PDC20269 error handling problem

PDC20269 error handling problem

[Manfred Spraul]

[-- Attachment #1: Type: text/plain, Size: 1112 bytes --]

I run into hard system hangs with my promise controller (kernel:
2.4.19-pre4-ac3)

Background:

* Harddisk with bad sectors.
* reading from one of the bad sectors usually causes error messages, but
sooner or later the kernel falls back from UDMA to pio.
* obviously pio doesn't work either, and ide_error is called.
* the fallback works if the harddisk is connected to a via VT82C686
south bridge.
* the fallback causes a total system hang if the drive is connected to
the promise controller.
* it hangs on the "rep;insw;" instruction in ata_input_data(), called by
try_to_flush_leftover_data().

Before someone replies that I should not try to read bad sectors: The
harddisk was used in a headless server, and I did not expect that sector
errors cause system crashes. And the crash defeats the purpose of
software raid1.

Any idea where I should continue to search? It seems that the pdc20269
controller doesn't handle the error fallback to pio. If I boot with
"ide=nodma", then pio works.

I've attached lspci and dmesg. If you have further questions, please
ask. I have a test setup with kdb.

--
    Manfred

[-- Attachment #2: lspci.txt --]
[-- Type: text/plain, Size: 750 bytes --]

00:0a.0 Unknown mass storage controller: Promise Technology, Inc.: Unknown device 4d69 (rev 02) (prog-if 85)
	Subsystem: Promise Technology, Inc.: Unknown device 4d68
	Flags: bus master, 66Mhz, slow devsel, latency 32, IRQ 12
	I/O ports at 6300 [size=8]
	I/O ports at 6400 [size=4]
	I/O ports at 6500 [size=8]
	I/O ports at 6600 [size=4]
	I/O ports at 6700 [size=16]
	Memory at e1000000 (32-bit, non-prefetchable) [size=16K]
	Expansion ROM at <unassigned> [disabled] [size=16K]
	Capabilities: [60] Power Management version 1
00: 5a 10 69 4d 07 00 30 04 02 85 80 01 08 20 00 00
10: 01 63 00 00 01 64 00 00 01 65 00 00 01 66 00 00
20: 01 67 00 00 00 00 00 e1 00 00 00 00 5a 10 68 4d
30: 00 00 00 00 60 00 00 00 00 00 00 00 0c 01 04 12


[-- Attachment #3: out.txt --]
[-- Type: text/plain, Size: 5525 bytes --]

