This lets modem get a bit closer to working

[   15.885735] ipa 1e40000.ipa: channel 4 limited to 256 TREs
[   15.888550] ipa 1e40000.ipa: IPA driver initialized
[   21.038968] ipa 1e40000.ipa: received modem starting event
[   23.097401] ipa 1e40000.ipa: received modem running event

Signed-off-by: Paul Sajna <sajattack@postmarketos.org>
---
 arch/arm64/boot/dts/qcom/sdm845-lg-common.dtsi | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/arm64/boot/dts/qcom/sdm845-lg-common.dtsi b/arch/arm64/boot/dts/qcom/sdm845-lg-common.dtsi
index f309e6ebc075a691c7a522238b4a93ba9e63f3c0..898f87818d3d4d028cc762d7ffdf33b3c2352c92 100644
--- a/arch/arm64/boot/dts/qcom/sdm845-lg-common.dtsi
+++ b/arch/arm64/boot/dts/qcom/sdm845-lg-common.dtsi
@@ -484,7 +484,7 @@ zap-shader {
 };
 
 &ipa {
-	qcom,gsi-loader = "modem";
+	qcom,gsi-loader = "self";
 	memory-region = <&ipa_fw_mem>;
 
 	status = "okay";

-- 
2.51.0

On Sun, Sep 28, 2025 at 10:05:34PM -0700, Paul Sajna wrote:
> This lets modem get a bit closer to working
> 
> [   15.885735] ipa 1e40000.ipa: channel 4 limited to 256 TREs
> [   15.888550] ipa 1e40000.ipa: IPA driver initialized
> [   21.038968] ipa 1e40000.ipa: received modem starting event
> [   23.097401] ipa 1e40000.ipa: received modem running event
> 
> Signed-off-by: Paul Sajna <sajattack@postmarketos.org>
> ---
>  arch/arm64/boot/dts/qcom/sdm845-lg-common.dtsi | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/arch/arm64/boot/dts/qcom/sdm845-lg-common.dtsi b/arch/arm64/boot/dts/qcom/sdm845-lg-common.dtsi
> index f309e6ebc075a691c7a522238b4a93ba9e63f3c0..898f87818d3d4d028cc762d7ffdf33b3c2352c92 100644
> --- a/arch/arm64/boot/dts/qcom/sdm845-lg-common.dtsi
> +++ b/arch/arm64/boot/dts/qcom/sdm845-lg-common.dtsi
> @@ -484,7 +484,7 @@ zap-shader {
>  };
>  
>  &ipa {
> -	qcom,gsi-loader = "modem";
> +	qcom,gsi-loader = "self";

The commit message might need to be improved. The patch itself LGTM.

>  	memory-region = <&ipa_fw_mem>;
>  
>  	status = "okay";
> 
> -- 
> 2.51.0
> 

-- 
With best wishes
Dmitry

On 9/29/25 7:05 AM, Paul Sajna wrote:
> This lets modem get a bit closer to working

What prevents it from working? Does it crash? Has rmtfs started?
Does `rmtfs -vP` print any warnings?

Konrad

October 1, 2025 at 9:11 AM, "Konrad Dybcio" <konrad.dybcio@oss.qualcomm.com mailto:konrad.dybcio@oss.qualcomm.com?to=%22Konrad%20Dybcio%22%20%3Ckonrad.dybcio%40oss.qualcomm.com%3E > wrote:


> 
> On 9/29/25 7:05 AM, Paul Sajna wrote:
> 
> > 
> > This lets modem get a bit closer to working
> > 
> What prevents it from working? Does it crash? Has rmtfs started?
> Does `rmtfs -vP` print any warnings?
> 
> Konrad
>

<msg> [modem0] state changed (unknown -> locked)
<wrn> [modem0] modem couldn't be initialized: Couldn't check unlock status: Couldn't get SIM lock status after 30 retries
<msg> [modem0] state changed (locked -> failed)
<wrn> [modem0] couldn't load SUPL server: Failed to receive indication with the current server settings
<wrn> [modem0] couldn't load supported assistance data types: Failed to receive indication with the predicted orbits data source
<wrn> [modem0] couldn't build device ids: Unknown device subsystem
<wrn> [device qcom-soc] error initializing: Modem in failed state: sim-missing

On 11/25/25 5:29 AM, Paul Sajna wrote:
> October 1, 2025 at 9:11 AM, "Konrad Dybcio" <konrad.dybcio@oss.qualcomm.com mailto:konrad.dybcio@oss.qualcomm.com?to=%22Konrad%20Dybcio%22%20%3Ckonrad.dybcio%40oss.qualcomm.com%3E > wrote:
> 
> 
>>
>> On 9/29/25 7:05 AM, Paul Sajna wrote:
>>
>>>
>>> This lets modem get a bit closer to working
>>>
>> What prevents it from working? Does it crash? Has rmtfs started?
>> Does `rmtfs -vP` print any warnings?
>>
>> Konrad
>>
> 
> <msg> [modem0] state changed (unknown -> locked)
> <wrn> [modem0] modem couldn't be initialized: Couldn't check unlock status: Couldn't get SIM lock status after 30 retries
> <msg> [modem0] state changed (locked -> failed)
> <wrn> [modem0] couldn't load SUPL server: Failed to receive indication with the current server settings
> <wrn> [modem0] couldn't load supported assistance data types: Failed to receive indication with the predicted orbits data source
> <wrn> [modem0] couldn't build device ids: Unknown device subsystem
> <wrn> [device qcom-soc] error initializing: Modem in failed state: sim-missing

IDK if it's super related, but some phones have additional mechanisms
for detecting (particulary the secondary) SIM - any chance yours is a
dual-SIM and you can try using the other slot?

Konrad

December 1, 2025 at 12:54 PM, "Konrad Dybcio" <konrad.dybcio@oss.qualcomm.com mailto:konrad.dybcio@oss.qualcomm.com?to=%22Konrad%20Dybcio%22%20%3Ckonrad.dybcio%40oss.qualcomm.com%3E > wrote:

> IDK if it's super related, but some phones have additional mechanisms
> for detecting (particulary the secondary) SIM - any chance yours is a
> dual-SIM and you can try using the other slot?
> 
> Konrad
>

No such luck. 

I think the device might have a second sim slot shared with the sdcard (if I put it sideways in the sdcard tray, there seems to be a bit of a groove there).

But the sim is detected by qmicli in the main slot I was using before, ModemManager just can't tell if it's locked or not.

The SIM is not detected at all in the sideways sdcard slot.

On 12/2/25 3:52 AM, Paul Sajna wrote:
> December 1, 2025 at 12:54 PM, "Konrad Dybcio" <konrad.dybcio@oss.qualcomm.com mailto:konrad.dybcio@oss.qualcomm.com?to=%22Konrad%20Dybcio%22%20%3Ckonrad.dybcio%40oss.qualcomm.com%3E > wrote:
> 
>> IDK if it's super related, but some phones have additional mechanisms
>> for detecting (particulary the secondary) SIM - any chance yours is a
>> dual-SIM and you can try using the other slot?
>>
>> Konrad
>>
> 
> No such luck. 
> 
> I think the device might have a second sim slot shared with the sdcard (if I put it sideways in the sdcard tray, there seems to be a bit of a groove there).
> 
> But the sim is detected by qmicli in the main slot I was using before, ModemManager just can't tell if it's locked or not.
> 
> The SIM is not detected at all in the sideways sdcard slot.

I'm afraid I can't help then. The modem is online and probably expects
some different handling from MM and/or vendor specific magic

Konrad