Hi,
2.4.19-pre4-ac[23] does not boot. I've not tested ac1 but vanilla pre4
works.
It stops there,
Mar 29 16:13:24 f5ibh kernel: ide1: BM-DMA at 0xd008-0xd00f, BIOS
settings: hdc:pio, hdd:pio
Mar 29 16:13:24 f5ibh kernel: hda: QUANTUM FIREBALLP LM30, ATA DISK
drive
Mar 29 16:13:24 f5ibh kernel: hdb: ST3491A, ATA DISK drive
Mar 29 16:13:24 f5ibh kernel: hdc: GoldStar CD-RW CED-8083B, ATAPI CDROM
drive
Mar 29 16:13:24 f5ibh kernel: hdd: CREATIVECD3621E, ATAPI CDROM drive
Mar 29 16:13:24 f5ibh kernel: ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
Mar 29 16:13:24 f5ibh kernel: ide1 at 0x170-0x177,0x376 on irq 15
Mar 29 16:13:24 f5ibh kernel: ALI15X3: Ultra DMA enabled
Mar 29 16:13:24 f5ibh kernel: hda: QUANTUM FIREBALLP LM30, 28629MB
w/1900kB Cache, CHS=3649/255/63, (U)DMA
Mar 29 16:13:24 f5ibh kernel: ALI15X3: MultiWord DMA enabled
Mar 29 16:13:24 f5ibh kernel: hdb: ST3491A, 408MB w/120kB Cache,
CHS=899/15/62, DMA
Mar 29 16:13:24 f5ibh kernel: Partition check:
Mar 29 16:13:24 f5ibh kernel: hda: hda1 hda2 hda3 hda4
Mar 29 16:13:24 f5ibh kernel: hdb:
There is only one partition on hdb.
Well, it is an ooold disk and I can survive without it.
----
regards
Jean-Luc
> 2.4.19-pre4-ac[23] does not boot. I've not tested ac1 but vanilla pre4
> works.
Can you try backing out the following two changes, one at a time. These are
the only ALi specific changes. So firstly I want to see if its an ALi or
core IDE bug
diff -u --new-file --recursive --exclude-from /usr/src/exclude linux.19p4/drivers/ide/alim15x3.c linux.19pre4-ac3/drivers/ide/alim15x3.c
--- linux.19p4/drivers/ide/alim15x3.c Mon Mar 25 17:47:11 2002
+++ linux.19pre4-ac3/drivers/ide/alim15x3.c Tue Mar 26 18:36:23 2002
@@ -248,7 +248,7 @@
byte s_clc, a_clc, r_clc;
unsigned long flags;
int bus_speed = system_bus_clock();
- int port = hwif->index ? 0x5c : 0x58;
+ int port = hwif->channel ? 0x5c : 0x58;
int portFIFO = hwif->channel ? 0x55 : 0x54;
byte cd_dma_fifo = 0;
@@ -442,6 +442,8 @@
ide_dma_action_t dma_func = ide_dma_on;
byte can_ultra_dma = ali15x3_can_ultra(drive);
+ (void) config_chipset_for_pio(drive);
+
if ((m5229_revision<=0x20) && (drive->media!=ide_disk))
return hwif->dmaproc(ide_dma_off_quietly, drive);
I've reverted the 1st change with the same result : the system hang
I've reverted the 2nd change and then the system boot fine
-------
Regards
Jean-Luc
Alan Cox wrote:
>
> > 2.4.19-pre4-ac[23] does not boot. I've not tested ac1 but vanilla pre4
> > works.
