Re: [PATCH v2 00/10] irqchip/gic-v5: Fix misc IRS and ITS host driver issues

From: Sascha Bischoff

Date: Wed Nov 18 2026 - 09:59:07 EST


Hi Fuad,

On 16/11/2026 14:06, Lorenzo Pieralisi wrote:
On Mon, Nov 16, 2026 at 01:28:56PM +0000, Fuad Tabba wrote:
Hi Lorenzo,

On Mon, 16 Nov 2026 12:14:50 +0000, Lorenzo Pieralisi
<lpieralisi@xxxxxxxxxx> wrote:
[...]
invalidate have the same shape. The question is, can an IRS miss the
poll, or is a timeout here only ever a dead IRS?

Thank you for your patches Fuad.

Seconded. Thanks a lot.

I'm more than happy to include "irqchip/gic-v5: Clean a level 2 device table before making it valid" in v3!


The latter, though freeing memory on timeout wasn't a good idea given
that there is no way to cancel the respective, pending, mapping.

Thanks, that answers it. If you'd like a patch that doesn't free the
memory on a timeout, let me know and I'll send one.

Thanks. Again, if that timeout expires I think we have more serious
problems than the IST mapping but I do agree that freeing memory
is not the answer; not urgent - please send a patch given that
you spotted it, can obviously do it too.

I've got some patches addressing the ITT and IST mapping issues in the host driver, which I can include in v3 if you haven't already started working on these. I'll make sure to include the appropriate Reported-by tags, of course!

I'll post a fix for the VMT separately as it doesn't belong in this series.



[...]
Notes:
Setting L1 on every invalidation turns each registration and
teardown into a range invalidate, both rare paths, and keeps the
function simple. Setting it only when registration allocates the
level 2 table would miss a registration that fails after the
allocation, which leaves the level 1 DTE valid.

I can't follow. I agree that we have to invalidate the L1 DTE on
L2 allocation but that's a one-off, I don't understand what issues
you are pointing out in the Notes above, if you can point out please
the code paths it refers to that'd help.

gicv5_its_device_register() can fail after
gicv5_its_devtab_get_dte_ref() has made the L1 DTE valid, and before
the gicv5_its_device_cache_inv() at the end. That happens if the
EventID bits check fails, or the ITT allocation does. The L1 DTE is
then already valid for the next device registered under it, so that
registration doesn't allocate a level 2 table.

The patch Notes were about setting ITS_INV_DEVICER.L1 in that
gicv5_its_device_cache_inv() call only when the registration allocated
the table. In that case the failed registration never reaches that
call and the next one reaches it with L1 clear, so this L1 DTE is
never invalidated. Invalidating in gicv5_its_alloc_l2_devtab(), right
after the L1 DTE write, is the one-off you describe. I'll rework the
patch to do that instead.

Yes, I think that's what we should do - invalidating in
gicv5_its_alloc_l2_devtab() once for all and leave everything else as-is
matches architecture guidelines.

Agreed. I've got a reworked fix for this too, which I can include in v3.


Thanks again.

Lorenzo

Thanks,
Sascha