URL
https://opencores.org/ocsvn/or1k_old/or1k_old/trunk
Subversion Repositories or1k_old
[/] [or1k_old/] [trunk/] [uclinux/] [uClinux-2.0.x/] [drivers/] [block/] [triton.c] - Rev 1782
Compare with Previous | Blame | View Log
/* * linux/drivers/block/triton.c Version 1.13 Aug 12, 1996 * Version 1.13a June 1998 - new chipsets * Version 1.13b July 1998 - DMA blacklist * * Copyright (c) 1995-1996 Mark Lord * May be copied or modified under the terms of the GNU General Public License */ /* * This module provides support for Bus Master IDE DMA functions in various * motherboard chipsets and PCI controller cards. * Please check /Documentation/ide.txt and /Documentation/udma.txt for details. */ #include <linux/config.h> #include <linux/types.h> #include <linux/kernel.h> #include <linux/timer.h> #include <linux/mm.h> #include <linux/ioport.h> #include <linux/interrupt.h> #include <linux/blkdev.h> #include <linux/hdreg.h> #include <linux/pci.h> #include <linux/bios32.h> #include <asm/io.h> #include <asm/dma.h> #include "ide.h" #undef DISPLAY_TRITON_TIMINGS /* define this to display timings */ #undef DISPLAY_APOLLO_TIMINGS /* define this for extensive debugging information */ #if defined(CONFIG_PROC_FS) && defined(DISPLAY_APOLLO_TIMINGS) #include <linux/stat.h> #include <linux/proc_fs.h> #include <linux/via_ide_dma.h> #endif /* * good_dma_drives() lists the model names (from "hdparm -i") * of drives which do not support mword2 DMA but which are * known to work fine with this interface under Linux. */ const char *good_dma_drives[] = {"Micropolis 2112A", "CONNER CTMA 4000", "CONNER CTT8000-A", NULL}; /* * bad_dma_drives() lists the model names (from "hdparm -i") * of drives which supposedly support (U)DMA but which are * known to corrupt data with this interface under Linux. * * Note: the list was generated by statistical analysis of problem * reports. It's not clear if there are problems with the drives, * or with some combination of drive/controller or what. * * You can forcibly override this if you wish. This is the kernel * 'Tread carefully' list. * * Finally see http://www.wdc.com/quality/err-rec.html if you have * one of the listed drives. */ const char *bad_dma_drives[] = {"WDC AC11000H", "WDC AC22100H", "WDC AC32500H", "WDC AC33100H", NULL}; /* * Our Physical Region Descriptor (PRD) table should be large enough * to handle the biggest I/O request we are likely to see. Since requests * can have no more than 256 sectors, and since the typical blocksize is * two sectors, we could get by with a limit of 128 entries here for the * usual worst case. Most requests seem to include some contiguous blocks, * further reducing the number of table entries required. * * The driver reverts to PIO mode for individual requests that exceed * this limit (possible with 512 byte blocksizes, eg. MSDOS f/s), so handling * 100% of all crazy scenarios here is not necessary. * * As it turns out though, we must allocate a full 4KB page for this, * so the two PRD tables (ide0 & ide1) will each get half of that, * allowing each to have about 256 entries (8 bytes each) from this. */ #define PRD_BYTES 8 #define PRD_ENTRIES (PAGE_SIZE / (2 * PRD_BYTES)) #define DEFAULT_BMIBA 0xe800 /* in case BIOS did not init it */ #define DEFAULT_BMCRBA 0xcc00 /* VIA's default value */ /* * dma_intr() is the handler for disk read/write DMA interrupts */ static void dma_intr (ide_drive_t *drive) { byte stat, dma_stat; int i; struct request *rq = HWGROUP(drive)->rq; unsigned short dma_base = HWIF(drive)->dma_base; outb(inb(dma_base)&~1, dma_base); /* stop DMA operation */ dma_stat = inb(dma_base+2); /* get DMA status */ stat = GET_STAT(); /* get drive status */ if (OK_STAT(stat,DRIVE_READY,drive->bad_wstat|DRQ_STAT)) { if ((dma_stat & 7) == 4) { /* verify good DMA status */ rq = HWGROUP(drive)->rq; for (i = rq->nr_sectors; i > 0;) { i -= rq->current_nr_sectors; ide_end_request(1, HWGROUP(drive)); } return; } printk("%s: bad DMA status: 0x%02x\n", drive->name, dma_stat); } sti(); ide_error(drive, "dma_intr", stat); } /* * build_dmatable() prepares a dma request. * Returns 0 if all went okay, returns 1 otherwise. */ static int build_dmatable (ide_drive_t *drive) { struct request *rq = HWGROUP(drive)->rq; struct buffer_head *bh = rq->bh; unsigned long size, addr, *table = HWIF(drive)->dmatable; unsigned int count = 0; do { /* * Determine addr and size of next buffer area. We assume that * individual virtual buffers are always composed linearly in * physical memory. For example, we assume that any 8kB buffer * is always composed of two adjacent physical 4kB pages rather * than two possibly non-adjacent physical 4kB pages. */ if (bh == NULL) { /* paging and tape requests have (rq->bh == NULL) */ addr = virt_to_bus (rq->buffer); #ifdef CONFIG_BLK_DEV_IDETAPE if (drive->media == ide_tape) size = drive->tape.pc->request_transfer; else #endif /* CONFIG_BLK_DEV_IDETAPE */ size = rq->nr_sectors << 9; } else { /* group sequential buffers into one large buffer */ addr = virt_to_bus (bh->b_data); size = bh->b_size; while ((bh = bh->b_reqnext) != NULL) { if ((addr + size) != virt_to_bus (bh->b_data)) break; size += bh->b_size; } } /* * Fill in the dma table, without crossing any 64kB boundaries. * We assume 16-bit alignment of all blocks. */ while (size) { if (++count >= PRD_ENTRIES) { printk("%s: DMA table too small\n", drive->name); return 1; /* revert to PIO for this request */ } else { unsigned long bcount = 0x10000 - (addr & 0xffff); if (bcount > size) bcount = size; *table++ = addr; *table++ = bcount & 0xffff; addr += bcount; size -= bcount; } } } while (bh != NULL); if (count) { *--table |= 0x80000000; /* set End-Of-Table (EOT) bit */ return 0; } printk("%s: empty DMA table?\n", drive->name); return 1; /* let the PIO routines handle this weirdness */ } /* * We will only enable drives with multi-word (mode2) (U)DMA capabilities, * and ignore the very rare cases of drives that can only do single-word * (modes 0 & 1) (U)DMA transfers. We also discard "blacklisted" hard disks. */ static int config_drive_for_dma (ide_drive_t *drive) { const char **list; struct hd_driveid *id = drive->id; if (id && (id->capability & 1)) { /* Consult the list of known "bad" drives */ list = bad_dma_drives; while (*list) { if (!strcmp(*list++,id->model)) { drive->using_dma = 0; /* no DMA */ printk("ide: Disabling DMA modes on %s drive (%s).\n", drive->name, id->model); return 1; /* DMA disabled */ } } /* Enable DMA on any drive that has mode 2 UltraDMA enabled */ if (id->field_valid & 4) /* UltraDMA */ if ((id->dma_ultra & 0x404) == 0x404) { drive->using_dma = 1; return 0; /* DMA enabled */ } /* Enable DMA on any drive that has mode2 DMA enabled */ if (id->field_valid & 2) /* regular DMA */ if ((id->dma_mword & 0x404) == 0x404) { drive->using_dma = 1; return 0; /* DMA enabled */ } /* Consult the list of known "good" drives */ list = good_dma_drives; while (*list) { if (!strcmp(*list++,id->model)) { drive->using_dma = 1; return 0; /* DMA enabled */ } } } return 1; /* DMA not enabled */ } /* * triton_dmaproc() initiates/aborts DMA read/write operations on a drive. * * The caller is assumed to have selected the drive and programmed