Hi Ashwin,
> My understanding of steps that I need to follow for adding a new hardware
> support to sigrok is as follows :
>
> *Step 1:* Use the new-driver script to generate skeleton files (api.c,
> protocol.c/.h)
> *Step 2: *Modify created files accordingly to support FX3.
> *Step 3:* I need to
> I would like to contribute, can I just submit pull request via github?
Yes, it is actually the preferred way these days.
Looking forward to your contribution, thank you!
signature.asc
Description: PGP signature
___
sigrok-devel mailing list
sigrok
Hi Soeren,
> Personally, my prefered solution would be to ask the user when uninstalling
> as he/she is the only one who can make the desired choice when using
> makefiles. Is there a way to add a yes/no confirmation? If so, this is the
> route I'd go.
Uh, please no. Nobody expects user interacti
Hi Gerhard,
> Minor make nit: Was to understand that $(RM) is more portable,
> and might be preferred. But could not find examples in existing
> rules within the minute that I spent searching. It's pre-set in
> GNU make, can't speak with certainty for automake on other
> platforms that don't use G
From: Wolfram Sang
Decoders were left when using 'uninstall' target.
Signed-off-by: Wolfram Sang
---
Found this while trying to remove outdated (because renamed) decoders
which triggered warnings. I hope this is proper automake.
Makefile.am | 2 ++
1 file changed, 2 insertion
Hi Merlijn,
> My question is -- what would the best way to integrate support in the
> korad-kap driver? Some flag to enable quircks for this device?
I'd go this way. If the rest of the protocol is largely the same, it
makes sense to me to have a quirk to fix up the in/out communication and
th
From: Wolfram Sang
- always say 'ID' when the ID command failed
- print hexdump of a faulty ID because on a stalled device we may get
0x00 bytes which would terminate the string early.
Introduce a new 'num_read' variable to avoid that in a future change
'ret' mi
_open(serial, ...)
... when != serial
- serial_flush(serial);
and then the results and the unmatched findings of serial_flush() were
audited.
Signed-off-by: Wolfram Sang
---
Changes since V1:
- do the generic flush *after* setting parameters
src/hardware/agilent-dmm/
On Sun, May 03, 2020 at 10:57:50PM +0200, Wolfram Sang wrote:
> Quite some drivers flush the serial port after opening it. And quite
> some don't although they should. Factor this out, so serial_open() will
> always flush the port. The removal in the drivers was done with this
>
On Sat, Apr 11, 2020 at 10:07:39PM +0200, Wolfram Sang wrote:
> Make sure that all buffers are empty after opening (on all supported
> systems).
>
> Signed-off-by: Wolfram Sang
> ---
>
> This came up while debugging bug #1025. It doesn't fix the bug but I
> t
From: Wolfram Sang
- always say 'ID' when the ID command failed
- print hexdump of a faulty ID because on a stalled device we may get 4
0x00 bytes which would terminate the string early.
Made debugging OLS startup issues a lot easier.
Signed-off-by: Wolfram Sang
---
src/hardware
> @@ -122,6 +122,10 @@ SR_PRIV int serial_open(struct sr_serial_dev_inst
> *serial, int flags)
> if (ret != SR_OK)
> return ret;
>
> + ret = serial_flush(serial);
> + if (ret != SR_OK)
> + return ret;
> +
> if (serial->serialcomm)
> re
> I really wish they had pointed out why the heck they disable it.
Searching for "stacktrace" in
https://gitweb.gentoo.org/repo/gentoo.git/log/dev-libs/boost
points to this commit:
https://gitweb.gentoo.org/repo/gentoo.git/commit/dev-libs/boost?id=2ee11719e251c9ae1a53cdcacc96ac94d7733862
which
> ...so... uh... why is it complaining even though it found a more
> recent version? Any idea?
The only idea I have is the obvious one: do you have the stacktrace
package for that boost version installed?
signature.asc
Description: PGP signature
___
_open(serial, ...)
... when != serial
- serial_flush(serial);
and then the results and the unmatched findings of serial_flush() were
audited.
Signed-off-by: Wolfram Sang
---
As discussed with Gerhard, this is the libsigrok counterpart of a
similar patch for libserialport.
Note the
---
Simple tracer still works, too. Again no testing on non-Linux platforms.
