From Bugzilla Helper: User-Agent: Mozilla/5.0 (X11; U; Linux i686 (x86_64); en-US; rv:1.8.1.1) Gecko/20061208 Firefox/2.0.0.1 Description of problem: I tried out kernel 2.6.19-1.2895 from updates and now I can't access SATA drives on my JMicron controller. Looks like new kernel incorrectly recognized JMB363 as JMB361. This is regression since 2.6.18-1.2856. Version-Release number of selected component (if applicable): kernel-2.6.19-1.2895.fc6.x86_64 How reproducible: Always Steps to Reproduce: Boot the system. Actual Results: From dmesg: JMB361: IDE controller at PCI slot 0000:04:00.0 ACPI: PCI Interrupt 0000:04:00.0[A] -> GSI 17 (level, low) -> IRQ 17 JMB361: chipset revision 2 JMB361: 100% native mode on irq 17 JMB361: dma_base is invalid ide0: JMB361 Bus-Master DMA disabled (BIOS) JMB361: dma_base is invalid ide1: JMB361 Bus-Master DMA disabled (BIOS) Probing IDE interface ide0... Probing IDE interface ide1... JMB361: IDE controller at PCI slot 0000:04:00.1 ACPI: PCI Interrupt 0000:04:00.1[B] -> GSI 18 (level, low) -> IRQ 18 JMB361: chipset revision 2 JMB361: 100% native mode on irq 18 ide2: BM-DMA at 0xa000-0xa007, BIOS settings: hde:pio, hdf:pio ide3: BM-DMA at 0xa008-0xa00f, BIOS settings: hdg:DMA, hdh:DMA As a result, AHCI driver ignores SATA drive on jmicron controller. Expected Results: With 2.6.18, it was JMB363: IDE controller at PCI slot 0000:04:00.1 GSI 18 sharing vector 0xE1 and IRQ 18 ACPI: PCI Interrupt 0000:04:00.1[B] -> GSI 18 (level, low) -> IRQ 225 JMB363: chipset revision 2 JMB363: 100% native mode on irq 225 ide2: BM-DMA at 0xa000-0xa007, BIOS settings: hde:pio, hdf:pio ide3: BM-DMA at 0xa008-0xa00f, BIOS settings: hdg:DMA, hdh:DMA AHCI finds another sata device and all works well. Additional info: JMB361 and 363 are very alike, the only difference is that JMB361 has only PATA port and JMB363 has PATA and two SATA ports. They have different PCI IDs - 197b:2363 for JMB363 and 197b:2361 for JMB361. I have Gigabyte GA-P965-DS4 MB, JMicron controller is on board. I have 7 SATA hard drives (6 on ICH8R and one on JMicron, one port on JMicron is free). Returning back to 2.6.18 kernel, since I can't work without that drive.
also confirmed on on x86_64
Lots of people echo this problem in fedoraForums. It affects owners of GA-965P-(D)S3 and Asus P5B owners, which are some of the most popular boards out there right now for Core 2 Duo. Looks like there is a fix in 2.6.20; is this coming to FC6?
This should be fixed in FC6 2911 kernels.
Hmmm... 2.6.19-2911 didn't work on my machine (GA-965P-S3) and I spent several hours on it with fellow Linux/Unix User Group members. Any hardware confirmations of this working, i.e. is it me or is it possible the 2911 kernel is still broken?
Did a 2x check - the troubles were with the most current kernel as of 11 Feb. I thought I remembered 2911, but maybe not, as 2911 was posted on 13 Feb: http://fedoranews.org/cms/node/2648 Would they have been available on the 11th?
kernel 2911 was posted to testing on Feb 12 and released on the 13th. If you want to test, please get kernel 2.6.19-1.2911.6.3 from testing. It has many more bug fixes: http://download.fedora.redhat.com/pub/fedora/linux/core/updates/testing/6/
I can verify this bug is present in 2.6.19-1.2911.fc6.i686. I'm trying to use an optical drive with the Jmicron controller on a Gigabyte 965p-ds3.rev2 motherboard. It work fine with 2.6.18-1.2798.fc6.i686. I triedd the testing kernel 2.6.19-1.2911.6.3.fc6.i686 and the bug is still present.