the drive's * sector address using CHS or LBA. All that remains is to prepare for DMA * and then issue the actual read/write DMA/PIO command to the drive. * * For ATAPI devices, we just prepare for DMA and return. The caller should * then issue the packet command to the drive and call us again with * ide_dma_begin afterwards. * * Returns 0 if all went well. * Returns 1 if DMA read/write could not be started, in which case * the caller should revert to PIO for the current request. */ static int triton_dmaproc (ide_dma_action_t func, ide_drive_t *drive) { unsigned long dma_base = HWIF(drive)->dma_base; unsigned int reading = (1 << 3); switch (func) { case ide_dma_abort: outb(inb(dma_base)&~1, dma_base); /* stop DMA */ return 0; case ide_dma_check: return config_drive_for_dma (drive); case ide_dma_write: reading = 0; case ide_dma_read: break; case ide_dma_status_bad: return ((inb(dma_base+2) & 7) != 4); /* verify good DMA status */ case ide_dma_transferred: #if 0 return (number of bytes actually transferred); #else return (0); #endif case ide_dma_begin: outb(inb(dma_base)|1, dma_base); /* begin DMA */ return 0; default: printk("triton_dmaproc: unsupported func: %d\n", func); return 1; } if (build_dmatable (drive)) return 1; outl(virt_to_bus (HWIF(drive)->dmatable), dma_base + 4); /* PRD table */ outb(reading, dma_base); /* specify r/w */ outb(inb(dma_base+2)|0x06, dma_base+2); /* clear status bits */ #ifdef CONFIG_BLK_DEV_IDEATAPI if (drive->media != ide_disk) return 0; #endif /* CONFIG_BLK_DEV_IDEATAPI */ ide_set_handler(drive, &dma_intr, WAIT_CMD); /* issue cmd to drive */ OUT_BYTE(reading ? WIN_READDMA : WIN_WRITEDMA, IDE_COMMAND_REG); outb(inb(dma_base)|1, dma_base); /* begin DMA */ return 0; } #ifdef DISPLAY_TRITON_TIMINGS /* * print_triton_drive_flags() displays the currently programmed options * in the i82371 (Triton) for a given drive. * * If fastDMA is "no", then slow ISA timings are used for DMA data xfers. * If fastPIO is "no", then slow ISA timings are used for PIO data xfers. * If IORDY is "no", then IORDY is assumed to always be asserted. * If PreFetch is "no", then data pre-fetch/post are not used. * * When "fastPIO" and/or "fastDMA" are "yes", then faster PCI timings and * back-to-back 16-bit data transfers are enabled, using the sample_CLKs * and recovery_CLKs (PCI clock cycles) timing parameters for that interface. */ static void print_triton_drive_flags (unsigned int unit, byte flags) { printk(" %s ", unit ? "slave :" : "master:"); printk( "fastDMA=%s", (flags&9) ? "on " : "off"); printk(" PreFetch=%s", (flags&4) ? "on " : "off"); printk(" IORDY=%s", (flags&2) ? "on " : "off"); printk(" fastPIO=%s\n", ((flags&9)==1) ? "on " : "off"); } #endif /* DISPLAY_TRITON_TIMINGS */ static void init_triton_dma (ide_hwif_t *hwif, unsigned short base) { static unsigned long dmatable = 0; printk(" %s: BM-DMA at 0x%04x-0x%04x", hwif->name, base, base+7); if (check_region(base, 8)) { printk(" -- ERROR, PORTS ALREADY IN USE"); } else { request_region(base, 8, "IDE DMA"); hwif->dma_base = base; if (!dmatable) { /* * The BM-DMA uses a full 32-bits, so we can * safely use __get_free_page() here instead * of __get_dma_pages() -- no ISA limitations. */ dmatable = __get_free_pages(GFP_KERNEL, 1, 0); } if (dmatable) { hwif->dmatable = (unsigned long *) dmatable; dmatable += (PRD_ENTRIES * PRD_BYTES); outl(virt_to_bus(hwif->dmatable), base + 4); hwif->dmaproc = &triton_dmaproc; } } printk("\n"); } /* * Set VIA Chipset Timings for (U)DMA modes enabled. */ static int set_via_timings (byte bus, byte fn, byte post, byte flush) { byte