klogd 1.4-0, log source = /proc/kmsg started.
Inspecting /boot/System.map-2.4.19-pre4-ac3
Loaded 16316 symbols from /boot/System.map-2.4.19-pre4-ac3.
Symbols match kernel version 2.4.19.
No module symbols loaded.
Linux version 2.4.19-pre4-ac3 (manfred@clmsdev.localdomain) (gcc version 2.96 20000731 (Red Hat Linux 7.1 2.96-98)) #21 SMP Fri Mar 29 21:32:52 CET 2002
BIOS-provided physical RAM map:
 BIOS-e820: 0000000000000000 - 00000000000a0000 (usable)
 BIOS-e820: 00000000000f0000 - 0000000000100000 (reserved)
 BIOS-e820: 0000000000100000 - 0000000003000000 (usable)
 BIOS-e820: 00000000ffff0000 - 0000000100000000 (reserved)
On node 0 totalpages: 12288
zone(0): 4096 pages.
zone(1): 8192 pages.
zone(2): 0 pages.
Kernel command line: auto BOOT_IMAGE=linux ro root=301 BOOT_FILE=/boot/vmlinuz
No local APIC present or hardware disabled
Initializing CPU#0
Console: colour VGA+ 80x25
Calibrating delay loop... 119.60 BogoMIPS
Memory: 45256k/49152k available (1262k kernel code, 3512k reserved, 839k data, 232k init, 0k highmem)
Checking if this processor honours the WP bit even in supervisor mode... Ok.
kdb version 2.1 by Scott Lurndal, Keith Owens. Copyright SGI, All Rights Reserved
Dentry cache hash table entries: 8192 (order: 4, 65536 bytes)
Inode cache hash table entries: 4096 (order: 3, 32768 bytes)
Mount cache hash table entries: 1024 (order: 1, 8192 bytes)
Buffer cache hash table entries: 1024 (order: 0, 4096 bytes)
Page-cache hash table entries: 16384 (order: 4, 65536 bytes)
Enabling CPUID on Cyrix processor.
Checking 'hlt' instruction... OK.
POSIX conformance testing by UNIFIX
mtrr: v1.40 (20010327) Richard Gooch (rgooch@atnf.csiro.au)
mtrr: detected mtrr type: Cyrix ARR
CPU0: Cyrix 6x86L 2x Core/Bus Clock stepping 02
SMP motherboard not detected.
Local APIC not detected. Using dummy APIC emulation.
PCI: PCI BIOS revision 2.10 entry at 0xfb610, last bus=0
PCI: Using configuration type 1
PCI: Probing PCI hardware
Activating ISA DMA hang workarounds.
Linux NET4.0 for Linux 2.4
Based upon Swansea University Computer Society NET3.039
Initializing RT netlink socket
Starting kswapd
Detected PS/2 Mouse Port.
pty: 256 Unix98 ptys configured
Serial driver version 5.05c (2001-07-08) with MANY_PORTS SHARE_IRQ SERIAL_PCI enabled
ttyS00 at 0x03f8 (irq = 4) is a 16550A
ttyS01 at 0x02f8 (irq = 3) is a 16550A
Real Time Clock Driver v1.10e
block: 80 slots per queue, batch=20
Uniform Multi-Platform E-IDE driver Revision: 6.31
ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx
VP_IDE: IDE controller on PCI bus 00 dev 39
VP_IDE: chipset revision 2
VP_IDE: not 100%% native mode: will probe irqs later
ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx
VP_IDE: VIA vt82c586 (rev 02) IDE MWDMA16 controller on pci00:07.1
    ide0: BM-DMA at 0x6000-0x6007, BIOS settings: hda:pio, hdb:pio
    ide1: BM-DMA at 0x6008-0x600f, BIOS settings: hdc:pio, hdd:pio
PDC20269: IDE controller on PCI bus 00 dev 50
PDC20269: chipset revision 2
PDC20269: not 100%% native mode: will probe irqs later
    ide2: BM-DMA at 0x6700-0x6707, BIOS settings: hde:pio, hdf:pio
reset_proc is 0h.
    ide3: BM-DMA at 0x6708-0x670f, BIOS settings: hdg:pio, hdh:pio
reset_proc is 0h.
hda: IBM-DCAA-34330, ATA DISK drive
hde: Maxtor 4G160J8, ATA DISK drive
hdg: Maxtor 4G160J8, ATA DISK drive
ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
ide2 at 0x6300-0x6307,0x6402 on irq 12
ide3 at 0x6500-0x6507,0x6602 on irq 12
hda: 8467200 sectors (4335 MB) w/96KiB Cache, CHS=527/255/63, DMA
hde: 320173056 sectors (163929 MB) w/2048KiB Cache, CHS=19929/255/63, UDMA(33)
hdg: 320173056 sectors (163929 MB) w/2048KiB Cache, CHS=19929/255/63, UDMA(33)
ide-floppy driver 0.99.newide
Partition check:
 hda: hda1 hda2 hda3
 hde: hde1
 hdg: hdg1
loop: loaded (max 8 devices)
ne2k-pci.c:v1.02 10/19/2000 D. Becker/P. Gortmaker
  http://www.scyld.com/network/ne2k-pci.html
eth0: RealTek RTL-8029 found at 0x6200, IRQ 10, 00:00:1C:30:58:85.
[drm] Initialized tdfx 1.0.0 20010216 on minor 0
ide-floppy driver 0.99.newide
NET4: Linux TCP/IP 1.0 for NET4.0
IP Protocols: ICMP, UDP, TCP
IP: routing cache hash table of 512 buckets, 4Kbytes
TCP: Hash tables configured (established 4096 bind 4096)
NET4: Unix domain sockets 1.0/SMP for Linux NET4.0.
VFS: Mounted root (ext2 filesystem) readonly.
Freeing unused kernel memory: 232k freed
Adding Swap: 313256k swap-space (priority -1)
Installing knfsd (copyright (C) 1996 okir@monad.swb.de).
hde: dma_intr: status=0x51 { DriveReady SeekComplete Error }
hde: dma_intr: error=0x01 { AddrMarkNotFound }, LBAsect=43141736, high=2, low=9587304, sector=43141736
hde: dma_intr: status=0x51 { DriveReady SeekComplete Error }
hde: dma_intr: error=0x01 { AddrMarkNotFound }, LBAsect=43141736, high=2, low=9587304, sector=43141736
hde: dma_intr: status=0x51 { DriveReady SeekComplete Error }
hde: dma_intr: error=0x01 { AddrMarkNotFound }, LBAsect=43141736, high=2, low=9587304, sector=43141736
hde: dma_intr: status=0x51 { DriveReady SeekComplete Error }
hde: dma_intr: error=0x01 { AddrMarkNotFound }, LBAsect=43141736, high=2, low=9587304, sector=43141736
hde: DMA disabled
ide2: reset: success

SYSTEM hang!!!
I don't have the remaining lines, they were something like

ide2: MULTI-READ assume all data transfered is bad status=??
hde: task_mulin_intr: ...
hde: task_mulin_intr: ...
/// And now try_to_flush_leftover_data()->ata_input_data()->'rep;insw;'->hang.

