]> sigrok.org Git - libsigrok.git/log
libsigrok.git
10 years agotests: Fix a few warnings.
Uwe Hermann [Fri, 17 Jan 2014 19:41:56 +0000 (20:41 +0100)]
tests: Fix a few warnings.

Fix a bunch of "no previous prototype for ..." warnings exposed
by -Wmissing-prototypes.

10 years agodemo: Bring analog square wave amplitude into line with other patterns.
Bert Vermeulen [Wed, 15 Jan 2014 00:51:04 +0000 (01:51 +0100)]
demo: Bring analog square wave amplitude into line with other patterns.

10 years agodemo: Add analog sawtooth pattern.
poljar (Damir Jelić) [Tue, 14 Jan 2014 22:03:10 +0000 (23:03 +0100)]
demo: Add analog sawtooth pattern.

10 years agodemo: Add triangle pattern.
poljar (Damir Jelić) [Tue, 14 Jan 2014 22:03:09 +0000 (23:03 +0100)]
demo: Add triangle pattern.

10 years agodemo: Implement sine wave pattern.
poljar (Damir Jelić) [Tue, 14 Jan 2014 22:03:08 +0000 (23:03 +0100)]
demo: Implement sine wave pattern.

This adds sine wave generation capabilities for the analog channels in the demo
driver. The frequency of the sine wave depends on the configured sample rate of
the demo device. The frequency of the sine wave is always 20 times smaller than
the sample rate, in other words we always have 20 samples per period.

10 years agouni-t-ut32x: Fix typo that prevents usage with multiple devices.
Bert Vermeulen [Fri, 17 Jan 2014 11:11:52 +0000 (12:11 +0100)]
uni-t-ut32x: Fix typo that prevents usage with multiple devices.

10 years agorigol-ds: Minor whitespace fixes.
Uwe Hermann [Fri, 17 Jan 2014 00:57:09 +0000 (01:57 +0100)]
rigol-ds: Minor whitespace fixes.

10 years agorigol-ds: Enable/disable LA pod when (de)selecting digital channels.
Martin Ling [Thu, 16 Jan 2014 23:25:59 +0000 (23:25 +0000)]
rigol-ds: Enable/disable LA pod when (de)selecting digital channels.

10 years agorigol-ds: Reset num_frames to zero in dev_acquisition_start.
Martin Ling [Thu, 16 Jan 2014 18:57:45 +0000 (18:57 +0000)]
rigol-ds: Reset num_frames to zero in dev_acquisition_start.

10 years agorigol-ds: Make sure to always send SR_DF_END at end of capture.
Martin Ling [Wed, 15 Jan 2014 17:06:46 +0000 (17:06 +0000)]
rigol-ds: Make sure to always send SR_DF_END at end of capture.

10 years agorigol-ds: Calculate effective samplerate.
Martin Ling [Tue, 14 Jan 2014 21:27:58 +0000 (21:27 +0000)]
rigol-ds: Calculate effective samplerate.

10 years agorigol-ds: Separate function to compute analog frame size.
Martin Ling [Tue, 14 Jan 2014 21:21:31 +0000 (21:21 +0000)]
rigol-ds: Separate function to compute analog frame size.

10 years agorigol-ds: Advertise SR_CONF_LIMIT_FRAMES support.
Martin Ling [Tue, 14 Jan 2014 20:55:20 +0000 (20:55 +0000)]
rigol-ds: Advertise SR_CONF_LIMIT_FRAMES support.

10 years agorigol-ds: Fix check for probe group validity.
Martin Ling [Tue, 14 Jan 2014 20:48:17 +0000 (20:48 +0000)]
rigol-ds: Fix check for probe group validity.

10 years agohameg-hmo: Fix two compiler warnings on Windows.
Uwe Hermann [Thu, 16 Jan 2014 23:37:58 +0000 (00:37 +0100)]
hameg-hmo: Fix two compiler warnings on Windows.