via_config = 0; int rc = 0; /* setting IDE read prefetch buffer and IDE post write buffer */ if ((rc = pcibios_read_config_byte(bus, fn, 0x41, &via_config))) return (1); if ((rc = pcibios_write_config_byte(bus, fn, 0x41, via_config | post))) return (1); /* setting Channel read and End-of-sector FIFO flush: */ if ((rc = pcibios_read_config_byte(bus, fn, 0x46, &via_config))) return (1); if ((rc = pcibios_write_config_byte(bus, fn, 0x46, via_config | flush))) return (1); return (0); } /* * ide_init_triton() prepares the IDE driver for DMA operation. * This routine is called once, from ide.c during driver initialization, * for each BM-DMA chipset which is found (rarely more than one). */ void ide_init_triton (byte bus, byte fn) { int rc = 0, h; int dma_enabled = 0; unsigned short io[6], count = 0, step_count = 0; unsigned short pcicmd, vendor, device, class; unsigned int bmiba, timings, reg, tmp; unsigned int addressbios = 0; #ifdef DISPLAY_APOLLO_TIMINGS bmide_bus = bus; bmide_fn = fn; #endif /* DISPLAY_APOLLO_TIMINGS */ /* * We pick up the vendor, device, and class info for selecting the correct * controller that is supported. Since we can access this routine more than * once with the use of onboard and off-board EIDE controllers, a method * of determining "who is who for what" is needed. */ pcibios_read_config_word (bus, fn, PCI_VENDOR_ID, &vendor); pcibios_read_config_word (bus, fn, PCI_DEVICE_ID, &device); pcibios_read_config_word (bus, fn, PCI_CLASS_DEVICE, &class); switch(vendor) { case PCI_VENDOR_ID_INTEL: printk("ide: Intel 82371 (single FIFO) DMA Bus Mastering IDE "); break; case PCI_VENDOR_ID_SI: printk("ide: SiS 5513 (dual FIFO) DMA Bus Mastering IDE "); break; case PCI_VENDOR_ID_VIA: printk("ide: VIA VT82C586B (split FIFO) UDMA Bus Mastering IDE "); break; case PCI_VENDOR_ID_PROMISE: /* PCI_CLASS_STORAGE_RAID == class */ /* * I have been able to make my Promise Ultra33 UDMA card change class. * It has reported as both PCI_CLASS_STORAGE_RAID and PCI_CLASS_STORAGE_IDE. * Since the PCI_CLASS_STORAGE_RAID mode should automatically mirror the * two halves of the PCI_CONFIG register data, but sometimes it forgets. * Thus we guarantee that they are identical, with a quick check and * correction if needed. * PDC20246 (primary) PDC20247 (secondary) IDE hwif's. * * Note that Promise "stories,fibs,..." about this device not being * capable of ATAPI and AT devices. */ if (PCI_CLASS_STORAGE_RAID == class) { unsigned char irq1 = 0, irq2 = 0; pcibios_read_config_byte (bus, fn, PCI_INTERRUPT_LINE, &irq1); pcibios_read_config_byte (bus, fn, (PCI_INTERRUPT_LINE)|0x80, &irq2); if (irq1 != irq2) { pcibios_write_config_byte(bus, fn, (PCI_INTERRUPT_LINE)|0x80, irq1); } } case PCI_VENDOR_ID_ARTOP: /* PCI_CLASS_STORAGE_SCSI == class */ /* * I have found that by stroking rom_enable_bit on both the AEC6210U/UF and * PDC20246 controller cards, the features desired are almost guaranteed * to be enabled and compatible. This ROM may not be registered in the * config data, but it can be turned on. Registration failure has only * been observed if and only if Linux sets up the pci_io_address in the * 0x6000 range. If they are setup in the 0xef00 range it is reported. * WHY??? got me......... */ printk("ide: %s UDMA Bus Mastering ", (vendor == PCI_VENDOR_ID_ARTOP) ? "AEC6210" : "PDC20246"); pcibios_read_config_dword(bus, fn, PCI_ROM_ADDRESS, &addressbios); if (addressbios) { pcibios_write_config_byte(bus, fn, PCI_ROM_ADDRESS, addressbios | PCI_ROM_ADDRESS_ENABLE); printk("with ROM enabled at 0x%08x", addressbios); } /* * This was stripped out of 2.1.XXX