Thanks to Soeren, Gerhard, and Uwe for support!
Looking forward to comments.
Wolfram Sang (2):
HACK: make pulseview crash when pressing "Run"
Add stacktrace support
CMakeLists.
Not-Signed-off-by: Wolfram Sang
---
pv/mainwindow.cpp | 1 +
1 file changed, 1 insertion(+)
diff --git a/pv/mainwindow.cpp b/pv/mainwindow.cpp
index ea9f01af..20b9ab22 100644
--- a/pv/mainwindow.cpp
+++ b/pv/mainwindow.cpp
@@ -710,6 +710,7 @@ void MainWindow::on_run_stop_clicked
Use backtrace when available, otherwise the basic stacktracer.
Signed-off-by: Wolfram Sang
---
CMakeLists.txt | 17 ++---
1 file changed, 14 insertions(+), 3 deletions(-)
diff --git a/CMakeLists.txt b/CMakeLists.txt
index ecaf6694..e157e7db 100644
--- a/CMakeLists.txt
+++ b
> For symbols outside pv, you might need to disable ASLR with
>
> $ sudo sysctl -w kernel.randomize_va_space=0
>
> or using 'setarch ... -R ...'
I think I should add some documentation containing the above (and some
more) in V2.
signature.asc
Description: PGP signature
__
> > list(APPEND PULSEVIEW_LINK_LIBS "-ldl")
> > + link_libraries("-no-pie -fno-pie")
> > endif()
>
> May I suggest you add another change "for good measure", and
> replace the open coded -ldl spec with ${CMAKE_DL_LIBS} which is
> more portable?
Sure, good catch, thanks! Will do that in V2
> Wait, is the "RelWithDebInfo" as good a choice, and already gets
> enforced in the absence of a user spec? And is accepting the user
> spec -- when provided -- the most appropriate activity? Does it
> mean that strictly speaking this specific change is not needed?
OK, my take on this was "enfor
On Thu, Apr 16, 2020 at 03:42:34PM +0200, Stefan Brüns wrote:
> On Thursday, 16 April 2020 15:22:06 CEST Wolfram Sang wrote:
> > This small series is the result of my first journey into the worlds of
> > CMake, Boost, and userspace stacktraces. I get useful output now on my
> &g
Not-Signed-off-by: Wolfram Sang
---
pv/session.cpp | 1 +
1 file changed, 1 insertion(+)
diff --git a/pv/session.cpp b/pv/session.cpp
index 6b879cf9..4a6498b6 100644
--- a/pv/session.cpp
+++ b/pv/session.cpp
@@ -180,6 +180,7 @@ shared_ptr Session::main_view() const
void Session::set_main_bar
Signed-off-by: Wolfram Sang
---
CMakeLists.txt | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/CMakeLists.txt b/CMakeLists.txt
index a1403c92..4881ca36 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -474,7 +474,7 @@ if(ENABLE_SIGNALS)
endif()
if(ENABLE_STACKTRACE
ks for me(tm) now. Looking forward to hear from other tests or
review of the CMake bits.
Happy hacking,
Wolfram
Wolfram Sang (4):
Add boost::stacktrace to CMakeList
With stacktrace, enfore 'Debug' build
For stacktraces, build with no-pie
HACK: make pulseview crash early
CMak
uests/2541/diffs
Signed-off-by: Wolfram Sang
---
CMakeLists.txt | 5 +
1 file changed, 5 insertions(+)
diff --git a/CMakeLists.txt b/CMakeLists.txt
index ecaf6694..2986ee94 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -123,6 +123,11 @@ if(ENABLE_TESTS)
endif()
if(ENABLE_
Signed-off-by: Wolfram Sang
---
CMakeLists.txt | 4
1 file changed, 4 insertions(+)
diff --git a/CMakeLists.txt b/CMakeLists.txt
index 2986ee94..a1403c92 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -58,6 +58,10 @@ if(WIN32)
set(ENABLE_SIGNALS FALSE)
endif()
+if
> Is the device driver's or the serial port's close an issue? Or
> one of them first and the other one as a consequence? Haven't
> checked the code base, so I'm talking from the top of my head
> here, and may miss something.
It's independent of serial ports. I am quite confident bugs #1151 and
#1
> Is it intended that potential sp_flush() errors get ignored?