  CC       libsigrok_hw_hameg_hmo_la-protocol.lo
api.c: In function 'auto_find_usb':
api.c:70:39: warning: unused parameter 'vendor_id' [-Wunused-parameter]
 static GSList *auto_find_usb(uint16_t vendor_id, uint16_t product_id)
                                       ^
api.c:70:59: warning: unused parameter 'product_id' [-Wunused-parameter]
 static GSList *auto_find_usb(uint16_t vendor_id, uint16_t product_id)
                                                           ^

10 years agousb: Fix two compiler warnings on Windows.
Uwe Hermann [Thu, 16 Jan 2014 23:26:30 +0000 (00:26 +0100)]
usb: Fix two compiler warnings on Windows.

  CC       libsigrok_hw_common_la-usb.lo
usb.c:241:18: warning: no previous prototype for 'usb_thread'
[-Wmissing-prototypes]
 SR_PRIV gpointer usb_thread(gpointer data)
                  ^
usb.c:256:13: warning: no previous prototype for 'usb_callback'
[-Wmissing-prototypes]
 SR_PRIV int usb_callback(int fd, int revents, void *cb_data)
             ^

10 years agohameg-hmo: Silence warning about format security.
poljar (Damir Jelić) [Thu, 16 Jan 2014 18:42:27 +0000 (19:42 +0100)]
hameg-hmo: Silence warning about format security.

The command for fetching the sample rate while in single shot mode doesn't take
an argument so the snprintf here is redundant.

10 years agohameg-hmo: Reset the number of acquired frames in acquisition stop.
poljar (Damir Jelić) [Thu, 16 Jan 2014 18:55:00 +0000 (19:55 +0100)]
hameg-hmo: Reset the number of acquired frames in acquisition stop.

10 years agohameg-hmo: Expose more options with config_get().
poljar (Damir Jelić) [Thu, 16 Jan 2014 14:28:58 +0000 (15:28 +0100)]
hameg-hmo: Expose more options with config_get().

This patch exposes the TRIGGER_SOURCE and COUPLING options via config_get().

10 years agohameg-hmo: Advertise the frame limit option.
poljar (Damir Jelić) [Thu, 16 Jan 2014 14:28:57 +0000 (15:28 +0100)]
hameg-hmo: Advertise the frame limit option.

10 years agohameg-hmo: Update the sample rate when the channel states or the timebase change.
poljar (Damir Jelić) [Thu, 16 Jan 2014 14:28:56 +0000 (15:28 +0100)]
hameg-hmo: Update the sample rate when the channel states or the timebase change.

The sample rate on the Hameg scopes changes depending on the number of channels
turned on and on the current timebase.

Update the sample rate if any of the above change.

10 years agohameg-hmo: Add support for sample rate fetching.
poljar (Damir Jelić) [Thu, 16 Jan 2014 14:28:55 +0000 (15:28 +0100)]
hameg-hmo: Add support for sample rate fetching.

10 years agoscpi: Add CMD_GET_SAMPLE_RATE to the command enum.
poljar (Damir Jelić) [Thu, 16 Jan 2014 14:28:54 +0000 (15:28 +0100)]
scpi: Add CMD_GET_SAMPLE_RATE to the command enum.

10 years agohameg-hmo: Change the way vdiv and timebase are saved.
poljar (Damir Jelić) [Thu, 16 Jan 2014 14:28:53 +0000 (15:28 +0100)]
hameg-hmo: Change the way vdiv and timebase are saved.

The current vertical division setting (per channel) and the timebase are stored
as a floating point number. This is suboptimal since clients expect us to send
this information to them in form of a rational number.

Store only the index of the current setting since all the supported settings are
already stored inside of an array.

10 years agohameg-hmo: Send the DF_END packet in acquisition_stop()
poljar (Damir Jelić) [Thu, 16 Jan 2014 14:28:52 +0000 (15:28 +0100)]
hameg-hmo: Send the DF_END packet in acquisition_stop()

The DF_END packet was send out after all configured frames were fetched, but
devices may stop the acquisition at any point in time and an DF_END will not be
send out in this case.

Send the DF_END packet inside of acquisition_stop() so this can't happen.

