This series is against the latest tip tree.
Currently multiple MSI mode is limited to a single vector per device (at
least on x86 and PPC). This series breathes life into pci_enable_msi_block()
and makes it possible to set interrupt affinity for multiple IRQs, similarly
to MSI-X. Yet, only for x86 and only when IOMMUs are present.
Although IRQ and PCI subsystems are modified, the current behaviour left
intact. The drivers could just start using multiple MSIs just by following
the existing documentation.
The AHCI device driver makes use of the new mode and a new function -
pci_enable_msi_block_auto() - patches 4,5.
Changes since v1:
4/5: based on Bjorn's review pci_enable_msi_block_auto() returns
the number of interrupts allocated or error code
5/5: based on Jeff's comments:
- AHCI specific code moved from libata-core.c to libahci.c;
- the host-wide lock concern addressed - ahci_interrupt() handler
split in two parts: ahci_hw_interrupt() hardware context handler
and ahci_thread_fn() threaded handler;
- host->lock statistics attached;
Host-wide lock statistics summary:
Taken using command 'if=/dev/sd{a,b,c} of=/dev/null' running in parallel
on three SATA HDDs:
ata1: SATA link up 1.5 Gbps (SStatus 113 SControl 300)
ata2: SATA link up 1.5 Gbps (SStatus 113 SControl 300)
ata3: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
ahci_interrupt()
Capabilities: [80] MSI: Enable+ Count=1/16 Maskable- 64bit-
Test holdtime-total waittime-total acquis. holdtime-avg waittime-avg
#
01. 22565801.27 93056.41 6377094 3.538571216 0.014592291
02. 20358324.12 60825.47 6411012 3.175524257 0.009487655
03. 21190730.38 63610.38 6384508 3.319085884 0.009963239
04. 22491064.54 80624.96 6398390 3.515113105 0.012600820
05. 21986193.10 78034.38 6379092 3.446602291 0.012232835
06. 20556437.70 62314.64 6287673 3.269323596 0.009910604
07. 20477137.55 66190.92 6388507 3.205308776 0.010360937
08. 21442240.03 69306.80 6402109 3.349246323 0.010825620
----------- -----------
avg 3.352346931 0.011246750
ahci_hw_interrupt()
Capabilities: [80] MSI: Enable+ Count=16/16 Maskable- 64bit-
Test holdtime-total waittime-total acquis. holdtime-avg waittime-avg
#
09. 8936643.32 54559.79 6505606 1.373683454 0.008386581
10. 8579972.36 51496.56 6496233 1.320761180 0.007927142
11. 8138708.47 54061.46 6494647 1.253141005 0.008324003
12. 8322068.81 60427.51 6509975 1.278356493 0.009282295
13. 8527745.18 65978.83 6515589 1.308821839 0.010126303
14. 8662461.99 57291.39 6510702 1.330495850 0.008799572
15. 8054911.26 69186.41 6514262 1.236504037 0.010620759
16. 9618631.17 39726.37 6517385 1.475842101 0.006095446
----------- -----------
avg 1.322200745 0.008695263
Attached patches:
1/5: x86, MSI: Support multiple MSIs in presense of IRQ remapping
2/5: x86, MSI: Allocate as many multiple IRQs as requested
3/5: x86, MSI: Minor readability fixes
4/5: PCI, MSI: Enable multiple MSIs with pci_enable_msi_block_auto()
5/5: AHCI: Support multiple MSIs
Documentation/PCI/MSI-HOWTO.txt | 37 ++++++-
arch/x86/kernel/apic/io_apic.c | 170 +++++++++++++++++++++++++++++--
