[RERESEND] [PATCH 1/2] Distinguish HSDPA, HSUPA or both as access technologies

[RERESEND] [PATCH 1/2] Distinguish HSDPA, HSUPA or both as access technologies

[=?unknown-8bit?q?R=C3=A9mi?= Denis-Courmont]

[-- Attachment #1: Type: text/plain, Size: 593 bytes --]

---
 src/common.c |    4 ++--
 1 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/src/common.c b/src/common.c
index b5b9a6f..22bf1c2 100644
--- a/src/common.c
+++ b/src/common.c
@@ -661,9 +661,9 @@ const char *registration_tech_to_string(int tech)
 	case ACCESS_TECHNOLOGY_GSM_EGPRS:
 		return "edge";
 	case ACCESS_TECHNOLOGY_UTRAN_HSDPA:
-		return "hspa";
+		return "hsdpa";
 	case ACCESS_TECHNOLOGY_UTRAN_HSUPA:
-		return "hspa";
+		return "hsupa";
 	case ACCESS_TECHNOLOGY_UTRAN_HSDPA_HSUPA:
 		return "hspa";
 	case ACCESS_TECHNOLOGY_EUTRAN:
-- 
1.7.1

[RERESEND] [PATCH 2/2] Update the network documentation

[=?unknown-8bit?q?R=C3=A9mi?= Denis-Courmont]

[-- Attachment #1: Type: text/plain, Size: 625 bytes --]

---
 doc/network-api.txt |    5 +++--
 1 files changed, 3 insertions(+), 2 deletions(-)

diff --git a/doc/network-api.txt b/doc/network-api.txt
index 083e918..9fb9974 100644
--- a/doc/network-api.txt
+++ b/doc/network-api.txt
@@ -125,8 +125,9 @@ Properties	string Mode [readonly]
 
 			Contains the technology of the current network.
 
-			The possible values are: "gsm", "edge", "umts", "hspa",
-							"lte"
+			The possible values are: "gsm", "edge", "umts",
+			"hsdpa", "hsupa", "hspa" (meaning both hsdpa and hsupa)
+			and "lte"
 
 			TODO: Values for CDMA and EVDO based networks.
 
-- 
1.7.1

Re: [RERESEND] [PATCH 2/2] Update the network documentation

[Marcel Holtmann]

[-- Attachment #1: Type: text/plain, Size: 856 bytes --]

Hi Remi,

>  doc/network-api.txt |    5 +++--
>  1 files changed, 3 insertions(+), 2 deletions(-)
> 
> diff --git a/doc/network-api.txt b/doc/network-api.txt
> index 083e918..9fb9974 100644
> --- a/doc/network-api.txt
> +++ b/doc/network-api.txt
> @@ -125,8 +125,9 @@ Properties	string Mode [readonly]
>  
>  			Contains the technology of the current network.
>  
> -			The possible values are: "gsm", "edge", "umts", "hspa",
> -							"lte"
> +			The possible values are: "gsm", "edge", "umts",
> +			"hsdpa", "hsupa", "hspa" (meaning both hsdpa and hsupa)
> +			and "lte"

so what is the intended usage for showing this difference? When we
discussed this the last time, it was clear that we need to show a
difference between 3G (umts) and 3.5G (hspa) and where HSPA would be
either HSDPA or HSUPA or both.

Regards

Marcel

Re: [RERESEND] [PATCH 2/2] Update the network documentation

[=?unknown-8bit?q?R=C3=A9mi?= Denis-Courmont]

[-- Attachment #1: Type: text/plain, Size: 1184 bytes --]

On Tuesday 23 November 2010 16:46:14 ext Marcel Holtmann, you wrote:
> > -			The possible values are: "gsm", "edge", "umts", "hspa",
> > -							"lte"
> > +			The possible values are: "gsm", "edge", "umts",
> > +			"hsdpa", "hsupa", "hspa" (meaning both hsdpa and hsupa)
> > +			and "lte"
> 
> so what is the intended usage for showing this difference? When we
> discussed this the last time, it was clear that we need to show a
> difference between 3G (umts) and 3.5G (hspa) and where HSPA would be
> either HSDPA or HSUPA or both.

In some UI scenarios, we want to know the difference between those three states 
as they show downlink and uplink separately. This gives the flexibility to do 
so.