10 years agohameg-hmo: Close the device after initial scan.
poljar (Damir Jelić) [Thu, 16 Jan 2014 14:28:51 +0000 (15:28 +0100)]
hameg-hmo: Close the device after initial scan.

The device is after a scan left open and clients that don't call unconditionally
dev_open() will never fetch the initial device state.

Close the device after the scan so clients know they need to open it.

10 years agohameg-hmo: Handle floating point numbers while ignoring the locale.
poljar (Damir Jelić) [Thu, 16 Jan 2014 14:28:50 +0000 (15:28 +0100)]
hameg-hmo: Handle floating point numbers while ignoring the locale.

10 years agoscpi: Use sr_atof_ascii() instead of sr_atof().
poljar (Damir Jelić) [Thu, 16 Jan 2014 14:28:49 +0000 (15:28 +0100)]
scpi: Use sr_atof_ascii() instead of sr_atof().

10 years agostrutil: Add function to parse floating point numbers while ignoring the locale.
poljar (Damir Jelić) [Thu, 16 Jan 2014 14:28:48 +0000 (15:28 +0100)]
strutil: Add function to parse floating point numbers while ignoring the locale.

Most of the supported gear uses the ANSI C locale for communication, and if the
client sets up an incompatible locale parsing would fail.

This function ignores the client's locale and parses floating point numbers
using the ANSI C locale. This function should be always used when parsing
floating point numbers coming from the instrument.

10 years agodevice: Pass sdi as an function argument to config_list in dev_has_option()
poljar (Damir Jelić) [Thu, 16 Jan 2014 01:53:41 +0000 (02:53 +0100)]
device: Pass sdi as an function argument to config_list in dev_has_option()

With drivers that support multiple devices we need to know the device model
while listing options. That information is most of the time saved in the private
part of the dev_inst structure.

Pass the pointer to the dev_inst structure as an function argument so we have
access to this information.

10 years agostrutil: Correctly parse floating point frequencies in parse_size_string().
poljar (Damir Jelić) [Thu, 16 Jan 2014 01:53:40 +0000 (02:53 +0100)]
strutil: Correctly parse floating point frequencies in parse_size_string().

parse_size_string() incorrectly parses a real number, e.g. 1.5 kHz ends up
being 1Hz.

This patch fixes parse_size_string() to take the fractional part as well into
account. The fractional part is parsed as an double precision floating point
number while ignoring the locale.

10 years agosysclk-lwla: Make use of the probe index again.
Daniel Elstner [Wed, 15 Jan 2014 10:09:06 +0000 (11:09 +0100)]
sysclk-lwla: Make use of the probe index again.

(configure_probes): sr_probe::index has been un-deprecated,
so use it instead of relying on the list order.

10 years agosysclk-lwla: Bypass divider in external clock mode.
Daniel Elstner [Wed, 15 Jan 2014 01:51:08 +0000 (02:51 +0100)]
sysclk-lwla: Bypass divider in external clock mode.

(lwla_setup_acquisition): Set the clock divider bypass
flag to 1 for the external clock modes as well.
(capture_setup): Set the clock divide count to 0 if an
external clock source is selected.

10 years agosysclk-lwla: Utility functions clean-up and semantic fixes.
Daniel Elstner [Wed, 15 Jan 2014 00:12:39 +0000 (01:12 +0100)]
sysclk-lwla: Utility functions clean-up and semantic fixes.

(lwla_send_bitstream): Unref the mapped file earlier in order
to simplify the error handling.
(lwla_receive_reply): Do not treat a reply buffer length of
zero as silent no-op.  That logic was left over from an earlier
iteration, before the distinction between reply buffer size and
expected read length was introduced.

10 years agosysclk-lwla: Remove unused dev_context::next_state.
Daniel Elstner [Tue, 14 Jan 2014 23:59:08 +0000 (00:59 +0100)]
sysclk-lwla: Remove unused dev_context::next_state.

10 years agosysclk-lwla: Avoid warning due to bogus range check.
Daniel Elstner [Tue, 14 Jan 2014 23:52:26 +0000 (00:52 +0100)]
sysclk-lwla: Avoid warning due to bogus range check.