>
> Can you try backing out the following two changes, one at a time. These are
> the only ALi specific changes. So firstly I want to see if its an ALi or
> core IDE bug
>
> diff -u --new-file --recursive --exclude-from /usr/src/exclude linux.19p4/drivers/ide/alim15x3.c linux.19pre4-ac3/drivers/ide/alim15x3.c
> --- linux.19p4/drivers/ide/alim15x3.c Mon Mar 25 17:47:11 2002
> +++ linux.19pre4-ac3/drivers/ide/alim15x3.c Tue Mar 26 18:36:23 2002
> @@ -248,7 +248,7 @@
> byte s_clc, a_clc, r_clc;
> unsigned long flags;
> int bus_speed = system_bus_clock();
> - int port = hwif->index ? 0x5c : 0x58;
> + int port = hwif->channel ? 0x5c : 0x58;
> int portFIFO = hwif->channel ? 0x55 : 0x54;
> byte cd_dma_fifo = 0;
>
> @@ -442,6 +442,8 @@
> ide_dma_action_t dma_func = ide_dma_on;
> byte can_ultra_dma = ali15x3_can_ultra(drive);
>
> + (void) config_chipset_for_pio(drive);
> +
> if ((m5229_revision<=0x20) && (drive->media!=ide_disk))
> return hwif->dmaproc(ide_dma_off_quietly, drive);
>
On Fri, 29 Mar 2002, Alan Cox wrote:
> > 2.4.19-pre4-ac[23] does not boot. I've not tested ac1 but vanilla pre4
> > works.
>
> Can you try backing out the following two changes, one at a time. These are
> the only ALi specific changes. So firstly I want to see if its an ALi or
> core IDE bug
>
> diff -u --new-file --recursive --exclude-from /usr/src/exclude linux.19p4/drivers/ide/alim15x3.c linux.19pre4-ac3/drivers/ide/alim15x3.c
> --- linux.19p4/drivers/ide/alim15x3.c Mon Mar 25 17:47:11 2002
> +++ linux.19pre4-ac3/drivers/ide/alim15x3.c Tue Mar 26 18:36:23 2002
> @@ -248,7 +248,7 @@
> byte s_clc, a_clc, r_clc;
> unsigned long flags;
> int bus_speed = system_bus_clock();
> - int port = hwif->index ? 0x5c : 0x58;
> + int port = hwif->channel ? 0x5c : 0x58;
> int portFIFO = hwif->channel ? 0x55 : 0x54;
> byte cd_dma_fifo = 0;
>
> @@ -442,6 +442,8 @@
> ide_dma_action_t dma_func = ide_dma_on;
> byte can_ultra_dma = ali15x3_can_ultra(drive);
>
> + (void) config_chipset_for_pio(drive);
> +
This is possible, however one of the problems encountered is the
following under several chipsets. If there is no pio timing set at all,
then we can run into host lock issues if the driver drops out of dma.
Thus, if it is going to lockup here it would/could lock up in other
places when one trys to program the host for PIO.
Regardsless this is bad, in one direction or another.
> if ((m5229_revision<=0x20) && (drive->media!=ide_disk))
> return hwif->dmaproc(ide_dma_off_quietly, drive);
>
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
>
Andre Hedrick
LAD Storage Consulting Group
> This is possible, however one of the problems encountered is the
> following under several chipsets. If there is no pio timing set at all,
> then we can run into host lock issues if the driver drops out of dma.
> Thus, if it is going to lockup here it would/could lock up in other
> places when one trys to program the host for PIO.
Well right at the moment the ALi locks up on boot reliably. That means a
fix has to be found, or the ALi changes reverted
On Fri, 29 Mar 2002, Alan Cox wrote:
> > This is possible, however one of the problems encountered is the
> > following under several chipsets. If there is no pio timing set at all,
> > then we can run into host lock issues if the driver drops out of dma.
> > Thus, if it is going to lockup here it would/could lock up in other
> > places when one trys to program the host for PIO.
>
> Well right at the moment the ALi locks up on boot reliably. That means a
> fix has to be found, or the ALi changes reverted
The real bug may be a rusult of improper setup and since there is no cable
detection bit in the XMETA Lifebook, but it needs the southbridge setup
calls, that snippet of code must be move to above function.
lines 570-590 and 611-620 or there abouts, with their if(version) test.
Rereading the code, docs, comments -- it is clear pci_init code for the
host.
Cheers,
Andre Hedrick
LAD Storage Consulting Group
On Fri, 29 Mar 2002, Alan Cox wrote:
> > This is possible, however one of the problems encountered is the
> > following under several chipsets. If there is no pio timing set at all,
> > then we can run into host lock issues if the driver drops out of dma.
> > Thus, if it is going to lockup here it would/could lock up in other
> > places when one trys to program the host for PIO.
