Thanks for the reply. It really pointed me in the right way.

The inmate didn't crash when issuing the first mmio access, but it was 
because I had overlapping memory regions.
My AXI GPIO address was the same as the communication area.
I changed the hardware design in Vivado to get different addresses and 
then, as you said, I needed the "map_range" call.
It all worked as expected now.

Thank you very much.

El sábado, 20 de julio de 2019, 4:44:56 (UTC+2), Jan Kiszka escribió:
>
> On 20.07.19 00:58, Alejandro Largacha wrote: 
> > I just tried and did not work. I don't know what I'm missing 
> > I attach the config and the demo code 
> > 
>
> Config looks good, but your inmate should crash when issuing the first 
> mmio 
> accesses because you didn't map them into the inmate address space. Try 
> adding 
>
> map_range(GPIO_SWITCH_BASE_ADDRESS, 0x2000, MAP_UNCACHED); 
>
> before the first access. 
>
> Jan 
>
> > El viernes, 19 de julio de 2019, 7:08:41 (UTC+2), Jan Kiszka escribió: 
> > 
> >     On 19.07.19 01:32, Alejandro Largacha wrote: 
> >     > Hello, 
> >     > 
> >     > Thank you very much for the reply. I tried adding the irqchip 
> entry to the 
> >     cell config but no success. I have a doubt about the pin_bitmap 
> entry. Is it 
> >     divided in 4 groups of 32 bits? Then, how should be the entry like 
> for the 
> >     irq num 136. Like this? 
> >     > 
> >     > .irqchips = { 
> >     >            /* GIC */ { 
> >     >                .address = 0xf9010000, 
> >     >                .pin_base = 32, 
> > 
> >     This means bit 0 in the first word of pin_bitmap encodes GIC 
> interrupt 32. 
> >     Then, 
> >     when that bit is set, permission is granted. 
> > 
> >     >                .pin_bitmap = { 
> >     >                    1 << (54 - 32), 
> > 
> >     So this one grants access to GIC int 54 - not sure if you need that. 
> > 
> >     >                    0, 
> >     >                    0, 
> >     >                    1 << (136 - 128) 
> > 
> >     And this permits access to int 136 because this 4th word controls 
> interrupts 
> >     128-159. 
> > 
> >     Did you already try out the config? 
> > 
> >     Jan 
> > 
> >     >                }, 
> >     >            }, 
> >     >        } 
> >     > 
> >     > 
> >     > I attach my new cell config file. 
> >     > 
> >     > El miércoles, 10 de julio de 2019, 7:20:56 (UTC+2), Jan Kiszka 
>  escribió: 
> >     >> On 10.07.19 06:57, Alejandro Largacha wrote: 
> >     >>> Hello, 
> >     >>> 
> >     >>> I have been paying with jailhouse in a UltraZed SoM with 
> AES-ZU-IOCC-G 
> >     carrier 
> >     >>> card from avnet. 
> >     >>> So far, I have been able to enable the root cell, create a 
> aremetal cell 
> >     where I 
> >     >>> was able to load the baremetal 
> >     >>> examples like gic-demo and a custom one where I was able to turn 
> on some 
> >     leds 
> >     >>> via axi gpio in PL side. 
> >     >>> 
> >     >>> Now I'm trying to run an example where I can catch the interrupt 
> from an 
> >     axi 
> >     >>> gpio in PL and I am not able to do that. 
> >     >>> I'm using PL to PS interrupts in my design and the signal is 
> being 
> >     generated. I 
> >     >>> also tested with Xilinx SDK. 
> >     >>> I tried with PS to PL group 0 and group 1. Irq numbers 121 and 
> 136 and no 
> >     >>> success. I don't know what I'm missing. 
> >     >>> I attach the cells and the demo source. 
> >     >>> 
> >     >> 
> >     >> Your baremetal cell config is not permitting the cell access to 
> the GIC 
> >     >> interrupt 136 you are using. Replicate the irqchip entry and 
> create bitmask 
> >     >> where only bit 136 - 32 (32 is the base) is set. See also other 
> examples 
> >     >> in-tree, including those for the zcu102 or the ultra96. Both 
> grant their 
> >     >> non-root linux cells access to certain interrupts. 
> >     >> 
> >     >> Jan 
> >     >> 
> >     >> -- 
> >     >> Siemens AG, Corporate Technology, CT RDA IOT SES-DE 
> >     >> Corporate Competence Center Embedded Linux 
> >     > 
> > 
> >     -- 
> >     Siemens AG, Corporate Technology, CT RDA IOT SES-DE 
> >     Corporate Competence Center Embedded Linux 
> > 
> > -- 
> > You received this message because you are subscribed to the Google 
> Groups 
> > "Jailhouse" group. 
> > To unsubscribe from this group and stop receiving emails from it, send 
> an email 
> > to [email protected] <javascript:> 
> > <mailto:[email protected] <javascript:>>. 
> > To view this discussion on the web visit 
> > 
> https://groups.google.com/d/msgid/jailhouse-dev/1fcf087a-14f5-4b96-b95b-bbbd9dafc9b8%40googlegroups.com
>  
> > <
> https://groups.google.com/d/msgid/jailhouse-dev/1fcf087a-14f5-4b96-b95b-bbbd9dafc9b8%40googlegroups.com?utm_medium=email&utm_source=footer>.
>  
>
>

-- 
You received this message because you are subscribed to the Google Groups 
"Jailhouse" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jailhouse-dev/b4070ff0-8c14-4153-9e35-f82884fb8ae4%40googlegroups.com.

Reply via email to