Is there anything in '2911.fc6.3 that would change the status of this bug from the file mentioned in #7?
same thing with 2925.fc6.i686. bug is still there. HDD on JMicron363 SATA + PATA OpticalDD on JMicron363 work in 2798.fc6.i686 but not with any recent kernels. For my case the following's proven not to solve problem : all BIOS SATA/IDE options + irqpoll/all-generic-ide kernel options. Tried even FC7-test1,-test2: they just don't see SATA HDD on JMB363. The only workaround for me - to connect SATA HDD to ICH8 SATA port; drawback is HDD becomes twice slower than on JMB363 and PATA OpticalDD invisibility.
FYI, I'm using kernel-rt-2.6.20-0119.rt8 from http://people.redhat.com/~mingo/realtime-preempt/ It doesn't have this problem. > HDD becomes twice slower than on JMB363 Something is definitely wrong with your system. I have a lot of hard drives on ICH8 and they work very fast, faster than on jmicron.
If I build a test kernel for this issue will someone test it? And can I get confirmation that JMB363 is still being reported as 361 with kernel 2925?
I haven't tested 2925, but I'd be willing to test a new kernel.
Test kernels (version 1.2937) for this issue are at: http://people.redhat.com/cebbert Please test and report back.
(In reply to comment #13) > Please test and report back. I have tested this kernel on a GA-965P-DS3. This is the first kernel that has booted correctly since the installation kernel.
(In reply to comment #14) > I have tested this kernel on a GA-965P-DS3. > This is the first kernel that has booted correctly since the installation kernel. Can you post contents of /var/log/dmesg after booting the new kernel?
Created attachment 151184 [details] dmesg output for 2.6.20-1.2937.fc6
I'd like to confirm this bug. Here are the details of my config: This motherboard has an intel P965 chipset (ICH8). All 4 SATA drives are always recognized by Linux. It also has a JMicro IDE controller, to which I have a PATA CD-Rom drive attached, and a single SATA drive. As of this posting, I am running the latest BIOS provided by Gigabyte, F10. I am currently running kernel 2.6.19-1.2911.6.5.fc6.i686 happily on my Gigabyte GA-P965-DS3 motherboard, with FIVE SATA hard drives and CD-ROM all detected. However, all newer kernels released so far (as of this posting) no longer recognize the SATA drive attached to my JMicro IDE port, however, it still recognizes the CDROM drive. It may be of interest to note possible BIOS settings, and how the affect detection of my IDE devices. The ICH8 managed ports have an option called AHCI which can be enabled or disabled. If I enable this option, Linux does not see any of them. Interestingly, during the BIOS screen, a 'secondary' Intel AHCI BIOS is booted and detects the IDE drives, as opposed to the main bios. The JMicro ports have an option of IDE, AHCI, or RAID. Linux 2.6.19-1.2911.6.5.fc6.i686 will detect my SATA drive on this controller AND the CD-ROM PATA drive if this is set to IDE. However, if it's set to anything other than IDE, only the CD-ROM is detected. Any FC6-released kernel thats newer than 2.6.19-1.2911.6.5.fc6.i686 behaves identically as noted above, however, no matter what setting is used in BIOS, the JMicro SATA drive is NOT detected...
kernel-2.6.20-1.2943.fc6 from the testing repo still behaves as noted in comment #17. It still does not recognize my SATA drive attached to the JMicro controller. Here is a snipit from the bootlog: Uniform Multi-Platform E-IDE driver Revision: 7.00alpha2 ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx jmicron_init_one: device = 2363, driver_data = 1 JMB363: IDE controller at PCI slot 0000:05:00.0 PCI: Device 0000:05:00.0 not available because of resource collisions ACPI: PCI Interrupt 0000:05:00.0[A] -> GSI 19 (level, low) -> IRQ 18 JMB363: BIOS configuration fixed. JMB363: chipset revision 2 JMB363: 100% native mode on irq 18 JMB363: dma_base is invalid ide0: JMB363 Bus-Master DMA disabled (BIOS) JMB363: dma_base is invalid ide1: JMB363 Bus-Master DMA disabled (BIOS) Probing IDE interface ide0... Probing IDE interface ide1... jmicron_init_one: device = 2363, driver_data = 1 JMB363: IDE controller at PCI slot 0000:05:00.1 PCI: Enabling device 0000:05:00.1 (0000 -> 0001) ACPI: PCI Interrupt 0000:05:00.1[B] -> GSI 16 (level, low) -> IRQ 16 JMB363: chipset revision 2 JMB363: 100% native mode on irq 16 PCI: Setting latency timer of device 0000:05:00.1 to 64 ide2: BM-DMA at 0x8000-0x8007, BIOS settings: hde:DMA, hdf:pio ide3: BM-DMA at 0x8008-0x800f, BIOS settings: hdg:DMA, hdh:pio Probing IDE interface ide2... hde: ASUS CD-S520/A4, ATAPI CD/DVD-ROM drive hde: hw_config=0000 hde: hw_config=0000 ide2 at 0x7000-0x7007,0x7402 on irq 16 Probing IDE interface ide3... Probing IDE interface ide0... Probing IDE interface ide1... Probing IDE interface ide3...
