Attention is currently required from: Angel Pons, Arthur Heymans, Felix Held, Krystian Hebel, Kyösti Mälkki, Michał Żygowski.
Patch set 17:Code-Review +1
11 comments:
File src/device/pciexp_device.c:
Patch Set #11, Line 598: awlays
Thanks for the update and your patience. […]
This ended up in the wrong thread, so closing here.
Patch Set #11, Line 630: root->max_payload_set = 1;
Ad.2. […]
Current implementation looks good to me. Krystian, Patrick, do you want to
have another look?
File src/device/pciexp_device.c:
Patch Set #15, Line 638: pciexp_dev_set_max_payload_size(child, max_payload);
Indeed, we only care about the MPS propagation upwards. Removed.
Acknowledged
if (max_payload != child_max_payload)
printk(BIOS_INFO, "%s: Max_Payload_Size adjusted to %d\n", dev_path(child),
(1 << (max_payload + 7)));
Removed.
Acknowledged
The whole function is not needed at all actually. […]
Acknowledged
Patch Set #15, Line 728: pciexp_dev_set_max_payload_size(dev, pciexp_dev_get_max_payload_size_cap(dev));
Yes, limited it to express endpoints and legacy endpoints.
Acknowledged
Patch Set #15, Line 740: pciexp_dev_set_max_payload_size(bus->dev, max_payload);
Right, not sure what I was thinking... Removed.
Acknowledged
File src/device/pciexp_device.c:
Nit, pciexp_dev_set_max_payload_size() might be a better place for this check.
Patch Set #17, Line 771: Paylaod
Payl*oa*d
Patch Set #17, Line 774: dvices
d*e*vices
Just a note: We could also have a special _scan_bus() function for root ports,
that would save us the `if`. I'm not sure how much work this would be and if
it would be really cleaner, so I'll leave it for me to look into later.
To view, visit change 77338. To unsubscribe, or for help writing mail filters, visit settings.