(lwla_set_clock_source): Checking whether an enum value is greater
than or equal to zero apparently results in a warning with some
compilers.  Assign the enum to an unsigned variable to avoid this,
and return SR_ERR_BUG if the range is exceeded, as this indicates
a bug in the driver code itself.

10 years agoconfigure: add -Wmissing-prototypes compiler option
Aurelien Jacobs [Tue, 14 Jan 2014 22:30:02 +0000 (23:30 +0100)]
configure: add -Wmissing-prototypes compiler option

This should avoid introduction SR_PRIV functions when static would be enough.

10 years agoremove unused static functions
Aurelien Jacobs [Tue, 14 Jan 2014 22:29:17 +0000 (23:29 +0100)]
remove unused static functions

10 years agochange a bunch of functions from SR_PRIV to static
Aurelien Jacobs [Tue, 14 Jan 2014 22:16:55 +0000 (23:16 +0100)]
change a bunch of functions from SR_PRIV to static

None of those functions are called across compilation units.

10 years agomove function declarations to the appropriate header
Aurelien Jacobs [Tue, 14 Jan 2014 22:38:23 +0000 (23:38 +0100)]
move function declarations to the appropriate header

10 years agoes519xx: add missing declarations for es519xx_2400_11b_altfn
Aurelien Jacobs [Tue, 14 Jan 2014 22:24:47 +0000 (23:24 +0100)]
es519xx: add missing declarations for es519xx_2400_11b_altfn

10 years agognuplot: proper error message when using gnuplot output without logic probe
Aurelien Jacobs [Tue, 14 Jan 2014 21:59:33 +0000 (22:59 +0100)]
gnuplot: proper error message when using gnuplot output without logic probe

10 years agoadd udev rule for the Rigol DS2000 series
Aurelien Jacobs [Tue, 14 Jan 2014 21:56:55 +0000 (22:56 +0100)]
add udev rule for the Rigol DS2000 series

10 years agosysclk-lwla: Fix probe name issue.
Uwe Hermann [Tue, 14 Jan 2014 18:52:26 +0000 (19:52 +0100)]
sysclk-lwla: Fix probe name issue.

The g_ascii_formatd() function expects the "format" argument to start
with a '%' character, e.g. it should be "%f" or such (this is not
clearly documented in the glib API docs, but visible from the source code).

The usage of "CH%f" for example will trigger an assertion and thus make the
LWLA device unusable in practice (e.g. in PulseView on Windows no probenames
would be shown, and sampling wouldn't work).

Example:
  GLib-CRITICAL **: g_ascii_formatd: assertion 'format[0] == '%'' failed

(not exposed in all glib versions or builds of glib on all distros
apparently, some may need G_MESSAGES_DEBUG=all or other measures)

From the glib g_ascii_formatd() code:
  g_return_val_if_fail (format[0] == '%', NULL);

We now use g_snprintf() instead for simplicity. This has been tested to
fix this specific issue (i.e. the probenames now do show up in PulseView).

This closes bug #270.

10 years agoconfigure.ac: Don't build sysclk-lwla if libusb-1.0 is not found.
Uwe Hermann [Tue, 14 Jan 2014 18:46:05 +0000 (19:46 +0100)]
configure.ac: Don't build sysclk-lwla if libusb-1.0 is not found.

10 years agoconfigure.ac: Move sysclk-lwla chunk to correct location.
Uwe Hermann [Tue, 14 Jan 2014 18:36:28 +0000 (19:36 +0100)]
configure.ac: Move sysclk-lwla chunk to correct location.

10 years agosysclk-lwla: Honor SR_CONF_CONN at scan time.
Daniel Elstner [Tue, 14 Jan 2014 00:43:02 +0000 (01:43 +0100)]
sysclk-lwla: Honor SR_CONF_CONN at scan time.

10 years agosysclk-lwla: Fix calculation of the running sample count.
Daniel Elstner [Tue, 14 Jan 2014 00:28:00 +0000 (01:28 +0100)]
sysclk-lwla: Fix calculation of the running sample count.