The latest Fedora-released kernel (2944) still has this bug. /dev/hdg is NOT recognized.. nor is it presented as a /dev/sdXX... So, no change. Uniform Multi-Platform E-IDE driver Revision: 7.00alpha2 ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx jmicron_init_one: device = 2363, driver_data = 1 JMB363: IDE controller at PCI slot 0000:05:00.0 PCI: Device 0000:05:00.0 not available because of resource 2 collisions ACPI: PCI Interrupt 0000:05:00.0[A] -> GSI 19 (level, low) -> IRQ 18 JMB363: BIOS configuration fixed. JMB363: chipset revision 2 JMB363: 100% native mode on irq 18 JMB363: dma_base is invalid ide0: JMB363 Bus-Master DMA disabled (BIOS) JMB363: dma_base is invalid ide1: JMB363 Bus-Master DMA disabled (BIOS) jmicron_init_one: device = 2363, driver_data = 1 JMB363: IDE controller at PCI slot 0000:05:00.1 PCI: Enabling device 0000:05:00.1 (0000 -> 0001) ACPI: PCI Interrupt 0000:05:00.1[B] -> GSI 16 (level, low) -> IRQ 16 JMB363: chipset revision 2 JMB363: 100% native mode on irq 16 ide2: BM-DMA at 0x8000-0x8007, BIOS settings: hde:DMA, hdf:pio ide3: BM-DMA at 0x8008-0x800f, BIOS settings: hdg:DMA, hdh:pio hde: ASUS CD-S520/A4, ATAPI CD/DVD-ROM drive hde: hw_config=0000 hde: hw_config=0000 ide2 at 0x7000-0x7007,0x7402 on irq 16
(In reply to comment #19) > PCI: Device 0000:05:00.0 not available because of resource 2 collisions The new debug message worked, anyway. Can you post output of the command 'lspci -vv' for working and non-working kernels? Also can you try some PCI boot options after that: pci=nobios or pci=conf1 or pci=conf2 Try them one at a time, not all at once. They go after the "root=<whatever>" line if /etc/grub.conf
Created attachment 153069 [details] lspci-vv on working 2.6.19-1.2911.6.5.fc6 Sorry it took me so long to get these to ya.
Created attachment 153075 [details] lspci -vv on NON-working 2.6.20-1.2944.fc6
Created attachment 153076 [details] bootlog of 2.6.20-1.2944 with pci=nobios (no change)
Created attachment 153077 [details] bootlog of 2.6.20-1.2944 with pci=conf1 (no change)
Created attachment 153078 [details] bootlog of 2.6.20-1.2944 with pci=conf2 (kernel panic!)