> (Assuming that DEBUG() is silent in regular applications. And at
> most only results in output but won't affect the return code.)
Yes, to avoid regressions. That shows I rate an error there as 'not
critical'. I have no strong opinio
s not idle.
Signed-off-by: Wolfram Sang
---
I am not sure if the added condition is good enough. Yet, I'd appreciate
comments to see if I am on the right track.
src/std.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/std.c b/src/std.c
index 581e2e46..09ebf1ed 100644
Make sure that all buffers are empty after opening (on all supported
systems).
Signed-off-by: Wolfram Sang
---
This came up while debugging bug #1025. It doesn't fix the bug but I
think it is still nice to have. Opinions?
serialport.c | 4
1 file changed, 4 insertions(+)
diff --
The above function must always return an 'enum sp_transport'. So, add a
'NONE' type and return it if no port ist present.
Fixes bug #1531.
Signed-off-by: Wolfram Sang
---
I added this also to Bugzilla, but I prefer discussions via mail :)
I think this is a viable fix. Bu
Hi Soeren!
> Thanks for your input :)
Very welcome!
> > The latter is called a void message and declared illegal
> > in the specs.
>
> So it would make sense to emit a warning annotation if
> this condition is encountered in FIND_ADDRESS?
This would be ideal.
All the best,
Wolfram
signa
Hi,
>elif self.state == 'FIND ACK':
> -# Wait for a data/ack bit: SCL = rising.
> -(scl, sda) = self.wait({0: 'r'})
> -self.get_ack(scl, sda)
> +# Wait for any of the following conditions (or combinations):
> +# a) a data/ack bit: SCL = rising.
> +# b) STO
Hi,
> SMBus is based on the I2C protocol, roughly with timeouts and differences in
> acknowledgements. Should a SMBus decoder work on logic signals or stack on
> top of the existing i2c decoder?
True about the timeout, but the differences in acknowledgements are in
the protocol layer. The I2C dec
> what are the next steps?
I guess you could send a pull-request via github if you haven't done so
already?
The IRC channel #sigrok on Freenode is usually a good place to get a
timely response from the core developers.
Thanks for your work!
signature.asc
Description: PGP signature
__
Hi,
> unfortunately I did not get any sort of answer and I am unsure if I was
> considered rude to ask the support for Hantek 1008C (I am sorry in this
> case), if nobody is there "responsible" for it or what the problem is.
Definately the latter. Questions are really welcome here, but currently
Hi Soeren,
I have a question about the variable suppress_zoom_to_fit_after_acq_:
I assume this is meant to suppress zoom-to-fit if settings were changed
while an acquisition is running? Or what is the use case?
I see it in effect when a session gets restored because of a commandline
parameter wh
only half the channels available at 200MHz. We never
handled this case and this series doesn't add it as well.
Wolfram Sang (12):
ols: remove unused define
ols: remove needless initialization
ols: use 32bit for handling sample counts
ols: drop wrong usage of macro
ols: simplify cal
Signed-off-by: Wolfram Sang
---
src/hardware/openbench-logic-sniffer/api.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/hardware/openbench-logic-sniffer/api.c
b/src/hardware/openbench-logic-sniffer/api.c
index 26490c05..5e959507 100644
--- a/src/hardware/openbench
Magnus (creator of the Pipistrello) confirmed that he mixed up the
register names. The code was doing it correctly nonetheless but was
confusing to read because of this. Fix it to make it easier to
comprehend.
Signed-off-by: Wolfram Sang
---
src/hardware/pipistrello-ols/api.c | 4 ++--
src
My Pepino needs 15ms instead of 10ms, so let's use 20ms to be on the
safe side.
Signed-off-by: Wolfram Sang
---
src/hardware/openbench-logic-sniffer/api.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/hardware/openbench-logic-sniffer/api.c
b/src/hardware/open
Makes adding new ones easier.
Signed-off-by: Wolfram Sang
---
src/hardware/openbench-logic-sniffer/protocol.h | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/src/hardware/openbench-logic-sniffer/protocol.h
b/src/hardware/openbench-logic-sniffer/protocol.h
index 1ec671a4
commit f51acd69 ("ols: combine demux samples") wrongly replaced the bit
pattern of 0x20 with the number of channels which just happens to be 32
as well. So, the code works but is confusing to read. Reword the
for-loop to make it more comprehensible.