Re: PDC20269 error handling problem

[Alan Cox]

> * the fallback causes a total system hang if the drive is connected to
> the promise controller.
> * it hangs on the "rep;insw;" instruction in ata_input_data(), called by
> try_to_flush_leftover_data().

That does sound like the drive didnt fall back properly - the rep insw would
then hang when the controller hangs the bus indefinitely waiting for a data
byte. Probably something got a little broken in the IDE updates

Alan

Re: PDC20269 error handling problem

[Manfred Spraul]

[-- Attachment #1: Type: text/plain, Size: 1108 bytes --]

The attached patch "fixes" the bug. How bad sectors cause just sector
errors, no more system hangs. The patch is not perfect, but definitively
an improvement:

<<<<<<<
hde: dma_intr: status=0x51 { DriveReady SeekComplete Error }
hde: dma_intr: error=0x01 { AddrMarkNotFound }, LBAsect=43141736,
high=2, low=9587304, sector=43141736
hde: dma_intr: status=0x51 { DriveReady SeekComplete Error }
hde: dma_intr: error=0x01 { AddrMarkNotFound }, LBAsect=43141736,
high=2, low=9587304, sector=43141736
hde: DMA disabled
PDC202XX: Primary channel reset.
ide2: reset: master: error (0x00?)
hde: task_in_intr: status=0x59 { DriveReady SeekComplete DataRequest
Error }
hde: task_in_intr: error=0x40 { UncorrectableError }, LBAsect=43141736,
high=2, low=9587304, sector=43141736
end_request: I/O error, dev 21:00 (hde), sector 43141736
hde: task_in_intr: status=0x59 { DriveReady SeekComplete DataRequest
Error }
hde: task_in_intr: error=0x40 { UncorrectableError }, LBAsect=43141736,
high=2, low=9587304, sector=43141736
<<<

I think "master: error (0x00?)" indicate another bug.
Andre, what do you think?

--
	Manfred

[-- Attachment #2: patch-pdc202 --]
[-- Type: text/plain, Size: 1962 bytes --]

--- 2.4/drivers/ide/pdc202xx.c	Fri Mar 29 14:34:13 2002
+++ build-2.4/drivers/ide/pdc202xx.c	Sun Mar 31 01:04:02 2002
@@ -729,12 +729,7 @@
 	byte mask		= hwif->channel ? 0x08 : 0x02;
 	unsigned short c_mask	= hwif->channel ? (1<<11) : (1<<10);
 
-	byte ultra_66		= ((id->dma_ultra & 0x0010) ||
-				   (id->dma_ultra & 0x0008)) ? 1 : 0;
-	byte ultra_100		= ((id->dma_ultra & 0x0020) ||
-				   (ultra_66)) ? 1 : 0;
-	byte ultra_133		= ((id->dma_ultra & 0x0040) ||
-				   (ultra_100)) ? 1 : 0;
+	byte fast_ultra		= id->dma_ultra & 0x0078;
 
 	switch(dev->device) {
 		case PCI_DEVICE_ID_PROMISE_20276:
@@ -792,7 +787,7 @@
 	 * parameters.
 	 */
 
-	if (((ultra_66) || (ultra_100) || (ultra_133)) && (cable)) {
+	if (fast_ultra && cable) {
 #ifdef DEBUG
 		printk("ULTRA66: %s channel of Ultra 66 requires an 80-pin cable for Ultra66 operation.\n", hwif->channel ? "Secondary" : "Primary");
 		printk("         Switching to Ultra33 mode.\n");
@@ -805,17 +800,13 @@
 		printk("%s reduced to Ultra33 mode.\n", drive->name);
 		udma_66 = 0; udma_100 = 0; udma_133 = 0;
 	} else {
-		if ((ultra_66) || (ultra_100) || (ultra_133)) {
+		if (fast_ultra) {
 			/*
 			 * check to make sure drive on same channel
 			 * is u66 capable
 			 */
 			if (hwif->drives[!(drive->dn%2)].id) {
-				if ((hwif->drives[!(drive->dn%2)].id->dma_ultra & 0x0040) ||
-				    (hwif->drives[!(drive->dn%2)].id->dma_ultra
-& 0x0020) ||
-				    (hwif->drives[!(drive->dn%2)].id->dma_ultra & 0x0010) ||
-				    (hwif->drives[!(drive->dn%2)].id->dma_ultra & 0x0008)) {
+				if (hwif->drives[!(drive->dn%2)].id->dma_ultra & 0x0078) {
 					if (!jumpbit)
 						OUT_BYTE(CLKSPD | mask, (high_16 + 0x11));
 				} else {
@@ -1075,6 +1066,8 @@
 	mdelay(1000);
 	printk("PDC202XX: %s channel reset.\n",
 		HWIF(drive)->channel ? "Secondary" : "Primary");
+	/* the reset clears the configuration, reinit to pio */
+	config_chipset_for_pio(drive, 5);
 }
 
 void pdc202xx_reset (ide_drive_t *drive)