Bug 242773 - ata driver as module, interrupt storm on atapci0 and low attachment speed for disks
Summary: ata driver as module, interrupt storm on atapci0 and low attachment speed for...
Status: New
Alias: None
Product: Base System
Classification: Unclassified
Component: kern (show other bugs)
Version: 11.3-STABLE
Hardware: Any Any
: --- Affects Only Me
Assignee: freebsd-bugs (Nobody)
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2019-12-21 16:15 UTC by emss
Modified: 2019-12-21 16:15 UTC (History)
0 users

See Also:

Attachments
11.3-STABLE dmesg for SE7525GP2 (7.80 KB, text/plain)
2019-12-21 16:15 UTC, emss 		no flags Details
View All Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description emss 2019-12-21 16:15:08 UTC 
Created attachment 210107 [details]
11.3-STABLE dmesg for SE7525GP2

Hi,

Intel SE7525GP2 MB (quite old, I know)

When using MINIMAL and trying to load atapci as module, atapci reports low speed for attached disks :

ada4 at ata2 bus 0 scbus4 target 0 lun 0
ada4: <SanDisk SSD PLUS 120GB UE4500RL> ACS-2 ATA SATA 3.x device
ada4: Serial Number 1841B1803114
ada4: 33.000MB/s transfers (UDMA2, PIO 512bytes)
ada4: 114480MB (234455040 512 byte sectors)

Kernel spits messages about interrupt storm on irq used by atapci.

When ata support is compiled into kernel, the reported speed is fine :

ada4 at ata2 bus 0 scbus4 target 0 lun 0
ada4: <SanDisk SSD PLUS 120GB UE4500RL> ACS-2 ATA SATA 3.x device
ada4: Serial Number 1841B1803114
ada4: 150.000MB/s transfers (SATA 1.x, UDMA5, PIO 512bytes)
ada4: 114480MB (234455040 512 byte sectors)

I've reproduced the same issue using a freshly compiled 12.1-STABLE kernel

Regards

Éric Masson