redhat9.0使用PromiseUltra100（IDE）扩展卡的怪问题？

wxxszzz

PromiseUltra100 芯片型号为pdc20267

当有几台客户机同时向服务器的这块IDE扩展卡的硬盘传输数据时

出现 dma timeouts
如下：
hde: timeout waiting for DMA

然后这块卡上的硬盘的数据就不能读写了。有时即使能读也是奇慢无比。

不知道是哪里出了问题。

有知道原因的朋友请解答一下，万分感谢。

dmesg信息如下:

1. 
   
2. ICH5: not 100% native mode: will probe irqs later
   
3.     ide0: BM-DMA at 0xf000-0xf007, BIOS settings: hda:DMA, hdb:DMA
   
4.     ide1: BM-DMA at 0xf008-0xf00f, BIOS settings: hdc:DMA, hdd:DMA
   
5. PDC20267: IDE controller at PCI slot 01:01.0
   
6. PDC20267: chipset revision 2
   
7. PDC20267: not 100% native mode: will probe irqs later
   
8. PDC20267: (U)DMA Burst Bit ENABLED Primary PCI Mode Secondary PCI Mode.
   
9.     ide2: BM-DMA at 0xb000-0xb007, BIOS settings: hde:pio, hdf:pio
   
10.     ide3: BM-DMA at 0xb008-0xb00f, BIOS settings: hdg:pio, hdh:pio
    
11. hda: ST340014A, ATA DISK drive
    
12. hdb: ST3120022A, ATA DISK drive
    
13. blk: queue c03b1580, I/O limit 4095Mb (mask 0xffffffff)
    
14. blk: queue c03b16d4, I/O limit 4095Mb (mask 0xffffffff)
    
15. hdc: ST3120022A, ATA DISK drive
    
16. hdd: ST3120022A, ATA DISK drive
    
17. blk: queue c03b1a00, I/O limit 4095Mb (mask 0xffffffff)
    
18. blk: queue c03b1b54, I/O limit 4095Mb (mask 0xffffffff)
    
19. hde: ST380011A, ATA DISK drive
    
20. blk: queue c03b1e80, I/O limit 4095Mb (mask 0xffffffff)
    
21. hdg: ST380011A, ATA DISK drive
    
22. blk: queue c03b2300, I/O limit 4095Mb (mask 0xffffffff)
    
23. ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
    
24. ide1 at 0x170-0x177,0x376 on irq 15
    
25. ide2 at 0xa000-0xa007,0xa402 on irq 21
    
26. ide3 at 0xa800-0xa807,0xac02 on irq 21
    
27. hda: host protected area => 1
    
28. hda: 78165360 sectors (40021 MB) w/2048KiB Cache, CHS=4865/255/63, UDMA(100)
    
29. hdb: host protected area => 1
    
30. hdb: 234441648 sectors (120034 MB) w/2048KiB Cache, CHS=14593/255/63, UDMA(100)
    
31. hdc: host protected area => 1
    
32. hdc: 234441648 sectors (120034 MB) w/2048KiB Cache, CHS=14593/255/63, UDMA(100)
    
33. hdd: host protected area => 1
    
34. hdd: 234441648 sectors (120034 MB) w/2048KiB Cache, CHS=14593/255/63, UDMA(100)
    
35. hde: host protected area => 1
    
36. hde: 156301488 sectors (80026 MB) w/2048KiB Cache, CHS=155061/16/63, UDMA(100)
    
37. hdg: host protected area => 1
    
38. hdg: 156301488 sectors (80026 MB) w/2048KiB Cache, CHS=155061/16/63, UDMA(100)
    
39. ide-floppy driver 0.99.newide
    
40. Partition check:
    
41. hda: hda1 hda2 hda3 hda4 < hda5 hda6 >
    
42. hdb: hdb1
    
43. hdc: hdc1
    
44. hdd: hdd1
    
45. hde: hde1
    
46. hdg: hdg1
    
47. ide-floppy driver 0.99.newide
    
48. LVM version 1.0.5+(22/07/2002)
    
49. NET4: Linux TCP/IP 1.0 for NET4.0
    
50. IP Protocols: ICMP, UDP, TCP, IGMP
    
51. IP: routing cache hash table of 8192 buckets, 64Kbytes
    
52. TCP: Hash tables configured (established 262144 bind 65536)
    
53. Linux IP multicast router 0.06 plus PIM-SM
    
54. NET4: Unix domain sockets 1.0/SMP for Linux NET4.0.
    
55. RAMDISK: Compressed image found at block 0
    
56. Freeing initrd memory: 107k freed
    
57. VFS: Mounted root (ext2 filesystem).
    
58. md: md driver 0.90.0 MAX_MD_DEVS=256, MD_SB_DISKS=27
    
59. md: raid0 personality registered as nr 2
    
60. EXT3-fs: INFO: recovery required on readonly filesystem.
    
61. EXT3-fs: write access will be enabled during recovery.
    
62. (recovery.c, 256): journal_recover: JBD: recovery, exit status 0, recovered tran
    
63. sactions 95737 to 95786
    
64. (recovery.c, 258): journal_recover: JBD: Replayed 207 and revoked 0/0 blocks
    
65. kjournald starting.  Commit interval 5 seconds
    
66. EXT3-fs: ide0(3,6): orphan cleanup on readonly fs
    
67. ext3_orphan_cleanup: deleting unreferenced inode 195652
    
68. ext3_orphan_cleanup: deleting unreferenced inode 195281
    

复制代码

[ 本帖最后由 wxxszzz 于 2005-12-26 21:52 编辑 ]