In fact, I have some cases where we'd need know whether HSDPA/HSUPA channels 
are actually allocated, not just whether the cell supports them. ISI does 
that, but there does not seem to be any standard way to get that informations 
out of the modem. In particular, I assume that both the CREG and CGREG 
technology indications are always equal :-(

Anyway, lets fix what's easy to fix...

-- 
Rémi Denis-Courmont
Nokia Devices R&D, Maemo Software, Helsinki

Re: [RERESEND] [PATCH 2/2] Update the network documentation

[Denis Kenzior]

[-- Attachment #1: Type: text/plain, Size: 699 bytes --]

Hi Remi,

> In fact, I have some cases where we'd need know whether HSDPA/HSUPA channels 
> are actually allocated, not just whether the cell supports them. ISI does 
> that, but there does not seem to be any standard way to get that informations 
> out of the modem. In particular, I assume that both the CREG and CGREG 
> technology indications are always equal :-(

Actually they're not.  Most AT modems report HSPA/HSDPA/HSUPA only on
the CGREG command.  CREG only reports umts/edge/gsm.

In fact we were planning to report the current bearer on the
ConnectionManager interface once we figured out the vendor commands that
can actually report this information.

Regards,
-Denis

Re: [RERESEND] [PATCH 2/2] Update the network documentation

[Aki Niemi]

[-- Attachment #1: Type: text/plain, Size: 763 bytes --]

Hi,

2010/11/23 Denis Kenzior <denkenz@gmail.com>:
> In fact we were planning to report the current bearer on the
> ConnectionManager interface once we figured out the vendor commands that
> can actually report this information.

Do you mean HSDPA/HSUPA channel allocations, or the tech reported by
CGREG? Because these can differ. For example, a Finnish operator's
cells will not indicate HS*PA support, but HSDPA channel will
nevertheless get allocated when data is moving. This is why for
instance the N900 will show a 3G icon when idle, but switch to 3.5G
when actively receiving data.

Also, we might need to expose CELL_DCH state at some point as well. I
was thinking this could be a property in the RadioSettings interface.

Cheers,
Aki

Re: [RERESEND] [PATCH 2/2] Update the network documentation

[Denis Kenzior]

[-- Attachment #1: Type: text/plain, Size: 815 bytes --]

Hi Aki,

On 11/23/2010 01:20 PM, Aki Niemi wrote:
> Hi,
> 
> 2010/11/23 Denis Kenzior <denkenz@gmail.com>:
>> In fact we were planning to report the current bearer on the
>> ConnectionManager interface once we figured out the vendor commands that
>> can actually report this information.
> 
> Do you mean HSDPA/HSUPA channel allocations, or the tech reported by
> CGREG? Because these can differ. For example, a Finnish operator's
> cells will not indicate HS*PA support, but HSDPA channel will
> nevertheless get allocated when data is moving. This is why for
> instance the N900 will show a 3G icon when idle, but switch to 3.5G
> when actively receiving data.
> 

I mean the actual bearer being used.  The 27.007 equivalent would be
+CPSB, not what is reported by +CGREG.

Regards,
-Denis

Re: [RERESEND] [PATCH 2/2] Update the network documentation

[=?unknown-8bit?q?R=C3=A9mi?= Denis-Courmont]

[-- Attachment #1: Type: text/plain, Size: 1203 bytes --]

On Tuesday 23 November 2010 23:45:07 ext Denis Kenzior, you wrote:
> On 11/23/2010 01:20 PM, Aki Niemi wrote:
> > Hi,
> > 
> > 2010/11/23 Denis Kenzior <denkenz@gmail.com>:
> >> In fact we were planning to report the current bearer on the
> >> ConnectionManager interface once we figured out the vendor commands that
> >> can actually report this information.
> > 
> > Do you mean HSDPA/HSUPA channel allocations, or the tech reported by
> > CGREG? Because these can differ. For example, a Finnish operator's
> > cells will not indicate HS*PA support, but HSDPA channel will
> > nevertheless get allocated when data is moving. This is why for
> > instance the N900 will show a 3G icon when idle, but switch to 3.5G
> > when actively receiving data.
> 
> I mean the actual bearer being used.  The 27.007 equivalent would be
> +CPSB, not what is reported by +CGREG.

I think we need both. I assume we should have two properties in the 
ConnectionManager, one for each.

But then it is questionable why the NetworkRegistration should exhibit 
anything other than 'gsm', 'umts' and 'lte' (and whatever-CDMA).

-- 
Rémi Denis-Courmont
Nokia Devices R&D, Maemo Software, Helsinki