From: Linus Torvalds torvalds@linux-foundation.org Date: Sat, 3 Jan 2015 16:25:19 -0800 Subject: [PATCH 4/4] Add EON Steel bookmarks, warnings and alarm events
I'd missed these because I'm blind. But mostly because they aren't as common.
Signed-off-by: Linus Torvalds torvalds@linux-foundation.org ---
There is no really good way to distinguish between "notification", "warning" and "alarm" in libdivecomputer. Sometimes there are sane middle grounds (ie a "notification" of a safety stop is would be the "voluntary" version, while a warning would be a "mandatory" one). But how to distinguish between a deep stop ceiling violation notification vs warning vs alarm?
Anyway, these translations really aren't wonderful, and some things aren'ty translated at all (ie they translate to SAMPLE_EVENT_NONE). But they make *some* amount of sense.
As usual, I'd be happier with a string-based model, rather than the crazy enumeration constants that don't make any sense and are too limited anyway.
But now I think I *really* have all the core EON Steel data translated.
src/suunto_eonsteel_parser.c | 101 +++++++++++++++++++++++++++++++++++++++++-- 1 file changed, 98 insertions(+), 3 deletions(-)
diff --git a/src/suunto_eonsteel_parser.c b/src/suunto_eonsteel_parser.c index a31c1a7e9b68..d5fe9f8ca7f2 100644 --- a/src/suunto_eonsteel_parser.c +++ b/src/suunto_eonsteel_parser.c @@ -229,6 +229,7 @@ struct sample_data { void *userdata; unsigned int time; unsigned char state_type, notify_type; + unsigned char warning_type, alarm_type; };
static void sample_time(struct sample_data *info, unsigned short time_delta) @@ -292,6 +293,17 @@ static void sample_cylinder_pressure(struct sample_data *info, unsigned char idx if (info->callback) info->callback(DC_SAMPLE_PRESSURE, sample, info->userdata); }
+static void sample_bookmark_event(struct sample_data *info, unsigned short idx) +{ + suunto_eonsteel_parser_t *eon = info->eon; + dc_sample_value_t sample = {0}; + + sample.event.type = SAMPLE_EVENT_BOOKMARK; + sample.event.value = idx; + + if (info->callback) info->callback(DC_SAMPLE_EVENT, sample, info->userdata); +} + static void sample_gas_switch_event(struct sample_data *info, unsigned short idx) { suunto_eonsteel_parser_t *eon = info->eon; @@ -312,8 +324,8 @@ static void sample_gas_switch_event(struct sample_data *info, unsigned short idx }
/* - * The EON Steel has two different sample events: "state" and "notification". - * Both end up having two fields: type and a boolean value. + * The EON Steel has four different sample events: "state", "notification", + * "warning" and "alarm". All end up having two fields: type and a boolean value. * * The type enumerations are available as part of the type descriptor, and we * *should* probably parse them dynamically, but this hardcodes the different @@ -363,7 +375,7 @@ static void sample_event_notify_value(struct sample_data *info, unsigned char va SAMPLE_EVENT_TISSUELEVEL, // 3=Tissue Level SAMPLE_EVENT_NONE, // 4=Deco SAMPLE_EVENT_NONE, // 5=Deco Window - SAMPLE_EVENT_SAFETYSTOP_MANDATORY, // 6=Safety Stop Ahead + SAMPLE_EVENT_SAFETYSTOP_VOLUNTARY, // 6=Safety Stop Ahead SAMPLE_EVENT_SAFETYSTOP, // 7=Safety Stop SAMPLE_EVENT_CEILING_SAFETYSTOP, // 8=Safety Stop Broken SAMPLE_EVENT_NONE, // 9=Deep Stop Ahead @@ -387,6 +399,74 @@ static void sample_event_notify_value(struct sample_data *info, unsigned char va }
+static void sample_event_warning_type(struct sample_data *info, unsigned char type) +{ + info->warning_type = type; +} + + +static void sample_event_warning_value(struct sample_data *info, unsigned char value) +{ + dc_sample_value_t sample = {0}; + static const enum parser_sample_event_t translate_warning[] = { + SAMPLE_EVENT_NONE, // 0=ICD Penalty ("Isobaric counterdiffusion") + SAMPLE_EVENT_VIOLATION, // 1=Deep Stop Penalty + SAMPLE_EVENT_SAFETYSTOP_MANDATORY, // 2=Mandatory Safety Stop + SAMPLE_EVENT_NONE, // 3=OTU250 + SAMPLE_EVENT_NONE, // 4=OTU300 + SAMPLE_EVENT_NONE, // 5=CNS80% + SAMPLE_EVENT_NONE, // 6=CNS100% + SAMPLE_EVENT_AIRTIME, // 7=Air Time + SAMPLE_EVENT_MAXDEPTH, // 8=Max.Depth + SAMPLE_EVENT_AIRTIME, // 