/* * CDDL HEADER START * * The contents of this file are subject to the terms of the * Common Development and Distribution License (the "License"). * You may not use this file except in compliance with the License. * * You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE * or http://www.opensolaris.org/os/licensing. * See the License for the specific language governing permissions * and limitations under the License. * * When distributing Covered Code, include this CDDL HEADER in each * file and include the License file at usr/src/OPENSOLARIS.LICENSE. * If applicable, add the following below this CDDL HEADER, with the * fields enclosed by brackets "[]" replaced with your own identifying * information: Portions Copyright [yyyy] [name of copyright owner] * * CDDL HEADER END */ /* * Copyright 2009 Sun Microsystems, Inc. All rights reserved. * Use is subject to license terms. */ /* * This module provides support for labeling operations for target * drivers. */ #include #include #include #include #include #include #include #include #include #include #include /* * Driver minor node structure and data table */ struct driver_minor_data { char *name; minor_t minor; int type; }; static struct driver_minor_data dk_minor_data[] = { {"a", 0, S_IFBLK}, {"b", 1, S_IFBLK}, {"c", 2, S_IFBLK}, {"d", 3, S_IFBLK}, {"e", 4, S_IFBLK}, {"f", 5, S_IFBLK}, {"g", 6, S_IFBLK}, {"h", 7, S_IFBLK}, #if defined(_SUNOS_VTOC_16) {"i", 8, S_IFBLK}, {"j", 9, S_IFBLK}, {"k", 10, S_IFBLK}, {"l", 11, S_IFBLK}, {"m", 12, S_IFBLK}, {"n", 13, S_IFBLK}, {"o", 14, S_IFBLK}, {"p", 15, S_IFBLK}, #endif /* defined(_SUNOS_VTOC_16) */ #if defined(_FIRMWARE_NEEDS_FDISK) {"q", 16, S_IFBLK}, {"r", 17, S_IFBLK}, {"s", 18, S_IFBLK}, {"t", 19, S_IFBLK}, {"u", 20, S_IFBLK}, #endif /* defined(_FIRMWARE_NEEDS_FDISK) */ {"a,raw", 0, S_IFCHR}, {"b,raw", 1, S_IFCHR}, {"c,raw", 2, S_IFCHR}, {"d,raw", 3, S_IFCHR}, {"e,raw", 4, S_IFCHR}, {"f,raw", 5, S_IFCHR}, {"g,raw", 6, S_IFCHR}, {"h,raw", 7, S_IFCHR}, #if defined(_SUNOS_VTOC_16) {"i,raw", 8, S_IFCHR}, {"j,raw", 9, S_IFCHR}, {"k,raw", 10, S_IFCHR}, {"l,raw", 11, S_IFCHR}, {"m,raw", 12, S_IFCHR}, {"n,raw", 13, S_IFCHR}, {"o,raw", 14, S_IFCHR}, {"p,raw", 15, S_IFCHR}, #endif /* defined(_SUNOS_VTOC_16) */ #if defined(_FIRMWARE_NEEDS_FDISK) {"q,raw", 16, S_IFCHR}, {"r,raw", 17, S_IFCHR}, {"s,raw", 18, S_IFCHR}, {"t,raw", 19, S_IFCHR}, {"u,raw", 20, S_IFCHR}, #endif /* defined(_FIRMWARE_NEEDS_FDISK) */ {0} }; static struct driver_minor_data dk_minor_data_efi[] = { {"a", 0, S_IFBLK}, {"b", 1, S_IFBLK}, {"c", 2, S_IFBLK}, {"d", 3, S_IFBLK}, {"e", 4, S_IFBLK}, {"f", 5, S_IFBLK}, {"g", 6, S_IFBLK}, {"wd", 7, S_IFBLK}, #if defined(_FIRMWARE_NEEDS_FDISK) {"q", 16, S_IFBLK}, {"r", 17, S_IFBLK}, {"s", 18, S_IFBLK}, {"t", 19, S_IFBLK}, {"u", 20, S_IFBLK}, #endif /* defined(_FIRMWARE_NEEDS_FDISK) */ {"a,raw", 0, S_IFCHR}, {"b,raw", 1, S_IFCHR}, {"c,raw", 2, S_IFCHR}, {"d,raw", 3, S_IFCHR}, {"e,raw", 4, S_IFCHR}, {"f,raw", 5, S_IFCHR}, {"g,raw", 6, S_IFCHR}, {"wd,raw", 7, S_IFCHR}, #if defined(_FIRMWARE_NEEDS_FDISK) {"q,raw", 16, S_IFCHR}, {"r,raw", 17, S_IFCHR}, {"s,raw", 18, S_IFCHR}, {"t,raw", 19, S_IFCHR}, {"u,raw", 20, S_IFCHR}, #endif /* defined(_FIRMWARE_NEEDS_FDISK) */ {0} }; /* * Declare the dynamic properties implemented in prop_op(9E) implementation * that we want to have show up in a di_init(3DEVINFO) device tree snapshot * of drivers that call cmlb_attach(). */ static i_ddi_prop_dyn_t cmlb_prop_dyn[] = { {"Nblocks", DDI_PROP_TYPE_INT64, S_IFBLK}, {"Size", DDI_PROP_TYPE_INT64, S_IFCHR}, {"device-nblocks", DDI_PROP_TYPE_INT64}, {"device-blksize", DDI_PROP_TYPE_INT}, {NULL} }; /* * External kernel interfaces */ extern struct mod_ops mod_miscops; extern int ddi_create_internal_pathname(dev_info_t *dip, char *name, int spec_type, minor_t minor_num); /* * Global buffer and mutex for debug logging */ static char cmlb_log_buffer[1024]; static kmutex_t cmlb_log_mutex; struct cmlb_lun *cmlb_debug_cl = NULL; uint_t cmlb_level_mask = 0x0; int cmlb_rot_delay = 4; /* default rotational delay */ static struct modlmisc modlmisc = { &mod_miscops, /* Type of module */ "Common Labeling module" }; static struct modlinkage modlinkage = { MODREV_1, (void *)&modlmisc, NULL }; /* Local function prototypes */ static dev_t cmlb_make_device(struct cmlb_lun *cl); static int cmlb_validate_geometry(struct cmlb_lun *cl, int forcerevalid, int flags, void *tg_cookie); static void cmlb_resync_geom_caches(struct cmlb_lun *cl, diskaddr_t capacity, void *tg_cookie); static int cmlb_read_fdisk(struct cmlb_lun *cl, diskaddr_t capacity, void *tg_cookie); static void cmlb_swap_efi_gpt(efi_gpt_t *e); static void cmlb_swap_efi_gpe(int nparts, efi_gpe_t *p); static int cmlb_validate_efi(efi_gpt_t *labp); static int cmlb_use_efi(struct cmlb_lun *cl, diskaddr_t capacity, int flags, void *tg_cookie); static void cmlb_build_default_label(struct cmlb_lun *cl, void *tg_cookie); static int cmlb_uselabel(struct cmlb_lun *cl, struct dk_label *l, int flags); #if defined(_SUNOS_VTOC_8) static void cmlb_build_user_vtoc(struct cmlb_lun *cl, struct vtoc *user_vtoc); #endif static int cmlb_build_label_vtoc(struct cmlb_lun *cl, struct vtoc *user_vtoc); static int cmlb_write_label(struct cmlb_lun *cl, void *tg_cookie); static int cmlb_set_vtoc(struct cmlb_lun *cl, struct dk_label *dkl, void *tg_cookie); static void cmlb_clear_efi(struct cmlb_lun *cl, void *tg_cookie); static void cmlb_clear_vtoc(struct cmlb_lun *cl, void *tg_cookie); static void cmlb_setup_default_geometry(struct cmlb_lun *cl, void *tg_cookie); static int cmlb_create_minor_nodes(struct cmlb_lun *cl); static int cmlb_check_update_blockcount(struct cmlb_lun *cl, void *tg_cookie); static int cmlb_check_efi_mbr(uchar_t *buf, int *is_mbr); #if defined(__i386) || defined(__amd64) static int cmlb_update_fdisk_and_vtoc(struct cmlb_lun *cl, void *tg_cookie); #endif #if defined(_FIRMWARE_NEEDS_FDISK) static int cmlb_has_max_chs_vals(struct ipart *fdp); #endif #if defined(_SUNOS_VTOC_16) static void cmlb_convert_geometry(diskaddr_t capacity, struct dk_geom *cl_g); #endif static int cmlb_dkio_get_geometry(struct cmlb_lun *cl, caddr_t arg, int flag, void *tg_cookie); static int cmlb_dkio_set_geometry(struct cmlb_lun *cl, caddr_t arg, int flag); static int cmlb_dkio_get_partition(struct cmlb_lun *cl, caddr_t arg, int flag, void *tg_cookie); static int cmlb_dkio_set_partition(struct cmlb_lun *cl, caddr_t arg, int flag); static int cmlb_dkio_get_efi(struct cmlb_lun *cl, caddr_t arg, int flag, void *tg_cookie); static int cmlb_dkio_set_efi(struct cmlb_lun *cl, dev_t dev, caddr_t arg, int flag, void *tg_cookie); static int cmlb_dkio_get_vtoc(struct cmlb_lun *cl, caddr_t arg, int flag, void *tg_cookie); static int cmlb_dkio_get_extvtoc(struct cmlb_lun *cl, caddr_t arg, int flag, void *tg_cookie); static int cmlb_dkio_set_vtoc(struct cmlb_lun *cl, dev_t dev, caddr_t arg, int flag, void *tg_cookie); static int cmlb_dkio_set_extvtoc(struct cmlb_lun *cl, dev_t dev, caddr_t arg, int flag, void *tg_cookie); static int cmlb_dkio_get_mboot(struct cmlb_lun *cl, caddr_t arg, int flag, void *tg_cookie); static int cmlb_dkio_set_mboot(struct cmlb_lun *cl, caddr_t arg, int flag, void *tg_cookie); static int cmlb_dkio_partition(struct cmlb_lun *cl, caddr_t arg, int flag, void *tg_cookie); #if defined(__i386) || defined(__amd64) static int cmlb_dkio_get_virtgeom(struct cmlb_lun *cl, caddr_t arg, int flag); static int cmlb_dkio_get_phygeom(struct cmlb_lun *cl, caddr_t arg, int flag); static int cmlb_dkio_partinfo(struct cmlb_lun *cl, dev_t dev, caddr_t arg, int flag); static int cmlb_dkio_extpartinfo(struct cmlb_lun *cl, dev_t dev, caddr_t arg, int flag); #endif static void cmlb_dbg(uint_t comp, struct cmlb_lun *cl, const char *fmt, ...); static void cmlb_v_log(dev_info_t *dev, char *label, uint_t level, const char *fmt, va_list ap); static void cmlb_log(dev_info_t *dev, char *label, uint_t level, const char *fmt, ...); int _init(void) { mutex_init(&cmlb_log_mutex, NULL, MUTEX_DRIVER, NULL); return (mod_install(&modlinkage)); } int _info(struct modinfo *modinfop) { return (mod_info(&modlinkage, modinfop)); } int _fini(void) { int err; if ((err = mod_remove(&modlinkage)) != 0) { return (err); } mutex_destroy(&cmlb_log_mutex); return (err); } /* * cmlb_dbg is used for debugging to log additional info * Level of output is controlled via cmlb_level_mask setting. */ static void cmlb_dbg(uint_t comp, struct cmlb_lun *cl, const char *fmt, ...) { va_list ap; dev_info_t *dev; uint_t level_mask = 0; ASSERT(cl != NULL); dev = CMLB_DEVINFO(cl); ASSERT(dev != NULL); /* * Filter messages based on the global component and level masks, * also print if cl matches the value of cmlb_debug_cl, or if * cmlb_debug_cl is set to NULL. */ if (comp & CMLB_TRACE) level_mask |= CMLB_LOGMASK_TRACE; if (comp & CMLB_INFO) level_mask |= CMLB_LOGMASK_INFO; if (comp & CMLB_ERROR) level_mask |= CMLB_LOGMASK_ERROR; if ((cmlb_level_mask & level_mask) && ((cmlb_debug_cl == NULL) || (cmlb_debug_cl == cl))) { va_start(ap, fmt); cmlb_v_log(dev, CMLB_LABEL(cl), CE_CONT, fmt, ap); va_end(ap); } } /* * cmlb_log is basically a duplicate of scsi_log. It is redefined here * so that this module does not depend on scsi module. */ static void cmlb_log(dev_info_t *dev, char *label, uint_t level, const char *fmt, ...) { va_list ap; va_start(ap, fmt); cmlb_v_log(dev, label, level, fmt, ap); va_end(ap); } static void cmlb_v_log(dev_info_t *dev, char *label, uint_t level, const char *fmt, va_list ap) { static char name[256]; int log_only = 0; int boot_only = 0; int console_only = 0; mutex_enter(&cmlb_log_mutex); if (dev) { if (level == CE_PANIC || level == CE_WARN || level == CE_NOTE) { (void) sprintf(name, "%s (%s%d):\n", ddi_pathname(dev, cmlb_log_buffer), label, ddi_get_instance(dev)); } else { name[0] = '\0'; } } else { (void) sprintf(name, "%s:", label); } (void) vsprintf(cmlb_log_buffer, fmt, ap); switch (cmlb_log_buffer[0]) { case '!': log_only = 1; break; case '?': boot_only = 1; break; case '^': console_only = 1; break; } switch (level) { case CE_NOTE: level = CE_CONT; /* FALLTHROUGH */ case CE_CONT: case CE_WARN: case CE_PANIC: if (boot_only) { cmn_err(level, "?%s\t%s", name, &cmlb_log_buffer[1]); } else if (console_only) { cmn_err(level, "^%s\t%s", name, &cmlb_log_buffer[1]); } else if (log_only) { cmn_err(level, "!%s\t%s", name, &cmlb_log_buffer[1]); } else { cmn_err(level, "%s\t%s", name, cmlb_log_buffer); } break; case CE_IGNORE: break; default: cmn_err(CE_CONT, "^DEBUG: %s\t%s", name, cmlb_log_buffer); break; } mutex_exit(&cmlb_log_mutex); } /* * cmlb_alloc_handle: * * Allocates a handle. * * Arguments: * cmlbhandlep pointer to handle * * Notes: * Allocates a handle and stores the allocated handle in the area * pointed to by cmlbhandlep * * Context: * Kernel thread only (can sleep). */ void cmlb_alloc_handle(cmlb_handle_t *cmlbhandlep) { struct cmlb_lun *cl; cl = kmem_zalloc(sizeof (struct cmlb_lun), KM_SLEEP); ASSERT(cmlbhandlep != NULL); cl->cl_state = CMLB_INITED; cl->cl_def_labeltype = CMLB_LABEL_UNDEF; mutex_init(CMLB_MUTEX(cl), NULL, MUTEX_DRIVER, NULL); *cmlbhandlep = (cmlb_handle_t)(cl); } /* * cmlb_free_handle * * Frees handle. * * Arguments: * cmlbhandlep pointer to handle */ void cmlb_free_handle(cmlb_handle_t *cmlbhandlep) { struct cmlb_lun *cl; cl = (struct cmlb_lun *)*cmlbhandlep; if (cl != NULL) { mutex_destroy(CMLB_MUTEX(cl)); kmem_free(cl, sizeof (struct cmlb_lun)); } } /* * cmlb_attach: * * Attach handle to device, create minor nodes for device. * * Arguments: * devi pointer to device's dev_info structure. * tgopsp pointer to array of functions cmlb can use to callback * to target driver. * * device_type Peripheral device type as defined in * scsi/generic/inquiry.h * * is_removable whether or not device is removable. * 0 non-removable, 1 removable. * * is_hotpluggable whether or not device is hotpluggable. * 0 non-hotpluggable, 1 hotpluggable. * * node_type minor node type (as used by ddi_create_minor_node) * * alter_behavior * bit flags: * * CMLB_CREATE_ALTSLICE_VTOC_16_DTYPE_DIRECT: create * an alternate slice for the default label, if * device type is DTYPE_DIRECT an architectures default * label type is VTOC16. * Otherwise alternate slice will no be created. * * * CMLB_FAKE_GEOM_LABEL_IOCTLS_VTOC8: report a default * geometry and label for DKIOCGGEOM and DKIOCGVTOC * on architecture with VTOC8 label types. * * CMLB_OFF_BY_ONE: do the workaround for legacy off-by- * one bug in obtaining capacity (in sd): * SCSI READ_CAPACITY command returns the LBA number of the * last logical block, but sd once treated this number as * disks' capacity on x86 platform. And LBAs are addressed * based 0. So the last block was lost on x86 platform. * * Now, we remove this workaround. In order for present sd * driver to work with disks which are labeled/partitioned * via previous sd, we add workaround as follows: * * 1) Locate backup EFI label: cmlb searches the next to * last * block for backup EFI label. If fails, it will * turn to the last block for backup EFI label; * * 2) Clear backup EFI label: cmlb first search the last * block for backup EFI label, and will search the * next to last block only if failed for the last * block. * * 3) Calculate geometry:refer to cmlb_convert_geometry() * If capacity increasing by 1 causes disks' capacity * to cross over the limits in geometry calculation, * geometry info will change. This will raise an issue: * In case that primary VTOC label is destroyed, format * commandline can restore it via backup VTOC labels. * And format locates backup VTOC labels by use of * geometry. So changing geometry will * prevent format from finding backup VTOC labels. To * eliminate this side effect for compatibility, * sd uses (capacity -1) to calculate geometry; * * 4) 1TB disks: some important data structures use * 32-bit signed long/int (for example, daddr_t), * so that sd doesn't support a disk with capacity * larger than 1TB on 32-bit platform. However, * for exactly 1TB disk, it was treated as (1T - 512)B * in the past, and could have valid Solaris * partitions. To workaround this, if an exactly 1TB * disk has Solaris fdisk partition, it will be allowed * to work with sd. * * * * CMLB_FAKE_LABEL_ONE_PARTITION: create s0 and s2 covering * the entire disk, if there is no valid partition info. * If there is a valid Solaris partition, s0 and s2 will * only cover the entire Solaris partition. * * * cmlbhandle cmlb handle associated with device * * tg_cookie cookie from target driver to be passed back to target * driver when we call back to it through tg_ops. * * Notes: * Assumes a default label based on capacity for non-removable devices. * If capacity > 1TB, EFI is assumed otherwise VTOC (default VTOC * for the architecture). * * For removable devices, default label type is assumed to be VTOC * type. Create minor nodes based on a default label type. * Label on the media is not validated. * minor number consists of: * if _SUNOS_VTOC_8 is defined * lowest 3 bits is taken as partition number * the rest is instance number * if _SUNOS_VTOC_16 is defined * lowest 6 bits is taken as partition number * the rest is instance number * * * Return values: * 0 Success * ENXIO creating minor nodes failed. * EINVAL invalid arg, unsupported tg_ops version */ int cmlb_attach(dev_info_t *devi, cmlb_tg_ops_t *tgopsp, int device_type, int is_removable, int is_hotpluggable, char *node_type, int alter_behavior, cmlb_handle_t cmlbhandle, void *tg_cookie) { struct cmlb_lun *cl = (struct cmlb_lun *)cmlbhandle; diskaddr_t cap; int status; if (tgopsp->tg_version < TG_DK_OPS_VERSION_1) return (EINVAL); mutex_enter(CMLB_MUTEX(cl)); CMLB_DEVINFO(cl) = devi; cl->cmlb_tg_ops = tgopsp; cl->cl_device_type = device_type; cl->cl_is_removable = is_removable; cl->cl_is_hotpluggable = is_hotpluggable; cl->cl_node_type = node_type; cl->cl_sys_blocksize = DEV_BSIZE; cl->cl_f_geometry_is_valid = FALSE; cl->cl_def_labeltype = CMLB_LABEL_VTOC; cl->cl_alter_behavior = alter_behavior; cl->cl_reserved = -1; cl->cl_msglog_flag |= CMLB_ALLOW_2TB_WARN; if (is_removable == 0) { mutex_exit(CMLB_MUTEX(cl)); status = DK_TG_GETCAP(cl, &cap, tg_cookie); mutex_enter(CMLB_MUTEX(cl)); if (status == 0 && cap > CMLB_EXTVTOC_LIMIT) { /* set default EFI if > 2TB */ cl->cl_def_labeltype = CMLB_LABEL_EFI; } } /* create minor nodes based on default label type */ cl->cl_last_labeltype = CMLB_LABEL_UNDEF; cl->cl_cur_labeltype = CMLB_LABEL_UNDEF; if (cmlb_create_minor_nodes(cl) != 0) { mutex_exit(CMLB_MUTEX(cl)); return (ENXIO); } /* Define the dynamic properties for devinfo spapshots. */ i_ddi_prop_dyn_driver_set(CMLB_DEVINFO(cl), cmlb_prop_dyn); cl->cl_state = CMLB_ATTACHED; mutex_exit(CMLB_MUTEX(cl)); return (0); } /* * cmlb_detach: * * Invalidate in-core labeling data and remove all minor nodes for * the device associate with handle. * * Arguments: * cmlbhandle cmlb handle associated with device. * * tg_cookie cookie from target driver to be passed back to target * driver when we call back to it through tg_ops. * */ /*ARGSUSED1*/ void cmlb_detach(cmlb_handle_t cmlbhandle, void *tg_cookie) { struct cmlb_lun *cl = (struct cmlb_lun *)cmlbhandle; mutex_enter(CMLB_MUTEX(cl)); cl->cl_def_labeltype = CMLB_LABEL_UNDEF; cl->cl_f_geometry_is_valid = FALSE; ddi_remove_minor_node(CMLB_DEVINFO(cl), NULL); i_ddi_prop_dyn_driver_set(CMLB_DEVINFO(cl), NULL); cl->cl_state = CMLB_INITED; mutex_exit(CMLB_MUTEX(cl)); } /* * cmlb_validate: * * Validates label. * * Arguments * cmlbhandle cmlb handle associated with device. * * flags operation flags. used for verbosity control * * tg_cookie cookie from target driver to be passed back to target * driver when we call back to it through tg_ops. * * * Notes: * If new label type is different from the current, adjust minor nodes * accordingly. * * Return values: * 0 success * Note: having fdisk but no solaris partition is assumed * success. * * ENOMEM memory allocation failed * EIO i/o errors during read or get capacity * EACCESS reservation conflicts * EINVAL label was corrupt, or no default label was assumed * ENXIO invalid handle */ int cmlb_validate(cmlb_handle_t cmlbhandle, int flags, void *tg_cookie) { struct cmlb_lun *cl = (struct cmlb_lun *)cmlbhandle; int rval; int ret = 0; /* * Temp work-around checking cl for NULL since there is a bug * in sd_detach calling this routine from taskq_dispatch * inited function. */ if (cl == NULL) return (ENXIO); mutex_enter(CMLB_MUTEX(cl)); if (cl->cl_state < CMLB_ATTACHED) { mutex_exit(CMLB_MUTEX(cl)); return (ENXIO); } rval = cmlb_validate_geometry((struct cmlb_lun *)cmlbhandle, 1, flags, tg_cookie); if (rval == ENOTSUP) { if (cl->cl_f_geometry_is_valid == TRUE) { cl->cl_cur_labeltype = CMLB_LABEL_EFI; ret = 0; } else { ret = EINVAL; } } else { ret = rval; if (ret == 0) cl->cl_cur_labeltype = CMLB_LABEL_VTOC; } if (ret == 0) (void) cmlb_create_minor_nodes(cl); mutex_exit(CMLB_MUTEX(cl)); return (ret); } /* * cmlb_invalidate: * Invalidate in core label data * * Arguments: * cmlbhandle cmlb handle associated with device. * tg_cookie cookie from target driver to be passed back to target * driver when we call back to it through tg_ops. */ /*ARGSUSED1*/ void cmlb_invalidate(cmlb_handle_t cmlbhandle, void *tg_cookie) { struct cmlb_lun *cl = (struct cmlb_lun *)cmlbhandle; if (cl == NULL) return; mutex_enter(CMLB_MUTEX(cl)); cl->cl_f_geometry_is_valid = FALSE; mutex_exit(CMLB_MUTEX(cl)); } /* * cmlb_is_valid * Get status on whether the incore label/geom data is valid * * Arguments: * cmlbhandle cmlb handle associated with device. * * Return values: * TRUE if incore label/geom data is valid. * FALSE otherwise. * */ int cmlb_is_valid(cmlb_handle_t cmlbhandle) { struct cmlb_lun *cl = (struct cmlb_lun *)cmlbhandle; if (cmlbhandle == NULL) return (FALSE); return (cl->cl_f_geometry_is_valid); } /* * cmlb_close: * * Close the device, revert to a default label minor node for the device, * if it is removable. * * Arguments: * cmlbhandle cmlb handle associated with device. * * tg_cookie cookie from target driver to be passed back to target * driver when we call back to it through tg_ops. * Return values: * 0 Success * ENXIO Re-creating minor node failed. */ /*ARGSUSED1*/ int cmlb_close(cmlb_handle_t cmlbhandle, void *tg_cookie) { struct cmlb_lun *cl = (struct cmlb_lun *)cmlbhandle; mutex_enter(CMLB_MUTEX(cl)); cl->cl_f_geometry_is_valid = FALSE; /* revert to default minor node for this device */ if (ISREMOVABLE(cl)) { cl->cl_cur_labeltype = CMLB_LABEL_UNDEF; (void) cmlb_create_minor_nodes(cl); } mutex_exit(CMLB_MUTEX(cl)); return (0); } /* * cmlb_get_devid_block: * get the block number where device id is stored. * * Arguments: * cmlbhandle cmlb handle associated with device. * devidblockp pointer to block number. * tg_cookie cookie from target driver to be passed back to target * driver when we call back to it through tg_ops. * * Notes: * It stores the block number of device id in the area pointed to * by devidblockp. * with the block number of device id. * * Return values: * 0 success * EINVAL device id does not apply to current label type. */ /*ARGSUSED2*/ int cmlb_get_devid_block(cmlb_handle_t cmlbhandle, diskaddr_t *devidblockp, void *tg_cookie) { daddr_t spc, blk, head, cyl; struct cmlb_lun *cl = (struct cmlb_lun *)cmlbhandle; mutex_enter(CMLB_MUTEX(cl)); if (cl->cl_state < CMLB_ATTACHED) { mutex_exit(CMLB_MUTEX(cl)); return (EINVAL); } if ((cl->cl_f_geometry_is_valid == FALSE) || (cl->cl_solaris_size < DK_LABEL_LOC)) { mutex_exit(CMLB_MUTEX(cl)); return (EINVAL); } if (cl->cl_cur_labeltype == CMLB_LABEL_EFI) { if (cl->cl_reserved != -1) { blk = cl->cl_map[cl->cl_reserved].dkl_cylno; } else { mutex_exit(CMLB_MUTEX(cl)); return (EINVAL); } } else { /* if the disk is unlabeled, don't write a devid to it */ if (cl->cl_label_from_media != CMLB_LABEL_VTOC) { mutex_exit(CMLB_MUTEX(cl)); return (EINVAL); } /* this geometry doesn't allow us to write a devid */ if (cl->cl_g.dkg_acyl < 2) { mutex_exit(CMLB_MUTEX(cl)); return (EINVAL); } /* * Subtract 2 guarantees that the next to last cylinder * is used */ cyl = cl->cl_g.dkg_ncyl + cl->cl_g.dkg_acyl - 2; spc = cl->cl_g.dkg_nhead * cl->cl_g.dkg_nsect; head = cl->cl_g.dkg_nhead - 1; blk = cl->cl_solaris_offset + (cyl * (spc - cl->cl_g.dkg_apc)) + (head * cl->cl_g.dkg_nsect) + 1; } *devidblockp = blk; mutex_exit(CMLB_MUTEX(cl)); return (0); } /* * cmlb_partinfo: * Get partition info for specified partition number. * * Arguments: * cmlbhandle cmlb handle associated with device. * part partition number * nblocksp pointer to number of blocks * startblockp pointer to starting block * partnamep pointer to name of partition * tagp pointer to tag info * tg_cookie cookie from target driver to be passed back to target * driver when we call back to it through tg_ops. * * * Notes: * If in-core label is not valid, this functions tries to revalidate * the label. If label is valid, it stores the total number of blocks * in this partition in the area pointed to by nblocksp, starting * block number in area pointed to by startblockp, pointer to partition * name in area pointed to by partnamep, and tag value in area * pointed by tagp. * For EFI labels, tag value will be set to 0. * * For all nblocksp, startblockp and partnamep, tagp, a value of NULL * indicates the corresponding info is not requested. * * * Return values: * 0 success * EINVAL no valid label or requested partition number is invalid. * */ int cmlb_partinfo(cmlb_handle_t cmlbhandle, int part, diskaddr_t *nblocksp, diskaddr_t *startblockp, char **partnamep, uint16_t *tagp, void *tg_cookie) { struct cmlb_lun *cl = (struct cmlb_lun *)cmlbhandle; int rval; ASSERT(cl != NULL); mutex_enter(CMLB_MUTEX(cl)); if (cl->cl_state < CMLB_ATTACHED) { mutex_exit(CMLB_MUTEX(cl)); return (EINVAL); } if (part < 0 || part >= MAXPART) { rval = EINVAL; } else { if (cl->cl_f_geometry_is_valid == FALSE) (void) cmlb_validate_geometry((struct cmlb_lun *)cl, 0, 0, tg_cookie); #if defined(_SUNOS_VTOC_16) if (((cl->cl_f_geometry_is_valid == FALSE) || (part < NDKMAP && cl->cl_solaris_size == 0)) && (part != P0_RAW_DISK)) { #else if ((cl->cl_f_geometry_is_valid == FALSE) || (part < NDKMAP && cl->cl_solaris_size == 0)) { #endif rval = EINVAL; } else { if (startblockp != NULL) *startblockp = (diskaddr_t)cl->cl_offset[part]; if (nblocksp != NULL) *nblocksp = (diskaddr_t) cl->cl_map[part].dkl_nblk; if (tagp != NULL) if (cl->cl_cur_labeltype == CMLB_LABEL_EFI) *tagp = V_UNASSIGNED; else *tagp = cl->cl_vtoc.v_part[part].p_tag; rval = 0; } /* consistent with behavior of sd for getting minor name */ if (partnamep != NULL) *partnamep = dk_minor_data[part].name; } mutex_exit(CMLB_MUTEX(cl)); return (rval); } /* * cmlb_efi_label_capacity: * Get capacity stored in EFI disk label. * * Arguments: * cmlbhandle cmlb handle associated with device. * capacity pointer to capacity stored in EFI disk label. * tg_cookie cookie from target driver to be passed back to target * driver when we call back to it through tg_ops. * * * Notes: * If in-core label is not valid, this functions tries to revalidate * the label. If label is valid and is an EFI label, it stores the capacity * in disk label in the area pointed to by capacity. * * * Return values: * 0 success * EINVAL no valid EFI label or capacity is NULL. * */ int cmlb_efi_label_capacity(cmlb_handle_t cmlbhandle, diskaddr_t *capacity, void *tg_cookie) { struct cmlb_lun *cl = (struct cmlb_lun *)cmlbhandle; int rval; ASSERT(cl != NULL); mutex_enter(CMLB_MUTEX(cl)); if (cl->cl_state < CMLB_ATTACHED) { mutex_exit(CMLB_MUTEX(cl)); return (EINVAL); } if (cl->cl_f_geometry_is_valid == FALSE) (void) cmlb_validate_geometry((struct cmlb_lun *)cl, 0, 0, tg_cookie); if ((cl->cl_f_geometry_is_valid == FALSE) || (capacity == NULL) || (cl->cl_cur_labeltype != CMLB_LABEL_EFI)) { rval = EINVAL; } else { *capacity = (diskaddr_t)cl->cl_map[WD_NODE].dkl_nblk; rval = 0; } mutex_exit(CMLB_MUTEX(cl)); return (rval); } /* Caller should make sure Test Unit Ready succeeds before calling this. */ /*ARGSUSED*/ int cmlb_ioctl(cmlb_handle_t cmlbhandle, dev_t dev, int cmd, intptr_t arg, int flag, cred_t *cred_p, int *rval_p, void *tg_cookie) { int err; struct cmlb_lun *cl; cl = (struct cmlb_lun *)cmlbhandle; ASSERT(cl != NULL); mutex_enter(CMLB_MUTEX(cl)); if (cl->cl_state < CMLB_ATTACHED) { mutex_exit(CMLB_MUTEX(cl)); return (EIO); } switch (cmd) { case DKIOCSEXTVTOC: case DKIOCSGEOM: case DKIOCSETEFI: case DKIOCSMBOOT: break; case DKIOCSVTOC: #if defined(__i386) || defined(__amd64) case DKIOCPARTINFO: #endif if (cl->cl_blockcount > CMLB_OLDVTOC_LIMIT) { mutex_exit(CMLB_MUTEX(cl)); return (EOVERFLOW); } break; default: (void) cmlb_validate_geometry(cl, 1, CMLB_SILENT, tg_cookie); switch (cmd) { case DKIOCGVTOC: case DKIOCGAPART: case DKIOCSAPART: if (cl->cl_label_from_media == CMLB_LABEL_EFI) { /* GPT label on disk */ mutex_exit(CMLB_MUTEX(cl)); return (ENOTSUP); } else if (cl->cl_blockcount > CMLB_OLDVTOC_LIMIT) { mutex_exit(CMLB_MUTEX(cl)); return (EOVERFLOW); } break; case DKIOCGGEOM: if (cl->cl_label_from_media == CMLB_LABEL_EFI) { /* GPT label on disk */ mutex_exit(CMLB_MUTEX(cl)); return (ENOTSUP); } break; default: break; } } mutex_exit(CMLB_MUTEX(cl)); switch (cmd) { case DKIOCGGEOM: cmlb_dbg(CMLB_TRACE, cl, "DKIOCGGEOM\n"); err = cmlb_dkio_get_geometry(cl, (caddr_t)arg, flag, tg_cookie); break; case DKIOCSGEOM: cmlb_dbg(CMLB_TRACE, cl, "DKIOCSGEOM\n"); err = cmlb_dkio_set_geometry(cl, (caddr_t)arg, flag); break; case DKIOCGAPART: cmlb_dbg(CMLB_TRACE, cl, "DKIOCGAPART\n"); err = cmlb_dkio_get_partition(cl, (caddr_t)arg, flag, tg_cookie); break; case DKIOCSAPART: cmlb_dbg(CMLB_TRACE, cl, "DKIOCSAPART\n"); err = cmlb_dkio_set_partition(cl, (caddr_t)arg, flag); break; case DKIOCGVTOC: cmlb_dbg(CMLB_TRACE, cl, "DKIOCGVTOC\n"); err = cmlb_dkio_get_vtoc(cl, (caddr_t)arg, flag, tg_cookie); break; case DKIOCGEXTVTOC: cmlb_dbg(CMLB_TRACE, cl, "DKIOCGVTOC\n"); err = cmlb_dkio_get_extvtoc(cl, (caddr_t)arg, flag, tg_cookie); break; case DKIOCGETEFI: cmlb_dbg(CMLB_TRACE, cl, "DKIOCGETEFI\n"); err = cmlb_dkio_get_efi(cl, (caddr_t)arg, flag, tg_cookie); break; case DKIOCPARTITION: cmlb_dbg(CMLB_TRACE, cl, "DKIOCPARTITION\n"); err = cmlb_dkio_partition(cl, (caddr_t)arg, flag, tg_cookie); break; case DKIOCSVTOC: cmlb_dbg(CMLB_TRACE, cl, "DKIOCSVTOC\n"); err = cmlb_dkio_set_vtoc(cl, dev, (caddr_t)arg, flag, tg_cookie); break; case DKIOCSEXTVTOC: cmlb_dbg(CMLB_TRACE, cl, "DKIOCSVTOC\n"); err = cmlb_dkio_set_extvtoc(cl, dev, (caddr_t)arg, flag, tg_cookie); break; case DKIOCSETEFI: cmlb_dbg(CMLB_TRACE, cl, "DKIOCSETEFI\n"); err = cmlb_dkio_set_efi(cl, dev, (caddr_t)arg, flag, tg_cookie); break; case DKIOCGMBOOT: cmlb_dbg(CMLB_TRACE, cl, "DKIOCGMBOOT\n"); err = cmlb_dkio_get_mboot(cl, (caddr_t)arg, flag, tg_cookie); break; case DKIOCSMBOOT: cmlb_dbg(CMLB_TRACE, cl, "DKIOCSMBOOT\n"); err = cmlb_dkio_set_mboot(cl, (caddr_t)arg, flag, tg_cookie); break; case DKIOCG_PHYGEOM: cmlb_dbg(CMLB_TRACE, cl, "DKIOCG_PHYGEOM\n"); #if defined(__i386) || defined(__amd64) err = cmlb_dkio_get_phygeom(cl, (caddr_t)arg, flag); #else err = ENOTTY; #endif break; case DKIOCG_VIRTGEOM: cmlb_dbg(CMLB_TRACE, cl, "DKIOCG_VIRTGEOM\n"); #if defined(__i386) || defined(__amd64) err = cmlb_dkio_get_virtgeom(cl, (caddr_t)arg, flag); #else err = ENOTTY; #endif break; case DKIOCPARTINFO: cmlb_dbg(CMLB_TRACE, cl, "DKIOCPARTINFO"); #if defined(__i386) || defined(__amd64) err = cmlb_dkio_partinfo(cl, dev, (caddr_t)arg, flag); #else err = ENOTTY; #endif break; case DKIOCEXTPARTINFO: cmlb_dbg(CMLB_TRACE, cl, "DKIOCPARTINFO"); #if defined(__i386) || defined(__amd64) err = cmlb_dkio_extpartinfo(cl, dev, (caddr_t)arg, flag); #else err = ENOTTY; #endif break; default: err = ENOTTY; } /* * An ioctl that succeeds and changed ('set') size(9P) information * needs to invalidate the cached devinfo snapshot to avoid having * old information being returned in a snapshots. * * NB: When available, call ddi_change_minor_node() to clear * SSIZEVALID in specfs vnodes via spec_size_invalidate(). */ if (err == 0) { switch (cmd) { case DKIOCSGEOM: case DKIOCSAPART: case DKIOCSVTOC: case DKIOCSEXTVTOC: case DKIOCSETEFI: i_ddi_prop_dyn_cache_invalidate(CMLB_DEVINFO(cl), i_ddi_prop_dyn_driver_get(CMLB_DEVINFO(cl))); } } return (err); } dev_t cmlb_make_device(struct cmlb_lun *cl) { return (makedevice(ddi_driver_major(CMLB_DEVINFO(cl)), ddi_get_instance(CMLB_DEVINFO(cl)) << CMLBUNIT_SHIFT)); } /* * Function: cmlb_check_update_blockcount * * Description: If current capacity value is invalid, obtains the * current capacity from target driver. * * Return Code: 0 success * EIO failure */ static int cmlb_check_update_blockcount(struct cmlb_lun *cl, void *tg_cookie) { int status; diskaddr_t capacity; uint32_t lbasize; ASSERT(mutex_owned(CMLB_MUTEX(cl))); if (cl->cl_f_geometry_is_valid == FALSE) { mutex_exit(CMLB_MUTEX(cl)); status = DK_TG_GETCAP(cl, &capacity, tg_cookie); if (status != 0) { mutex_enter(CMLB_MUTEX(cl)); return (EIO); } status = DK_TG_GETBLOCKSIZE(cl, &lbasize, tg_cookie); mutex_enter(CMLB_MUTEX(cl)); if (status != 0) return (EIO); if ((capacity != 0) && (lbasize != 0)) { cl->cl_blockcount = capacity; cl->cl_tgt_blocksize = lbasize; return (0); } else return (EIO); } else return (0); } static int cmlb_create_minor(dev_info_t *dip, char *name, int spec_type, minor_t minor_num, char *node_type, int flag, boolean_t internal) { if (internal) return (ddi_create_internal_pathname(dip, name, spec_type, minor_num)); else return (ddi_create_minor_node(dip, name, spec_type, minor_num, node_type, flag)); } /* * Function: cmlb_create_minor_nodes * * Description: Create or adjust the minor device nodes for the instance. * Minor nodes are created based on default label type, * current label type and last label type we created * minor nodes based on. * * * Arguments: cl - driver soft state (unit) structure * * Return Code: 0 success * ENXIO failure. * * Context: Kernel thread context */ static int cmlb_create_minor_nodes(struct cmlb_lun *cl) { struct driver_minor_data *dmdp; int instance; char name[48]; cmlb_label_t newlabeltype; boolean_t internal; ASSERT(cl != NULL); ASSERT(mutex_owned(CMLB_MUTEX(cl))); internal = ((cl->cl_alter_behavior & (CMLB_INTERNAL_MINOR_NODES)) != 0); /* check the most common case */ if (cl->cl_cur_labeltype != CMLB_LABEL_UNDEF && cl->cl_last_labeltype == cl->cl_cur_labeltype) { /* do nothing */ return (0); } if (cl->cl_def_labeltype == CMLB_LABEL_UNDEF) { /* we should never get here */ return (ENXIO); } if (cl->cl_last_labeltype == CMLB_LABEL_UNDEF) { /* first time during attach */ newlabeltype = cl->cl_def_labeltype; instance = ddi_get_instance(CMLB_DEVINFO(cl)); /* Create all the minor nodes for this target. */ dmdp = (newlabeltype == CMLB_LABEL_EFI) ? dk_minor_data_efi : dk_minor_data; while (dmdp->name != NULL) { (void) sprintf(name, "%s", dmdp->name); if (cmlb_create_minor(CMLB_DEVINFO(cl), name, dmdp->type, (instance << CMLBUNIT_SHIFT) | dmdp->minor, cl->cl_node_type, NULL, internal) == DDI_FAILURE) { /* * Clean up any nodes that may have been * created, in case this fails in the middle * of the loop. */ ddi_remove_minor_node(CMLB_DEVINFO(cl), NULL); return (ENXIO); } dmdp++; } cl->cl_last_labeltype = newlabeltype; return (0); } /* Not first time */ if (cl->cl_cur_labeltype == CMLB_LABEL_UNDEF) { if (cl->cl_last_labeltype != cl->cl_def_labeltype) { /* close time, revert to default. */ newlabeltype = cl->cl_def_labeltype; } else { /* * do nothing since the type for which we last created * nodes matches the default */ return (0); } } else { if (cl->cl_cur_labeltype != cl->cl_last_labeltype) { /* We are not closing, use current label type */ newlabeltype = cl->cl_cur_labeltype; } else { /* * do nothing since the type for which we last created * nodes matches the current label type */ return (0); } } instance = ddi_get_instance(CMLB_DEVINFO(cl)); /* * Currently we only fix up the s7 node when we are switching * label types from or to EFI. This is consistent with * current behavior of sd. */ if (newlabeltype == CMLB_LABEL_EFI && cl->cl_last_labeltype != CMLB_LABEL_EFI) { /* from vtoc to EFI */ ddi_remove_minor_node(CMLB_DEVINFO(cl), "h"); ddi_remove_minor_node(CMLB_DEVINFO(cl), "h,raw"); (void) cmlb_create_minor(CMLB_DEVINFO(cl), "wd", S_IFBLK, (instance << CMLBUNIT_SHIFT) | WD_NODE, cl->cl_node_type, NULL, internal); (void) cmlb_create_minor(CMLB_DEVINFO(cl), "wd,raw", S_IFCHR, (instance << CMLBUNIT_SHIFT) | WD_NODE, cl->cl_node_type, NULL, internal); } else { /* from efi to vtoc */ ddi_remove_minor_node(CMLB_DEVINFO(cl), "wd"); ddi_remove_minor_node(CMLB_DEVINFO(cl), "wd,raw"); (void) cmlb_create_minor(CMLB_DEVINFO(cl), "h", S_IFBLK, (instance << CMLBUNIT_SHIFT) | WD_NODE, cl->cl_node_type, NULL, internal); (void) cmlb_create_minor(CMLB_DEVINFO(cl), "h,raw", S_IFCHR, (instance << CMLBUNIT_SHIFT) | WD_NODE, cl->cl_node_type, NULL, internal); } cl->cl_last_labeltype = newlabeltype; return (0); } /* * Function: cmlb_validate_geometry * * Description: Read the label from the disk (if present). Update the unit's * geometry and vtoc information from the data in the label. * Verify that the label is valid. * * Arguments: * cl driver soft state (unit) structure * * forcerevalid force revalidation even if we are already valid. * flags operation flags from target driver. Used for verbosity * control at this time. * tg_cookie cookie from target driver to be passed back to target * driver when we call back to it through tg_ops. * * Return Code: 0 - Successful completion * EINVAL - Invalid value in cl->cl_tgt_blocksize or * cl->cl_blockcount; or label on disk is corrupted * or unreadable. * EACCES - Reservation conflict at the device. * ENOMEM - Resource allocation error * ENOTSUP - geometry not applicable * * Context: Kernel thread only (can sleep). */ static int cmlb_validate_geometry(struct cmlb_lun *cl, int forcerevalid, int flags, void *tg_cookie) { int label_error = 0; diskaddr_t capacity; int count; ASSERT(mutex_owned(CMLB_MUTEX(cl))); if ((cl->cl_f_geometry_is_valid == TRUE) && (forcerevalid == 0)) { if (cl->cl_cur_labeltype == CMLB_LABEL_EFI) return (ENOTSUP); return (0); } if (cmlb_check_update_blockcount(cl, tg_cookie) != 0) return (EIO); capacity = cl->cl_blockcount; #if defined(_SUNOS_VTOC_16) /* * Set up the "whole disk" fdisk partition; this should always * exist, regardless of whether the disk contains an fdisk table * or vtoc. */ cl->cl_map[P0_RAW_DISK].dkl_cylno = 0; cl->cl_offset[P0_RAW_DISK] = 0; /* * note if capacity > int32_max(1TB) we are in 64bit environment * so no truncation happens */ cl->cl_map[P0_RAW_DISK].dkl_nblk = capacity; #endif /* * Refresh the logical and physical geometry caches. * (data from MODE SENSE format/rigid disk geometry pages, * and scsi_ifgetcap("geometry"). */ cmlb_resync_geom_caches(cl, capacity, tg_cookie); cl->cl_label_from_media = CMLB_LABEL_UNDEF; label_error = cmlb_use_efi(cl, capacity, flags, tg_cookie); if (label_error == 0) { /* found a valid EFI label */ cmlb_dbg(CMLB_TRACE, cl, "cmlb_validate_geometry: found EFI label\n"); /* * solaris_size and geometry_is_valid are set in * cmlb_use_efi */ return (ENOTSUP); } /* NO EFI label found */ if (capacity > CMLB_EXTVTOC_LIMIT) { if (label_error == ESRCH) { /* * they've configured a LUN over 2TB, but used * format.dat to restrict format's view of the * capacity to be under 2TB in some earlier Solaris * release. */ /* i.e > 2TB with a VTOC < 2TB */ if (!(flags & CMLB_SILENT) && (cl->cl_msglog_flag & CMLB_ALLOW_2TB_WARN)) { cmlb_log(CMLB_DEVINFO(cl), CMLB_LABEL(cl), CE_NOTE, "!Disk (%s%d) is limited to 2 TB " "due to VTOC label. To use the full " "capacity of the disk, use format(1M) to " "relabel the disk with EFI/GPT label.\n", CMLB_LABEL(cl), ddi_get_instance(CMLB_DEVINFO(cl))); cl->cl_msglog_flag &= ~CMLB_ALLOW_2TB_WARN; } } else { return (ENOTSUP); } } label_error = 0; /* * at this point it is either labeled with a VTOC or it is * under 1TB (<= 1TB actually for off-by-1) */ /* * Only DIRECT ACCESS devices will have Scl labels. * CD's supposedly have a Scl label, too */ if (cl->cl_device_type == DTYPE_DIRECT || ISREMOVABLE(cl)) { struct dk_label *dkl; offset_t label_addr; int rval; size_t buffer_size; /* * Note: This will set up cl->cl_solaris_size and * cl->cl_solaris_offset. */ rval = cmlb_read_fdisk(cl, capacity, tg_cookie); if ((rval != 0) && !ISCD(cl)) { ASSERT(mutex_owned(CMLB_MUTEX(cl))); return (rval); } if (cl->cl_solaris_size <= DK_LABEL_LOC) { /* * Found fdisk table but no Solaris partition entry, * so don't call cmlb_uselabel() and don't create * a default label. */ label_error = 0; cl->cl_f_geometry_is_valid = TRUE; goto no_solaris_partition; } label_addr = (daddr_t)(cl->cl_solaris_offset + DK_LABEL_LOC); buffer_size = sizeof (struct dk_label); cmlb_dbg(CMLB_TRACE, cl, "cmlb_validate_geometry: " "label_addr: 0x%x allocation size: 0x%x\n", label_addr, buffer_size); if ((dkl = kmem_zalloc(buffer_size, KM_NOSLEEP)) == NULL) return (ENOMEM); mutex_exit(CMLB_MUTEX(cl)); rval = DK_TG_READ(cl, dkl, label_addr, buffer_size, tg_cookie); mutex_enter(CMLB_MUTEX(cl)); switch (rval) { case 0: /* * cmlb_uselabel will establish that the geometry * is valid. */ if (cmlb_uselabel(cl, (struct dk_label *)(uintptr_t)dkl, flags) != CMLB_LABEL_IS_VALID) { label_error = EINVAL; } else cl->cl_label_from_media = CMLB_LABEL_VTOC; break; case EACCES: label_error = EACCES; break; default: label_error = EINVAL; break; } kmem_free(dkl, buffer_size); } /* * If a valid label was not found, AND if no reservation conflict * was detected, then go ahead and create a default label (4069506). * * Note: currently, for VTOC_8 devices, the default label is created * for removables and hotpluggables only. For VTOC_16 devices, the * default label will be created for all devices. * (see cmlb_build_default_label) */ #if defined(_SUNOS_VTOC_8) if ((ISREMOVABLE(cl) || ISHOTPLUGGABLE(cl)) && (label_error != EACCES)) { #elif defined(_SUNOS_VTOC_16) if (label_error != EACCES) { #endif if (cl->cl_f_geometry_is_valid == FALSE) { cmlb_build_default_label(cl, tg_cookie); } label_error = 0; } no_solaris_partition: #if defined(_SUNOS_VTOC_16) /* * If we have valid geometry, set up the remaining fdisk partitions. * Note that dkl_cylno is not used for the fdisk map entries, so * we set it to an entirely bogus value. */ for (count = 0; count < FD_NUMPART; count++) { cl->cl_map[FDISK_P1 + count].dkl_cylno = UINT16_MAX; cl->cl_map[FDISK_P1 + count].dkl_nblk = cl->cl_fmap[count].fmap_nblk; cl->cl_offset[FDISK_P1 + count] = cl->cl_fmap[count].fmap_start; } #endif for (count = 0; count < NDKMAP; count++) { #if defined(_SUNOS_VTOC_8) struct dk_map *lp = &cl->cl_map[count]; cl->cl_offset[count] = cl->cl_g.dkg_nhead * cl->cl_g.dkg_nsect * lp->dkl_cylno; #elif defined(_SUNOS_VTOC_16) struct dkl_partition *vp = &cl->cl_vtoc.v_part[count]; cl->cl_offset[count] = vp->p_start + cl->cl_solaris_offset; #else #error "No VTOC format defined." #endif } return (label_error); } #if defined(_SUNOS_VTOC_16) /* * Function: cmlb_convert_geometry * * Description: Convert physical geometry into a dk_geom structure. In * other words, make sure we don't wrap 16-bit values. * e.g. converting from geom_cache to dk_geom * * Context: Kernel thread only */ static void cmlb_convert_geometry(diskaddr_t capacity, struct dk_geom *cl_g) { /* Unlabeled SCSI floppy device */ if (capacity <= 0x1000) { cl_g->dkg_nhead = 2; cl_g->dkg_ncyl = 80; cl_g->dkg_nsect = capacity / (cl_g->dkg_nhead * cl_g->dkg_ncyl); return; } /* * For all devices we calculate cylinders using the heads and sectors * we assign based on capacity of the device. The algorithm is * designed to be compatible with the way other operating systems * lay out fdisk tables for X86 and to insure that the cylinders never * exceed 65535 to prevent problems with X86 ioctls that report * geometry. * For some smaller disk sizes we report geometry that matches those * used by X86 BIOS usage. For larger disks, we use SPT that are * multiples of 63, since other OSes that are not limited to 16-bits * for cylinders stop at 63 SPT we make do by using multiples of 63 SPT. * * The following table (in order) illustrates some end result * calculations: * * Maximum number of blocks nhead nsect * * 2097152 (1GB) 64 32 * 16777216 (8GB) 128 32 * 1052819775 (502.02GB) 255 63 * 2105639550 (0.98TB) 255 126 * 3158459325 (1.47TB) 255 189 * 4211279100 (1.96TB) 255 252 * 5264098875 (2.45TB) 255 315 * ... */ if (capacity <= 0x200000) { cl_g->dkg_nhead = 64; cl_g->dkg_nsect = 32; } else if (capacity <= 0x01000000) { cl_g->dkg_nhead = 128; cl_g->dkg_nsect = 32; } else { cl_g->dkg_nhead = 255; /* make nsect be smallest multiple of 63 */ cl_g->dkg_nsect = ((capacity + (UINT16_MAX * 255 * 63) - 1) / (UINT16_MAX * 255 * 63)) * 63; if (cl_g->dkg_nsect == 0) cl_g->dkg_nsect = (UINT16_MAX / 63) * 63; } } #endif /* * Function: cmlb_resync_geom_caches * * Description: (Re)initialize both geometry caches: the virtual geometry * information is extracted from the HBA (the "geometry" * capability), and the physical geometry cache data is * generated by issuing MODE SENSE commands. * * Arguments: * cl driver soft state (unit) structure * capacity disk capacity in #blocks * tg_cookie cookie from target driver to be passed back to target * driver when we call back to it through tg_ops. * * Context: Kernel thread only (can sleep). */ static void cmlb_resync_geom_caches(struct cmlb_lun *cl, diskaddr_t capacity, void *tg_cookie) { struct cmlb_geom pgeom; struct cmlb_geom lgeom; struct cmlb_geom *pgeomp = &pgeom; unsigned short nhead; unsigned short nsect; int spc; int ret; ASSERT(cl != NULL); ASSERT(mutex_owned(CMLB_MUTEX(cl))); /* * Ask the controller for its logical geometry. * Note: if the HBA does not support scsi_ifgetcap("geometry"), * then the lgeom cache will be invalid. */ mutex_exit(CMLB_MUTEX(cl)); bzero(&lgeom, sizeof (struct cmlb_geom)); ret = DK_TG_GETVIRTGEOM(cl, &lgeom, tg_cookie); mutex_enter(CMLB_MUTEX(cl)); bcopy(&lgeom, &cl->cl_lgeom, sizeof (cl->cl_lgeom)); /* * Initialize the pgeom cache from lgeom, so that if MODE SENSE * doesn't work, DKIOCG_PHYSGEOM can return reasonable values. */ if (ret != 0 || cl->cl_lgeom.g_nsect == 0 || cl->cl_lgeom.g_nhead == 0) { /* * Note: Perhaps this needs to be more adaptive? The rationale * is that, if there's no HBA geometry from the HBA driver, any * guess is good, since this is the physical geometry. If MODE * SENSE fails this gives a max cylinder size for non-LBA access */ nhead = 255; nsect = 63; } else { nhead = cl->cl_lgeom.g_nhead; nsect = cl->cl_lgeom.g_nsect; } if (ISCD(cl)) { pgeomp->g_nhead = 1; pgeomp->g_nsect = nsect * nhead; } else { pgeomp->g_nhead = nhead; pgeomp->g_nsect = nsect; } spc = pgeomp->g_nhead * pgeomp->g_nsect; pgeomp->g_capacity = capacity; pgeomp->g_ncyl = pgeomp->g_capacity / spc; pgeomp->g_acyl = 0; /* * Retrieve fresh geometry data from the hardware, stash it * here temporarily before we rebuild the incore label. * * We want to use the MODE SENSE commands to derive the * physical geometry of the device, but if either command * fails, the logical geometry is used as the fallback for * disk label geometry. */ mutex_exit(CMLB_MUTEX(cl)); (void) DK_TG_GETPHYGEOM(cl, pgeomp, tg_cookie); mutex_enter(CMLB_MUTEX(cl)); /* * Now update the real copy while holding the mutex. This * way the global copy is never in an inconsistent state. */ bcopy(pgeomp, &cl->cl_pgeom, sizeof (cl->cl_pgeom)); cmlb_dbg(CMLB_INFO, cl, "cmlb_resync_geom_caches: " "(cached from lgeom)\n"); cmlb_dbg(CMLB_INFO, cl, " ncyl: %ld; acyl: %d; nhead: %d; nsect: %d\n", cl->cl_pgeom.g_ncyl, cl->cl_pgeom.g_acyl, cl->cl_pgeom.g_nhead, cl->cl_pgeom.g_nsect); cmlb_dbg(CMLB_INFO, cl, " lbasize: %d; capacity: %ld; " "intrlv: %d; rpm: %d\n", cl->cl_pgeom.g_secsize, cl->cl_pgeom.g_capacity, cl->cl_pgeom.g_intrlv, cl->cl_pgeom.g_rpm); } /* * Function: cmlb_read_fdisk * * Description: utility routine to read the fdisk table. * * Arguments: * cl driver soft state (unit) structure * capacity disk capacity in #blocks * tg_cookie cookie from target driver to be passed back to target * driver when we call back to it through tg_ops. * * Return Code: 0 for success (includes not reading for no_fdisk_present case * errnos from tg_rw if failed to read the first block. * * Context: Kernel thread only (can sleep). */ /*ARGSUSED*/ static int cmlb_read_fdisk(struct cmlb_lun *cl, diskaddr_t capacity, void *tg_cookie) { #if defined(_NO_FDISK_PRESENT) cl->cl_solaris_offset = 0; cl->cl_solaris_size = capacity; bzero(cl->cl_fmap, sizeof (struct fmap) * FD_NUMPART); return (0); #elif defined(_FIRMWARE_NEEDS_FDISK) struct ipart *fdp; struct mboot *mbp; struct ipart fdisk[FD_NUMPART]; int i; char sigbuf[2]; caddr_t bufp; int uidx; int rval; int lba = 0; uint_t solaris_offset; /* offset to solaris part. */ daddr_t solaris_size; /* size of solaris partition */ uint32_t blocksize; ASSERT(cl != NULL); ASSERT(mutex_owned(CMLB_MUTEX(cl))); /* * Start off assuming no fdisk table */ solaris_offset = 0; solaris_size = capacity; blocksize = cl->cl_tgt_blocksize; bufp = kmem_zalloc(blocksize, KM_SLEEP); mutex_exit(CMLB_MUTEX(cl)); rval = DK_TG_READ(cl, bufp, 0, blocksize, tg_cookie); mutex_enter(CMLB_MUTEX(cl)); if (rval != 0) { cmlb_dbg(CMLB_ERROR, cl, "cmlb_read_fdisk: fdisk read err\n"); bzero(cl->cl_fmap, sizeof (struct fmap) * FD_NUMPART); goto done; } mbp = (struct mboot *)bufp; /* * The fdisk table does not begin on a 4-byte boundary within the * master boot record, so we copy it to an aligned structure to avoid * alignment exceptions on some processors. */ bcopy(&mbp->parts[0], fdisk, sizeof (fdisk)); /* * Check for lba support before verifying sig; sig might not be * there, say on a blank disk, but the max_chs mark may still * be present. * * Note: LBA support and BEFs are an x86-only concept but this * code should work OK on SPARC as well. */ /* * First, check for lba-access-ok on root node (or prom root node) * if present there, don't need to search fdisk table. */ if (ddi_getprop(DDI_DEV_T_ANY, ddi_root_node(), 0, "lba-access-ok", 0) != 0) { /* All drives do LBA; don't search fdisk table */ lba = 1; } else { /* Okay, look for mark in fdisk table */ for (fdp = fdisk, i = 0; i < FD_NUMPART; i++, fdp++) { /* accumulate "lba" value from all partitions */ lba = (lba || cmlb_has_max_chs_vals(fdp)); } } if (lba != 0) { dev_t dev = cmlb_make_device(cl); if (ddi_getprop(dev, CMLB_DEVINFO(cl), DDI_PROP_DONTPASS, "lba-access-ok", 0) == 0) { /* not found; create it */ if (ddi_prop_create(dev, CMLB_DEVINFO(cl), 0, "lba-access-ok", (caddr_t)NULL, 0) != DDI_PROP_SUCCESS) { cmlb_dbg(CMLB_ERROR, cl, "cmlb_read_fdisk: Can't create lba " "property for instance %d\n", ddi_get_instance(CMLB_DEVINFO(cl))); } } } bcopy(&mbp->signature, sigbuf, sizeof (sigbuf)); /* * Endian-independent signature check */ if (((sigbuf[1] & 0xFF) != ((MBB_MAGIC >> 8) & 0xFF)) || (sigbuf[0] != (MBB_MAGIC & 0xFF))) { cmlb_dbg(CMLB_ERROR, cl, "cmlb_read_fdisk: no fdisk\n"); bzero(cl->cl_fmap, sizeof (struct fmap) * FD_NUMPART); goto done; } #ifdef CMLBDEBUG if (cmlb_level_mask & CMLB_LOGMASK_INFO) { fdp = fdisk; cmlb_dbg(CMLB_INFO, cl, "cmlb_read_fdisk:\n"); cmlb_dbg(CMLB_INFO, cl, " relsect " "numsect sysid bootid\n"); for (i = 0; i < FD_NUMPART; i++, fdp++) { cmlb_dbg(CMLB_INFO, cl, " %d: %8d %8d 0x%08x 0x%08x\n", i, fdp->relsect, fdp->numsect, fdp->systid, fdp->bootid); } } #endif /* * Try to find the unix partition */ uidx = -1; solaris_offset = 0; solaris_size = 0; for (fdp = fdisk, i = 0; i < FD_NUMPART; i++, fdp++) { uint32_t relsect; uint32_t numsect; if (fdp->numsect == 0) { cl->cl_fmap[i].fmap_start = 0; cl->cl_fmap[i].fmap_nblk = 0; continue; } /* * Data in the fdisk table is little-endian. */ relsect = LE_32(fdp->relsect); numsect = LE_32(fdp->numsect); cl->cl_fmap[i].fmap_start = relsect; cl->cl_fmap[i].fmap_nblk = numsect; if (fdp->systid != SUNIXOS && fdp->systid != SUNIXOS2 && fdp->systid != EFI_PMBR) { continue; } /* * use the last active solaris partition id found * (there should only be 1 active partition id) * * if there are no active solaris partition id * then use the first inactive solaris partition id */ if ((uidx == -1) || (fdp->bootid == ACTIVE)) { uidx = i; solaris_offset = relsect; solaris_size = numsect; } } cmlb_dbg(CMLB_INFO, cl, "fdisk 0x%x 0x%lx", cl->cl_solaris_offset, cl->cl_solaris_size); done: /* * Clear the VTOC info, only if the Solaris partition entry * has moved, changed size, been deleted, or if the size of * the partition is too small to even fit the label sector. */ if ((cl->cl_solaris_offset != solaris_offset) || (cl->cl_solaris_size != solaris_size) || solaris_size <= DK_LABEL_LOC) { cmlb_dbg(CMLB_INFO, cl, "fdisk moved 0x%x 0x%lx", solaris_offset, solaris_size); bzero(&cl->cl_g, sizeof (struct dk_geom)); bzero(&cl->cl_vtoc, sizeof (struct dk_vtoc)); bzero(&cl->cl_map, NDKMAP * (sizeof (struct dk_map))); cl->cl_f_geometry_is_valid = FALSE; } cl->cl_solaris_offset = solaris_offset; cl->cl_solaris_size = solaris_size; kmem_free(bufp, blocksize); return (rval); #else /* #elif defined(_FIRMWARE_NEEDS_FDISK) */ #error "fdisk table presence undetermined for this platform." #endif /* #if defined(_NO_FDISK_PRESENT) */ } static void cmlb_swap_efi_gpt(efi_gpt_t *e) { _NOTE(ASSUMING_PROTECTED(*e)) e->efi_gpt_Signature = LE_64(e->efi_gpt_Signature); e->efi_gpt_Revision = LE_32(e->efi_gpt_Revision); e->efi_gpt_HeaderSize = LE_32(e->efi_gpt_HeaderSize); e->efi_gpt_HeaderCRC32 = LE_32(e->efi_gpt_HeaderCRC32); e->efi_gpt_MyLBA = LE_64(e->efi_gpt_MyLBA); e->efi_gpt_AlternateLBA = LE_64(e->efi_gpt_AlternateLBA); e->efi_gpt_FirstUsableLBA = LE_64(e->efi_gpt_FirstUsableLBA); e->efi_gpt_LastUsableLBA = LE_64(e->efi_gpt_LastUsableLBA); UUID_LE_CONVERT(e->efi_gpt_DiskGUID, e->efi_gpt_DiskGUID); e->efi_gpt_PartitionEntryLBA = LE_64(e->efi_gpt_PartitionEntryLBA); e->efi_gpt_NumberOfPartitionEntries = LE_32(e->efi_gpt_NumberOfPartitionEntries); e->efi_gpt_SizeOfPartitionEntry = LE_32(e->efi_gpt_SizeOfPartitionEntry); e->efi_gpt_PartitionEntryArrayCRC32 = LE_32(e->efi_gpt_PartitionEntryArrayCRC32); } static void cmlb_swap_efi_gpe(int nparts, efi_gpe_t *p) { int i; _NOTE(ASSUMING_PROTECTED(*p)) for (i = 0; i < nparts; i++) { UUID_LE_CONVERT(p[i].efi_gpe_PartitionTypeGUID, p[i].efi_gpe_PartitionTypeGUID); p[i].efi_gpe_StartingLBA = LE_64(p[i].efi_gpe_StartingLBA); p[i].efi_gpe_EndingLBA = LE_64(p[i].efi_gpe_EndingLBA); /* PartitionAttrs */ } } static int cmlb_validate_efi(efi_gpt_t *labp) { if (labp->efi_gpt_Signature != EFI_SIGNATURE) return (EINVAL); /* at least 96 bytes in this version of the spec. */ if (sizeof (efi_gpt_t) - sizeof (labp->efi_gpt_Reserved2) > labp->efi_gpt_HeaderSize) return (EINVAL); /* this should be 128 bytes */ if (labp->efi_gpt_SizeOfPartitionEntry != sizeof (efi_gpe_t)) return (EINVAL); return (0); } /* * This function returns FALSE if there is a valid MBR signature and no * partition table entries of type EFI_PMBR (0xEE). Otherwise it returns TRUE. * * The EFI spec (1.10 and later) requires having a Protective MBR (PMBR) to * recognize the disk as GPT partitioned. However, some other OS creates an MBR * where a PMBR entry is not the only one. Also, if the first block has been * corrupted, currently best attempt to allow data access would be to try to * check for GPT headers. Hence in case of more than one partition entry, but * at least one EFI_PMBR partition type or no valid magic number, the function * returns TRUE to continue with looking for GPT header. */ static int cmlb_check_efi_mbr(uchar_t *buf, int *is_mbr) { struct ipart *fdp; struct mboot *mbp = (struct mboot *)buf; struct ipart fdisk[FD_NUMPART]; int i; if (is_mbr != NULL) *is_mbr = TRUE; if (LE_16(mbp->signature) != MBB_MAGIC) { if (is_mbr != NULL) is_mbr = FALSE; return (TRUE); } bcopy(&mbp->parts[0], fdisk, sizeof (fdisk)); for (fdp = fdisk, i = 0; i < FD_NUMPART; i++, fdp++) { if (fdp->systid == EFI_PMBR) return (TRUE); } return (FALSE); } static int cmlb_use_efi(struct cmlb_lun *cl, diskaddr_t capacity, int flags, void *tg_cookie) { int i; int rval = 0; efi_gpe_t *partitions; uchar_t *buf; uint_t lbasize; /* is really how much to read */ diskaddr_t cap = 0; uint_t nparts; diskaddr_t gpe_lba; diskaddr_t alternate_lba; int iofailed = 0; struct uuid uuid_type_reserved = EFI_RESERVED; #if defined(_FIRMWARE_NEEDS_FDISK) int is_mbr; #endif ASSERT(mutex_owned(CMLB_MUTEX(cl))); if (cl->cl_tgt_blocksize != cl->cl_sys_blocksize) { rval = EINVAL; goto done_err1; } lbasize = cl->cl_sys_blocksize; cl->cl_reserved = -1; mutex_exit(CMLB_MUTEX(cl)); buf = kmem_zalloc(EFI_MIN_ARRAY_SIZE, KM_SLEEP); rval = DK_TG_READ(cl, buf, 0, lbasize, tg_cookie); if (rval) { iofailed = 1; goto done_err; } if (((struct dk_label *)buf)->dkl_magic == DKL_MAGIC) { /* not ours */ rval = ESRCH; goto done_err; } #if defined(_FIRMWARE_NEEDS_FDISK) if (cmlb_check_efi_mbr(buf, &is_mbr) == FALSE) { if (is_mbr == TRUE) rval = ESRCH; else rval = EINVAL; goto done_err; } #else if (cmlb_check_efi_mbr(buf, NULL) == FALSE) { rval = EINVAL; goto done_err; } #endif rval = DK_TG_READ(cl, buf, 1, lbasize, tg_cookie); if (rval) { iofailed = 1; goto done_err; } cmlb_swap_efi_gpt((efi_gpt_t *)buf); if ((rval = cmlb_validate_efi((efi_gpt_t *)buf)) != 0) { /* * Couldn't read the primary, try the backup. Our * capacity at this point could be based on CHS, so * check what the device reports. */ rval = DK_TG_GETCAP(cl, &cap, tg_cookie); if (rval) { iofailed = 1; goto done_err; } /* * CMLB_OFF_BY_ONE case, we check the next to last block first * for backup GPT header, otherwise check the last block. */ if ((rval = DK_TG_READ(cl, buf, cap - ((cl->cl_alter_behavior & CMLB_OFF_BY_ONE) ? 2 : 1), lbasize, tg_cookie)) != 0) { iofailed = 1; goto done_err; } cmlb_swap_efi_gpt((efi_gpt_t *)buf); if ((rval = cmlb_validate_efi((efi_gpt_t *)buf)) != 0) { if (!(cl->cl_alter_behavior & CMLB_OFF_BY_ONE)) goto done_err; if ((rval = DK_TG_READ(cl, buf, cap - 1, lbasize, tg_cookie)) != 0) goto done_err; cmlb_swap_efi_gpt((efi_gpt_t *)buf); if ((rval = cmlb_validate_efi((efi_gpt_t *)buf)) != 0) goto done_err; } if (!(flags & CMLB_SILENT)) cmlb_log(CMLB_DEVINFO(cl), CMLB_LABEL(cl), CE_WARN, "primary label corrupt; using backup\n"); } nparts = ((efi_gpt_t *)buf)->efi_gpt_NumberOfPartitionEntries; gpe_lba = ((efi_gpt_t *)buf)->efi_gpt_PartitionEntryLBA; alternate_lba = ((efi_gpt_t *)buf)->efi_gpt_AlternateLBA; rval = DK_TG_READ(cl, buf, gpe_lba, EFI_MIN_ARRAY_SIZE, tg_cookie); if (rval) { iofailed = 1; goto done_err; } partitions = (efi_gpe_t *)buf; if (nparts > MAXPART) { nparts = MAXPART; } cmlb_swap_efi_gpe(nparts, partitions); mutex_enter(CMLB_MUTEX(cl)); /* Fill in partition table. */ for (i = 0; i < nparts; i++) { if (partitions->efi_gpe_StartingLBA != 0 || partitions->efi_gpe_EndingLBA != 0) { cl->cl_map[i].dkl_cylno = partitions->efi_gpe_StartingLBA; cl->cl_map[i].dkl_nblk = partitions->efi_gpe_EndingLBA - partitions->efi_gpe_StartingLBA + 1; cl->cl_offset[i] = partitions->efi_gpe_StartingLBA; } if (cl->cl_reserved == -1) { if (bcmp(&partitions->efi_gpe_PartitionTypeGUID, &uuid_type_reserved, sizeof (struct uuid)) == 0) { cl->cl_reserved = i; } } if (i == WD_NODE) { /* * minor number 7 corresponds to the whole disk * if the disk capacity is expanded after disk is * labeled, minor number 7 represents the capacity * indicated by the disk label. */ cl->cl_map[i].dkl_cylno = 0; if (alternate_lba == 1) { /* * We are using backup label. Since we can * find a valid label at the end of disk, * the disk capacity is not expanded. */ cl->cl_map[i].dkl_nblk = capacity; } else { cl->cl_map[i].dkl_nblk = alternate_lba + 1; } cl->cl_offset[i] = 0; } partitions++; } cl->cl_solaris_offset = 0; cl->cl_solaris_size = capacity; cl->cl_label_from_media = CMLB_LABEL_EFI; cl->cl_f_geometry_is_valid = TRUE; /* clear the vtoc label */ bzero(&cl->cl_vtoc, sizeof (struct dk_vtoc)); kmem_free(buf, EFI_MIN_ARRAY_SIZE); return (0); done_err: kmem_free(buf, EFI_MIN_ARRAY_SIZE); mutex_enter(CMLB_MUTEX(cl)); done_err1: /* * if we didn't find something that could look like a VTOC * and the disk is over 1TB, we know there isn't a valid label. * Otherwise let cmlb_uselabel decide what to do. We only * want to invalidate this if we're certain the label isn't * valid because cmlb_prop_op will now fail, which in turn * causes things like opens and stats on the partition to fail. */ if ((capacity > CMLB_EXTVTOC_LIMIT) && (rval != ESRCH) && !iofailed) { cl->cl_f_geometry_is_valid = FALSE; } return (rval); } /* * Function: cmlb_uselabel * * Description: Validate the disk label and update the relevant data (geometry, * partition, vtoc, and capacity data) in the cmlb_lun struct. * Marks the geometry of the unit as being valid. * * Arguments: cl: unit struct. * dk_label: disk label * * Return Code: CMLB_LABEL_IS_VALID: Label read from disk is OK; geometry, * partition, vtoc, and capacity data are good. * * CMLB_LABEL_IS_INVALID: Magic number or checksum error in the * label; or computed capacity does not jibe with capacity * reported from the READ CAPACITY command. * * Context: Kernel thread only (can sleep). */ static int cmlb_uselabel(struct cmlb_lun *cl, struct dk_label *labp, int flags) { short *sp; short sum; short count; int label_error = CMLB_LABEL_IS_VALID; int i; diskaddr_t label_capacity; uint32_t part_end; diskaddr_t track_capacity; #if defined(_SUNOS_VTOC_16) struct dkl_partition *vpartp; #endif ASSERT(cl != NULL); ASSERT(mutex_owned(CMLB_MUTEX(cl))); /* Validate the magic number of the label. */ if (labp->dkl_magic != DKL_MAGIC) { #if defined(__sparc) if (!ISREMOVABLE(cl) && !ISHOTPLUGGABLE(cl)) { if (!(flags & CMLB_SILENT)) cmlb_log(CMLB_DEVINFO(cl), CMLB_LABEL(cl), CE_WARN, "Corrupt label; wrong magic number\n"); } #endif return (CMLB_LABEL_IS_INVALID); } /* Validate the checksum of the label. */ sp = (short *)labp; sum = 0; count = sizeof (struct dk_label) / sizeof (short); while (count--) { sum ^= *sp++; } if (sum != 0) { #if defined(_SUNOS_VTOC_16) if (!ISCD(cl)) { #elif defined(_SUNOS_VTOC_8) if (!ISREMOVABLE(cl) && !ISHOTPLUGGABLE(cl)) { #endif if (!(flags & CMLB_SILENT)) cmlb_log(CMLB_DEVINFO(cl), CMLB_LABEL(cl), CE_WARN, "Corrupt label - label checksum failed\n"); } return (CMLB_LABEL_IS_INVALID); } /* * Fill in geometry structure with data from label. */ bzero(&cl->cl_g, sizeof (struct dk_geom)); cl->cl_g.dkg_ncyl = labp->dkl_ncyl; cl->cl_g.dkg_acyl = labp->dkl_acyl; cl->cl_g.dkg_bcyl = 0; cl->cl_g.dkg_nhead = labp->dkl_nhead; cl->cl_g.dkg_nsect = labp->dkl_nsect; cl->cl_g.dkg_intrlv = labp->dkl_intrlv; #if defined(_SUNOS_VTOC_8) cl->cl_g.dkg_gap1 = labp->dkl_gap1; cl->cl_g.dkg_gap2 = labp->dkl_gap2; cl->cl_g.dkg_bhead = labp->dkl_bhead; #endif #if defined(_SUNOS_VTOC_16) cl->cl_dkg_skew = labp->dkl_skew; #endif #if defined(__i386) || defined(__amd64) cl->cl_g.dkg_apc = labp->dkl_apc; #endif /* * Currently we rely on the values in the label being accurate. If * dkl_rpm or dkl_pcly are zero in the label, use a default value. * * Note: In the future a MODE SENSE may be used to retrieve this data, * although this command is optional in SCSI-2. */ cl->cl_g.dkg_rpm = (labp->dkl_rpm != 0) ? labp->dkl_rpm : 3600; cl->cl_g.dkg_pcyl = (labp->dkl_pcyl != 0) ? labp->dkl_pcyl : (cl->cl_g.dkg_ncyl + cl->cl_g.dkg_acyl); /* * The Read and Write reinstruct values may not be valid * for older disks. */ cl->cl_g.dkg_read_reinstruct = labp->dkl_read_reinstruct; cl->cl_g.dkg_write_reinstruct = labp->dkl_write_reinstruct; /* Fill in partition table. */ #if defined(_SUNOS_VTOC_8) for (i = 0; i < NDKMAP; i++) { cl->cl_map[i].dkl_cylno = labp->dkl_map[i].dkl_cylno; cl->cl_map[i].dkl_nblk = labp->dkl_map[i].dkl_nblk; } #endif #if defined(_SUNOS_VTOC_16) vpartp = labp->dkl_vtoc.v_part; track_capacity = labp->dkl_nhead * labp->dkl_nsect; /* Prevent divide by zero */ if (track_capacity == 0) { if (!(flags & CMLB_SILENT)) cmlb_log(CMLB_DEVINFO(cl), CMLB_LABEL(cl), CE_WARN, "Corrupt label - zero nhead or nsect value\n"); return (CMLB_LABEL_IS_INVALID); } for (i = 0; i < NDKMAP; i++, vpartp++) { cl->cl_map[i].dkl_cylno = vpartp->p_start / track_capacity; cl->cl_map[i].dkl_nblk = vpartp->p_size; } #endif /* Fill in VTOC Structure. */ bcopy(&labp->dkl_vtoc, &cl->cl_vtoc, sizeof (struct dk_vtoc)); #if defined(_SUNOS_VTOC_8) /* * The 8-slice vtoc does not include the ascii label; save it into * the device's soft state structure here. */ bcopy(labp->dkl_asciilabel, cl->cl_asciilabel, LEN_DKL_ASCII); #endif /* Now look for a valid capacity. */ track_capacity = (cl->cl_g.dkg_nhead * cl->cl_g.dkg_nsect); label_capacity = (cl->cl_g.dkg_ncyl * track_capacity); if (cl->cl_g.dkg_acyl) { #if defined(__i386) || defined(__amd64) /* we may have > 1 alts cylinder */ label_capacity += (track_capacity * cl->cl_g.dkg_acyl); #else label_capacity += track_capacity; #endif } /* * Force check here to ensure the computed capacity is valid. * If capacity is zero, it indicates an invalid label and * we should abort updating the relevant data then. */ if (label_capacity == 0) { if (!