drivers/ata/ahci.c | 14 ++-
drivers/ata/ahci.h | 5 +
drivers/ata/libahci.c | 211 +++++++++++++++++++++++++++++++++++++-
drivers/pci/msi.c | 36 ++++++-
include/linux/irq.h | 6 +
include/linux/msi.h | 1 +
include/linux/pci.h | 7 ++
kernel/irq/chip.c | 30 ++++--
kernel/irq/irqdesc.c | 31 ++++++
11 files changed, 509 insertions(+), 39 deletions(-)
--
1.7.7.6
--
Regards,
Alexander Gordeev
agordeev@xxxxxxxxxx
ahci_interrupt()
Capabilities: [80] MSI: Enable+ Count=1/16 Maskable- 64bit-
Test holdtime-total waittime-total acquis. holdtime-avg waittime-avg
#
01. 22565801.27 93056.41 6377094 3.538571216 0.014592291
02. 20358324.12 60825.47 6411012 3.175524257 0.009487655
03. 21190730.38 63610.38 6384508 3.319085884 0.009963239
04. 22491064.54 80624.96 6398390 3.515113105 0.012600820
05. 21986193.10 78034.38 6379092 3.446602291 0.012232835
06. 20556437.70 62314.64 6287673 3.269323596 0.009910604
07. 20477137.55 66190.92 6388507 3.205308776 0.010360937
08. 21442240.03 69306.80 6402109 3.349246323 0.010825620
----------- -----------
avg 3.352346931 0.011246750
ahci_hw_interrupt()
Capabilities: [80] MSI: Enable+ Count=16/16 Maskable- 64bit-
Test holdtime-total waittime-total acquis. holdtime-avg waittime-avg
#
09. 8936643.32 54559.79 6505606 1.373683454 0.008386581
10. 8579972.36 51496.56 6496233 1.320761180 0.007927142
11. 8138708.47 54061.46 6494647 1.253141005 0.008324003
12. 8322068.81 60427.51 6509975 1.278356493 0.009282295
13. 8527745.18 65978.83 6515589 1.308821839 0.010126303
14. 8662461.99 57291.39 6510702 1.330495850 0.008799572
15. 8054911.26 69186.41 6514262 1.236504037 0.010620759
16. 9618631.17 39726.37 6517385 1.475842101 0.006095446
----------- -----------
avg 1.322200745 0.008695263
class name con-bounces contentions waittime-min waittime-max waittime-total acq-bounces acquisitions holdtime-min holdtime-max holdtime-total
Test #1
&(&host->lock)->rlock: 41558 41558 0.13 126.84 93056.41 5141195 6377094 0.42 197.27 22565801.27
---------------------
&(&host->lock)->rlock 36643 [<ffffffff81416b03>] ata_scsi_queuecmd+0x33/0x290
&(&host->lock)->rlock 4915 [<ffffffff814283e4>] ahci_interrupt+0x64/0x120
---------------------
&(&host->lock)->rlock 25017 [<ffffffff81416b03>] ata_scsi_queuecmd+0x33/0x290
&(&host->lock)->rlock 16541 [<ffffffff814283e4>] ahci_interrupt+0x64/0x120
Test #2
&(&host->lock)->rlock: 30714 30714 0.14 104.50 60825.47 4398165 6411012 0.23 176.84 20358324.12
---------------------
&(&host->lock)->rlock 25858 [<ffffffff81416b03>] ata_scsi_queuecmd+0x33/0x290
&(&host->lock)->rlock 4856 [<ffffffff814283e4>] ahci_interrupt+0x64/0x120
---------------------
&(&host->lock)->rlock 18199 [<ffffffff81416b03>] ata_scsi_queuecmd+0x33/0x290
&(&host->lock)->rlock 12515 [<ffffffff814283e4>] ahci_interrupt+0x64/0x120
Test #3
&(&host->lock)->rlock: 32869 32870 0.14 106.30 63610.38 4655399 6384508 0.43 214.18 21190730.38
---------------------
&(&host->lock)->rlock 27494 [<ffffffff81416b03>] ata_scsi_queuecmd+0x33/0x290
&(&host->lock)->rlock 5376 [<ffffffff814283e4>] ahci_interrupt+0x64/0x120
