tech-kern archive
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]
Re: biowait for seconds
On Fri, Apr 18, 2025 at 04:22:09PM +0100, Robert Swindells wrote:
> Maybe read the suggestion as "find out why it isn't using msi".
I am back with some results with this.
The nvme is available from a BOSS-N1 RAID controller. It is confgurable
from the BIOS and is able to export one RAID made of one or two NVME
(two in my case, but I tested one, no difference). It is not able to
export the two NVME as-is. Removing the RAID set from the controller
cause a nvme0ns1 (0 bytes) device to appear, and the biowait lags are
gone (but the NVME cannot be used, of course).
I added many printfs to understand what is going on. MSI is not enabled
because the host bridge does not support it happens in
sys/arch/x86/pci/pci_msi_machdep.c:pci_msix_alloc_common()
We have:
ppb0 at pci0 dev 10 function 0: Intel product 1bba (rev. 0x11)
pci1 at ppb0 bus 1
nvme0 at pci1 dev 0 function 0: Marvell product 2241 (rev. 0x20)
nvme0: NVMe 1.3
nvme0: interrupting at ioapic0 pin 16
nvme0: Dell BOSS-N1, firmware 11131069, serial VN0WW56VFCV004A801TY
ld0 at nvme0 nsid 1
pcictl pci0 dump -b 0 -d 10 -f 0 says:
(...)
Capability register at 0x90
type: 0x05 (MSI)
(...)
PCI Message Signaled Interrupt
Message Control register: 0x0100
MSI Enabled: off
Hence the question is now how to enable MSI?
--
Emmanuel Dreyfus
manu%netbsd.org@localhost
Home |
Main Index |
Thread Index |
Old Index