highpoint driver problem, 2.4.21-ac4
A message that is not in the archive (sent off-list or not yet received): [email protected]
A message that is not in the archive (sent off-list or not yet received): [email protected]
A message that is not in the archive (sent off-list or not yet received): [email protected]
Quoted 16 lines · Wil Reichert
Quoted 7 lines
Quoted 6 lines
The on-board Highpoint controller (HPT372A) on my DFI NF2 is having issue. Loading the hptraid module results in a 'No such device' message while the hpt366 module segfaults and leaves an oops in my logs. These errors occur regardless of the disk/raid
configuration in the hpt BIOS. Following are the oops trace, an
lsmod, the .config and a lspci -vvv.Quoted 8 lines
The crash occurs in the hpt366 module. Loading hptraid will not work because it depends on the kernel to claim the disks of the raid volume (that is what hpt366 would do). I will add autoloading of the ide-controller module in the next raid-driver release. However, I do not know why the kernel oopses. You might want to try to build the hpt366 code into the kernel instead of a module. If it works it would probably mean that "
ide_hwif_t*hwif" was not properly initalized.
Quoted 4 lines
I initially had all the hpt modules built into the kernel, but that would also produce an oops and die immediately after ID'ing the two drives I have on attached. Would any more information be of use to you?
2.4.21-ac4 contains several fixes for the hpt37x. Please try this patch out.
bye,
Wilfried
A message that is not in the archive (sent off-list or not yet received): [email protected]
Quoted 1 line · Wil Reichert
Thats the kernel I'm using.
could you try the attachted patch, and report if this changes something?
bye,
wilfried
Quoted 33 lines
Wil
Quoted 26 lines · On Thu, 03 Jul 2003 21:21:30 +0200 Wilfried Weissmann…
Quoted 16 lines · Wil Reichert
Quoted 7 lines
Quoted 6 lines
The on-board Highpoint controller (HPT372A) on my DFI NF2 is having issue. Loading the hptraid module results in a 'No such device' message while the hpt366 module segfaults and leaves an oops in my logs. These errors occur regardless of the disk/raid
configuration in the hpt BIOS. Following are the oops trace, an
lsmod, the .config and a lspci -vvv.Quoted 8 lines
The crash occurs in the hpt366 module. Loading hptraid will not work because it depends on the kernel to claim the disks of the raid volume (that is what hpt366 would do). I will add autoloading of the ide-controller module in the next raid-driver release. However, I do not know why the kernel oopses. You might want to try to build the hpt366 code into the kernel instead of a module. If it works it would probably mean that "
ide_hwif_t*hwif" was not properly initalized.Quoted 4 lines
I initially had all the hpt modules built into the kernel, but that would also produce an oops and die immediately after ID'ing the two drives I have on attached. Would any more information be of use to you?
2.4.21-ac4 contains several fixes for the hpt37x. Please try this patch out.
bye,
Wilfried
drivers/ide/setup-pci.c renamed from drivers/ide/setup-pci.c.orig +1 -1
@@ -172,7 +172,7 @@ static int ide_setup_pci_baseregs (struc* is already in DMA mode we check and enforce IDE simplex rules. */ -static unsigned long __init ide_get_or_set_dma_base (ide_hwif_t *hwif) +static unsigned long ide_get_or_set_dma_base (ide_hwif_t *hwif) {unsigned long dma_base =0 ;struct pci_dev *dev = hwif->pci_dev;
A message that is not in the archive (sent off-list or not yet received): [email protected]
Quoted 11 lines · Wil Reichert
Quoted 2 lines
could you try the attachted patch, and report if this changes something?
Applied patch & rebuilt with hpt366 as a module. No more oops, dmesg prints the following:
HPT372A: IDE controller at PCI slot 01:0b.0 HPT372A: chipset revision 2 HPT372A: not 100% native mode: will probe irqs later hpt: HPT372N detected, using 372N timing. FREQ: 126 PLL: 45 hpt: no known IDE timings, disabling DMA. hpt: no known IDE timings, disabling DMA.
it looks like the controller is detected as a HPT372N instead of a HPT372A. the attached patch disables this check.
Quoted 6 lines
It has 2 drives attached to it, neither seems to be found.
Other things: the 2.5.xx seems to work ok and my board supports some 'RAID 1.5' which seems to be nothing more than PR crap and some firmware hacks. Could that cause problems?
raid 15 is a mirrored raid 5, which should provide high availability and good performance (even in the case of a disk failure). but in this case it is nothing more than a software solution. ;)
Quoted 2 lines
Wil
bye,
wilfried
drivers/ide/pci/hpt366.c renamed from drivers/ide/pci/hpt366.c.orig +2 -0
@@ -889,8 +889,10 @@ static int __init init_hpt37x(struct pci did =inb (dmabase +0x22 ); rid =inb (dmabase +0x28 ); +#if 0 if ((did ==4 && rid ==6 ) || (did ==5 && rid >1 )) is_372n =1 ; +#endif }/*
Quoted 12 lines · Wil Reichert
Quoted 2 lines
could you try the attachted patch, and report if this changes something?
Applied patch & rebuilt with hpt366 as a module. No more oops, dmesg prints the following:
HPT372A: IDE controller at PCI slot 01:0b.0 HPT372A: chipset revision 2 HPT372A: not 100% native mode: will probe irqs later hpt: HPT372N detected, using 372N timing. FREQ: 126 PLL: 45 hpt: no known IDE timings, disabling DMA. hpt: no known IDE timings, disabling DMA.
Quoted 7 lines
It has 2 drives attached to it, neither seems to be found.
Other things: the 2.5.xx seems to work ok and my board supports some 'RAID 1.5' which seems to be nothing more than PR crap and some firmware hacks. Could that cause problems?
Wil
ide_get_or_set_dma_base() is may be called via ide_setup_pci_device() when an ide controller module is loaded. however, ide_get_or_set_dma_base() is defined with __init and the function is called when its memory is already deallocated.
greetings,
Wilfried
drivers/ide/setup-pci.c renamed from drivers/ide/setup-pci.c.orig +1 -1
@@ -172,7 +172,7 @@ static int ide_setup_pci_baseregs (struc* is already in DMA mode we check and enforce IDE simplex rules. */ -static unsigned long __init ide_get_or_set_dma_base (ide_hwif_t *hwif) +static unsigned long ide_get_or_set_dma_base (ide_hwif_t *hwif) {unsigned long dma_base =0 ;struct pci_dev *dev = hwif->pci_dev;