>
> Well right at the moment the ALi locks up on boot reliably. That means a
> fix has to be found, or the ALi changes reverted
Pull out the GOOF-UP of mine :-/
autotune is enabled and does the same thing, Gurrr....
Cheers,
Andre Hedrick
LAD Storage Consulting Group
> Pull out the GOOF-UP of mine :-/
> autotune is enabled and does the same thing, Gurrr....
Ok will do
Mar 29 18:32:43 Apollo kernel: PIIX4: not 100%% native mode: will probe irqs later
Mar 29 18:32:43 Apollo kernel: ide0: BM-DMA at 0xf000-0xf007, BIOS settings: hda:DMA, hdb:DMA
Mar 29 18:32:43 Apollo kernel: ide1: BM-DMA at 0xf008-0xf00f, BIOS settings: hdc:DMA, hdd:pio
Mar 29 18:32:43 Apollo xinetd[780]: Service ntalk: missing '{' [line=5]
Mar 29 18:32:43 Apollo kernel: HPT366: onboard version of chipset, pin1=1 pin2=2
Mar 29 18:32:43 Apollo kernel: HPT366: IDE controller on PCI bus 00 dev 98
Mar 29 18:32:43 Apollo kernel: HPT366: chipset revision 1
Mar 29 18:32:43 Apollo kernel: HPT366: not 100%% native mode: will probe irqs later
Mar 29 18:32:43 Apollo kernel: ide2: BM-DMA at 0xb800-0xb807, BIOS settings: hde:DMA, hdf:DMA
Mar 29 18:32:43 Apollo kernel: HPT366: IDE controller on PCI bus 00 dev 99
Mar 29 18:32:43 Apollo kernel: HPT366: chipset revision 1
Mar 29 18:32:43 Apollo kernel: HPT366: not 100%% native mode: will probe irqs later
Mar 29 18:32:43 Apollo kernel: ide3: BM-DMA at 0xc400-0xc407, BIOS settings: hdg:DMA, hdh:pio
Mar 29 18:32:44 Apollo kernel: hda: WDC AC31200F, ATA DISK drive
Mar 29 18:32:44 Apollo kernel: hdb: Maxtor 88400D8, ATA DISK drive
Mar 29 18:32:44 Apollo kernel: hdc: ATAPI 48X CDROM, ATAPI CD/DVD-ROM drive
Mar 29 18:32:44 Apollo kernel: hde: Maxtor 34098H4, ATA DISK drive
Mar 29 18:32:44 Apollo kernel: hdf: Maxtor 5T060H6, ATA DISK drive
Mar 29 18:32:44 Apollo kernel: hdg: ST320413A, ATA DISK drive
Mar 29 18:32:44 Apollo kernel: ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
Mar 29 18:32:44 Apollo kernel: ide1 at 0x170-0x177,0x376 on irq 15
Mar 29 18:32:44 Apollo kernel: ide2 at 0xb000-0xb007,0xb402 on irq 18
Mar 29 18:32:44 Apollo kernel: ide3 at 0xbc00-0xbc07,0xc002 on irq 18
Mar 29 18:32:44 Apollo kernel: hda: 2503872 sectors (1282 MB) w/64KiB Cache, CHS=621/64/63, DMA
Mar 29 18:32:44 Apollo kernel: hdb: 16408224 sectors (8401 MB) w/256KiB Cache, CHS=1021/255/63, UDMA(33)
Mar 29 18:32:44 Apollo kernel: hde: 80043264 sectors (40982 MB) w/2048KiB Cache, CHS=79408/16/63, UDMA(66)
Mar 29 18:32:44 Apollo kernel: hdf: 120103200 sectors (61493 MB) w/2048KiB Cache, CHS=119150/16/63, UDMA(66)
Mar 29 18:32:44 Apollo kernel: hdg: 39102336 sectors (20020 MB) w/512KiB Cache, CHS=38792/16/63, UDMA(66)
Mar 29 18:32:44 Apollo kernel: hdc: ATAPI 48X CD-ROM drive, 128kB Cache, UDMA(33)
Mar 29 18:32:44 Apollo xinetd: xinetd startup succeeded
Mar 29 18:32:44 Apollo kernel: Uniform CD-ROM driver Revision: 3.12
Mar 29 18:32:44 Apollo kernel: Partition check:
Mar 29 18:32:44 Apollo kernel: hda: hda1 hda2
Mar 29 18:32:44 Apollo kernel: hdb:
Mar 29 18:32:44 Apollo kernel: hde: [EZD] [remap 0->1] [4982/255/63] hde1
I just dusted off the my BP6 with original BIOS's and it hangs too :-/
So off to rebuild the driver froms scratch ...