Field 7 of the status response is actually a duration in
milliseconds at all samplerates but 125 MHz.

10 years agosysclk-lwla: Implement support for the LWLA1034.
Daniel Elstner [Mon, 13 Jan 2014 21:57:59 +0000 (22:57 +0100)]
sysclk-lwla: Implement support for the LWLA1034.

10 years agosysclk-lwla: Initial driver skeleton.
Daniel Elstner [Mon, 13 Jan 2014 21:49:55 +0000 (22:49 +0100)]
sysclk-lwla: Initial driver skeleton.

10 years agorigol-ds: Use correct digital channel numbers when fetching config.
Martin Ling [Tue, 14 Jan 2014 18:25:08 +0000 (18:25 +0000)]
rigol-ds: Use correct digital channel numbers when fetching config.

10 years agocem-dt-885x: Don't use char as signed type.
Bert Vermeulen [Mon, 13 Jan 2014 23:05:24 +0000 (00:05 +0100)]
cem-dt-885x: Don't use char as signed type.

10 years agoscpi: properly check for HAVE_RPC (which is always defined)
Aurelien Jacobs [Mon, 13 Jan 2014 23:17:44 +0000 (00:17 +0100)]
scpi: properly check for HAVE_RPC (which is always defined)

10 years agorigol-ds: remove lonely break
Aurelien Jacobs [Mon, 13 Jan 2014 22:40:55 +0000 (23:40 +0100)]
rigol-ds: remove lonely break

10 years agorigol-ds: prevent config_list() to return empty g_variant
Aurelien Jacobs [Mon, 13 Jan 2014 22:38:32 +0000 (23:38 +0100)]
rigol-ds: prevent config_list() to return empty g_variant

This fixes the following glib message:
$ ./sigrok-cli -d rigol-ds --show
[...]
g_variant_builder_end: assertion '!GVSB(builder)->uniform_item_types || GVSB(builder)->prev_item_type != NULL || g_variant_type_is_definite (GVSB(builder)->type)' failed

10 years agoUse PRIu64 for format in pretty-printer.
Bert Vermeulen [Mon, 13 Jan 2014 21:22:40 +0000 (22:22 +0100)]
Use PRIu64 for format in pretty-printer.

Thanks to Marcus Comstedt for the fix.

10 years agorigol-ds: does not depend on libserialport
Aurelien Jacobs [Mon, 13 Jan 2014 21:49:57 +0000 (22:49 +0100)]
rigol-ds: does not depend on libserialport

10 years agoscpi_usbtmc: does not depend on libserialport
Aurelien Jacobs [Mon, 13 Jan 2014 21:49:21 +0000 (22:49 +0100)]
scpi_usbtmc: does not depend on libserialport

10 years agoudev: Add the Rigol DG4000 series to the udev rules
poljar (Damir Jelić) [Mon, 13 Jan 2014 11:54:24 +0000 (12:54 +0100)]
udev: Add the Rigol DG4000 series to the udev rules

10 years agoatten-pps3xxx: Fix options reporting.
Bert Vermeulen [Mon, 13 Jan 2014 01:10:11 +0000 (02:10 +0100)]
atten-pps3xxx: Fix options reporting.

10 years agoatten-pps3xxx: Fix output channel mode get.
Bert Vermeulen [Mon, 13 Jan 2014 01:05:39 +0000 (02:05 +0100)]
atten-pps3xxx: Fix output channel mode get.

10 years agodemo: Properly handle logic vs. analog when setting the pattern.
Bert Vermeulen [Sun, 12 Jan 2014 22:31:23 +0000 (23:31 +0100)]
demo: Properly handle logic vs. analog when setting the pattern.

10 years agodemo: Split supported device options by probe group.
Bert Vermeulen [Sun, 12 Jan 2014 21:36:39 +0000 (22:36 +0100)]
demo: Split supported device options by probe group.