(flags & CMLB_SILENT)) cmlb_log(CMLB_DEVINFO(cl), CMLB_LABEL(cl), CE_WARN, "Corrupt label - no valid capacity could be " "retrieved\n"); return (CMLB_LABEL_IS_INVALID); } /* Mark the geometry as valid. */ cl->cl_f_geometry_is_valid = TRUE; /* * if we got invalidated when mutex exit and entered again, * if blockcount different than when we came in, need to * retry from beginning of cmlb_validate_geometry. * revisit this on next phase of utilizing this for * sd. */ if (label_capacity <= cl->cl_blockcount) { #if defined(_SUNOS_VTOC_8) /* * We can't let this happen on drives that are subdivided * into logical disks (i.e., that have an fdisk table). * The cl_blockcount field should always hold the full media * size in sectors, period. This code would overwrite * cl_blockcount with the size of the Solaris fdisk partition. */ cmlb_dbg(CMLB_ERROR, cl, "cmlb_uselabel: Label %d blocks; Drive %d blocks\n", label_capacity, cl->cl_blockcount); cl->cl_solaris_size = label_capacity; #endif /* defined(_SUNOS_VTOC_8) */ goto done; } if (ISCD(cl)) { /* For CDROMs, we trust that the data in the label is OK. */ #if defined(_SUNOS_VTOC_8) for (i = 0; i < NDKMAP; i++) { part_end = labp->dkl_nhead * labp->dkl_nsect * labp->dkl_map[i].dkl_cylno + labp->dkl_map[i].dkl_nblk - 1; if ((labp->dkl_map[i].dkl_nblk) && (part_end > cl->cl_blockcount)) { cl->cl_f_geometry_is_valid = FALSE; break; } } #endif #if defined(_SUNOS_VTOC_16) vpartp = &(labp->dkl_vtoc.v_part[0]); for (i = 0; i < NDKMAP; i++, vpartp++) { part_end = vpartp->p_start + vpartp->p_size; if ((vpartp->p_size > 0) && (part_end > cl->cl_blockcount)) { cl->cl_f_geometry_is_valid = FALSE; break; } } #endif } else { /* label_capacity > cl->cl_blockcount */ if (!(flags & CMLB_SILENT)) { cmlb_log(CMLB_DEVINFO(cl), CMLB_LABEL(cl), CE_WARN, "Corrupt label - bad geometry\n"); cmlb_log(CMLB_DEVINFO(cl), CMLB_LABEL(cl), CE_CONT, "Label says %llu blocks; Drive says %llu blocks\n", label_capacity, cl->cl_blockcount); } cl->cl_f_geometry_is_valid = FALSE; label_error = CMLB_LABEL_IS_INVALID; } done: cmlb_dbg(CMLB_INFO, cl, "cmlb_uselabel: (label geometry)\n"); cmlb_dbg(CMLB_INFO, cl, " ncyl: %d; acyl: %d; nhead: %d; nsect: %d\n", cl->cl_g.dkg_ncyl, cl->cl_g.dkg_acyl, cl->cl_g.dkg_nhead, cl->cl_g.dkg_nsect); cmlb_dbg(CMLB_INFO, cl, " label_capacity: %d; intrlv: %d; rpm: %d\n", cl->cl_blockcount, cl->cl_g.dkg_intrlv, cl->cl_g.dkg_rpm); cmlb_dbg(CMLB_INFO, cl, " wrt_reinstr: %d; rd_reinstr: %d\n", cl->cl_g.dkg_write_reinstruct, cl->cl_g.dkg_read_reinstruct); ASSERT(mutex_owned(CMLB_MUTEX(cl))); return (label_error); } /* * Function: cmlb_build_default_label * * Description: Generate a default label for those devices that do not have * one, e.g., new media, removable cartridges, etc.. * * Context: Kernel thread only */ /*ARGSUSED*/ static void cmlb_build_default_label(struct cmlb_lun *cl, void *tg_cookie) { #if defined(_SUNOS_VTOC_16) uint_t phys_spc; uint_t disksize; struct dk_geom cl_g; diskaddr_t capacity; #endif ASSERT(cl != NULL); ASSERT(mutex_owned(CMLB_MUTEX(cl))); #if defined(_SUNOS_VTOC_8) /* * Note: This is a legacy check for non-removable devices on VTOC_8 * only. This may be a valid check for VTOC_16 as well. * Once we understand why there is this difference between SPARC and * x86 platform, we could remove this legacy check. */ if (!ISREMOVABLE(cl) && !ISHOTPLUGGABLE(cl)) { return; } #endif bzero(&cl->cl_g, sizeof (struct dk_geom)); bzero(&cl->cl_vtoc, sizeof (struct dk_vtoc)); bzero(&cl->cl_map, NDKMAP * (sizeof (struct dk_map))); #if defined(_SUNOS_VTOC_8) /* * It's a REMOVABLE media, therefore no label (on sparc, anyway). * But it is still necessary to set up various geometry information, * and we are doing this here. */ /* * For the rpm, we use the minimum for the disk. For the head, cyl, * and number of sector per track, if the capacity <= 1GB, head = 64, * sect = 32. else head = 255, sect 63 Note: the capacity should be * equal to C*H*S values. This will cause some truncation of size due * to round off errors. For CD-ROMs, this truncation can have adverse * side effects, so returning ncyl and nhead as 1. The nsect will * overflow for most of CD-ROMs as nsect is of type ushort. (4190569) */ cl->cl_solaris_size = cl->cl_blockcount; if (ISCD(cl)) { tg_attribute_t tgattribute; int is_writable; /* * Preserve the old behavior for non-writable * medias. Since dkg_nsect is a ushort, it * will lose bits as cdroms have more than * 65536 sectors. So if we recalculate * capacity, it will become much shorter. * But the dkg_* information is not * used for CDROMs so it is OK. But for * Writable CDs we need this information * to be valid (for newfs say). So we * make nsect and nhead > 1 that way * nsect can still stay within ushort limit * without losing any bits. */ bzero(&tgattribute, sizeof (tg_attribute_t)); mutex_exit(CMLB_MUTEX(cl)); is_writable = (DK_TG_GETATTRIBUTE(cl, &tgattribute, tg_cookie) == 0) ? tgattribute.media_is_writable : 1; mutex_enter(CMLB_MUTEX(cl)); if (is_writable) { cl->cl_g.dkg_nhead = 64; cl->cl_g.dkg_nsect = 32; cl->cl_g.dkg_ncyl = cl->cl_blockcount / (64 * 32); cl->cl_solaris_size = (diskaddr_t)cl->cl_g.dkg_ncyl * cl->cl_g.dkg_nhead * cl->cl_g.dkg_nsect; } else { cl->cl_g.dkg_ncyl = 1; cl->cl_g.dkg_nhead = 1; cl->cl_g.dkg_nsect = cl->cl_blockcount; } } else { if (cl->cl_blockcount <= 0x1000) { /* unlabeled SCSI floppy device */ cl->cl_g.dkg_nhead = 2; cl->cl_g.dkg_ncyl = 80; cl->cl_g.dkg_nsect = cl->cl_blockcount / (2 * 80); } else if (cl->cl_blockcount <= 0x200000) { cl->cl_g.dkg_nhead = 64; cl->cl_g.dkg_nsect = 32; cl->cl_g.dkg_ncyl = cl->cl_blockcount / (64 * 32); } else { cl->cl_g.dkg_nhead = 255; cl->cl_g.dkg_nsect = ((cl->cl_blockcount + (UINT16_MAX * 255 * 63) - 1) / (UINT16_MAX * 255 * 63)) * 63; if (cl->cl_g.dkg_nsect == 0) cl->cl_g.dkg_nsect = (UINT16_MAX / 63) * 63; cl->cl_g.dkg_ncyl = cl->cl_blockcount / (255 * cl->cl_g.dkg_nsect); } cl->cl_solaris_size = (diskaddr_t)cl->cl_g.dkg_ncyl * cl->cl_g.dkg_nhead * cl->cl_g.dkg_nsect; } cl->cl_g.dkg_acyl = 0; cl->cl_g.dkg_bcyl = 0; cl->cl_g.dkg_rpm = 200; cl->cl_asciilabel[0] = '\0'; cl->cl_g.dkg_pcyl = cl->cl_g.dkg_ncyl; cl->cl_map[0].dkl_cylno = 0; cl->cl_map[0].dkl_nblk = cl->cl_solaris_size; cl->cl_map[2].dkl_cylno = 0; cl->cl_map[2].dkl_nblk = cl->cl_solaris_size; #elif defined(_SUNOS_VTOC_16) if (cl->cl_solaris_size == 0) { /* * Got fdisk table but no solaris entry therefore * don't create a default label */ cl->cl_f_geometry_is_valid = TRUE; return; } /* * For CDs we continue to use the physical geometry to calculate * number of cylinders. All other devices must convert the * physical geometry (cmlb_geom) to values that will fit * in a dk_geom structure. */ if (ISCD(cl)) { phys_spc = cl->cl_pgeom.g_nhead * cl->cl_pgeom.g_nsect; } else { /* Convert physical geometry to disk geometry */ bzero(&cl_g, sizeof (struct dk_geom)); /* * Refer to comments related to off-by-1 at the * header of this file. * Before calculating geometry, capacity should be * decreased by 1. */ if (cl->cl_alter_behavior & CMLB_OFF_BY_ONE) capacity = cl->cl_blockcount - 1; else capacity = cl->cl_blockcount; cmlb_convert_geometry(capacity, &cl_g); bcopy(&cl_g, &cl->cl_g, sizeof (cl->cl_g)); phys_spc = cl->cl_g.dkg_nhead * cl->cl_g.dkg_nsect; } ASSERT(phys_spc != 0); cl->cl_g.dkg_pcyl = cl->cl_solaris_size / phys_spc; if (cl->cl_alter_behavior & CMLB_FAKE_LABEL_ONE_PARTITION) { /* disable devid */ cl->cl_g.dkg_ncyl = cl->cl_g.dkg_pcyl; disksize = cl->cl_solaris_size; } else { cl->cl_g.dkg_acyl = DK_ACYL; cl->cl_g.dkg_ncyl = cl->cl_g.dkg_pcyl - DK_ACYL; disksize = cl->cl_g.dkg_ncyl * phys_spc; } if (ISCD(cl)) { /* * CD's don't use the "heads * sectors * cyls"-type of * geometry, but instead use the entire capacity of the media. */ disksize = cl->cl_solaris_size; cl->cl_g.dkg_nhead = 1; cl->cl_g.dkg_nsect = 1; cl->cl_g.dkg_rpm = (cl->cl_pgeom.g_rpm == 0) ? 200 : cl->cl_pgeom.g_rpm; cl->cl_vtoc.v_part[0].p_start = 0; cl->cl_vtoc.v_part[0].p_size = disksize; cl->cl_vtoc.v_part[0].p_tag = V_BACKUP; cl->cl_vtoc.v_part[0].p_flag = V_UNMNT; cl->cl_map[0].dkl_cylno = 0; cl->cl_map[0].dkl_nblk = disksize; cl->cl_offset[0] = 0; } else { /* * Hard disks and removable media cartridges */ cl->cl_g.dkg_rpm = (cl->cl_pgeom.g_rpm == 0) ? 3600: cl->cl_pgeom.g_rpm; cl->cl_vtoc.v_sectorsz = cl->cl_sys_blocksize; /* Add boot slice */ cl->cl_vtoc.v_part[8].p_start = 0; cl->cl_vtoc.v_part[8].p_size = phys_spc; cl->cl_vtoc.v_part[8].p_tag = V_BOOT; cl->cl_vtoc.v_part[8].p_flag = V_UNMNT; cl->cl_map[8].dkl_cylno = 0; cl->cl_map[8].dkl_nblk = phys_spc; cl->cl_offset[8] = 0; if ((cl->cl_alter_behavior & CMLB_CREATE_ALTSLICE_VTOC_16_DTYPE_DIRECT) && cl->cl_device_type == DTYPE_DIRECT) { cl->cl_vtoc.v_part[9].p_start = phys_spc; cl->cl_vtoc.v_part[9].p_size = 2 * phys_spc; cl->cl_vtoc.v_part[9].p_tag = V_ALTSCTR; cl->cl_vtoc.v_part[9].p_flag = 0; cl->cl_map[9].dkl_cylno = 1; cl->cl_map[9].dkl_nblk = 2 * phys_spc; cl->cl_offset[9] = phys_spc; } } cl->cl_g.dkg_apc = 0; /* Add backup slice */ cl->cl_vtoc.v_part[2].p_start = 0; cl->cl_vtoc.v_part[2].p_size = disksize; cl->cl_vtoc.v_part[2].p_tag = V_BACKUP; cl->cl_vtoc.v_part[2].p_flag = V_UNMNT; cl->cl_map[2].dkl_cylno = 0; cl->cl_map[2].dkl_nblk = disksize; cl->cl_offset[2] = 0; /* * single slice (s0) covering the entire disk */ if (cl->cl_alter_behavior & CMLB_FAKE_LABEL_ONE_PARTITION) { cl->cl_vtoc.v_part[0].p_start = 0; cl->cl_vtoc.v_part[0].p_tag = V_UNASSIGNED; cl->cl_vtoc.v_part[0].p_flag = 0; cl->cl_vtoc.v_part[0].p_size = disksize; cl->cl_map[0].dkl_cylno = 0; cl->cl_map[0].dkl_nblk = disksize; cl->cl_offset[0] = 0; } (void) sprintf(cl->cl_vtoc.v_asciilabel, "DEFAULT cyl %d alt %d" " hd %d sec %d", cl->cl_g.dkg_ncyl, cl->cl_g.dkg_acyl, cl->cl_g.dkg_nhead, cl->cl_g.dkg_nsect); #else #error "No VTOC format defined." #endif cl->cl_g.dkg_read_reinstruct = 0; cl->cl_g.dkg_write_reinstruct = 0; cl->cl_g.dkg_intrlv = 1; cl->cl_vtoc.v_sanity = VTOC_SANE; cl->cl_vtoc.v_nparts = V_NUMPAR; cl->cl_vtoc.v_version = V_VERSION; cl->cl_f_geometry_is_valid = TRUE; cl->cl_label_from_media = CMLB_LABEL_UNDEF; cmlb_dbg(CMLB_INFO, cl, "cmlb_build_default_label: Default label created: " "cyl: %d\tacyl: %d\tnhead: %d\tnsect: %d\tcap: %d\n", cl->cl_g.dkg_ncyl, cl->cl_g.dkg_acyl, cl->cl_g.dkg_nhead, cl->cl_g.dkg_nsect, cl->cl_blockcount); } #if defined(_FIRMWARE_NEEDS_FDISK) /* * Max CHS values, as they are encoded into bytes, for 1022/254/63 */ #define LBA_MAX_SECT (63 | ((1022 & 0x300) >> 2)) #define LBA_MAX_CYL (1022 & 0xFF) #define LBA_MAX_HEAD (254) /* * Function: cmlb_has_max_chs_vals * * Description: Return TRUE if Cylinder-Head-Sector values are all at maximum. * * Arguments: fdp - ptr to CHS info * * Return Code: True or false * * Context: Any. */ static int cmlb_has_max_chs_vals(struct ipart *fdp) { return ((fdp->begcyl == LBA_MAX_CYL) && (fdp->beghead == LBA_MAX_HEAD) && (fdp->begsect == LBA_MAX_SECT) && (fdp->endcyl == LBA_MAX_CYL) && (fdp->endhead == LBA_MAX_HEAD) && (fdp->endsect == LBA_MAX_SECT)); } #endif /* * Function: cmlb_dkio_get_geometry * * Description: This routine is the driver entry point for handling user * requests to get the device geometry (DKIOCGGEOM). * * Arguments: * arg pointer to user provided dk_geom structure specifying * the controller's notion of the current geometry. * * flag this argument is a pass through to ddi_copyxxx() * directly from the mode argument of ioctl(). * * tg_cookie cookie from target driver to be passed back to target * driver when we call back to it through tg_ops. * * Return Code: 0 * EFAULT * ENXIO * EIO */ static int cmlb_dkio_get_geometry(struct cmlb_lun *cl, caddr_t arg, int flag, void *tg_cookie) { struct dk_geom *tmp_geom = NULL; int rval = 0; /* * cmlb_validate_geometry does not spin a disk up * if it was spcl down. We need to make sure it * is ready. */ mutex_enter(CMLB_MUTEX(cl)); rval = cmlb_validate_geometry(cl, 1, 0, tg_cookie); #if defined(_SUNOS_VTOC_8) if (rval == EINVAL && cl->cl_alter_behavior & CMLB_FAKE_GEOM_LABEL_IOCTLS_VTOC8) { /* * This is to return a default label geometry even when we * do not really assume a default label for the device. * dad driver utilizes this. */ if (cl->cl_blockcount <= CMLB_OLDVTOC_LIMIT) { cmlb_setup_default_geometry(cl, tg_cookie); rval = 0; } } #endif if (rval) { mutex_exit(CMLB_MUTEX(cl)); return (rval); } #if defined(__i386) || defined(__amd64) if (cl->cl_solaris_size == 0) { mutex_exit(CMLB_MUTEX(cl)); return (EIO); } #endif /* * Make a local copy of the soft state geometry to avoid some potential * race conditions associated with holding the mutex and updating the * write_reinstruct value */ tmp_geom = kmem_zalloc(sizeof (struct dk_geom), KM_SLEEP); bcopy(&cl->cl_g, tmp_geom, sizeof (struct dk_geom)); if (tmp_geom->dkg_write_reinstruct == 0) { tmp_geom->dkg_write_reinstruct = (int)((int)(tmp_geom->dkg_nsect * tmp_geom->dkg_rpm * cmlb_rot_delay) / (int)60000); } mutex_exit(CMLB_MUTEX(cl)); rval = ddi_copyout(tmp_geom, (void *)arg, sizeof (struct dk_geom), flag); if (rval != 0) { rval = EFAULT; } kmem_free(tmp_geom, sizeof (struct dk_geom)); return (rval); } /* * Function: cmlb_dkio_set_geometry * * Description: This routine is the driver entry point for handling user * requests to set the device geometry (DKIOCSGEOM). The actual * device geometry is not updated, just the driver "notion" of it. * * Arguments: * arg pointer to user provided dk_geom structure used to set * the controller's notion of the current geometry. * * flag this argument is a pass through to ddi_copyxxx() * directly from the mode argument of ioctl(). * * tg_cookie cookie from target driver to be passed back to target * driver when we call back to it through tg_ops. * * Return Code: 0 * EFAULT * ENXIO * EIO */ static int cmlb_dkio_set_geometry(struct cmlb_lun *cl, caddr_t arg, int flag) { struct dk_geom *tmp_geom; struct dk_map *lp; int rval = 0; int i; #if defined(__i386) || defined(__amd64) if (cl->cl_solaris_size == 0) { return (EIO); } #endif /* * We need to copy the user specified geometry into local * storage and then update the softstate. We don't want to hold * the mutex and copyin directly from the user to the soft state */ tmp_geom = (struct dk_geom *) kmem_zalloc(sizeof (struct dk_geom), KM_SLEEP); rval = ddi_copyin(arg, tmp_geom, sizeof (struct dk_geom), flag); if (rval != 0) { kmem_free(tmp_geom, sizeof (struct dk_geom)); return (EFAULT); } mutex_enter(CMLB_MUTEX(cl)); bcopy(tmp_geom, &cl->cl_g, sizeof (struct dk_geom)); for (i = 0; i < NDKMAP; i++) { lp = &cl->cl_map[i]; cl->cl_offset[i] = cl->cl_g.dkg_nhead * cl->cl_g.dkg_nsect * lp->dkl_cylno; #if defined(__i386) || defined(__amd64) cl->cl_offset[i] += cl->cl_solaris_offset; #endif } cl->cl_f_geometry_is_valid = FALSE; mutex_exit(CMLB_MUTEX(cl)); kmem_free(tmp_geom, sizeof (struct dk_geom)); return (rval); } /* * Function: cmlb_dkio_get_partition * * Description: This routine is the driver entry point for handling user * requests to get the partition table (DKIOCGAPART). * * Arguments: * arg pointer to user provided dk_allmap structure specifying * the controller's notion of the current partition table. * * flag this argument is a pass through to ddi_copyxxx() * directly from the mode argument of ioctl(). * * tg_cookie cookie from target driver to be passed back to target * driver when we call back to it through tg_ops. * * Return Code: 0 * EFAULT * ENXIO * EIO */ static int cmlb_dkio_get_partition(struct cmlb_lun *cl, caddr_t arg, int flag, void *tg_cookie) { int rval = 0; int size; /* * Make sure the geometry is valid before getting the partition * information. */ mutex_enter(CMLB_MUTEX(cl)); if ((rval = cmlb_validate_geometry(cl, 1, 