Signed-off-by: Wolfram Sang
---
sr
ch is the removal of the NUM_CHANNELS macro.
The number of channels needs to be dealt with at runtime.
Signed-off-by: Wolfram Sang
---
src/hardware/openbench-logic-sniffer/protocol.c | 5 +++--
src/hardware/openbench-logic-sniffer/protocol.h | 1 -
2 files changed, 3 insertions(+), 3 dele
Somewhen we probably want DIV_ROUND_UP from the Kernel.
Signed-off-by: Wolfram Sang
---
src/hardware/openbench-logic-sniffer/api.c | 8 ++--
1 file changed, 2 insertions(+), 6 deletions(-)
diff --git a/src/hardware/openbench-logic-sniffer/api.c
b/src/hardware/openbench-logic-sniffer/api.c
channel_mask is recalculated before it gets accessed, so this
initialization is not needed.
Signed-off-by: Wolfram Sang
---
src/hardware/openbench-logic-sniffer/protocol.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/src/hardware/openbench-logic-sniffer/protocol.c
b/src/hardware
Signed-off-by: Wolfram Sang
---
src/hardware/openbench-logic-sniffer/protocol.h | 1 -
1 file changed, 1 deletion(-)
diff --git a/src/hardware/openbench-logic-sniffer/protocol.h
b/src/hardware/openbench-logic-sniffer/protocol.h
index cfe98cb4..e8e7574c 100644
--- a/src/hardware/openbench-logic
We needs this twice so put it into a seperate function, so updates to it
will automatically handled for both callers.
Signed-off-by: Wolfram Sang
---
.../openbench-logic-sniffer/protocol.c| 19 ---
1 file changed, 12 insertions(+), 7 deletions(-)
diff --git a/src
this) and to prepare support for devices with
even more memory (Pepino).
Signed-off-by: Wolfram Sang
---
src/hardware/openbench-logic-sniffer/api.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/hardware/openbench-logic-sniffer/api.c
b/src/hardware/openbench-logic-sniffer
ding on the device name later.
Signed-off-by: Wolfram Sang
---
src/hardware/openbench-logic-sniffer/api.c| 28 +++
.../openbench-logic-sniffer/protocol.h| 2 ++
2 files changed, 24 insertions(+), 6 deletions(-)
diff --git a/src/hardware/openbench-logic-sniffer/a
> +#define CMD_CAPTURE_COUNT 0x83 /* extension for Pepino
> */
> +#define CMD_CAPTURE_DELAY 0x84 /* extension for Pepino
> */
I just agreed with Magnus (creator of the Pepino) that these registers
in deed need to be renamed.
signature.asc
Descripti
> - .longname = "Openbench Logic Sniffer",
> + .longname = "Openbench Logic Sniffer & Compatibles",
Probably even better "& SUMP compatibles".
signature.asc
Description: PGP signature
___
sigrok-devel mailing list
sigrok-devel@lists.sourcefor
Somewhen we probably want DIV_ROUND_UP from the Kernel.
Signed-off-by: Wolfram Sang
---
src/hardware/openbench-logic-sniffer/api.c | 8 ++--
1 file changed, 2 insertions(+), 6 deletions(-)
diff --git a/src/hardware/openbench-logic-sniffer/api.c
b/src/hardware/openbench-logic-sniffer/api.c
Add support for the Pepino-style of accessing >256K of memory. For that,
we introduce 'feature_flag' which can be set according to the device.
Signed-off-by: Wolfram Sang
---
src/hardware/openbench-logic-sniffer/api.c| 28 +++
.../openbench-logic-sniff
commit f51acd69 ("ols: combine demux samples") wrongly replaced the bit
pattern of 0x20 with the number of channels which just happens to be 32
as well. So, the code works but is confusing to read. Reword the
for-loop to make it more comprehensible.
Signed-off-by: Wolfram Sang
---
sr
My Pepino needs 15ms instead of 10ms, so let's use 20ms to be on the
safe side.
Signed-off-by: Wolfram Sang
---
src/hardware/openbench-logic-sniffer/api.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/hardware/openbench-logic-sniffer/api.c
b/src/hardware/open
ch is the removal of the NUM_CHANNELS macro.