10 years agoscpi_vxi: ensure the vxi link was properly opened before closing it
Aurelien Jacobs [Sun, 12 Jan 2014 17:46:24 +0000 (18:46 +0100)]
scpi_vxi: ensure the vxi link was properly opened before closing it

10 years agoscpi_tcp: split into scpi_tcp_raw and scpi_tcp_rigol
Aurelien Jacobs [Sat, 11 Jan 2014 23:38:08 +0000 (00:38 +0100)]
scpi_tcp: split into scpi_tcp_raw and scpi_tcp_rigol

The current implementation is renamed to tcp-rigol as it seems to be
a Rigol proprietary protocol used only on Rigol VS5000 series.

A new tcp-raw implementation is introduced which simply carries raw SCPI
commands over TCP. It is probably a much more common protocol and it is
at least available on Rigol DS2000 series on port 5555.

10 years agoscpi: make the scpi_dev_inst_new more generic
Aurelien Jacobs [Sat, 11 Jan 2014 23:05:02 +0000 (00:05 +0100)]
scpi: make the scpi_dev_inst_new more generic

10 years agoscpi: factorize dev_inst_new calls out of individual drivers
Aurelien Jacobs [Sat, 11 Jan 2014 16:56:15 +0000 (17:56 +0100)]
scpi: factorize dev_inst_new calls out of individual drivers

10 years agoscpi: add VXI transport support
Aurelien Jacobs [Thu, 9 Jan 2014 22:44:35 +0000 (23:44 +0100)]
scpi: add VXI transport support

10 years agoimport VXI RPC Language description
Aurelien Jacobs [Tue, 7 Jan 2014 22:28:28 +0000 (23:28 +0100)]
import VXI RPC Language description

along with the corresponding rpcgen generated source code files

10 years agoconfigure.ac: Disable atten-pps3xxx driver if libserialport is not found.
Marc Schink [Sat, 11 Jan 2014 12:55:26 +0000 (13:55 +0100)]
configure.ac: Disable atten-pps3xxx driver if libserialport is not found.

10 years agostd: use #ifdef rather than #if where the constant may not be defined
Aurelien Jacobs [Sat, 11 Jan 2014 16:48:53 +0000 (17:48 +0100)]
std: use #ifdef rather than #if where the constant may not be defined

10 years agoatten-pps3xxx: Push configured settings even without acquisition.
Bert Vermeulen [Fri, 10 Jan 2014 01:10:51 +0000 (02:10 +0100)]
atten-pps3xxx: Push configured settings even without acquisition.

10 years agoatten-pps3xxx: Use configured values to construct packet.
Bert Vermeulen [Fri, 10 Jan 2014 01:09:49 +0000 (02:09 +0100)]
atten-pps3xxx: Use configured values to construct packet.

This used the values previously received from the device, not very useful.

10 years agoatten-pps3xxx: Full support for the PPS3203T-3S.
Bert Vermeulen [Wed, 8 Jan 2014 20:18:19 +0000 (21:18 +0100)]
atten-pps3xxx: Full support for the PPS3203T-3S.

10 years agoAdd config keys for programmable power supplies.
Bert Vermeulen [Wed, 8 Jan 2014 20:15:45 +0000 (21:15 +0100)]
Add config keys for programmable power supplies.

10 years agoatten-pps3xxx: Initial driver skeleton.
Bert Vermeulen [Sun, 5 Jan 2014 12:12:01 +0000 (13:12 +0100)]
atten-pps3xxx: Initial driver skeleton.

10 years agochronovu-la8: Publish SR_CONF_MAX_UNCOMPRESSED_SAMPLES.
Bert Vermeulen [Thu, 9 Jan 2014 15:28:29 +0000 (16:28 +0100)]
chronovu-la8: Publish SR_CONF_MAX_UNCOMPRESSED_SAMPLES.

10 years agoikalogic-scanalogic2: List all keys in device options.
Bert Vermeulen [Thu, 9 Jan 2014 15:27:56 +0000 (16:27 +0100)]
ikalogic-scanalogic2: List all keys in device options.