---------------------
&(&host->lock)->rlock 19632 [<ffffffff81416b03>] ata_scsi_queuecmd+0x33/0x290
&(&host->lock)->rlock 13238 [<ffffffff814283e4>] ahci_interrupt+0x64/0x120
Test #4
&(&host->lock)->rlock: 37426 37426 0.13 104.95 80624.96 5009173 6398390 0.38 189.14 22491064.54
---------------------
&(&host->lock)->rlock 6162 [<ffffffff814283e4>] ahci_interrupt+0x64/0x120
&(&host->lock)->rlock 31264 [<ffffffff81416b03>] ata_scsi_queuecmd+0x33/0x290
---------------------
&(&host->lock)->rlock 22271 [<ffffffff81416b03>] ata_scsi_queuecmd+0x33/0x290
&(&host->lock)->rlock 15155 [<ffffffff814283e4>] ahci_interrupt+0x64/0x120
Test #5
&(&host->lock)->rlock: 38154 38154 0.14 119.50 78034.38 4923881 6379092 0.24 183.30 21986193.10
---------------------
&(&host->lock)->rlock 32396 [<ffffffff81416b03>] ata_scsi_queuecmd+0x33/0x290
&(&host->lock)->rlock 5758 [<ffffffff814283e4>] ahci_interrupt+0x64/0x120
---------------------
&(&host->lock)->rlock 23014 [<ffffffff81416b03>] ata_scsi_queuecmd+0x33/0x290
&(&host->lock)->rlock 15140 [<ffffffff814283e4>] ahci_interrupt+0x64/0x120
Test #6
&(&host->lock)->rlock: 33812 33812 0.14 98.56 62314.64 4479447 6287673 0.33 183.96 20556437.70
---------------------
&(&host->lock)->rlock 28301 [<ffffffff81416b03>] ata_scsi_queuecmd+0x33/0x290
&(&host->lock)->rlock 5511 [<ffffffff814283e4>] ahci_interrupt+0x64/0x120
---------------------
&(&host->lock)->rlock 21378 [<ffffffff81416b03>] ata_scsi_queuecmd+0x33/0x290
&(&host->lock)->rlock 12434 [<ffffffff814283e4>] ahci_interrupt+0x64/0x120
Test #7
&(&host->lock)->rlock: 35336 35336 0.14 105.05 66190.92 4642703 6388507 0.40 179.67 20477137.55
---------------------
&(&host->lock)->rlock 30348 [<ffffffff81416b03>] ata_scsi_queuecmd+0x33/0x290
&(&host->lock)->rlock 4988 [<ffffffff814283e4>] ahci_interrupt+0x64/0x120
---------------------
&(&host->lock)->rlock 21983 [<ffffffff81416b03>] ata_scsi_queuecmd+0x33/0x290
&(&host->lock)->rlock 13353 [<ffffffff814283e4>] ahci_interrupt+0x64/0x120
Test #8
&(&host->lock)->rlock: 37715 37715 0.14 122.00 69306.80 4827790 6402109 0.42 192.10 21442240.03
---------------------
&(&host->lock)->rlock 31913 [<ffffffff81416b03>] ata_scsi_queuecmd+0x33/0x290
&(&host->lock)->rlock 5802 [<ffffffff814283e4>] ahci_interrupt+0x64/0x120
---------------------
&(&host->lock)->rlock 23613 [<ffffffff81416b03>] ata_scsi_queuecmd+0x33/0x290
&(&host->lock)->rlock 14102 [<ffffffff814283e4>] ahci_interrupt+0x64/0x120
Test #9
&(&host->lock)->rlock: 24517 24519 0.11 114.67 54559.79 5550859 6505606 0.06 153.24 8936643.32
---------------------
&(&host->lock)->rlock 10797 [<ffffffff814280d8>] ahci_thread_fn+0x48/0xa0
&(&host->lock)->rlock 13722 [<ffffffff81428164>] ahci_hw_interrupt+0x34/0x130
---------------------
&(&host->lock)->rlock 18255 [<ffffffff81428164>] ahci_hw_interrupt+0x34/0x130
&(&host->lock)->rlock 6264 [<ffffffff814280d8>] ahci_thread_fn+0x48/0xa0
Test #10
&(&host->lock)->rlock: 27503 27505 0.13 143.29 51496.56 5659327 6496233 0.06 196.87 8579972.36
---------------------