For now if you need it to go copy over the old on from 2.4.18, but then a
bunch of hosts 370/370A/372/374 break but the can be added back in..
On Fri, 29 Mar 2002, Art Wagner wrote:
> Hi;
> I seem to have the identical problem with system hangs on boot. The
> problem I see occurs on a Abit BP6
> on the first access of the first drive on the HPT-366 interface. The
> problem occurs on all -ac-x revisions
> 1 through 3. The attached log is from an boot on 2.4.19-pre4 which did
> not hang but the log is identical
> except for the fact that the hang occurs with only the "hde:" portion of
> the last line displayed.
> If I can provide any further information pleas let me know via the list
> or direct email.
> Art Wagner
>
> Andre Hedrick wrote:
>
> >On Fri, 29 Mar 2002, Alan Cox wrote:
> >
> >>>This is possible, however one of the problems encountered is the
> >>>following under several chipsets. If there is no pio timing set at all,
> >>>then we can run into host lock issues if the driver drops out of dma.
> >>>Thus, if it is going to lockup here it would/could lock up in other
> >>>places when one trys to program the host for PIO.
> >>>
> >>Well right at the moment the ALi locks up on boot reliably. That means a
> >>fix has to be found, or the ALi changes reverted
> >>
> >
> >Pull out the GOOF-UP of mine :-/
> >
> >autotune is enabled and does the same thing, Gurrr....
> >
> >Cheers,
> >
> >Andre Hedrick
> >LAD Storage Consulting Group
> >
> >-
> >To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> >the body of a message to [email protected]
> >More majordomo info at http://vger.kernel.org/majordomo-info.html
> >Please read the FAQ at http://www.tux.org/lkml/
> >
> >
>
>
Andre Hedrick
LAD Storage Consulting Group
Fixed now!
Code drop tonight!
On Fri, 29 Mar 2002, Art Wagner wrote:
> Hi;
> I seem to have the identical problem with system hangs on boot. The
> problem I see occurs on a Abit BP6
> on the first access of the first drive on the HPT-366 interface. The
> problem occurs on all -ac-x revisions
> 1 through 3. The attached log is from an boot on 2.4.19-pre4 which did
> not hang but the log is identical
> except for the fact that the hang occurs with only the "hde:" portion of
> the last line displayed.
> If I can provide any further information pleas let me know via the list
> or direct email.
> Art Wagner
>
> Andre Hedrick wrote:
>
> >On Fri, 29 Mar 2002, Alan Cox wrote:
> >
> >>>This is possible, however one of the problems encountered is the
> >>>following under several chipsets. If there is no pio timing set at all,
> >>>then we can run into host lock issues if the driver drops out of dma.
> >>>Thus, if it is going to lockup here it would/could lock up in other
> >>>places when one trys to program the host for PIO.
> >>>
> >>Well right at the moment the ALi locks up on boot reliably. That means a
> >>fix has to be found, or the ALi changes reverted
> >>
> >
> >Pull out the GOOF-UP of mine :-/
> >
> >autotune is enabled and does the same thing, Gurrr....
> >
> >Cheers,
> >
> >Andre Hedrick
> >LAD Storage Consulting Group
> >
> >-
> >To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> >the body of a message to [email protected]
> >More majordomo info at http://vger.kernel.org/majordomo-info.html
> >Please read the FAQ at http://www.tux.org/lkml/
> >
> >
>
>
Andre Hedrick
LAD Storage Consulting Group
Hi Alan and others,
It is fine for me with 2.4.19-pre4-ac4
--------
Regards
Jean-Luc