Distribution:Any(tried Gentoo/Debian/Fedora/Ubuntu with latest kernels) Hardware Environment:Sis motherboard K8S MX, AMD64(2800+), SATA HDD Software Environment:GNU/Linux kernels 2.6.8 & 2.6.9 Problem Description: I have a AMD64 +asus k8s mx motherboard(SIS chipset 760GX & sis 965L) + samsung 80GB sata. I am not able to install GNU/Linux on my system as Linux does not detect my sata drive. I have tried debian, gentoo , slackware, ubuntu, fedora, mepis(all above 2.6.8 kernels) etc. without success. All the above mentined distros do not have sata module built into installer so i did ALT+F2 modprobe sata_sis. It loads libata and sata_sis but i do not see any hdd drives under /proc/scsi/scsi.Neither is the drive listed in /dev/hd* or /dev/sd* lspci -n shows sata as 0000:00:05 Unknown device 0182. Windows XP 64 bit works well on this system so there is no hardware problem. There are no drivers available from asus or sis for linux. Sata_sis module upto 2.6.10 currently supports sis964 but mine is sis965L. I have done a lot of googling and found only open ended questions without any solution. Some answers on google suggest a workaround to do install on IDE hdd and then rebuild the kernel to include sata support and then boot that kernel from SATA drive ---this seems to be a bit time consuming and i dont have a IDE drive currently. Also if there is no sis965L support there is no use of this workaround to use IDE. Please let me know if there is any other solution or/and if someone has info that the future kernels will support this chipset. Additional info: i have even inserted all sata* scsi* raid* modules(even sd_mod). Also did noapic noprob at bootup. I am really keen to run GNU/Linux on my pc as i have bought it to use blender3D with high redering speed of 64 bit AMD. Thank you in anticipation. Regards; Rituraj Steps to reproduce: Installing GNU/Linux any distro doesnt detect media for partitioning.
Created attachment 4542 [details] SiS's submission of SATA support for PCI ID 0x182 SiS submitted a patch for this, unfortunately it needs a lot of work before it can be merged.
Is this in 2.6.12-rc3? or did it made it into 2.6.11?
Has this been fixed?
The 2.6.12 kernel released today also doesnt include a patch for this. How long should I use Windows OS on my PC? Will the patch be integrated in future releases? Please update me.
I suppose it has not been fixed in 2.6.12.1 either? I have ASUS K8S-MX with sis965L and the sata_sis module doesn't work. I tried the linux driver module from sis homepage, got me a kernel panic. It was for 2.6.10 kernel tho, but built fine in 2.6.12.1. Any known workarounds?
Added Kernel Version-->"all 2.6 series upto 2.6.12"
Almost same configuration I too have. Tried installing SuSE 9.3 but not able to install. In which version I can install. I wanted to use Linux desperately.
Does the Patch work with the 2.6.12.3 Kernel?
Created attachment 5463 [details] updated SiS's support for PCI ID 0x182 to 2.6.11 I managed to update the SiS patch for 2.6.11. Working like a charm for me, hope it will help some people until the driver is fixed in the official sources...
Problems with the proposed patch: 1) duplicates SATA phy reset 2) abuses infrastructure to support PATA, rather than doing it properly. doing it properly involves an approach similar to that found in the 'promise-sata-pata' branch of libata-dev.git. Same problem as Promise SATA+PATA, with the same solution. 3) duplicates ATA bus reset, except, does it poorly 4) duplicates ata_busy_sleep() 5) appears to do strange things with PATA devices, when one uses the ->scr_write() and ->scr_read() hooks -- hooks used to talk to SATA PHYs (never PATA devices). 6) [maybe] sets DMA/PIO timings even for SATA devices. This -may- be needed, depending on PATA<->SATA bridge presence in the host controller 7) Pads DMA to 32-bit boundary. Should be done in libata core, this is needed for all host controllers. 8) The DMA pad code is very buggy. It uses the dma_map_single() to map a buffer, but never synchronizes nor flushes the buffer. This can and will lead to data corruption, particularly on x86-64 platform.
Is there any chance this patch can be cleaned up enough to be usable? I need this driver desparately for a client, and I'm afraid to set up their hardware with the sis supplied driver for 2.6.10 if it could cause data corruption. If I had C and device driver skills I would jump on it, but I don't... I can test and report though!
Created attachment 6133 [details] Mandriva_sis182_sata_sis_patch
I have seen, that Mandriva have added their own patch to the current Mandriva Linux 2006 RC2. http://www.mandrivalinux.com/en/mandrivalinux-2006-beta.php3 kernel 2.6.12-12mdk * 2.6.12.6 * Added sata_sil24 module, SiS182? minimal support, SiS190? driver, support of the IDE chipset of the SiS965L? boards. I'm currently using this patch with the 2.6.13 x86_64 kernel (gentoo) and it's working (but i don't know if it is safe too use it) libata version 1.12 loaded. ACPI: PCI Interrupt 0000:00:05.0[A] -> GSI 17 (level, low) -> IRQ 217 sata_sis: Detected SiS 182 chipset ata1: SATA max UDMA/133 cmd 0xE880 ctl 0xE802 bmdma 0xE080 irq 217 ata2: SATA max UDMA/133 cmd 0xE480 ctl 0xE402 bmdma 0xE088 irq 217 ata1: dev 0 cfg 49:2f00 82:346b 83:7d01 84:4003 85:3468 86:3c01 87:4003 88:007f ata1: dev 0 ATA, max UDMA/133, 312581808 sectors: lba48 ata1: dev 0 configured for UDMA/133 scsi0 : sata_sis ata2: no device found (phy stat 00000001) scsi1 : sata_sis Vendor: ATA Model: ST3160827AS Rev: 3.42 Type: Direct-Access ANSI SCSI revision: 05 SCSI device sda: 312581808 512-byte hdwr sectors (160042 MB) SCSI device sda: drive cache: write back SCSI device sda: 312581808 512-byte hdwr sectors (160042 MB) SCSI device sda: drive cache: write back sda: sda1 sda2 sda3 sda4 < sda5 sda6 sda7 sda8 sda9 > Attached scsi disk sda at scsi0, channel 0, id 0, lun 0
Ralf, afaik the patch made is way in 2.6.14-rc2