The number of channels needs to be dealt with at runtime.
Signed-off-by: Wolfram Sang
---
src/hardware/openbench-logic-sniffer/protocol.c | 5 +++--
src/hardware/openbench-logic-sniffer/protocol.h | 1 -
2 files changed, 3 insertions(+), 3 dele
so it needs another look. Will happily share that patch in case
someone else is interested.
* OLS/Pepino have only half the channels available at 200MHz. We never
handled this case and this series doesn't add it as well.
Wolfram Sang (11):
ols: remove unused define
ols: remove needl
Makes adding new ones easier.
Signed-off-by: Wolfram Sang
---
src/hardware/openbench-logic-sniffer/protocol.h | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/src/hardware/openbench-logic-sniffer/protocol.h
b/src/hardware/openbench-logic-sniffer/protocol.h
index 1ec671a4
We needs this twice so put it into a seperate function, so updates to it
will automatically handled for both callers.
Signed-off-by: Wolfram Sang
---
.../openbench-logic-sniffer/protocol.c| 19 ---
1 file changed, 12 insertions(+), 7 deletions(-)
diff --git a/src
this) and to prepare support for devices with
even more memory (Pepino).
Signed-off-by: Wolfram Sang
---
src/hardware/openbench-logic-sniffer/api.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/hardware/openbench-logic-sniffer/api.c
b/src/hardware/openbench-logic-sniffer
channel_mask is recalculated before it gets accessed, so this
initialization is not needed.
Signed-off-by: Wolfram Sang
---
src/hardware/openbench-logic-sniffer/protocol.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/src/hardware/openbench-logic-sniffer/protocol.c
b/src/hardware
Signed-off-by: Wolfram Sang
---
src/hardware/openbench-logic-sniffer/protocol.h | 1 -
1 file changed, 1 deletion(-)
diff --git a/src/hardware/openbench-logic-sniffer/protocol.h
b/src/hardware/openbench-logic-sniffer/protocol.h
index cfe98cb4..e8e7574c 100644
--- a/src/hardware/openbench-logic
Signed-off-by: Wolfram Sang
---
src/hardware/openbench-logic-sniffer/api.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/hardware/openbench-logic-sniffer/api.c
b/src/hardware/openbench-logic-sniffer/api.c
index 26490c05..351f7954 100644
--- a/src/hardware/openbench
Hi,
> 1. When first loaded the number of samples is uninitialized, and "RUN" sends
> a request for 0x samples. This seems to cause Pulseview to stop
> responding, maybe because it is waiting for data it will never get. After
> the number of samples drop down is changed once it requests the
> Using an FPGA as a front-end and RAM interface to shovel the data
> into, then pushing the data over a slow serial (UART or USB2)
> interface into the PC when the RAM is full and acquisition has stopped
> is a concept that worked well for years but it's totally outdated
> now.
But still, for ho
Hi!
> > with an Openbench Logic Sniffer (open hardware/firmware), a Saanlima
> > Pipistrello (open hardware/firmware)
>
> Unfortunately the Pipistrello with 64 MB DRAM appear to be discontinued:
Yes :( I plan to update all OLS based wiki entries, hopefully in
February. Such information will be a
Hi Soeren,
> insist that DreamsSource labs is not an ethical company since they
> treated (and still treat) the sigrok project like shit.
Didn't know that, but really good to know! Uwe Bonnes pointed to the
same direction. Any pointers for details?
I submitted a talk about measuring with free HW
> Both look like pretty good ideas, but I am really after something in a
> case that I could chuck into a tool bag if needed.
I don't have this one, but maybe someone else can add experiences with
it?
https://sigrok.org/wiki/DreamSourceLab_DSLogic
In their Kickstarter, they claimed "OpenHardwar
> I am often going to be looking at 4+ channels at over 10MHz per
> channel and really want the hardware to be Open Source/from a decent,
> ethical company that will last a long time (so I can recommend them to
> colleagues).
I do such kind of measurements with an Openbench Logic Sniffer.
Reasona
Hi,
just wanted to let you know that I will be giving a talk and a showcase
including sigrok at ELC-E in Prague next week.