0, tg_cookie)) != 0) { mutex_exit(CMLB_MUTEX(cl)); return (rval); } mutex_exit(CMLB_MUTEX(cl)); #if defined(__i386) || defined(__amd64) if (cl->cl_solaris_size == 0) { return (EIO); } #endif #ifdef _MULTI_DATAMODEL switch (ddi_model_convert_from(flag & FMODELS)) { case DDI_MODEL_ILP32: { struct dk_map32 dk_map32[NDKMAP]; int i; for (i = 0; i < NDKMAP; i++) { dk_map32[i].dkl_cylno = cl->cl_map[i].dkl_cylno; dk_map32[i].dkl_nblk = cl->cl_map[i].dkl_nblk; } size = NDKMAP * sizeof (struct dk_map32); rval = ddi_copyout(dk_map32, (void *)arg, size, flag); if (rval != 0) { rval = EFAULT; } break; } case DDI_MODEL_NONE: size = NDKMAP * sizeof (struct dk_map); rval = ddi_copyout(cl->cl_map, (void *)arg, size, flag); if (rval != 0) { rval = EFAULT; } break; } #else /* ! _MULTI_DATAMODEL */ size = NDKMAP * sizeof (struct dk_map); rval = ddi_copyout(cl->cl_map, (void *)arg, size, flag); if (rval != 0) { rval = EFAULT; } #endif /* _MULTI_DATAMODEL */ return (rval); } /* * Function: cmlb_dkio_set_partition * * Description: This routine is the driver entry point for handling user * requests to set the partition table (DKIOCSAPART). The actual * device partition is not updated. * * Arguments: * arg - pointer to user provided dk_allmap structure used to set * the controller's notion of the partition table. * flag - this argument is a pass through to ddi_copyxxx() * directly from the mode argument of ioctl(). * * Return Code: 0 * EINVAL * EFAULT * ENXIO * EIO */ static int cmlb_dkio_set_partition(struct cmlb_lun *cl, caddr_t arg, int flag) { struct dk_map dk_map[NDKMAP]; struct dk_map *lp; int rval = 0; int size; int i; #if defined(_SUNOS_VTOC_16) struct dkl_partition *vp; #endif /* * Set the map for all logical partitions. We lock * the priority just to make sure an interrupt doesn't * come in while the map is half updated. */ _NOTE(DATA_READABLE_WITHOUT_LOCK(cmlb_lun::cl_solaris_size)) mutex_enter(CMLB_MUTEX(cl)); if (cl->cl_blockcount > CMLB_OLDVTOC_LIMIT) { mutex_exit(CMLB_MUTEX(cl)); return (ENOTSUP); } mutex_exit(CMLB_MUTEX(cl)); if (cl->cl_solaris_size == 0) { return (EIO); } #ifdef _MULTI_DATAMODEL switch (ddi_model_convert_from(flag & FMODELS)) { case DDI_MODEL_ILP32: { struct dk_map32 dk_map32[NDKMAP]; size = NDKMAP * sizeof (struct dk_map32); rval = ddi_copyin((void *)arg, dk_map32, size, flag); if (rval != 0) { return (EFAULT); } for (i = 0; i < NDKMAP; i++) { dk_map[i].dkl_cylno = dk_map32[i].dkl_cylno; dk_map[i].dkl_nblk = dk_map32[i].dkl_nblk; } break; } case DDI_MODEL_NONE: size = NDKMAP * sizeof (struct dk_map); rval = ddi_copyin((void *)arg, dk_map, size, flag); if (rval != 0) { return (EFAULT); } break; } #else /* ! _MULTI_DATAMODEL */ size = NDKMAP * sizeof (struct dk_map); rval = ddi_copyin((void *)arg, dk_map, size, flag); if (rval != 0) { return (EFAULT); } #endif /* _MULTI_DATAMODEL */ mutex_enter(CMLB_MUTEX(cl)); /* Note: The size used in this bcopy is set based upon the data model */ bcopy(dk_map, cl->cl_map, size); #if defined(_SUNOS_VTOC_16) vp = (struct dkl_partition *)&(cl->cl_vtoc); #endif /* defined(_SUNOS_VTOC_16) */ for (i = 0; i < NDKMAP; i++) { lp = &cl->cl_map[i]; cl->cl_offset[i] = cl->cl_g.dkg_nhead * cl->cl_g.dkg_nsect * lp->dkl_cylno; #if defined(_SUNOS_VTOC_16) vp->p_start = cl->cl_offset[i]; vp->p_size = lp->dkl_nblk; vp++; #endif /* defined(_SUNOS_VTOC_16) */ #if defined(__i386) || defined(__amd64) cl->cl_offset[i] += cl->cl_solaris_offset; #endif } mutex_exit(CMLB_MUTEX(cl)); return (rval); } /* * Function: cmlb_dkio_get_vtoc * * Description: This routine is the driver entry point for handling user * requests to get the current volume table of contents * (DKIOCGVTOC). * * Arguments: * arg pointer to user provided vtoc structure specifying * the current vtoc. * * flag this argument is a pass through to ddi_copyxxx() * directly from the mode argument of ioctl(). * * tg_cookie cookie from target driver to be passed back to target * driver when we call back to it through tg_ops. * * Return Code: 0 * EFAULT * ENXIO * EIO */ static int cmlb_dkio_get_vtoc(struct cmlb_lun *cl, caddr_t arg, int flag, void *tg_cookie) { #if defined(_SUNOS_VTOC_8) struct vtoc user_vtoc; #endif /* defined(_SUNOS_VTOC_8) */ int rval = 0; mutex_enter(CMLB_MUTEX(cl)); if (cl->cl_blockcount > CMLB_OLDVTOC_LIMIT) { mutex_exit(CMLB_MUTEX(cl)); return (EOVERFLOW); } rval = cmlb_validate_geometry(cl, 1, 0, tg_cookie); #if defined(_SUNOS_VTOC_8) if (rval == EINVAL && (cl->cl_alter_behavior & CMLB_FAKE_GEOM_LABEL_IOCTLS_VTOC8)) { /* * This is to return a default label even when we do not * really assume a default label for the device. * dad driver utilizes this. */ if (cl->cl_blockcount <= CMLB_OLDVTOC_LIMIT) { cmlb_setup_default_geometry(cl, tg_cookie); rval = 0; } } #endif if (rval) { mutex_exit(CMLB_MUTEX(cl)); return (rval); } #if defined(_SUNOS_VTOC_8) cmlb_build_user_vtoc(cl, &user_vtoc); mutex_exit(CMLB_MUTEX(cl)); #ifdef _MULTI_DATAMODEL switch (ddi_model_convert_from(flag & FMODELS)) { case DDI_MODEL_ILP32: { struct vtoc32 user_vtoc32; vtoctovtoc32(user_vtoc, user_vtoc32); if (ddi_copyout(&user_vtoc32, (void *)arg, sizeof (struct vtoc32), flag)) { return (EFAULT); } break; } case DDI_MODEL_NONE: if (ddi_copyout(&user_vtoc, (void *)arg, sizeof (struct vtoc), flag)) { return (EFAULT); } break; } #else /* ! _MULTI_DATAMODEL */ if (ddi_copyout(&user_vtoc, (void *)arg, sizeof (struct vtoc), flag)) { return (EFAULT); } #endif /* _MULTI_DATAMODEL */ #elif defined(_SUNOS_VTOC_16) mutex_exit(CMLB_MUTEX(cl)); #ifdef _MULTI_DATAMODEL /* * The cl_vtoc structure is a "struct dk_vtoc" which is always * 32-bit to maintain compatibility with existing on-disk * structures. Thus, we need to convert the structure when copying * it out to a datamodel-dependent "struct vtoc" in a 64-bit * program. If the target is a 32-bit program, then no conversion * is necessary. */ /* LINTED: logical expression always true: op "||" */ ASSERT(sizeof (cl->cl_vtoc) == sizeof (struct vtoc32)); switch (ddi_model_convert_from(flag & FMODELS)) { case DDI_MODEL_ILP32: if (ddi_copyout(&(cl->cl_vtoc), (void *)arg, sizeof (cl->cl_vtoc), flag)) { return (EFAULT); } break; case DDI_MODEL_NONE: { struct vtoc user_vtoc; vtoc32tovtoc(cl->cl_vtoc, user_vtoc); if (ddi_copyout(&user_vtoc, (void *)arg, sizeof (struct vtoc), flag)) { return (EFAULT); } break; } } #else /* ! _MULTI_DATAMODEL */ if (ddi_copyout(&(cl->cl_vtoc), (void *)arg, sizeof (cl->cl_vtoc), flag)) { return (EFAULT); } #endif /* _MULTI_DATAMODEL */ #else #error "No VTOC format defined." #endif return (rval); } /* * Function: cmlb_dkio_get_extvtoc */ static int cmlb_dkio_get_extvtoc(struct cmlb_lun *cl, caddr_t arg, int flag, void *tg_cookie) { struct extvtoc ext_vtoc; #if defined(_SUNOS_VTOC_8) struct vtoc user_vtoc; #endif /* defined(_SUNOS_VTOC_8) */ int rval = 0; bzero(&ext_vtoc, sizeof (struct extvtoc)); mutex_enter(CMLB_MUTEX(cl)); rval = cmlb_validate_geometry(cl, 1, 0, tg_cookie); #if defined(_SUNOS_VTOC_8) if (rval == EINVAL && (cl->cl_alter_behavior & CMLB_FAKE_GEOM_LABEL_IOCTLS_VTOC8)) { /* * This is to return a default label even when we do not * really assume a default label for the device. * dad driver utilizes this. */ if (cl->cl_blockcount <= CMLB_OLDVTOC_LIMIT) { cmlb_setup_default_geometry(cl, tg_cookie); rval = 0; } } #endif if (rval) { mutex_exit(CMLB_MUTEX(cl)); return (rval); } #if defined(_SUNOS_VTOC_8) cmlb_build_user_vtoc(cl, &user_vtoc); mutex_exit(CMLB_MUTEX(cl)); /* * Checking callers data model does not make much sense here * since extvtoc will always be equivalent to 64bit vtoc. * What is important is whether the kernel is in 32 or 64 bit */ #ifdef _LP64 if (ddi_copyout(&user_vtoc, (void *)arg, sizeof (struct extvtoc), flag)) { return (EFAULT); } #else vtoc32tovtoc(user_vtoc, ext_vtoc); if (ddi_copyout(&ext_vtoc, (void *)arg, sizeof (struct extvtoc), flag)) { return (EFAULT); } #endif #elif defined(_SUNOS_VTOC_16) /* * The cl_vtoc structure is a "struct dk_vtoc" which is always * 32-bit to maintain compatibility with existing on-disk * structures. Thus, we need to convert the structure when copying * it out to extvtoc */ vtoc32tovtoc(cl->cl_vtoc, ext_vtoc); mutex_exit(CMLB_MUTEX(cl)); if (ddi_copyout(&ext_vtoc, (void *)arg, sizeof (struct extvtoc), flag)) return (EFAULT); #else #error "No VTOC format defined." #endif return (rval); } static int cmlb_dkio_get_efi(struct cmlb_lun *cl, caddr_t arg, int flag, void *tg_cookie) { dk_efi_t user_efi; int rval = 0; void *buffer; diskaddr_t tgt_lba; if (ddi_copyin(arg, &user_efi, sizeof (dk_efi_t), flag)) return (EFAULT); user_efi.dki_data = (void *)(uintptr_t)user_efi.dki_data_64; tgt_lba = user_efi.dki_lba; mutex_enter(CMLB_MUTEX(cl)); if ((cmlb_check_update_blockcount(cl, tg_cookie) != 0) || (cl->cl_tgt_blocksize == 0)) { mutex_exit(CMLB_MUTEX(cl)); return (EINVAL); } if (cl->cl_tgt_blocksize != cl->cl_sys_blocksize) tgt_lba = tgt_lba * cl->cl_tgt_blocksize / cl->cl_sys_blocksize; mutex_exit(CMLB_MUTEX(cl)); buffer = kmem_alloc(user_efi.dki_length, KM_SLEEP); rval = DK_TG_READ(cl, buffer, tgt_lba, user_efi.dki_length, tg_cookie); if (rval == 0 && ddi_copyout(buffer, user_efi.dki_data, user_efi.dki_length, flag) != 0) rval = EFAULT; kmem_free(buffer, user_efi.dki_length); return (rval); } #if defined(_SUNOS_VTOC_8) /* * Function: cmlb_build_user_vtoc * * Description: This routine populates a pass by reference variable with the * current volume table of contents. * * Arguments: cl - driver soft state (unit) structure * user_vtoc - pointer to vtoc structure to be populated */ static void cmlb_build_user_vtoc(struct cmlb_lun *cl, struct vtoc *user_vtoc) { struct dk_map2 *lpart; struct dk_map *lmap; struct partition *vpart; uint32_t nblks; int i; ASSERT(mutex_owned(CMLB_MUTEX(cl))); /* * Return vtoc structure fields in the provided VTOC area, addressed * by *vtoc. */ bzero(user_vtoc, sizeof (struct vtoc)); user_vtoc->v_bootinfo[0] = cl->cl_vtoc.v_bootinfo[0]; user_vtoc->v_bootinfo[1] = cl->cl_vtoc.v_bootinfo[1]; user_vtoc->v_bootinfo[2] = cl->cl_vtoc.v_bootinfo[2]; user_vtoc->v_sanity = VTOC_SANE; user_vtoc->v_version = cl->cl_vtoc.v_version; bcopy(cl->cl_vtoc.v_volume, user_vtoc->v_volume, LEN_DKL_VVOL); user_vtoc->v_sectorsz = cl->cl_sys_blocksize; user_vtoc->v_nparts = cl->cl_vtoc.v_nparts; for (i = 0; i < 10; i++) user_vtoc->v_reserved[i] = cl->cl_vtoc.v_reserved[i]; /* * Convert partitioning information. * * Note the conversion from starting cylinder number * to starting sector number. */ lmap = cl->cl_map; lpart = (struct dk_map2 *)cl->cl_vtoc.v_part; vpart = user_vtoc->v_part; nblks = cl->cl_g.dkg_nsect * cl->cl_g.dkg_nhead; for (i = 0; i < V_NUMPAR; i++) { vpart->p_tag = lpart->p_tag; vpart->p_flag = lpart->p_flag; vpart->p_start = lmap->dkl_cylno * nblks; vpart->p_size = lmap->dkl_nblk; lmap++; lpart++; vpart++; /* (4364927) */ user_vtoc->timestamp[i] = (time_t)cl->cl_vtoc.v_timestamp[i]; } bcopy(cl->cl_asciilabel, user_vtoc->v_asciilabel, LEN_DKL_ASCII); } #endif static int cmlb_dkio_partition(struct cmlb_lun *cl, caddr_t arg, int flag, void *tg_cookie) { struct partition64 p64; int rval = 0; uint_t nparts; efi_gpe_t *partitions; efi_gpt_t *buffer; diskaddr_t gpe_lba; if (ddi_copyin((const void *)arg, &p64, sizeof (struct partition64), flag)) { return (EFAULT); } buffer = kmem_alloc(EFI_MIN_ARRAY_SIZE, KM_SLEEP); rval = DK_TG_READ(cl, buffer, 1, DEV_BSIZE, tg_cookie); if (rval != 0) goto done_error; cmlb_swap_efi_gpt(buffer); if ((rval = cmlb_validate_efi(buffer)) != 0) goto done_error; nparts = buffer->efi_gpt_NumberOfPartitionEntries; gpe_lba = buffer->efi_gpt_PartitionEntryLBA; if (p64.p_partno > nparts) { /* couldn't find it */ rval = ESRCH; goto done_error; } /* * if we're dealing with a partition that's out of the normal * 16K block, adjust accordingly */ gpe_lba += p64.p_partno / sizeof (efi_gpe_t); rval = DK_TG_READ(cl, buffer, gpe_lba, EFI_MIN_ARRAY_SIZE, tg_cookie); if (rval) { goto done_error; } partitions = (efi_gpe_t *)buffer; cmlb_swap_efi_gpe(nparts, partitions); partitions += p64.p_partno; bcopy(&partitions->efi_gpe_PartitionTypeGUID, &p64.p_type, sizeof (struct uuid)); p64.p_start = partitions->efi_gpe_StartingLBA; p64.p_size = partitions->efi_gpe_EndingLBA - p64.p_start + 1; if (ddi_copyout(&p64, (void *)arg, sizeof (struct partition64), flag)) rval = EFAULT; done_error: kmem_free(buffer, EFI_MIN_ARRAY_SIZE); return (rval); } /* * Function: cmlb_dkio_set_vtoc * * Description: This routine is the driver entry point for handling user * requests to set the current volume table of contents * (DKIOCSVTOC). * * Arguments: * dev the device number * arg pointer to user provided vtoc structure used to set the * current vtoc. * * flag this argument is a pass through to ddi_copyxxx() * directly from the mode argument of ioctl(). * * tg_cookie cookie from target driver to be passed back to target * driver when we call back to it through tg_ops. * * Return Code: 0 * EFAULT * ENXIO * EINVAL * ENOTSUP */ static int cmlb_dkio_set_vtoc(struct cmlb_lun *cl, dev_t dev, caddr_t arg, int flag, void *tg_cookie) { struct vtoc user_vtoc; int rval = 0; boolean_t internal; internal = ((cl->cl_alter_behavior & (CMLB_INTERNAL_MINOR_NODES)) != 0); #ifdef _MULTI_DATAMODEL switch (ddi_model_convert_from(flag & FMODELS)) { case DDI_MODEL_ILP32: { struct vtoc32 user_vtoc32; if (ddi_copyin((const void *)arg, &user_vtoc32, sizeof (struct vtoc32), flag)) { return (EFAULT); } vtoc32tovtoc(user_vtoc32, user_vtoc); break; } case DDI_MODEL_NONE: if (ddi_copyin((const void *)arg, &user_vtoc, sizeof (struct vtoc), flag)) { return (EFAULT); } break; } #else /* ! _MULTI_DATAMODEL */ if (ddi_copyin((const void *)arg, &user_vtoc, sizeof (struct vtoc), flag)) { return (EFAULT); } #endif /* _MULTI_DATAMODEL */ mutex_enter(CMLB_MUTEX(cl)); if (cl->cl_blockcount > CMLB_OLDVTOC_LIMIT) { mutex_exit(CMLB_MUTEX(cl)); return (EOVERFLOW); } #if defined(__i386) || defined(__amd64) if (cl->cl_tgt_blocksize != cl->cl_sys_blocksize) { mutex_exit(CMLB_MUTEX(cl)); return (EINVAL); } #endif if (cl->cl_g.dkg_ncyl == 0) { mutex_exit(CMLB_MUTEX(cl)); return (EINVAL); } mutex_exit(CMLB_MUTEX(cl)); cmlb_clear_efi(cl, tg_cookie); ddi_remove_minor_node(CMLB_DEVINFO(cl), "wd"); ddi_remove_minor_node(CMLB_DEVINFO(cl), "wd,raw"); /* * cmlb_dkio_set_vtoc creates duplicate minor nodes when * relabeling an SMI disk. To avoid that we remove them * before creating. * It should be OK to remove a non-existed minor node. */ ddi_remove_minor_node(CMLB_DEVINFO(cl), "h"); ddi_remove_minor_node(CMLB_DEVINFO(cl), "h,raw"); (void) cmlb_create_minor(CMLB_DEVINFO(cl), "h", S_IFBLK, (CMLBUNIT(dev) << CMLBUNIT_SHIFT) | WD_NODE, cl->cl_node_type, NULL, internal); (void) cmlb_create_minor(CMLB_DEVINFO(cl), "h,raw", S_IFCHR, (CMLBUNIT(dev) << CMLBUNIT_SHIFT) | WD_NODE, cl->cl_node_type, NULL, internal); mutex_enter(CMLB_MUTEX(cl)); if ((rval = cmlb_build_label_vtoc(cl, &user_vtoc)) == 0) { if ((rval = cmlb_write_label(cl, tg_cookie)) == 0) { if (cmlb_validate_geometry(cl, 1, 0, tg_cookie) != 0) { cmlb_dbg(CMLB_ERROR, cl, "cmlb_dkio_set_vtoc: " "Failed validate geometry\n"); } cl->cl_msglog_flag |= CMLB_ALLOW_2TB_WARN; } } mutex_exit(CMLB_MUTEX(cl)); return (rval); } /* * Function: cmlb_dkio_set_extvtoc */ static int cmlb_dkio_set_extvtoc(struct cmlb_lun *cl, dev_t dev, caddr_t arg, int flag, void *tg_cookie) { int rval = 0; struct vtoc user_vtoc; /* * Checking callers data model does not make much sense here * since extvtoc will always be equivalent to 64bit vtoc. * What is important is whether the kernel is in 32 or 64 bit */ #ifdef _LP64 if (ddi_copyin((const void *)arg, &user_vtoc, sizeof (struct extvtoc), flag)) { return (EFAULT); } #else struct extvtoc user_extvtoc; if (ddi_copyin((const void *)arg, &user_extvtoc, sizeof (struct extvtoc), flag)) { return (EFAULT); } vtoctovtoc32(user_extvtoc, user_vtoc); #endif mutex_enter(CMLB_MUTEX(cl)); #if defined(__i386) || defined(__amd64) if (cl->cl_tgt_blocksize != cl->cl_sys_blocksize) { mutex_exit(CMLB_MUTEX(cl)); return (EINVAL); } #endif if (cl->cl_g.dkg_ncyl == 0) { mutex_exit(CMLB_MUTEX(cl)); return (EINVAL); } mutex_exit(CMLB_MUTEX(cl)); cmlb_clear_efi(cl, tg_cookie); ddi_remove_minor_node(CMLB_DEVINFO(cl), "wd"); ddi_remove_minor_node(CMLB_DEVINFO(cl), "wd,raw"); (void) ddi_create_minor_node(CMLB_DEVINFO(cl), "h", S_IFBLK, (CMLBUNIT(dev) << CMLBUNIT_SHIFT) | WD_NODE, cl->cl_node_type, NULL); (void) ddi_create_minor_node(CMLB_DEVINFO(cl), "h,raw", S_IFCHR, (CMLBUNIT(dev) << CMLBUNIT_SHIFT) | WD_NODE, cl->cl_node_type, NULL); mutex_enter(CMLB_MUTEX(cl)); if ((rval = cmlb_build_label_vtoc(cl, &user_vtoc)) == 0) { if ((rval = cmlb_write_label(cl, tg_cookie)) == 0) { if (cmlb_validate_geometry(cl, 1, 0, tg_cookie) != 0) { cmlb_dbg(CMLB_ERROR, cl, "cmlb_dkio_set_vtoc: " "Failed validate geometry\n"); } } } mutex_exit(CMLB_MUTEX(cl)); return (rval); } /* * Function: cmlb_build_label_vtoc * * Description: This routine updates the driver soft state current volume table * of contents based on a user specified vtoc. * * Arguments: cl - driver soft state (unit) structure * user_vtoc - pointer to vtoc structure specifying vtoc to be used * to update the driver soft state. * * Return Code: 0 * EINVAL */ static int cmlb_build_label_vtoc(struct cmlb_lun *cl, struct vtoc *user_vtoc) { struct dk_map *lmap; struct partition *vpart; uint_t nblks; #if defined(_SUNOS_VTOC_8) int ncyl; struct dk_map2 *lpart; #endif /* defined(_SUNOS_VTOC_8) */ int i; ASSERT(mutex_owned(CMLB_MUTEX(cl))); /* Sanity-check the vtoc */ if (user_vtoc->v_sanity != VTOC_SANE || user_vtoc->v_sectorsz != cl->cl_sys_blocksize || user_vtoc->v_nparts != V_NUMPAR) { cmlb_dbg(CMLB_INFO, cl, "cmlb_build_label_vtoc: vtoc not valid\n"); return (EINVAL); } nblks = cl->cl_g.dkg_nsect * cl->cl_g.dkg_nhead; if (nblks == 0) { cmlb_dbg(CMLB_INFO, cl, "cmlb_build_label_vtoc: geom nblks is 0\n"); return (EINVAL); } #if defined(_SUNOS_VTOC_8) vpart = user_vtoc->v_part; for (i = 0; i < V_NUMPAR; i++) { if (((unsigned)vpart->p_start % nblks) != 0) { cmlb_dbg(CMLB_INFO, cl, "cmlb_build_label_vtoc: p_start not multiply of" "nblks part %d p_start %d nblks %d\n", i, vpart->p_start, nblks); return (EINVAL); } ncyl = (unsigned)vpart->p_start / nblks; ncyl += (unsigned)vpart->p_size / nblks; if (((unsigned)vpart->p_size % nblks) != 0) { ncyl++; } if (ncyl > (int)cl->cl_g.dkg_ncyl) { cmlb_dbg(CMLB_INFO, cl, "cmlb_build_label_vtoc: ncyl %d > dkg_ncyl %d" "p_size %ld p_start %ld nblks %d part number %d" "tag %d\n", ncyl, cl->cl_g.dkg_ncyl, vpart->p_size, vpart->p_start, nblks, i, vpart->p_tag); return (EINVAL); } vpart++; } #endif /* defined(_SUNOS_VTOC_8) */ /* Put appropriate vtoc structure fields into the disk label */ #if defined(_SUNOS_VTOC_16) /* * The vtoc is always a 32bit data structure to maintain the * on-disk format. Convert "in place" instead of doing bcopy. */ vtoctovtoc32((*user_vtoc), (*((struct vtoc32 *)&(cl->cl_vtoc)))); /* * in the 16-slice vtoc, starting sectors are expressed in * numbers *relative* to the start of the Solaris fdisk partition. */ lmap = cl->cl_map; vpart = user_vtoc->v_part; for (i = 0; i < (int)user_vtoc->v_nparts; i++, lmap++, vpart++) { lmap->dkl_cylno = (unsigned)vpart->p_start / nblks; lmap->dkl_nblk = (unsigned)vpart->p_size; } #elif defined(_SUNOS_VTOC_8) cl->cl_vtoc.v_bootinfo[0] = (uint32_t)user_vtoc->v_bootinfo[0]; cl->cl_vtoc.v_bootinfo[1] = (uint32_t)user_vtoc->v_bootinfo[1]; cl->cl_vtoc.v_bootinfo[2] = (uint32_t)user_vtoc->v_bootinfo[2]; cl->cl_vtoc.v_sanity = (uint32_t)user_vtoc->v_sanity; cl->cl_vtoc.v_version = (uint32_t)user_vtoc->v_version; bcopy(user_vtoc->v_volume, cl->cl_vtoc.v_volume, LEN_DKL_VVOL); cl->cl_vtoc.v_nparts = user_vtoc->v_nparts; for (i = 0; i < 10; i++) cl->cl_vtoc.v_reserved[i] = user_vtoc->v_reserved[i]; /* * Note the conversion from starting sector number * to starting cylinder number. * Return error if division results in a remainder. */ lmap = cl->cl_map; lpart = cl->cl_vtoc.v_part; vpart = user_vtoc->v_part; for (i = 0; i < (int)user_vtoc->v_nparts; i++) { lpart->p_tag = vpart->p_tag; lpart->p_flag = vpart->p_flag; lmap->dkl_cylno = (unsigned)vpart->p_start / nblks; lmap->dkl_nblk = (unsigned)vpart->p_size; lmap++; lpart++; vpart++; /* (4387723) */ #ifdef _LP64 if (user_vtoc->timestamp[i] > TIME32_MAX) { cl->cl_vtoc.v_timestamp[i] = TIME32_MAX; } else { cl->cl_vtoc.v_timestamp[i] = user_vtoc->timestamp[i]; } #else cl->cl_vtoc.v_timestamp[i] = user_vtoc->timestamp[i]; #endif } bcopy(user_vtoc->v_asciilabel, cl->cl_asciilabel, LEN_DKL_ASCII); #else #error "No VTOC format defined." #endif return (0); } /* * Function: cmlb_clear_efi * * Description: This routine clears all EFI labels. * * Arguments: * cl driver soft state (unit) structure * * tg_cookie cookie from target driver to be passed back to target * driver when we call back to it through tg_ops. * Return Code: void */ static void cmlb_clear_efi(struct cmlb_lun *cl, void *tg_cookie) { efi_gpt_t *gpt; diskaddr_t cap; int rval; ASSERT(!mutex_owned(CMLB_MUTEX(cl))); mutex_enter(CMLB_MUTEX(cl)); cl->cl_reserved = -1; mutex_exit(CMLB_MUTEX(cl)); gpt = kmem_alloc(sizeof (efi_gpt_t), KM_SLEEP); if (DK_TG_READ(cl, gpt, 1, DEV_BSIZE, tg_cookie) != 0) { goto done; } cmlb_swap_efi_gpt(gpt); rval = cmlb_validate_efi(gpt); if (rval == 0) { /* clear primary */ bzero(gpt, sizeof (efi_gpt_t)); if (rval = DK_TG_WRITE(cl, gpt, 1, EFI_LABEL_SIZE, tg_cookie)) { cmlb_dbg(CMLB_INFO, cl, "cmlb_clear_efi: clear primary label failed\n"); } } /* the backup */ rval = DK_TG_GETCAP(cl, &cap, tg_cookie); if (rval) { goto done; } if ((rval = DK_TG_READ(cl, gpt, cap - 1, EFI_LABEL_SIZE, tg_cookie)) != 0) { goto done; } cmlb_swap_efi_gpt(gpt); rval = cmlb_validate_efi(gpt); if (rval == 0) { /* clear backup */ cmlb_dbg(CMLB_TRACE, cl, "cmlb_clear_efi clear backup@%lu\n", cap - 1); bzero(gpt, sizeof (efi_gpt_t)); if ((rval = DK_TG_WRITE(cl, gpt, cap - 1, EFI_LABEL_SIZE, tg_cookie))) { cmlb_dbg(CMLB_INFO, cl, "cmlb_clear_efi: clear backup label failed\n"); } } else { /* * Refer to comments related to off-by-1 at the * header of this file */ if ((rval = DK_TG_READ(cl, gpt, cap - 2, EFI_LABEL_SIZE, tg_cookie)) != 0) { goto done; } cmlb_swap_efi_gpt(gpt); rval = cmlb_validate_efi(gpt); if (rval == 0) { /* clear legacy backup EFI label */ cmlb_dbg(CMLB_TRACE, cl, "cmlb_clear_efi clear legacy backup@%lu\n", cap - 2); bzero(gpt, sizeof (efi_gpt_t)); if ((rval = DK_TG_WRITE(cl, gpt, cap - 2, EFI_LABEL_SIZE, tg_cookie))) { cmlb_dbg(CMLB_INFO, cl, "cmlb_clear_efi: clear legacy backup label " "failed\n"); } } } done: kmem_free(gpt, sizeof (efi_gpt_t)); } /* * Function: cmlb_set_vtoc * * Description: This routine writes data to the appropriate positions * * Arguments: * cl driver soft state (unit) structure * * dkl the data to be written * * tg_cookie cookie from target driver to be passed back to target * driver when we call back to it through tg_ops. * * Return: void */ static int cmlb_set_vtoc(struct cmlb_lun *cl, struct dk_label *dkl, void *tg_cookie) { uint_t label_addr; int sec; diskaddr_t blk; int head; int cyl; int rval; #if defined(__i386) || defined(__amd64) label_addr = cl->cl_solaris_offset + DK_LABEL_LOC; #else /* Write the primary label at block 0 of the solaris partition. */ label_addr = 0; #endif rval = DK_TG_WRITE(cl, dkl, label_addr, cl->cl_sys_blocksize, tg_cookie); if (rval != 0) { return (rval); } /* * Calculate where the backup labels go. They are always on * the last alternate cylinder, but some older drives put them * on head 2 instead of the last head. They are always on the * first 5 odd sectors of the appropriate track. * * We have no choice at this point, but to believe that the * disk label is valid. Use the geometry of the disk * as described in the label. */ cyl = dkl->dkl_ncyl + dkl->dkl_acyl - 1; head = dkl->dkl_nhead - 1; /* * Write and verify the backup labels. Make sure we don't try to * write past the last cylinder. */ for (sec = 1; ((sec < 5 * 2 + 1) && (sec < dkl->dkl_nsect)); sec += 2) { blk = (diskaddr_t)( (cyl * ((dkl->dkl_nhead * dkl->dkl_nsect) - dkl->dkl_apc)) + (head * dkl->dkl_nsect) + sec); #if defined(__i386) || defined(__amd64) blk += cl->cl_solaris_offset; #endif rval = DK_TG_WRITE(cl, dkl, blk, cl->cl_sys_blocksize, tg_cookie); cmlb_dbg(CMLB_INFO, cl, "cmlb_set_vtoc: wrote backup label %llx\n", blk); if (rval != 0) { goto exit; } } exit: return (rval); } /* * Function: cmlb_clear_vtoc * * Description: This routine clears out the VTOC labels. * * Arguments: * cl driver soft state (unit) structure * * tg_cookie cookie from target driver to be passed back to target * driver when we call back to it through tg_ops. * * Return: void */ static void cmlb_clear_vtoc(struct cmlb_lun *cl, void *tg_cookie) { struct dk_label *dkl; mutex_exit(CMLB_MUTEX(cl)); dkl = kmem_zalloc(sizeof (struct dk_label), KM_SLEEP); mutex_enter(CMLB_MUTEX(cl)); /* * cmlb_set_vtoc uses these fields in order to figure out * where to overwrite the backup labels */ dkl->dkl_apc = cl->cl_g.dkg_apc; dkl->dkl_ncyl = cl->cl_g.dkg_ncyl; dkl->dkl_acyl = cl->cl_g.dkg_acyl; dkl->dkl_nhead = cl->cl_g.dkg_nhead; dkl->dkl_nsect = cl->cl_g.dkg_nsect; mutex_exit(CMLB_MUTEX(cl)); (void) cmlb_set_vtoc(cl, dkl, tg_cookie); kmem_free(dkl, sizeof (struct dk_label)); mutex_enter(CMLB_MUTEX(cl)); } /* * Function: cmlb_write_label * * Description: This routine will validate and write the driver soft state vtoc * contents to the device. * * Arguments: * cl cmlb handle * * tg_cookie cookie from target driver to be passed back to target * driver when we call back to it through tg_ops. * * * Return Code: the code returned by cmlb_send_scsi_cmd() * 0 * EINVAL * ENXIO * ENOMEM */ static int cmlb_write_label(struct cmlb_lun *cl, void *tg_cookie) { struct dk_label *dkl; short sum; short *sp; int i; int rval; ASSERT(mutex_owned(CMLB_MUTEX(cl))); mutex_exit(CMLB_MUTEX(cl)); dkl = kmem_zalloc(sizeof (struct dk_label), KM_SLEEP); mutex_enter(CMLB_MUTEX(cl)); bcopy(&cl->cl_vtoc, &dkl->dkl_vtoc, sizeof (struct dk_vtoc)); dkl->dkl_rpm = cl->cl_g.dkg_rpm; dkl->dkl_pcyl = cl->cl_g.dkg_pcyl; dkl->dkl_apc = cl->cl_g.dkg_apc; dkl->dkl_intrlv = cl->cl_g.dkg_intrlv; dkl->dkl_ncyl = cl->cl_g.dkg_ncyl; dkl->dkl_acyl = cl->cl_g.dkg_acyl; dkl->dkl_nhead = cl->cl_g.dkg_nhead; dkl->dkl_nsect = cl->cl_g.dkg_nsect; #if defined(_SUNOS_VTOC_8) dkl->dkl_obs1 = cl->cl_g.dkg_obs1; dkl->dkl_obs2 = cl->cl_g.dkg_obs2; dkl->dkl_obs3 = cl->cl_g.dkg_obs3; for (i = 0; i < NDKMAP; i++) { dkl->dkl_map[i].dkl_cylno = cl->cl_map[i].dkl_cylno; dkl->dkl_map[i].dkl_nblk = cl->cl_map[i].dkl_nblk; } bcopy(cl->cl_asciilabel, dkl->dkl_asciilabel, LEN_DKL_ASCII); #elif defined(_SUNOS_VTOC_16) dkl->dkl_skew = cl->cl_dkg_skew; #else #error "No VTOC format defined." #endif dkl->dkl_magic = DKL_MAGIC; dkl->dkl_write_reinstruct = cl->cl_g.dkg_write_reinstruct; dkl->dkl_read_reinstruct = cl->cl_g.dkg_read_reinstruct; /* Construct checksum for the new disk label */ sum = 0; sp = (short *)dkl; i = sizeof (struct dk_label) / sizeof (short); while (i--) { sum ^= *sp++; } dkl->dkl_cksum = sum; mutex_exit(CMLB_MUTEX(cl)); rval = cmlb_set_vtoc(cl, dkl, tg_cookie); exit: kmem_free(dkl, sizeof (struct dk_label)); mutex_enter(CMLB_MUTEX(cl)); return (rval); } static int cmlb_dkio_set_efi(struct cmlb_lun *cl, dev_t dev, caddr_t arg, int flag, void *tg_cookie) { dk_efi_t user_efi; int rval = 0; void *buffer; diskaddr_t tgt_lba; boolean_t internal; if (ddi_copyin(arg, &user_efi, sizeof (dk_efi_t), flag)) return (EFAULT); internal = ((cl->cl_alter_behavior & (CMLB_INTERNAL_MINOR_NODES)) != 0); user_efi.dki_data = (void *)(uintptr_t)user_efi.dki_data_64; buffer = kmem_alloc(user_efi.dki_length, KM_SLEEP); if (ddi_copyin(user_efi.dki_data, buffer, user_efi.dki_length, flag)) { rval = EFAULT; } else { /* * let's clear the vtoc labels and clear the softstate * vtoc. */ mutex_enter(CMLB_MUTEX(cl)); if (cl->cl_vtoc.v_sanity == VTOC_SANE) { cmlb_dbg(CMLB_TRACE, cl, "cmlb_dkio_set_efi: CLEAR VTOC\n"); if (cl->cl_label_from_media == CMLB_LABEL_VTOC) cmlb_clear_vtoc(cl, tg_cookie); bzero(&cl->cl_vtoc, sizeof (struct dk_vtoc)); mutex_exit(CMLB_MUTEX(cl)); ddi_remove_minor_node(CMLB_DEVINFO(cl), "h"); ddi_remove_minor_node(CMLB_DEVINFO(cl), "h,raw"); (void) cmlb_create_minor(CMLB_DEVINFO(cl), "wd", S_IFBLK, (CMLBUNIT(dev) << CMLBUNIT_SHIFT) | WD_NODE, cl->cl_node_type, NULL, internal); (void) cmlb_create_minor(CMLB_DEVINFO(cl), "wd,raw", S_IFCHR, (CMLBUNIT(dev) << CMLBUNIT_SHIFT) | WD_NODE, cl->cl_node_type, NULL, internal); } else mutex_exit(CMLB_MUTEX(cl)); tgt_lba = user_efi.dki_lba; mutex_enter(CMLB_MUTEX(cl)); if ((cmlb_check_update_blockcount(cl, tg_cookie) != 0) || (cl->cl_tgt_blocksize == 0)) { kmem_free(buffer, user_efi.dki_length); mutex_exit(CMLB_MUTEX(cl)); return (EINVAL); } if (cl->cl_tgt_blocksize != cl->cl_sys_blocksize) tgt_lba = tgt_lba * cl->cl_tgt_blocksize / cl->cl_sys_blocksize; mutex_exit(CMLB_MUTEX(cl)); rval = DK_TG_WRITE(cl, buffer, tgt_lba, user_efi.dki_length, tg_cookie); if (rval == 0) { mutex_enter(CMLB_MUTEX(cl)); cl->cl_f_geometry_is_valid = FALSE; mutex_exit(CMLB_MUTEX(cl)); } } kmem_free(buffer, user_efi.dki_length); return (rval); } /* * Function: cmlb_dkio_get_mboot * * Description: This routine is the driver entry point for handling user * requests to get the current device mboot (DKIOCGMBOOT) * * Arguments: * arg pointer to user provided mboot structure specifying * the current mboot. * * flag this argument is a pass through to ddi_copyxxx() * directly from the mode argument of ioctl(). * * tg_cookie cookie from target driver to be passed back to target * driver when we call back to it through tg_ops. * * Return Code: 0 * EINVAL * EFAULT * ENXIO */ static int cmlb_dkio_get_mboot(struct cmlb_lun *cl, caddr_t arg, int flag, void *tg_cookie) { struct mboot *mboot; int rval; size_t buffer_size; #if defined(_SUNOS_VTOC_8) if ((!ISREMOVABLE(cl) && !ISHOTPLUGGABLE(cl)) || (arg == NULL)) { #elif defined(_SUNOS_VTOC_16) if (arg == NULL) { #endif return (EINVAL); } /* * Read the mboot block, located at absolute block 0 on the target. */ buffer_size = sizeof (struct mboot); cmlb_dbg(CMLB_TRACE, cl, "cmlb_dkio_get_mboot: allocation size: 0x%x\n", buffer_size); mboot = kmem_zalloc(buffer_size, KM_SLEEP); if ((rval = DK_TG_READ(cl, mboot, 0, buffer_size, tg_cookie)) == 0) { if (ddi_copyout(mboot, (void *)arg, sizeof (struct mboot), flag) != 0) { rval = EFAULT; } } kmem_free(mboot, buffer_size); return (rval); } /* * Function: cmlb_dkio_set_mboot * * Description: This routine is the driver entry point for handling user * requests to validate and set the device master boot * (DKIOCSMBOOT). * * Arguments: * arg pointer to user provided mboot structure used to set the * master boot. * * flag this argument is a pass through to ddi_copyxxx() * directly from the mode argument of ioctl(). * * tg_cookie cookie from target driver to be passed back to target * driver when we call back to it through tg_ops. * * Return Code: 0 * EINVAL * EFAULT * ENXIO */ static int cmlb_dkio_set_mboot(struct cmlb_lun *cl, caddr_t arg, int flag, void *tg_cookie) { struct mboot *mboot = NULL; int rval; ushort_t magic; ASSERT(!mutex_owned(CMLB_MUTEX(cl))); #if defined(_SUNOS_VTOC_8) if (!ISREMOVABLE(cl) && !ISHOTPLUGGABLE(cl)) { return (EINVAL); } #endif if (arg == NULL) { return (EINVAL); } mboot = kmem_zalloc(sizeof (struct mboot), KM_SLEEP); if (ddi_copyin((const void *)arg, mboot, sizeof (struct mboot), flag) != 0) { kmem_free(mboot, (size_t)(sizeof (struct mboot))); return (EFAULT); } /* Is this really a master boot record? */ magic = LE_16(mboot->signature); if (magic != MBB_MAGIC) { kmem_free(mboot, (size_t)(sizeof (struct mboot))); return (EINVAL); } rval = DK_TG_WRITE(cl, mboot, 0, cl->cl_sys_blocksize, tg_cookie); mutex_enter(CMLB_MUTEX(cl)); #if defined(__i386) || defined(__amd64) if (rval == 0) { /* * mboot has been written successfully. * update the fdisk and vtoc tables in memory */ rval = cmlb_update_fdisk_and_vtoc(cl, tg_cookie); if ((cl->cl_f_geometry_is_valid == FALSE) || (rval != 0)) { mutex_exit(CMLB_MUTEX(cl)); kmem_free(mboot, (size_t)(sizeof (struct mboot))); return (rval); } } #ifdef __lock_lint cmlb_setup_default_geometry(cl, tg_cookie); #endif #else if (rval == 0) { /* * mboot has been written successfully. * set up the default geometry and VTOC */ if (cl->cl_blockcount <= CMLB_EXTVTOC_LIMIT) cmlb_setup_default_geometry(cl, tg_cookie); } #endif cl->cl_msglog_flag |= CMLB_ALLOW_2TB_WARN; mutex_exit(CMLB_MUTEX(cl)); kmem_free(mboot, (size_t)(sizeof (struct mboot))); return (rval); } /* * Function: cmlb_setup_default_geometry * * Description: This local utility routine sets the default geometry as part of * setting the device mboot. * * Arguments: * cl driver soft state (unit) structure * * tg_cookie cookie from target driver to be passed back to target * driver when we call back to it through tg_ops. * * * Note: This may be redundant with cmlb_build_default_label. */ static void cmlb_setup_default_geometry(struct cmlb_lun *cl, void *tg_cookie) { struct cmlb_geom pgeom; struct cmlb_geom *pgeomp = &pgeom; int ret; int geom_base_cap = 1; ASSERT(mutex_owned(CMLB_MUTEX(cl))); /* zero out the soft state geometry and partition table. */ bzero(&cl->cl_g, sizeof (struct dk_geom)); bzero(&cl->cl_vtoc, sizeof (struct dk_vtoc)); bzero(cl->cl_map, NDKMAP * (sizeof (struct dk_map))); /* * For the rpm, we use the minimum for the disk. * For the head, cyl and number of sector per track, * if the capacity <= 1GB, head = 64, sect = 32. * else head = 255, sect 63 * Note: the capacity should be equal to C*H*S values. * This will cause some truncation of size due to * round off errors. For CD-ROMs, this truncation can * have adverse side effects, so returning ncyl and * nhead as 1. The nsect will overflow for most of * CD-ROMs as nsect is of type ushort. */ if (cl->cl_alter_behavior & CMLB_FAKE_GEOM_LABEL_IOCTLS_VTOC8) { /* * newfs currently can not handle 255 ntracks for SPARC * so get the geometry from target driver instead of coming up * with one based on capacity. */ mutex_exit(CMLB_MUTEX(cl)); ret = DK_TG_GETPHYGEOM(cl, pgeomp, tg_cookie); mutex_enter(CMLB_MUTEX(cl)); if (ret == 0) { geom_base_cap = 0; } else { cmlb_dbg(CMLB_ERROR, cl, "cmlb_setup_default_geometry: " "tg_getphygeom failed %d\n", ret); /* do default setting, geometry based on capacity */ } } if (geom_base_cap) { if (ISCD(cl)) { cl->cl_g.dkg_ncyl = 1; cl->cl_g.dkg_nhead = 1; cl->cl_g.dkg_nsect = cl->cl_blockcount; } else if (cl->cl_blockcount <= 0x1000) { /* Needed for unlabeled SCSI floppies. */ cl->cl_g.dkg_nhead = 2; cl->cl_g.dkg_ncyl = 80; cl->cl_g.dkg_pcyl = 80; cl->cl_g.dkg_nsect = cl->cl_blockcount / (2 * 80); } else if (cl->cl_blockcount <= 0x200000) { cl->cl_g.dkg_nhead = 64; cl->cl_g.dkg_nsect = 32; cl->cl_g.dkg_ncyl = cl->cl_blockcount / (64 * 32); } else { cl->cl_g.dkg_nhead = 255; cl->cl_g.dkg_nsect = ((cl->cl_blockcount + (UINT16_MAX * 255 * 63) - 1) / (UINT16_MAX * 255 * 63)) * 63; if (cl->cl_g.dkg_nsect == 0) cl->cl_g.dkg_nsect = (UINT16_MAX / 63) * 63; cl->cl_g.dkg_ncyl = cl->cl_blockcount / (255 * cl->cl_g.dkg_nsect); } cl->cl_g.dkg_acyl = 0; cl->cl_g.dkg_bcyl = 0; cl->cl_g.dkg_intrlv = 1; cl->cl_g.dkg_rpm = 200; if (cl->cl_g.dkg_pcyl == 0) cl->cl_g.dkg_pcyl = cl->cl_g.dkg_ncyl + cl->cl_g.dkg_acyl; } else { cl->cl_g.dkg_ncyl = (short)pgeomp->g_ncyl; cl->cl_g.dkg_acyl = pgeomp->g_acyl; cl->cl_g.dkg_nhead = pgeomp->g_nhead; cl->cl_g.dkg_nsect = pgeomp->g_nsect; cl->cl_g.dkg_intrlv = pgeomp->g_intrlv; cl->cl_g.dkg_rpm = pgeomp->g_rpm; cl->cl_g.dkg_pcyl = cl->cl_g.dkg_ncyl + cl->cl_g.dkg_acyl; } cl->cl_g.dkg_read_reinstruct = 0; cl->cl_g.dkg_write_reinstruct = 0; cl->cl_solaris_size = cl->cl_g.dkg_ncyl * cl->cl_g.dkg_nhead * cl->cl_g.dkg_nsect; cl->cl_map['a'-'a'].dkl_cylno = 0; cl->cl_map['a'-'a'].dkl_nblk = cl->cl_solaris_size; cl->cl_map['c'-'a'].dkl_cylno = 0; cl->cl_map['c'-'a'].dkl_nblk = cl->cl_solaris_size; cl->cl_vtoc.v_part[2].p_tag = V_BACKUP; cl->cl_vtoc.v_part[2].p_flag = V_UNMNT; cl->cl_vtoc.v_nparts = V_NUMPAR; cl->cl_vtoc.v_version = V_VERSION; (void) sprintf((char *)cl->cl_asciilabel, "DEFAULT cyl %d alt %d" " hd %d sec %d", cl->cl_g.dkg_ncyl, cl->cl_g.dkg_acyl, cl->cl_g.dkg_nhead, cl->cl_g.dkg_nsect); cl->cl_f_geometry_is_valid = FALSE; } #if defined(__i386) || defined(__amd64) /* * Function: cmlb_update_fdisk_and_vtoc * * Description: This local utility routine updates the device fdisk and vtoc * as part of setting the device mboot. * * Arguments: * cl driver soft state (unit) structure * * tg_cookie cookie from target driver to be passed back to target * driver when we call back to it through tg_ops. * * * Return Code: 0 for success or errno-type return code. * * Note:x86: This looks like a duplicate of cmlb_validate_geometry(), but * these did exist separately in x86 sd.c. */ static int cmlb_update_fdisk_and_vtoc(struct cmlb_lun *cl, void *tg_cookie) { int count; int label_rc = 0; int fdisk_rval; diskaddr_t capacity; ASSERT(mutex_owned(CMLB_MUTEX(cl))); if (cmlb_check_update_blockcount(cl, tg_cookie) != 0) return (EINVAL); #if defined(_SUNOS_VTOC_16) /* * Set up the "whole disk" fdisk partition; this should always * exist, regardless of whether the disk contains an fdisk table * or vtoc. */ cl->cl_map[P0_RAW_DISK].dkl_cylno = 0; cl->cl_map[P0_RAW_DISK].dkl_nblk = cl->cl_blockcount; #endif /* defined(_SUNOS_VTOC_16) */ /* * copy the lbasize and capacity so that if they're * reset while we're not holding the CMLB_MUTEX(cl), we will * continue to use valid values after the CMLB_MUTEX(cl) is * reacquired. */ capacity = cl->cl_blockcount; /* * refresh the logical and physical geometry caches. * (data from mode sense format/rigid disk geometry pages, * and scsi_ifgetcap("geometry"). */ cmlb_resync_geom_caches(cl, capacity, tg_cookie); /* * Only DIRECT ACCESS devices will have Scl labels. * CD's supposedly have a Scl label, too */ if (cl->cl_device_type == DTYPE_DIRECT || ISREMOVABLE(cl)) { fdisk_rval = cmlb_read_fdisk(cl, capacity, tg_cookie); if (fdisk_rval != 0) { ASSERT(mutex_owned(CMLB_MUTEX(cl))); return (fdisk_rval); } if (cl->cl_solaris_size <= DK_LABEL_LOC) { /* * Found fdisk table but no Solaris partition entry, * so don't call cmlb_uselabel() and don't create * a default label. */ label_rc = 0; cl->cl_f_geometry_is_valid = TRUE; goto no_solaris_partition; } } else if (capacity < 0) { ASSERT(mutex_owned(CMLB_MUTEX(cl))); return (EINVAL); } /* * For Removable media We reach here if we have found a * SOLARIS PARTITION. * If cl_f_geometry_is_valid is FALSE it indicates that the SOLARIS * PARTITION has changed from the previous one, hence we will setup a * default VTOC in this case. */ if (cl->cl_f_geometry_is_valid == FALSE) { /* if we get here it is writable */ /* we are called from SMBOOT, and after a write of fdisk */ cmlb_build_default_label(cl, tg_cookie); label_rc = 0; } no_solaris_partition: #if defined(_SUNOS_VTOC_16) /* * If we have valid geometry, set up the remaining fdisk partitions. * Note that dkl_cylno is not used for the fdisk map entries, so * we set it to an entirely bogus value. */ for (count = 0; count < FD_NUMPART; count++) { cl->cl_map[FDISK_P1 + count].dkl_cylno = UINT32_MAX; cl->cl_map[FDISK_P1 + count].dkl_nblk = cl->cl_fmap[count].fmap_nblk; cl->cl_offset[FDISK_P1 + count] = cl->cl_fmap[count].fmap_start; } #endif for (count = 0; count < NDKMAP; count++) { #if defined(_SUNOS_VTOC_8) struct dk_map *lp = &cl->cl_map[count]; cl->cl_offset[count] = cl->cl_g.dkg_nhead * cl->cl_g.dkg_nsect * lp->dkl_cylno; #elif defined(_SUNOS_VTOC_16) struct dkl_partition *vp = &cl->cl_vtoc.v_part[count]; cl->cl_offset[count] = vp->p_start + cl->cl_solaris_offset; #else #error "No VTOC format defined." #endif } ASSERT(mutex_owned(CMLB_MUTEX(cl))); return (label_rc); } #endif #if defined(__i386) || defined(__amd64) static int cmlb_dkio_get_virtgeom(struct cmlb_lun *cl, caddr_t arg, int flag) { int err = 0; /* Return the driver's notion of the media's logical geometry */ struct dk_geom disk_geom; struct dk_geom *dkgp = &disk_geom; mutex_enter(CMLB_MUTEX(cl)); /* * If there is no HBA geometry available, or * if the HBA returned us something that doesn't * really fit into an Int 13/function 8 geometry * result, just fail the ioctl. See PSARC 1998/313. */ if (cl->cl_lgeom.g_nhead == 0 || cl->cl_lgeom.g_nsect == 0 || cl->cl_lgeom.g_ncyl > 1024) { mutex_exit(CMLB_MUTEX(cl)); err = EINVAL; } else { dkgp->dkg_ncyl = cl->cl_lgeom.g_ncyl; dkgp->dkg_acyl = cl->cl_lgeom.g_acyl; dkgp->dkg_pcyl = dkgp->dkg_ncyl + dkgp->dkg_acyl; dkgp->dkg_nhead = cl->cl_lgeom.g_nhead; dkgp->dkg_nsect = cl->cl_lgeom.g_nsect; mutex_exit(CMLB_MUTEX(cl)); if (ddi_copyout(dkgp, (void *)arg, sizeof (struct dk_geom), flag)) { err = EFAULT; } else { err = 0; } } return (err); } #endif #if defined(__i386) || defined(__amd64) static int cmlb_dkio_get_phygeom(struct cmlb_lun *cl, caddr_t arg, int flag) { int err = 0; diskaddr_t capacity; /* Return the driver's notion of the media physical geometry */ struct dk_geom disk_geom; struct dk_geom *dkgp = &disk_geom; mutex_enter(CMLB_MUTEX(cl)); if (cl->cl_g.dkg_nhead != 0 && cl->cl_g.dkg_nsect != 0) { /* * We succeeded in getting a geometry, but * right now it is being reported as just the * Solaris fdisk partition, just like for * DKIOCGGEOM. We need to change that to be * correct for the entire disk now. */ bcopy(&cl->cl_g, dkgp, sizeof (*dkgp)); dkgp->dkg_acyl = 0; dkgp->dkg_ncyl = cl->cl_blockcount / (dkgp->dkg_nhead * dkgp->dkg_nsect); } else { bzero(dkgp, sizeof (struct dk_geom)); /* * This disk does not have a Solaris VTOC * so we must present a physical geometry * that will remain consistent regardless * of how the disk is used. This will ensure * that the geometry does not change regardless * of the fdisk partition type (ie. EFI, FAT32, * Solaris, etc). */ if (ISCD(cl)) { dkgp->dkg_nhead = cl->cl_pgeom.g_nhead; dkgp->dkg_nsect = cl->cl_pgeom.g_nsect; dkgp->dkg_ncyl = cl->cl_pgeom.g_ncyl; dkgp->dkg_acyl = cl->cl_pgeom.g_acyl; } else { /* * Invalid cl_blockcount can generate invalid * dk_geom and may result in division by zero * system failure. Should make sure blockcount * is valid before using it here. */ if (cl->cl_blockcount == 0) { mutex_exit(CMLB_MUTEX(cl)); err = EIO; return (err); } /* * Refer to comments related to off-by-1 at the * header of this file */ if (cl->cl_alter_behavior & CMLB_OFF_BY_ONE) capacity = cl->cl_blockcount - 1; else capacity = cl->cl_blockcount; cmlb_convert_geometry(capacity, dkgp); dkgp->dkg_acyl = 0; dkgp->dkg_ncyl = capacity / (dkgp->dkg_nhead * dkgp->dkg_nsect); } } dkgp->dkg_pcyl = dkgp->dkg_ncyl + dkgp->dkg_acyl; mutex_exit(CMLB_MUTEX(cl)); if (ddi_copyout(dkgp, (void *)arg, sizeof (struct dk_geom), flag)) err = EFAULT; return (err); } #endif #if defined(__i386) || defined(__amd64) static int cmlb_dkio_partinfo(struct cmlb_lun *cl, dev_t dev, caddr_t arg, int flag) { int err = 0; /* * Return parameters describing the selected disk slice. * Note: this ioctl is for the intel platform only */ int part; part = CMLBPART(dev); mutex_enter(CMLB_MUTEX(cl)); /* don't check cl_solaris_size for pN */ if (part < P0_RAW_DISK && cl->cl_solaris_size == 0) { err = EIO; mutex_exit(CMLB_MUTEX(cl)); } else { struct part_info p; p.p_start = (daddr_t)cl->cl_offset[part]; p.p_length = (int)cl->cl_map[part].dkl_nblk; mutex_exit(CMLB_MUTEX(cl)); #ifdef _MULTI_DATAMODEL switch (ddi_model_convert_from(flag & FMODELS)) { case DDI_MODEL_ILP32: { struct part_info32 p32; p32.p_start = (daddr32_t)p.p_start; p32.p_length = p.p_length; if (ddi_copyout(&p32, (void *)arg, sizeof (p32), flag)) err = EFAULT; break; } case DDI_MODEL_NONE: { if (ddi_copyout(&p, (void *)arg, sizeof (p), flag)) err = EFAULT; break; } } #else /* ! _MULTI_DATAMODEL */ if (ddi_copyout(&p, (void *)arg, sizeof (p), flag)) err = EFAULT; #endif /* _MULTI_DATAMODEL */ } return (err); } static int cmlb_dkio_extpartinfo(struct cmlb_lun *cl, dev_t dev, caddr_t arg, int flag) { int err = 0; /* * Return parameters describing the selected disk slice. * Note: this ioctl is for the intel platform only */ int part; part = CMLBPART(dev); mutex_enter(CMLB_MUTEX(cl)); /* don't check cl_solaris_size for pN */ if (part < P0_RAW_DISK && cl->cl_solaris_size == 0) { err = EIO; mutex_exit(CMLB_MUTEX(cl)); } else { struct extpart_info p; p.p_start = (diskaddr_t)cl->cl_offset[part]; p.p_length = (diskaddr_t)cl->cl_map[part].dkl_nblk; mutex_exit(CMLB_MUTEX(cl)); if (ddi_copyout(&p, (void *)arg, sizeof (p), flag)) err = EFAULT; } return (err); } #endif int cmlb_prop_op(cmlb_handle_t cmlbhandle, dev_t dev, dev_info_t *dip, ddi_prop_op_t prop_op, int mod_flags, char *name, caddr_t valuep, int *lengthp, int part, void *tg_cookie) { struct cmlb_lun *cl; diskaddr_t capacity; uint32_t lbasize; enum dp { DP_NBLOCKS, DP_BLKSIZE } dp; int callers_length; caddr_t buffer; uint64_t nblocks64; uint_t dblk; /* Always fallback to ddi_prop_op... */ cl = (struct cmlb_lun *)cmlbhandle; if (cl == NULL) { fallback: return (ddi_prop_op(dev, dip, prop_op, mod_flags, name, valuep, lengthp)); } /* Pick up capacity and blocksize information. */ capacity = cl->cl_blockcount; if (capacity == 0) goto fallback; lbasize = cl->cl_tgt_blocksize; if (lbasize == 0) lbasize = DEV_BSIZE; /* 0 -> DEV_BSIZE units */ /* Check for dynamic property of whole device. */ if (dev == DDI_DEV_T_ANY) { /* Fallback to ddi_prop_op if we don't understand. */ if (strcmp(name, "device-nblocks") == 0) dp = DP_NBLOCKS; else if (strcmp(name, "device-blksize") == 0) dp = DP_BLKSIZE; else goto fallback; /* get callers length, establish length of our dynamic prop */ callers_length = *lengthp; if (dp == DP_NBLOCKS) *lengthp = sizeof (uint64_t); else if (dp == DP_BLKSIZE) *lengthp = sizeof (uint32_t); /* service request for the length of the property */ if (prop_op == PROP_LEN) return (DDI_PROP_SUCCESS); switch (prop_op) { case PROP_LEN_AND_VAL_ALLOC: if ((buffer = kmem_alloc(*lengthp, (mod_flags & DDI_PROP_CANSLEEP) ? KM_SLEEP : KM_NOSLEEP)) == NULL) return (DDI_PROP_NO_MEMORY); *(caddr_t *)valuep = buffer; /* set callers buf */ break; case PROP_LEN_AND_VAL_BUF: /* the length of the prop and the request must match */ if (callers_length != *lengthp) return (DDI_PROP_INVAL_ARG); buffer = valuep; /* get callers buf */ break; default: return (DDI_PROP_INVAL_ARG); } /* transfer the value into the buffer */ if (dp == DP_NBLOCKS) *((uint64_t *)buffer) = capacity; else if (dp == DP_BLKSIZE) *((uint32_t *)buffer) = lbasize; return (DDI_PROP_SUCCESS); } /* * Support dynamic size oriented properties of partition. Requests * issued under conditions where size is valid are passed to * ddi_prop_op_nblocks with the size information, otherwise the * request is passed to ddi_prop_op. Size depends on valid geometry. */ if (!cmlb_is_valid(cmlbhandle)) goto fallback; /* Get partition nblocks value. */ (void) cmlb_partinfo(cmlbhandle, part, (diskaddr_t *)&nblocks64, NULL, NULL, NULL, tg_cookie); /* * Assume partition information is in DEV_BSIZE units, compute * divisor for size(9P) property representation. */ dblk = lbasize / DEV_BSIZE; /* Now let ddi_prop_op_nblocks_blksize() handle the request. */ return (ddi_prop_op_nblocks_blksize(dev, dip, prop_op, mod_flags, name, valuep, lengthp, nblocks64 / dblk, lbasize)); }