9=Tank Pressure + SAMPLE_EVENT_CEILING_SAFETYSTOP, // 10=Safety Stop Broken + SAMPLE_EVENT_CEILING_SAFETYSTOP, // 11=Deep Stop Broken + SAMPLE_EVENT_CEILING, // 12=Ceiling Broken + SAMPLE_EVENT_PO2, // 13=PO2 High + }; + + if (info->warning_type > 13) + return; + + sample.event.type = translate_warning[info->warning_type]; + if (sample.event.type == SAMPLE_EVENT_NONE) + return; + + sample.event.value = value ? SAMPLE_FLAGS_BEGIN : SAMPLE_FLAGS_END; + if (info->callback) info->callback(DC_SAMPLE_EVENT, sample, info->userdata); +} + +static void sample_event_alarm_type(struct sample_data *info, unsigned char type) +{ + info->alarm_type = type; +} + + +static void sample_event_alarm_value(struct sample_data *info, unsigned char value) +{ + dc_sample_value_t sample = {0}; + static const enum parser_sample_event_t translate_alarm[] = { + SAMPLE_EVENT_CEILING_SAFETYSTOP, // 0=Mandatory Safety Stop Broken + SAMPLE_EVENT_ASCENT, // 1=Ascent Speed + SAMPLE_EVENT_NONE, // 2=Diluent Hyperoxia + SAMPLE_EVENT_VIOLATION, // 3=Violated Deep Stop + SAMPLE_EVENT_CEILING, // 4=Ceiling Broken + SAMPLE_EVENT_PO2, // 5=PO2 High + SAMPLE_EVENT_PO2, // 6=PO2 Low + }; + + if (info->alarm_type > 6) + return; + + sample.event.type = translate_alarm[info->alarm_type]; + if (sample.event.type == SAMPLE_EVENT_NONE) + return; + + sample.event.value = value ? SAMPLE_FLAGS_BEGIN : SAMPLE_FLAGS_END; + if (info->callback) info->callback(DC_SAMPLE_EVENT, sample, info->userdata); +} + + static int traverse_samples(unsigned short type, const struct type_desc *desc, const unsigned char *data, int len, void *user) { struct sample_data *info = (struct sample_data *) user; @@ -419,6 +499,21 @@ static int traverse_samples(unsigned short type, const struct type_desc *desc, c case 0x0016: sample_event_notify_value(info, data[0]); break; + case 0x0017: + sample_event_warning_type(info, data[0]); + break; + case 0x0018: + sample_event_warning_value(info, data[0]); + break; + case 0x0019: + sample_event_warning_type(info, data[0]); + break; + case 0x001a: + sample_event_warning_value(info, data[0]); + break; + case 0x001c: + sample_bookmark_event(info, array_uint16_le(data)); + break; case 0x001d: sample_gas_switch_event(info, array_uint16_le(data)); break;
On 04-01-15 01:34, Linus Torvalds wrote:
+static void sample_event_alarm_type(struct sample_data *info, unsigned char type) +{
- info->alarm_type = type;
+}
+static void sample_event_alarm_value(struct sample_data *info, unsigned char value) +{
- dc_sample_value_t sample = {0};
- static const enum parser_sample_event_t translate_alarm[] = {
SAMPLE_EVENT_CEILING_SAFETYSTOP, // 0=Mandatory Safety Stop BrokenSAMPLE_EVENT_ASCENT, // 1=Ascent SpeedSAMPLE_EVENT_NONE, // 2=Diluent HyperoxiaSAMPLE_EVENT_VIOLATION, // 3=Violated Deep StopSAMPLE_EVENT_CEILING, // 4=Ceiling BrokenSAMPLE_EVENT_PO2, // 5=PO2 HighSAMPLE_EVENT_PO2, // 6=PO2 Low- };
- if (info->alarm_type > 6)
return;- sample.event.type = translate_alarm[info->alarm_type];
- if (sample.event.type == SAMPLE_EVENT_NONE)
return;- sample.event.value = value ? SAMPLE_FLAGS_BEGIN : SAMPLE_FLAGS_END;
- if (info->callback) info->callback(DC_SAMPLE_EVENT, sample, info->userdata);
+}
These functions aren't used anywhere. Is that on purpose, or is there something missing in the patch?
Once this has been clarified, I'll commit your patches. Except for the dive mode and personal adjustment patches, because these two rely on the string api which isn't available yet.
Jef
On Jan 6, 2015 11:47 AM, "Jef Driesen" jef@libdivecomputer.org wrote:
These functions aren't used anywhere. Is that on purpose, or is there
something missing in the patch?
Not missing, no. Just too much cut-and-paste. The case statement with the actual callbacks does the "warning" callbacks twice. The second set should be "alarm" (ie cases 0x0019 and 0x001a).
Why didn't I notice the compiler warning? Hmm. Probably because the build system is so verbose. There is a reason why the kernel (and subsurface) builds try to be quiet. Or maybe the warning flags aren't complete enough. I'm mobile, so can't test.