Talk, Monday 12:05:
https://osseu17.sched.com/event/ByYP/more-robust-i2c-designs-with-a-new-fault-injection-driver-wolfram-sang-renesas
Showcase, Tuesday 17:35
https
From: Wolfram Sang
Signed-off-by: Wolfram Sang
---
This is already against the decoder changes Uwe merged yesterday.
decoder/test/sdcard_sd/R-Car-H2_cmd23_cmd18.output | 286 +
decoder/test/sdcard_sd/test.conf | 5 +
2 files changed, 291 insertions
Hi Uwe,
> Merged all 3 patches, thanks!
Thanks!
> > +self.last_cmd = cmd;
>
> Dropped the semicolon here, not needed.
Ooops, that shows where I come from :)
Regards,
Wolfram
signature.asc
Description: PGP signature
---
From: Wolfram Sang
Replace the hardcoded value with the last cmd used. I'd think the whole
ACMD handling could be simplified if we make further use of this new
variable. But for now, let's keep it simple and make it work.
Signed-off-by: Wolfram Sang
---
decoders/sdcard_sd/
From: Wolfram Sang
Probably a typo, currently 'cmd0' gets pushed out.
Signed-off-by: Wolfram Sang
---
decoders/sdcard_sd/pd.py | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/decoders/sdcard_sd/pd.py b/decoders/sdcard_sd/pd.py
index 7e486fe..2d88096 100644
---
Signed-off-by: Wolfram Sang
---
decoder/test/sdcard_sd/imx6_quad_working_snippet.output | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/decoder/test/sdcard_sd/imx6_quad_working_snippet.output
b/decoder/test/sdcard_sd/imx6_quad_working_snippet.output
index 4487b9b
Hi,
> Anyway, I think it makes sense bringing back gnuplot output until CSV
> stabilizes.
Thanks to your report I had a look at the CSV driver and reviewed the
process_logic() function. It has multiple problems for which I sent out
patches a second ago. Would you kindly re-test with the patches o
'i' was iterating in steps of unitsize. However, the destination array
was also indexed with it, but it is of u8 type. Let 'i' run bytewise and
only multiply with unitsize when we need it.
Signed-off-by: Wolfram Sang
---
src/output/csv.c | 4 ++--
1 file changed, 2 insert
Signed-off-by: Wolfram Sang
---
src/output/csv.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/src/output/csv.c b/src/output/csv.c
index 3cf19191..fd43f00d 100644
--- a/src/output/csv.c
+++ b/src/output/csv.c
@@ -608,6 +608,7 @@ static int cleanup(struct sr_output *o
'j' is the loop variable for channels, not 'ch'.
Signed-off-by: Wolfram Sang
---
src/output/csv.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/output/csv.c b/src/output/csv.c
index fd43f00d..5aa83458 100644
--- a/src/output/csv.c
+++ b/src/ou
indly re-test if it works for you now? It seems to work
with the data I captured recently.
Wolfram Sang (4):
output/csv: fix segfault when naming logic channels
output/csv: fix segfault with logic channels
output/csv: fix a minor leak
output/csv: get proper index for input channels
src/o
'j' is the loop variable for channels, not 'i'.
Reported-by: Maxim Sloyko
Signed-off-by: Wolfram Sang
---
src/output/csv.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/output/csv.c b/src/output/csv.c
index c39c29a7..8d1bcd98 100644
--- a/src/ou
From: Wolfram Sang
Signed-off-by: Wolfram Sang
---
decoder/pdtest | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/decoder/pdtest b/decoder/pdtest
index 1c24e6c..da548e3 100755
--- a/decoder/pdtest
+++ b/decoder/pdtest
@@ -62,7 +62,7 @@ def usage(msg=None):
-l List test
> -def putr(self, cmd, desc):
> -self.putt([cmd, ['Reply: %s' % desc]])
> +def putr(self, desc):
> +self.putt([self.cmd, ['Reply: %s' % desc]])
Hmmm, this doesn't handle ACMDs nicely... needs some more thought.
signature.asc
Description: PGP signature
--
From: Wolfram Sang
putr2 is unused, so let it go.
Signed-off-by: Wolfram Sang
---
Maybe I am missing something, but I even can't see an intended but
currently unimplemented use-case?
decoders/sdcard_sd/pd.py | 6 --
1 file changed, 6 deletions(-)
diff --git a/decoders/sdcard_sd/pd
From: Wolfram Sang
Replace the hardcoded value with the latest cmd used.