kernel code and parts from a patch called * promise_update. This finds the PCI_BASE_ADDRESS spaces and makes them * available for configuration later. * PCI_BASE_ADDRESS_0 hwif0->io_base * PCI_BASE_ADDRESS_1 hwif0->ctl_port * PCI_BASE_ADDRESS_2 hwif1->io_base * PCI_BASE_ADDRESS_3 hwif1->ctl_port * PCI_BASE_ADDRESS_4 bmiba */ memset(io, 0, 6 * sizeof(unsigned short)); for (reg = PCI_BASE_ADDRESS_0; reg <= PCI_BASE_ADDRESS_5; reg += 4) { pcibios_read_config_dword(bus, fn, reg, &tmp); if (tmp & PCI_BASE_ADDRESS_SPACE_IO) io[count++] = tmp & PCI_BASE_ADDRESS_IO_MASK; } break; default: return; } printk("\n Controller on PCI bus %d function %d\n", bus, fn); /* * See if IDE and BM-DMA features are enabled: */ if ((rc = pcibios_read_config_word(bus, fn, PCI_COMMAND, &pcicmd))) goto quit; if ((pcicmd & 1) == 0) { printk("ide: ports are not enabled (BIOS)\n"); goto quit; } if ((pcicmd & 4) == 0) { printk("ide: BM-DMA feature is not enabled (BIOS)\n"); } else { /* * Get the bmiba base address */ int try_again = 1; do { if ((rc = pcibios_read_config_dword(bus, fn, PCI_BASE_ADDRESS_4, &bmiba))) goto quit; bmiba &= 0xfff0; /* extract port base address */ if (bmiba) { dma_enabled = 1; break; } else { printk("ide: BM-DMA base register is invalid (0x%04x, PnP BIOS problem)\n", bmiba); if (inb(((vendor == PCI_VENDOR_ID_VIA) ? DEFAULT_BMCRBA : DEFAULT_BMIBA)) != 0xff || !try_again) break; printk("ide: setting BM-DMA base register to 0x%04x\n", ((vendor == PCI_VENDOR_ID_VIA) ? DEFAULT_BMCRBA : DEFAULT_BMIBA)); if ((rc = pcibios_write_config_word(bus, fn, PCI_COMMAND, pcicmd&~1))) goto quit; rc = pcibios_write_config_dword(bus, fn, 0x20, ((vendor == PCI_VENDOR_ID_VIA) ? DEFAULT_BMCRBA : DEFAULT_BMIBA)|1); if (pcibios_write_config_word(bus, fn, PCI_COMMAND, pcicmd|5) || rc) goto quit; } } while (try_again--); } /* * See if ide port(s) are enabled */ if ((rc = pcibios_read_config_dword(bus, fn, (vendor == PCI_VENDOR_ID_PROMISE) ? 0x50 : (vendor == PCI_VENDOR_ID_ARTOP) ? 0x54 : 0x40, &timings))) goto quit; /* * We do a vendor check since the Ultra33 and AEC6210 * holds their timings in a different location. */ printk("ide: timings == %08x\n", timings); /* * The switch preserves some stuff that was original. */ switch(vendor) { case PCI_VENDOR_ID_INTEL: if (!(timings & 0x80008000)) { printk("ide: INTEL: neither port is enabled\n"); goto quit; } break; case PCI_VENDOR_ID_VIA: if(!(timings & 0x03)) { printk("ide: VIA: neither port is enabled\n"); goto quit; } break; case PCI_VENDOR_ID_SI: case PCI_VENDOR_ID_PROMISE: case PCI_VENDOR_ID_ARTOP: default: break; } /* * Save the dma_base port addr for each interface */ for (h = 0; h < MAX_HWIFS; ++h) { ide_hwif_t *hwif = &ide_hwifs[h]; /* * This prevents the first contoller from accidentally * initalizing the hwif's that it does not use and block * an off-board ide-pci from getting in the game. */ if (step_count >= 2) { goto quit; } #ifdef CONFIG_BLK_DEV_OFFBOARD /* * This is a forced override for the onboard ide controller * to be enabled, if one chooses to have an offboard ide-pci * card as the primary booting device. This beasty is * for offboard UDMA upgrades with hard disks, but saving * the onboard DMA2 controllers for CDROMS, TAPES, ZIPS, etc... */ if ((vendor == PCI_VENDOR_ID_INTEL) || (vendor == PCI_VENDOR_ID_SI) || (vendor == PCI_VENDOR_ID_VIA)) { if (h == 2) { hwif->io_base = 0x1f0; hwif->ctl_port = 0x3f6; hwif->irq = 14; hwif->noprobe = 0; } if (h == 3) { hwif->io_base = 0x170; hwif->ctl_port = 0x376; hwif->irq = 15; hwif->noprobe = 0; } } #endif /* CONFIG_BLK_DEV_OFFBOARD */ /* * If the chipset