&(&host->lock)->rlock 10952 [<ffffffff814280d8>] ahci_thread_fn+0x48/0xa0
&(&host->lock)->rlock 16553 [<ffffffff81428164>] ahci_hw_interrupt+0x34/0x130
---------------------
&(&host->lock)->rlock 6553 [<ffffffff814280d8>] ahci_thread_fn+0x48/0xa0
&(&host->lock)->rlock 20952 [<ffffffff81428164>] ahci_hw_interrupt+0x34/0x130
Test #11
&(&host->lock)->rlock: 25655 25658 0.12 118.67 54061.46 5690296 6494647 0.07 210.10 8138708.47
---------------------
&(&host->lock)->rlock 16681 [<ffffffff814280d8>] ahci_thread_fn+0x48/0xa0
&(&host->lock)->rlock 8977 [<ffffffff81428164>] ahci_hw_interrupt+0x34/0x130
---------------------
&(&host->lock)->rlock 11899 [<ffffffff814280d8>] ahci_thread_fn+0x48/0xa0
&(&host->lock)->rlock 13759 [<ffffffff81428164>] ahci_hw_interrupt+0x34/0x130
Test #12
&(&host->lock)->rlock: 36741 36742 0.12 113.75 60427.51 5624197 6509975 0.06 226.06 8322068.81
---------------------
&(&host->lock)->rlock 24149 [<ffffffff814280d8>] ahci_thread_fn+0x48/0xa0
&(&host->lock)->rlock 12593 [<ffffffff81428164>] ahci_hw_interrupt+0x34/0x130
---------------------
&(&host->lock)->rlock 18710 [<ffffffff814280d8>] ahci_thread_fn+0x48/0xa0
&(&host->lock)->rlock 18032 [<ffffffff81428164>] ahci_hw_interrupt+0x34/0x130
Test #13
&(&host->lock)->rlock: 25348 25350 0.16 129.52 65978.83 5632012 6515589 0.06 154.50 8527745.18
---------------------
&(&host->lock)->rlock 10507 [<ffffffff814280d8>] ahci_thread_fn+0x48/0xa0
&(&host->lock)->rlock 14843 [<ffffffff81428164>] ahci_hw_interrupt+0x34/0x130
---------------------
&(&host->lock)->rlock 20301 [<ffffffff81428164>] ahci_hw_interrupt+0x34/0x130
&(&host->lock)->rlock 5049 [<ffffffff814280d8>] ahci_thread_fn+0x48/0xa0
Test #14
&(&host->lock)->rlock: 33124 33125 0.11 117.73 57291.39 5621960 6510702 0.06 209.74 8662461.99
---------------------
&(&host->lock)->rlock 14406 [<ffffffff814280d8>] ahci_thread_fn+0x48/0xa0
&(&host->lock)->rlock 18719 [<ffffffff81428164>] ahci_hw_interrupt+0x34/0x130
---------------------
&(&host->lock)->rlock 9999 [<ffffffff814280d8>] ahci_thread_fn+0x48/0xa0
&(&host->lock)->rlock 23126 [<ffffffff81428164>] ahci_hw_interrupt+0x34/0x130
Test #15
&(&host->lock)->rlock: 24801 24801 0.14 130.18 69186.41 5468958 6514262 0.07 160.17 8054911.26
---------------------
&(&host->lock)->rlock 16013 [<ffffffff814280d8>] ahci_thread_fn+0x48/0xa0
&(&host->lock)->rlock 8788 [<ffffffff81428164>] ahci_hw_interrupt+0x34/0x130
---------------------
&(&host->lock)->rlock 10132 [<ffffffff814280d8>] ahci_thread_fn+0x48/0xa0
&(&host->lock)->rlock 14669 [<ffffffff81428164>] ahci_hw_interrupt+0x34/0x130
Test #16
&(&host->lock)->rlock: 29420 29422 0.15 112.77 39726.37 5609359 6517385 0.06 160.30 9618631.17
---------------------
&(&host->lock)->rlock 14040 [<ffffffff814280d8>] ahci_thread_fn+0x48/0xa0
&(&host->lock)->rlock 15382 [<ffffffff81428164>] ahci_hw_interrupt+0x34/0x130
---------------------
&(&host->lock)->rlock 10237 [<ffffffff814280d8>] ahci_thread_fn+0x48/0xa0
&(&host->lock)->rlock 19185 [<ffffffff81428164>] ahci_hw_interrupt+0x34/0x130