Signed-off-by: Wolfram Sang
---
I assume it is safe to use self.cmd because the state machine always starts
with 'GET COMMAND TOKEN' first.
decoders/sdcard_sd/pd.py | 14 +++---
1 file changed, 7
> Basically -f will force-overwrite / generate the files;
Thanks!
> but you should
> generally check if the contents look reasonable and as expected
Did that and found another issue in the decoder ;) Patches coming
in a minute...
> graphically via PulseView. Otherwise there's a risk of introdu
Hi Uwe,
> the sigrok-test repo accordingly. Btw, if you could provide some more
> .sr files for sigrok-dumps (and/or tests for sigrok-test) that would be
> great!
Just posted the dump. How do I generate the reference output to match
against for the test?
Thanks,
Wolfram
signature.asc
Desc
From: Wolfram Sang
First dump: proof CMD23 before CMD18. Other dumps will follow.
---
sdcard/sd_mode/rcar-h2/README | 25 +
sdcard/sd_mode/rcar-h2/cmd23_cmd18.sr | Bin 0 -> 636 bytes
2 files changed, 25 insertions(+)
create mode 100644 sdcard/sd_mode/rcar
gt; correct behavior.
>
> According to http://mygizmos.org/ols/Logic-Sniffer-FPGA-Spec.pdf section 2.3.1
> p.8, the openbench logic sniffer must be reset before each arm command.
>
> This patch may fix bug #809
It still does :) For both patches:
Tested-by: Wolfram Sang
And
Hi,
> With trigger enabled, and with pulseview, a second (or more) acquisition start
> immediatly instead of blocking. It's mandatory to try several times to have a
> correct behavior.
>
> According to http://mygizmos.org/ols/Logic-Sniffer-FPGA-Spec.pdf section 2.3.1
> p.8, the openbench logic sn
From: Wolfram Sang
Now that self.arg is populated, we can finally use it.
Signed-off-by: Wolfram Sang
---
decoders/sdcard_sd/pd.py | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/decoders/sdcard_sd/pd.py b/decoders/sdcard_sd/pd.py
index 1ad9cb6..835a544 100644
--- a
From: Wolfram Sang
I just did some SD card measurements and came up with these 3 little patches
for the sdcard decoder.
I also got finally around to build pulseview from git after using the Debian
package for a long time. That felt like a quantum leap, thanks for all the
work on that!
Kind
From: Wolfram Sang
There is no handler for CMD4 yet.
Signed-off-by: Wolfram Sang
---
decoders/sdcard_sd/pd.py | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/decoders/sdcard_sd/pd.py b/decoders/sdcard_sd/pd.py
index 02e78d7..b544b37 100644
--- a/decoders/sdcard_sd/pd.py
From: Wolfram Sang
This is especially useful with the default handler, so you can better
understand the transfer if you are already familiar with CMD numbers and
their arguments. It makes it also easier to compare with the debug
output of the e.g. Linux Kernel.
Signed-off-by: Wolfram Sang
Hi,
I noticed that there sadly won't be a sigrok table at 33C3. But who is
going nonetheless? I'll be around...
Best wishes,
Wolfram
signature.asc
Description: PGP signature
--
Check out the vibrant tech community
According to the infos I have, VCD files should be plain ASCII, but we
got report of a version adding a UTF8 BOM at the beginning of the file,
so we need to skip it.
This fixes bug #755.
Signed-off-by: Wolfram Sang
---
Thanks to Aurelien Jacobs for the criticism.
src/input/vcd.c | 4
1
> > + /* Skip UTF8 BOM */
> > + if (buf->str[0] == '\xef' && buf->str[1] == '\xbb' && buf->str[2] ==
> > '\xbf')
>
> You could use strncmp() here. And for good mesure, you could check
> buf->len. Something like this:
I first intentionally decided against strncmp, but now I can't recall
why.
According to the infos I have, VCD files should be plain ASCII, but
nonetheless DSView on Windows seems to add a UTF8 BOM at the beginning
of the file, so we need to skip it.
This fixes bug #755.
Signed-off-by: Wolfram Sang
---
src/input/vcd.c | 4
1 file changed, 4 insertions(+)
diff
1 - 100 of 128 matches
Mail list logo
