65c1
Adam Wujek authored Oct 19, 2015
```
Signed-off-by: Adam Wujek <adam.wujek@cern.ch>
```
ed6f65c1
tools: use ar from env variable · d40ca8ed
Adam Wujek authored Oct 16, 2015
```
Signed-off-by: Adam Wujek <adam.wujek@cern.ch>
```
d40ca8ed
kernel: remove move files when cleaning wr_nic · f05235e5
Adam Wujek authored Oct 16, 2015
```
files wr_nic/.*.o.cmd were not removed
Signed-off-by: Adam Wujek <adam.wujek@cern.ch>
```
f05235e5

Makefile: remove loop · 48acbb74

Adam Wujek authored Oct 16, 2015

This allows parallel build.
Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

48acbb74

Makefile: add support for specific build environment · 572a7877

Adam Wujek authored Oct 16, 2015

Include parent_common.mk from up directory. Which may contain various
environment definitions such compiler, linker, etc. If such file is not
present it will not affect build.
Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

572a7877

kernel: Make build working · 9a1d7b79

Adam Wujek authored Oct 16, 2015

--export FMC_DRV
--correct passing GIT_VERSION
Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

9a1d7b79

kernel: split Makefile into Makefile and Kbuild · 040c0cdb
Adam Wujek authored Oct 16, 2015
```
Signed-off-by: Adam Wujek <adam.wujek@cern.ch>
```
040c0cdb

kernel: add missing vmalloc header · 22af1794

Federico Vaga authored Nov 13, 2015

We were using vmalloc() assunign that its library was already included by some
other libraries. Today, this is not the case anymore, so we must include it
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

22af1794

25 Aug, 2015 1 commit
- fmc-bus: do not use kernel headers, use local ones · b3cfb156
  Federico Vaga authored Aug 25, 2015
```
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>
```
  b3cfb156
07 May, 2015 2 commits

do not use anymore check-fmc-bus · d2488fc1

Federico Vaga authored Apr 07, 2015

Mainly for compatibility (simplification) with the CERN super module.
Insted of using a script to guess which fmc-bus define, assume always
that parent project define the FMC_BUS to use.
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

d2488fc1

kernel: allow userspace to program FPGA · 156b2ea2

Federico Vaga authored Apr 07, 2015

For the time being you must write the bitstream in a single write. For
example:

    dd if=/home/fvaga/spec-fmc-adc-v4.0.bin of=/dev/spec-0700 obs=5000000
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

156b2ea2

27 Apr, 2015 1 commit
- submodule: update fmc-bus · 12dc2b34
  Federico Vaga authored Apr 27, 2015
```
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>
```
  12dc2b34
20 Apr, 2015 3 commits

tools: add missing define when distro doesn't provide it · da2cf7b6
Federico Vaga authored Apr 20, 2015
```
Typically old distro releases
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>
```
da2cf7b6
tools: fix print data type · 0289f968
Federico Vaga authored Apr 20, 2015
```
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>
```
0289f968

tools: remove useless condition · 1f032b3b

Federico Vaga authored Apr 20, 2015

a channel cannot be less than 0 due to data type (uint16_t). If someone
write a negative number this become something greater than 4 so the
control still works
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

1f032b3b

24 Feb, 2015 1 commit

wr-nic-gpio: do not check on gpiochip_remove · 2b85fda2

Federico Vaga authored Feb 24, 2015

Since kernel 3.18 the gpiochip_remove function return void and not an
error code as before. In case the gpiochip is removed while there is a
requested gpio, the kernel will print an error message.

By applying this patch without any control on the kernel version we are
going to loose the error message in case of trouble with old kernels
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>

2b85fda2

20 Jan, 2015 1 commit
- kernel: pci_enable_msi_{block|exact} according to version · da456dfc
  Federico Vaga authored Jan 20, 2015
```
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>
```
  da456dfc
18 Sep, 2014 1 commit
- kernel: remove unused variable · aed186b9
  Federico Vaga authored Sep 18, 2014
```
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>
```
  aed186b9
13 Aug, 2014 4 commits
- kernel: always reset FPGA after programming · 969ee716
  Federico Vaga authored Aug 01, 2014
```
There should not be any side effect on resetting the FGPA every time
we program it. Confirmed by Tomasz Wlostowski <tomasz.wlostowski@cern.ch>
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>
```
  969ee716
- vic: helper to know if a source is managed · c1db3e8a
  Federico Vaga authored Aug 04, 2014
```
Signed-nof -by: Federico Vaga <federico.vaga@cern.ch>
```
  c1db3e8a
- submodule: fmc-bus update · 15fa14fb
  Federico Vaga authored Aug 13, 2014
```
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>
```
  15fa14fb
- speclib: allow to write FPGA with a given buffer · c641779b
  Federico Vaga authored Jul 30, 2014
```
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>
```
  c641779b