Searched hist:bc9402abdd11630ced33cbecb79b0d72f203f08a (Results 1 – 3 of 3) sorted by relevance
| /freebsd/sys/dev/igc/ |
| H A D | igc_txrx.c | diff bc9402abdd11630ced33cbecb79b0d72f203f08a Fri Oct 11 03:20:13 CEST 2024 Kevin Bowling <kbowling@FreeBSD.org> igc: Add AIM
igc is derived from igb and has never had an AIM implementation. The
same algorithm from e1000 is appropriate here.
Upon more detailed study of the Linux driver which has a newer AIM
implementation, it finally became clear to me this is actually a
holdoff timer and not an interrupt limit as it is conventionally
(statically) programmed and displayed as an interrupt rate. The data
sheets also make this somewhat clear.
Thus, AIM accomplishes two beneficial things for a wide variety of
workloads[1]:
1. At low throughput/packet rates, it will significantly lower latency
(by counter-intuitively "increasing" the interrupt rate.. better
thought of as decreasing the holdoff timer because you will modulate
down before coming anywhere near these interrupt rates).
2. At bulk data rates, it is tuned to achieve a lower interrupt rate
(by increasing the holdoff timer) than the current static 8000/s. This
decreases processing overhead and yields more headroom for other work
such as packet filters or userland.
For a single NIC this might be worth a few sys% on common CPUs, but may
be meaningful when multiplied such as if_lagg, if_bridge and forwarding
setups.
The AIM algorithm was re-introduced from the older igb or out of tree
driver, and then modernized with permission to use Intel code from other
drivers.
[1]: http://iommu.com/datasheets/ethernet/controllers-nics/intel/e1000/gbe-controllers-interrupt-moderation-appl-note.pdf
MFC after: 1 week
Relnotes: yes
Sponsored by: Rubicon Communications, LLC ("Netgate")
Sponsored by: BBOX.io
Differential Revision: https://reviews.freebsd.org/D47053
|
| H A D | if_igc.h | diff bc9402abdd11630ced33cbecb79b0d72f203f08a Fri Oct 11 03:20:13 CEST 2024 Kevin Bowling <kbowling@FreeBSD.org> igc: Add AIM
igc is derived from igb and has never had an AIM implementation. The
same algorithm from e1000 is appropriate here.
Upon more detailed study of the Linux driver which has a newer AIM
implementation, it finally became clear to me this is actually a
holdoff timer and not an interrupt limit as it is conventionally
(statically) programmed and displayed as an interrupt rate. The data
sheets also make this somewhat clear.
Thus, AIM accomplishes two beneficial things for a wide variety of
workloads[1]:
1. At low throughput/packet rates, it will significantly lower latency
(by counter-intuitively "increasing" the interrupt rate.. better
thought of as decreasing the holdoff timer because you will modulate
down before coming anywhere near these interrupt rates).
2. At bulk data rates, it is tuned to achieve a lower interrupt rate
(by increasing the holdoff timer) than the current static 8000/s. This
decreases processing overhead and yields more headroom for other work
such as packet filters or userland.
For a single NIC this might be worth a few sys% on common CPUs, but may
be meaningful when multiplied such as if_lagg, if_bridge and forwarding
setups.
The AIM algorithm was re-introduced from the older igb or out of tree
driver, and then modernized with permission to use Intel code from other
drivers.
[1]: http://iommu.com/datasheets/ethernet/controllers-nics/intel/e1000/gbe-controllers-interrupt-moderation-appl-note.pdf
MFC after: 1 week
Relnotes: yes
Sponsored by: Rubicon Communications, LLC ("Netgate")
Sponsored by: BBOX.io
Differential Revision: https://reviews.freebsd.org/D47053
|
| H A D | if_igc.c | diff bc9402abdd11630ced33cbecb79b0d72f203f08a Fri Oct 11 03:20:13 CEST 2024 Kevin Bowling <kbowling@FreeBSD.org> igc: Add AIM
igc is derived from igb and has never had an AIM implementation. The
same algorithm from e1000 is appropriate here.
Upon more detailed study of the Linux driver which has a newer AIM
implementation, it finally became clear to me this is actually a
holdoff timer and not an interrupt limit as it is conventionally
(statically) programmed and displayed as an interrupt rate. The data
sheets also make this somewhat clear.
Thus, AIM accomplishes two beneficial things for a wide variety of
workloads[1]:
1. At low throughput/packet rates, it will significantly lower latency
(by counter-intuitively "increasing" the interrupt rate.. better
thought of as decreasing the holdoff timer because you will modulate
down before coming anywhere near these interrupt rates).
2. At bulk data rates, it is tuned to achieve a lower interrupt rate
(by increasing the holdoff timer) than the current static 8000/s. This
decreases processing overhead and yields more headroom for other work
such as packet filters or userland.
For a single NIC this might be worth a few sys% on common CPUs, but may
be meaningful when multiplied such as if_lagg, if_bridge and forwarding
setups.
The AIM algorithm was re-introduced from the older igb or out of tree
driver, and then modernized with permission to use Intel code from other
drivers.
[1]: http://iommu.com/datasheets/ethernet/controllers-nics/intel/e1000/gbe-controllers-interrupt-moderation-appl-note.pdf
MFC after: 1 week
Relnotes: yes
Sponsored by: Rubicon Communications, LLC ("Netgate")
Sponsored by: BBOX.io
Differential Revision: https://reviews.freebsd.org/D47053
|