Just tried the new kernel-2.6.20-1.2948.fc6.. Still does not recognize my hard drive. So, I'm STILL running on kernel-2.6.19-1.2911.6.5.fc6 which has a buggy sky2 driver :( Uniform Multi-Platform E-IDE driver Revision: 7.00alpha2 ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx jmicron_init_one: device = 2363, driver_data = 1 JMB363: IDE controller at PCI slot 0000:05:00.0 PCI: Device 0000:05:00.0 not available because of resource 2 collisions ACPI: PCI Interrupt 0000:05:00.0[A] -> GSI 19 (level, low) -> IRQ 18 JMB363: BIOS configuration fixed. JMB363: chipset revision 2 JMB363: 100% native mode on irq 18 JMB363: dma_base is invalid ide0: JMB363 Bus-Master DMA disabled (BIOS) JMB363: dma_base is invalid ide1: JMB363 Bus-Master DMA disabled (BIOS) Probing IDE interface ide0... Probing IDE interface ide1... jmicron_init_one: device = 2363, driver_data = 1 JMB363: IDE controller at PCI slot 0000:05:00.1 PCI: Enabling device 0000:05:00.1 (0000 -> 0001) ACPI: PCI Interrupt 0000:05:00.1[B] -> GSI 16 (level, low) -> IRQ 16 JMB363: chipset revision 2 JMB363: 100% native mode on irq 16 PCI: Setting latency timer of device 0000:05:00.1 to 64 ide2: BM-DMA at 0x8000-0x8007, BIOS settings: hde:DMA, hdf:pio ide3: BM-DMA at 0x8008-0x800f, BIOS settings: hdg:DMA, hdh:pio Probing IDE interface ide2... hde: ASUS CD-S520/A4, ATAPI CD/DVD-ROM drive hde: hw_config=0000 hde: hw_config=0000 ide2 at 0x7000-0x7007,0x7402 on irq 16 Probing IDE interface ide3... Probing IDE interface ide0... Probing IDE interface ide1... Probing IDE interface ide3...
Huh? If it's recognized as JMB363, it SHOULD work! The problem was with JMB363 recognized as JMB361. As a matter or fact, I'm using 2.6.20-1.2944.fc6 which has sky2 fix for a while and have no problems with jmicron, both sata and ide channels work perfectly. jmicron_init_one: device = 2363, driver_data = 1 JMB363: IDE controller at PCI slot 0000:06:00.1 ACPI: PCI Interrupt 0000:06:00.1[B] -> GSI 18 (level, low) -> IRQ 18 JMB363: chipset revision 2 JMB363: 100% native mode on irq 18 ide0: BM-DMA at 0xa000-0xa007, BIOS settings: hda:pio, hdb:pio ide1: BM-DMA at 0xa008-0xa00f, BIOS settings: hdc:DMA, hdd:DMA Probing IDE interface ide0... hda: TOSHIBA ODD-DVD SD-M1802, ATAPI CD/DVD-ROM drive hdb: PIONEER DVD-RW DVR-110D, ATAPI CD/DVD-ROM drive hda: hw_config=603f hda: hw_config=603f hdb: hw_config=6b00 hdb: hw_config=6b00 ide0 at 0x9000-0x9007,0x9402 on irq 18 Probing IDE interface ide1... Probing IDE interface ide1... later on: ahci 0000:06:00.0: AHCI 0001.0000 32 slots 2 ports 3 Gbps 0x3 impl SATA mode ahci 0000:06:00.0: flags: 64bit ncq pm led clo pmp pio slum part ata7: SATA max UDMA/133 cmd 0xFFFFC20000020100 ctl 0x0 bmdma 0x0 irq 17 ata8: SATA max UDMA/133 cmd 0xFFFFC20000020180 ctl 0x0 bmdma 0x0 irq 17 scsi6 : ahci ata7: SATA link up 3.0 Gbps (SStatus 123 SControl 300) ata7.00: ATA-7, max UDMA7, 488395055 sectors: LBA48 NCQ (depth 31/32) ata7.00: ata7: dev 0 multi count 16 ata7.00: configured for UDMA/133 scsi7 : ahci ata8: SATA link down (SStatus 0 SControl 300) scsi 6:0:0:0: Direct-Access ATA SAMSUNG SP2504C VT10 PQ: 0 ANSI: 5 SCSI device sde: 488395055 512-byte hdwr sectors (250058 MB) sde: Write Protect is off sde: Mode Sense: 00 3a 00 00 SCSI device sde: write cache: enabled, read cache: enabled, doesn't support DPO or FUA SCSI device sde: 488395055 512-byte hdwr sectors (250058 MB) sde: Write Protect is off sde: Mode Sense: 00 3a 00 00 SCSI device sde: write cache: enabled, read cache: enabled, doesn't support DPO or FUA sde: sde1 sde2 < sde5 sde6 sde7 sde8 sde9 sde10 sde11 sde12 > sd 6:0:0:0: Attached scsi disk sde ata7/ata8 are sata devices on jmicron (06:00): $ /sbin/lspci |grep 06:00 06:00.0 SATA controller: JMicron Technologies, Inc. JMicron 20360/20363 AHCI Controller (rev 02) 06:00.1 IDE interface: JMicron Technologies, Inc. JMicron 20360/20363 AHCI Controller (rev 02) Can you post "/sbin/lspci|grep JMicron" output? Also, are you sure your jmicron is in ahci mode? I think this bug should be closed.