is listed as "ide_unknown", lets get a * hwif while they last. This does the first check on * the current availability of the ide_hwifs[h] in question. */ if (hwif->chipset != ide_unknown) { continue; } else if (vendor == PCI_VENDOR_ID_INTEL) { unsigned short time; #ifdef DISPLAY_TRITON_TIMINGS byte s_clks, r_clks; unsigned short devid; #endif /* DISPLAY_TRITON_TIMINGS */ if (hwif->io_base == 0x1f0) { time = timings & 0xffff; if ((time & 0x8000) == 0) /* interface enabled? */ continue; hwif->chipset = ide_triton; if (dma_enabled) init_triton_dma(hwif, bmiba); step_count++; } else if (hwif->io_base == 0x170) { time = timings >> 16; if ((time & 0x8000) == 0) /* interface enabled? */ continue; hwif->chipset = ide_triton; if (dma_enabled) init_triton_dma(hwif, bmiba + 8); step_count++; } else { continue; } #ifdef DISPLAY_TRITON_TIMINGS s_clks = ((~time >> 12) & 3) + 2; r_clks = ((~time >> 8) & 3) + 1; printk(" %s timing: (0x%04x) sample_CLKs=%d, recovery_CLKs=%d\n", hwif->name, time, s_clks, r_clks); if ((time & 0x40) && !pcibios_read_config_word(bus, fn, PCI_DEVICE_ID, &devid) && devid == PCI_DEVICE_ID_INTEL_82371SB_1) { byte stime; if (pcibios_read_config_byte(bus, fn, 0x44, &stime)) { if (hwif->io_base == 0x1f0) { s_clks = ~stime >> 6; r_clks = ~stime >> 4; } else { s_clks = ~stime >> 2; r_clks = ~stime; } s_clks = (s_clks & 3) + 2; r_clks = (r_clks & 3) + 1; printk(" slave: sample_CLKs=%d, recovery_CLKs=%d\n", s_clks, r_clks); } } print_triton_drive_flags (0, time & 0xf); print_triton_drive_flags (1, (time >> 4) & 0xf); #endif /* DISPLAY_TRITON_TIMINGS */ } else if (vendor == PCI_VENDOR_ID_SI) { if (hwif->io_base == 0x1f0) { hwif->chipset = ide_triton; if (dma_enabled) init_triton_dma(hwif, bmiba); step_count++; } else if (hwif->io_base == 0x170) { hwif->chipset = ide_triton; if (dma_enabled) init_triton_dma(hwif, bmiba + 8); step_count++; } else { continue; } } else if(vendor == PCI_VENDOR_ID_VIA) { if (hwif->io_base == 0x1f0) { if((timings & 0x02) == 0) continue; hwif->chipset = ide_triton; if (dma_enabled) init_triton_dma(hwif, bmiba); if (set_via_timings(bus, fn, 0xc0, 0xa0)) goto quit; #ifdef DISPLAY_APOLLO_TIMINGS proc_register_dynamic(&proc_root, &via_proc_entry); #endif /* DISPLAY_APOLLO_TIMINGS */ step_count++; } else if (hwif->io_base == 0x170) { if((timings & 0x01) == 0) continue; hwif->chipset = ide_triton; if (dma_enabled) init_triton_dma(hwif, bmiba + 8); if (set_via_timings(bus, fn, 0x30, 0x50)) goto quit; step_count++; } else { continue; } } else if ((vendor == PCI_VENDOR_ID_PROMISE) || (vendor == PCI_VENDOR_ID_ARTOP)) { /* * This silly tmp = h routine allows an off-board ide-pci card to * be booted as primary hwifgroup, provided that the onboard * controllers are disabled. If they are active, then we wait our * turn for hwif assignment. */ unsigned char irq = 0; pcibios_read_config_byte (bus, fn, PCI_INTERRUPT_LINE, &irq); if ((h == 0) || (h == 1)) { tmp = h * 2; } else { tmp = (h - 2) * 2; } hwif->io_base = io[tmp]; hwif->ctl_port = io[tmp + 1] + 2; hwif->irq = irq; hwif->noprobe = 0; if (vendor == PCI_VENDOR_ID_ARTOP) { hwif->serialized = 1; } if (dma_enabled) { if (!check_region(bmiba, 8)) { hwif->chipset = ide_udma; init_triton_dma(hwif, bmiba); step_count++; } else if (!check_region((bmiba + 0x08), 8)) { if ((vendor == PCI_VENDOR_ID_PROMISE) && (!check_region(bmiba+16, 16))) { request_region(bmiba+16, 16, "PDC20246"); } hwif->chipset = ide_udma; init_triton_dma(hwif, bmiba + 8); step_count++; } else { continue; } } } } quit: if (rc) printk("ide: pcibios access failed - %s\n", pcibios_strerror(rc)); }