10 years agozeroplus-logic-cube: List all keys in device options.
Bert Vermeulen [Thu, 9 Jan 2014 15:26:58 +0000 (16:26 +0100)]
zeroplus-logic-cube: List all keys in device options.

10 years agozeroplus-logic-cube: Publish SR_CONF_MAX_UNCOMPRESSED_SAMPLES.
Bert Vermeulen [Thu, 9 Jan 2014 13:30:20 +0000 (14:30 +0100)]
zeroplus-logic-cube: Publish SR_CONF_MAX_UNCOMPRESSED_SAMPLES.

10 years agoikalogic-scanalogic2: Publish SR_CONF_MAX_UNCOMPRESSED_SAMPLES.
Bert Vermeulen [Thu, 9 Jan 2014 13:00:53 +0000 (14:00 +0100)]
ikalogic-scanalogic2: Publish SR_CONF_MAX_UNCOMPRESSED_SAMPLES.

10 years agoAdd support for SR_CONF_MAX_UNCOMPRESSED_SAMPLES.
Bert Vermeulen [Thu, 9 Jan 2014 12:08:32 +0000 (13:08 +0100)]
Add support for SR_CONF_MAX_UNCOMPRESSED_SAMPLES.

10 years agoAdd SR_CONF_MAX_UNCOMPRESSED_SAMPLES.
Bert Vermeulen [Thu, 9 Jan 2014 11:36:05 +0000 (12:36 +0100)]
Add SR_CONF_MAX_UNCOMPRESSED_SAMPLES.

10 years agoAdd sr_session_save_init().
Bert Vermeulen [Tue, 7 Jan 2014 11:52:23 +0000 (12:52 +0100)]
Add sr_session_save_init().

This allows a frontend to initialize a session file, providing the
required samplerate and probe names, without having a proper
struct sr_dev_inst handy.

sr_session_append() is then used to add captured data to the session
file, as usual.

The existing sr_session_save() function works the same way as always.

10 years agoserial-dmm: Drop unused DMM_COUNT.
Uwe Hermann [Mon, 6 Jan 2014 19:53:12 +0000 (20:53 +0100)]
serial-dmm: Drop unused DMM_COUNT.

10 years agohardware/common/dmm: Fix debug output level.
Uwe Hermann [Mon, 6 Jan 2014 19:44:44 +0000 (20:44 +0100)]
hardware/common/dmm: Fix debug output level.

Most messages from the DMM parsers are not hard errors, lower to
sr_dbg() so that the sigrok-cli output doesn't get cluttered (by default)
with debug output such as:

 P1: 0.001100 V DC AUTO
 sr: fs9721: Sync nibble in byte 0 (0x00) is invalid.
 P1: 0.001100 V DC AUTO

(using -l 4 or -l 5 will still allow the user to see such messages)

10 years agoserial-dmm: Support the Tenma 72-7745 via UT-D02 cable.
Uwe Hermann [Mon, 6 Jan 2014 17:08:55 +0000 (18:08 +0100)]
serial-dmm: Support the Tenma 72-7745 via UT-D02 cable.

(it was already supported in uni-t-dmm via UT-D04 cable)

10 years agoAdd support for the Tenma 72-7750 (UNI-T UT60G rebadge).
Uwe Hermann [Mon, 6 Jan 2014 16:20:35 +0000 (17:20 +0100)]
Add support for the Tenma 72-7750 (UNI-T UT60G rebadge).

10 years agoAdd initial support for the UNI-T UT60G.
Uwe Hermann [Fri, 3 Jan 2014 21:32:16 +0000 (22:32 +0100)]
Add initial support for the UNI-T UT60G.

10 years agoREADME.devices: Cosmetics, consistency fixes, updates.
Uwe Hermann [Fri, 3 Jan 2014 23:18:29 +0000 (00:18 +0100)]
README.devices: Cosmetics, consistency fixes, updates.

10 years agoserial: Improved docs.
Matthias Heidbrink [Fri, 3 Jan 2014 13:37:39 +0000 (14:37 +0100)]
serial: Improved docs.