Please see comment #17, it has details that you may be interested in. The bottom line: If I boot kernel-2.6.19-1.2911.6.5.fc6, my drive is there and works happily. If I boot any newer kernel, it is not there.
I don't understand what exactly do you mean by "SATA drive attached to my JMicro IDE port,". And like I said, I have two PATA devices (ODD) and one SATA HDD attached jmicron, they all work with newer 2.6.20 kernels. Jmicron is in AHCI mode, of course - there is absolutely no reason to use anything else when AHCI is available. I don't understand your problem, you are not providing enough details. Please turn on AHCI, boot system with 2.6.20-1.2944.fc6 (I guess kernel-2.6.20-1.2948.fc6 should be fine too) and post COMPLETE dmesg (in attachment). Also please post /sbin/lspci output. I don't like PCI: Device 0000:05:00.0 not available because of resource 2 collisions line in your output, but I'd like to see if it appears when AHCI is turned on. Also, have you tried turning off unused devices? Like internal soundcard/firewire/com1/com2/lpt (or setting lpt to spp mode)? Whatever this resource collision thing is, it surely doesn't sound nice.
I stand corrected. I just set the JMicro conroller to AHCI mode in my bios, and now my drive is recognized by kernel-2.6.20-1.2948.fc6. Please understand, that this is the first kernel since kernel-2.6.19-1.2911.6.5.fc6 that this has worked. (I've tried AHCI mode on every kernel release.) However, it appears to be working happily. now if I can just get sbp2 working with xen :)
Obviously some workaround is in place, but... as you can see from the output, bus mastering is disabled. This is on an Asus P5B-V. Yep... duo 64bit cpu. Linux version 2.6.20-1.2948.fc6 micron_init_one: device = 2363, driver_data = 1 JMB363: IDE controller at PCI slot 0000:02:00.0 PCI: Device 0000:02:00.0 not available because of resource 2 collisions ACPI: PCI Interrupt 0000:02:00.0[A] -> GSI 16 (level, low) -> IRQ 16 JMB363: BIOS configuration fixed. JMB363: chipset revision 2 JMB363: 100% native mode on irq 16 JMB363: dma_base is invalid ide0: JMB363 Bus-Master DMA disabled (BIOS) JMB363: dma_base is invalid ide1: JMB363 Bus-Master DMA disabled (BIOS) Probing IDE interface ide0... Probing IDE interface ide1... jmicron_init_one: device = 2363, driver_data = 1 JMB363: IDE controller at PCI slot 0000:02:00.1 PCI: Enabling device 0000:02:00.1 (0000 -> 0001) ACPI: PCI Interrupt 0000:02:00.1[B] -> GSI 17 (level, low) -> IRQ 17 JMB363: chipset revision 2 JMB363: 100% native mode on irq 17 PCI: Setting latency timer of device 0000:02:00.1 to 64 ide2: BM-DMA at 0xa400-0xa407, BIOS settings: hde:pio, hdf:pio ide3: BM-DMA at 0xa408-0xa40f, BIOS settings: hdg:pio, hdh:pio Probing IDE interface ide2... hde: ASUS DVD-E616P2, ATAPI CD/DVD-ROM drive hde: hw_config=0000 hde: hw_config=0000 ide2 at 0xac00-0xac07,0xa882 on irq 17
Dunno, works for me. Have you tried things I suggested (turning on AHCI, turning off unused devices)?
The later kernels will boot for me, but I get frequent soft lockups and error messages from the HDs: May 27 04:30:34 localhost kernel: hda: status error: status=0x59 { DriveReady SeekComplete DataRequest Error } May 27 04:30:34 localhost kernel: hda: status error: error=0x44 { AbortedCommand LastFailedSense=0x04 } May 27 04:30:34 localhost kernel: ide: failed opcode was: unknown May 27 04:30:34 localhost kernel: hda: drive not ready for command The last kernel which has worked without these problems is 2.6.18-1.2869.fc6
Are you sure your HDD is OK? I suggest checking its SMART status, also you should check interface cable. Try unplugging/plugging it again, try replacing it.