Skip to contentSkip to conversation

its_alloc_tables() - with all BASER marked none

DISCUSSION Mario Smarduch 2 participants 5 messages Oct 15, 2015 KVM arm64

lore
Share & tools
lore
#0 lore raw

If its_init()/its_probe()/its_alloc_tables() finds all GITS_BASER type none, it continues with ITS initialization.

Is there a reason to continue?

Started to look through this code recently - little confused here.

Thanks,
mario

#1 lore raw
Quoted 4 lines · Mario Smarduch

If its_init()/its_probe()/its_alloc_tables() finds all GITS_BASER type none, it continues with ITS initialization.

Is there a reason to continue?

Of course. There is nothing that *mandates* the ITS to request memory from the operating system. The HW could perfectly come with its own memory and not need anything at all.

Quoted 1 line

Started to look through this code recently - little confused here.

Only the beginning! ;-)

	M.
Signature · 1 line
Jazz is not dead. It just smells funny...
#2 lore raw
Quoted 6 lines · Marc Zyngier
Quoted 4 lines · Mario Smarduch

If its_init()/its_probe()/its_alloc_tables() finds all GITS_BASER type none, it continues with ITS initialization.

Is there a reason to continue?

Quoted 7 lines

Of course. There is nothing that *mandates* the ITS to request memory from the operating system. The HW could perfectly come with its own memory and not need anything at all.

Quoted 1 line

Started to look through this code recently - little confused here.

Only the beginning! ;-)

Well at this time the beginning of console freeze close to "ITS: using cache flushing for cmd queue". Poked around CBASER and noticed Inner cache bits are updated but not sticky so it got me wondering.

Appears like I need to look elsewhere, like maybe my environment.

Thanks for clarifying.

- Mario

Quoted 3 lines
	M.
#3 lore raw
Quoted 15 lines · Mario Smarduch
Quoted 12 lines · Marc Zyngier
Quoted 4 lines · Mario Smarduch

If its_init()/its_probe()/its_alloc_tables() finds all GITS_BASER type none, it continues with ITS initialization.

Is there a reason to continue?

Of course. There is nothing that *mandates* the ITS to request memory from the operating system. The HW could perfectly come with its own memory and not need anything at all.

Quoted 1 line

Started to look through this code recently - little confused here.

Quoted 7 lines
Quoted 1 line

Only the beginning! ;-)

Well at this time the beginning of console freeze close to "ITS: using cache flushing for cmd queue". Poked around CBASER and noticed Inner cache bits are updated but not sticky so it got me wondering.

Appears like I need to look elsewhere, like maybe my environment.

Out of curiosity, is that on a FastModel? Or on some actual HW?

	M.
Signature · 1 line
Jazz is not dead. It just smells funny...
#4 lore raw
Quoted 18 lines · Marc Zyngier
Quoted 17 lines · Mario Smarduch
Quoted 13 lines · Marc Zyngier
Quoted 4 lines · Mario Smarduch

If its_init()/its_probe()/its_alloc_tables() finds all GITS_BASER type none, it continues with ITS initialization.

Is there a reason to continue?

Of course. There is nothing that *mandates* the ITS to request memory from the operating system. The HW could perfectly come with its own memory and not need anything at all.

Quoted 1 line

Started to look through this code recently - little confused here.

Only the beginning! ;-)

Quoted 7 lines
Quoted 5 lines

Well at this time the beginning of console freeze close to "ITS: using cache flushing for cmd queue". Poked around CBASER and noticed Inner cache bits are updated but not sticky so it got me wondering.

Appears like I need to look elsewhere, like maybe my environment.

Out of curiosity, is that on a FastModel? Or on some actual HW?

FastModel, license ran out on FVP.

Quoted 3 lines
	M.