10 years agoscpi_usbtmc: fix reading of blocks bigger than the 2048 bytes buffer
Aurelien Jacobs [Fri, 3 Jan 2014 16:50:16 +0000 (17:50 +0100)]
scpi_usbtmc: fix reading of blocks bigger than the 2048 bytes buffer

10 years agorigol-ds: fix waveform reception on DS2000 series
Aurelien Jacobs [Wed, 1 Jan 2014 17:24:45 +0000 (18:24 +0100)]
rigol-ds: fix waveform reception on DS2000 series

The ":WAV:DATA?" scpi command must be sent before calling
sr_scpi_read_begin().

10 years agorigol-ds: add a few more DS2000 models to the supported list
Aurelien Jacobs [Wed, 1 Jan 2014 17:23:48 +0000 (18:23 +0100)]
rigol-ds: add a few more DS2000 models to the supported list

10 years agozeroplus: Add voltage threshold support
Russ Dill [Tue, 3 Dec 2013 07:29:04 +0000 (23:29 -0800)]
zeroplus: Add voltage threshold support

It doesn't currently mesh well with libsigrok, but at least the support is there.

10 years agozeroplus: Major rework of sample buffer processing
Russ Dill [Mon, 30 Dec 2013 15:12:49 +0000 (07:12 -0800)]
zeroplus: Major rework of sample buffer processing

The sample buffer is a still a bit of a mystery, but this should help.
The variables in play:

triggerbar/ramsize_trigger - These two variables added together indicate
how many samples we want captured. ramsize_trigger - triggerbar
indicades how many samples must be captured. The ratio between the two
is determined by capture ratio.

memory_size - This indicates the number of samples in the circular
capture buffer. stop_address, now_address, and trigger_address are
pointers within the zeroplus that wrap based on this size.

now_address - The address that the zeroplus was about to write to when
it finished capturing, and now the address that will be read from when
reads are done from the capture buffer

stop_address - The address that the zeroplus last wrote to when it
completed capture.

trigger_address - The sample address for which the trigger occured.

status - This one is a bit tricky. Some testing has shown that if the
zeroplus has captured memory_size or less samples, the STATUS_READY bit
is set. For all captures generated with more samples than this,
STATUS_READY was cleared. However, boundary conditions are difficult to
test and values such as, memory_size + 1 have not been tested. We use
this to determine if the capture has wrapped through the sample buffer.

More testing is required, but this improves behavior in a number of
cases, specifically capturing sample amounts that are not a power of 2
of the sample buffer size. Before, random data was passed to libsigrok.

Signed-off-by: Russ Dill <redacted>
10 years agozeroplus: Add getters for memory configuration
Russ Dill [Mon, 30 Dec 2013 15:04:42 +0000 (07:04 -0800)]
zeroplus: Add getters for memory configuration

This is needed at capture readback time to determine how many samples to read
in.

Signed-off-by: Russ Dill <redacted>
10 years agozeroplus: Modify analyzer_read_start to just prep for bulk reads.
Russ Dill [Wed, 1 Jan 2014 00:25:51 +0000 (16:25 -0800)]
zeroplus: Modify analyzer_read_start to just prep for bulk reads.

Let the capture loop manage which samples are thrown out rather
than throwing out two here.

Signed-off-by: Russ Dill <redacted>
10 years agozeroplus: Always set DONT_CARE_TRIGGERBAR to 1
Russ Dill [Mon, 30 Dec 2013 14:59:55 +0000 (06:59 -0800)]
zeroplus: Always set DONT_CARE_TRIGGERBAR to 1

Experimentation with the windows driver has found no situation where
this is set to anything other than 1. The zerominus software also never
sets this to anything other than one. Revert the code change made in
0ab0cb942f.

Signed-off-by: Russ Dill <redacted>
10 years agozeroplus: Ignore capture ratio if there is no trigger
Russ Dill [Tue, 3 Dec 2013 06:00:30 +0000 (22:00 -0800)]
zeroplus: Ignore capture ratio if there is no trigger

If there is no trigger, don't try to capture anything before it. There
won't be any because we trigger immediately.

Signed-off-by: Russ Dill <redacted>