Linus
On Tue, Jan 6, 2015 at 12:00 PM, Linus Torvalds torvalds@linux-foundation.org wrote:
Why didn't I notice the compiler warning? Hmm. Probably because the build system is so verbose. There is a reason why the kernel (and subsurface) builds try to be quiet. Or maybe the warning flags aren't complete enough. I'm mobile, so can't test.
Hmm. It's a combination of both. There's no warning, but it's also so verbose that had there been a warning I could easily see having missed it:
make all-recursive make[1]: Entering directory '/home/torvalds/src/libdc' Making all in include make[2]: Entering directory '/home/torvalds/src/libdc/include' Making all in libdivecomputer make[3]: Entering directory '/home/torvalds/src/libdc/include/libdivecomputer' make[3]: Nothing to be done for 'all'. make[3]: Leaving directory '/home/torvalds/src/libdc/include/libdivecomputer' make[3]: Entering directory '/home/torvalds/src/libdc/include' make[3]: Nothing to be done for 'all-am'. make[3]: Leaving directory '/home/torvalds/src/libdc/include' make[2]: Leaving directory '/home/torvalds/src/libdc/include' Making all in src make[2]: Entering directory '/home/torvalds/src/libdc/src' make all-am make[3]: Entering directory '/home/torvalds/src/libdc/src' CC suunto_eonsteel_parser.lo CCLD libdivecomputer.la make[3]: Leaving directory '/home/torvalds/src/libdc/src' make[2]: Leaving directory '/home/torvalds/src/libdc/src' Making all in examples make[2]: Entering directory '/home/torvalds/src/libdc/examples' CCLD universal CCLD ostc-fwupdate make[2]: Leaving directory '/home/torvalds/src/libdc/examples' make[2]: Entering directory '/home/torvalds/src/libdc' make[2]: Leaving directory '/home/torvalds/src/libdc' make[1]: Leaving directory '/home/torvalds/src/libdc'
It's hard seeing the important stuff with all that noise. But no, there was no actual warning about unused functions either.
I don't do autoconf (tool of the devil, as far as I'm concerned), so I have no idea hoqw to get rid of the annoying directory entry crap, and the appatent lack of -Wall for he compiler. Oh well.
Linus
On 06 January, 2015 - Linus Torvalds wrote:
It's hard seeing the important stuff with all that noise. But no, there was no actual warning about unused functions either.
I don't do autoconf (tool of the devil, as far as I'm concerned), so I have no idea hoqw to get rid of the annoying directory entry crap, and the appatent lack of -Wall for he compiler. Oh well.
Does this look better? : $ make make -s all-recursive Making all in include Making all in libdivecomputer Making all in src GEN revision.h CC version.lo CC suunto_eonsteel.lo CCLD libdivecomputer.la Making all in examples CCLD universal CCLD ostc-fwupdate
( its actually automake, but anyway.. )
A patch to make this happen is coming.
//Anton
On Tue, Jan 6, 2015 at 1:33 PM, Anton Lundin glance@acc.umu.se wrote:
Does this look better? :
Yes. The big thing is to make the normal non-warning lines short, so that warnings and errors (that are typically 40+ characters) stand out visually without having to actually read things. Your proposed output looks fine to me.
( its actually automake, but anyway.. )
Heh. I'm not tool-racist - I dislike all the auto-tools equally.
A patch to make this happen is coming.
Can we get "-Wall" added to CFLAGS by default too when using gcc? I didn't look how horrible the resulting warning mess is, but "-Wall" is normally a reasonably good default. I didn't even realize that it wasn't on until this whole thing came up, because I'm so used to it that I almost *expect* it to be enabled..
Linus
On 06 January, 2015 - Linus Torvalds wrote:
On Tue, Jan 6, 2015 at 1:33 PM, Anton Lundin glance@acc.umu.se wrote:
Does this look better? :
Yes. The big thing is to make the normal non-warning lines short, so that warnings and errors (that are typically 40+ characters) stand out visually without having to actually read things. Your proposed output looks fine to me.
( its actually automake, but anyway.. )
Heh. I'm not tool-racist - I dislike all the auto-tools equally.
A patch to make this happen is coming.
Can we get "-Wall" added to CFLAGS by default too when using gcc? I didn't look how horrible the resulting warning mess is, but "-Wall" is normally a reasonably good default. I didn't even realize that it wasn't on until this whole thing came up, because I'm so used to it that I almost *expect* it to be enabled..
You asked and here it comes.
//Anton
On 2015-01-06 21:00, Linus Torvalds wrote:
On Jan 6, 2015 11:47 AM, "Jef Driesen" jef@libdivecomputer.org wrote:
These functions aren't used anywhere. Is that on purpose, or is there something missing in the patch?
Not missing, no. Just too much cut-and-paste. The case statement with the actual callbacks does the "warning" callbacks twice. The second set should be "alarm" (ie cases 0x0019 and 0x001a).
I fixed the mistake and pushed the patches.
Jef
-
Anton Lundin -
Jef Driesen -
Linus Torvalds