LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
* [PATCH] Add iSCSI iBFT support (v0.4.7)
@ 2008-02-08 19:47 Konrad Rzeszutek
  2008-02-09  3:10 ` REPOST: " Konrad Rzeszutek
  0 siblings, 1 reply; 8+ messages in thread
From: Konrad Rzeszutek @ 2008-02-08 19:47 UTC (permalink / raw)
  To: Andrew Morton, linux-kernel, Greg KH, michaelc, James Bottomley
  Cc: dwm, darnok, pjones, konradr, konradr, randy.dunlap, hpa, lenb,
	mike.anderson, dwm, arjan, Andy Whitcroft

This patch (v0.4.7) adds /sysfs/firmware/ibft/[initiator|targetX|ethernetX]
directories along with text properties which export the the iSCSI
Boot Firmware Table (iBFT) structure.

What is iSCSI Boot Firmware Table? It is a mechanism for the iSCSI
tools to extract from the machine NICs the iSCSI connection information
so that they can automagically mount the iSCSI share/target. Currently
the iSCSI information is hard-coded in the initrd. The /sysfs entries
are read-only one-name-and-value fields.

The usual set of data exposed is:

# for a in `find /sys/firmware/ibft/ -type f -print`; do  echo -n "$a: ";  cat $a; done
/sys/firmware/ibft/target0/target-name: iqn.2007.com.intel-sbx44:storage-10gb
/sys/firmware/ibft/target0/nic-assoc: 0
/sys/firmware/ibft/target0/chap-type: 0
/sys/firmware/ibft/target0/lun: 00000000
/sys/firmware/ibft/target0/port: 3260
/sys/firmware/ibft/target0/ip-addr: 192.168.79.116
/sys/firmware/ibft/target0/flags: 3
/sys/firmware/ibft/target0/index: 0
/sys/firmware/ibft/ethernet0/mac: 00:11:25:9d:8b:01
/sys/firmware/ibft/ethernet0/vlan: 0
/sys/firmware/ibft/ethernet0/gateway: 192.168.79.254
/sys/firmware/ibft/ethernet0/origin: 0
/sys/firmware/ibft/ethernet0/subnet-mask: 255.255.252.0
/sys/firmware/ibft/ethernet0/ip-addr: 192.168.77.41
/sys/firmware/ibft/ethernet0/flags: 7
/sys/firmware/ibft/ethernet0/index: 0
/sys/firmware/ibft/initiator/initiator-name: iqn.2007-07.com:konrad.initiator
/sys/firmware/ibft/initiator/flags: 3
/sys/firmware/ibft/initiator/index: 0

For full details of the IBFT structure please take a look at:
ftp://ftp.software.ibm.com/systems/support/system_x_pdf/ibm_iscsi_boot_firmware_table_v1.02.pdf

Signed-off-by: Konrad Rzeszutek <konradr@linux.vnet.ibm.com>
Cc: Mike Christie <michaelc@cs.wisc.edu>
Cc: Peter Jones <pjones@redhat.com>
Cc: James Bottomley <James.Bottomley@hansenpartnership.com>
Cc: Greg KH <greg@kroah.com>

 Documentation/ABI/testing/sysfs-ibft |   23 
 arch/x86/kernel/setup_32.c           |    3 
 arch/x86/kernel/setup_64.c           |    4 
 drivers/firmware/Kconfig             |   20 
 drivers/firmware/Makefile            |    2 
 drivers/firmware/iscsi_ibft.c        |  985 +++++++++++++++++++++++++++++++++++
 drivers/firmware/iscsi_ibft_find.c   |   83 ++
 include/linux/iscsi_ibft.h           |   38 +
 8 files changed, 1158 insertions(+)


diff --git a/Documentation/ABI/testing/sysfs-ibft b/Documentation/ABI/testing/sysfs-ibft
new file mode 100644
index 0000000..062c009
--- /dev/null
+++ b/Documentation/ABI/testing/sysfs-ibft
@@ -0,0 +1,23 @@
+What:		/sys/firmware/ibft/initiator
+Date:		November 2007
+Contact:	Konrad Rzeszutek <konradr@linux.vnet.ibm.com>
+Description:	The /sys/firmware/ibft/initiator directory will contain
+		files that expose the iSCSI Boot Firmware Table initiator data.
+		Usually this contains the Initiator name.
+
+What:		/sys/firmware/ibft/targetX
+Date:		November 2007
+Contact:	Konrad Rzeszutek <konradr@linux.vnet.ibm.com>
+Description:	The /sys/firmware/ibft/targetX directory will contain
+		files that expose the iSCSI Boot Firmware Table target data.
+		Usually this contains the target's IP address, boot LUN,
+		target name, and what NIC it is associated with. It can also
+		contain the CHAP name (and password), the reverse CHAP
+		name (and password)
+
+What:		/sys/firmware/ibft/ethernetX
+Date:		November 2007
+Contact:	Konrad Rzeszutek <konradr@linux.vnet.ibm.com>
+Description:	The /sys/firmware/ibft/ethernetX directory will contain
+		files that expose the iSCSI Boot Firmware Table NIC data.
+		This can this can the IP address, MAC, and gateway of the NIC.
diff --git a/arch/x86/kernel/setup_32.c b/arch/x86/kernel/setup_32.c
index 62adc5f..83a05cf 100644
--- a/arch/x86/kernel/setup_32.c
+++ b/arch/x86/kernel/setup_32.c
@@ -39,6 +39,7 @@
 #include <linux/efi.h>
 #include <linux/init.h>
 #include <linux/edd.h>
+#include <linux/iscsi_ibft.h>
 #include <linux/nodemask.h>
 #include <linux/kexec.h>
 #include <linux/crash_dump.h>
@@ -649,6 +650,8 @@ void __init setup_bootmem_allocator(void)
 #endif
 	numa_kva_reserve();
 	reserve_crashkernel();
+
+	reserve_ibft_region();
 }
 
 /*
diff --git a/arch/x86/kernel/setup_64.c b/arch/x86/kernel/setup_64.c
index c8939df..9c72fe1 100644
--- a/arch/x86/kernel/setup_64.c
+++ b/arch/x86/kernel/setup_64.c
@@ -34,6 +34,7 @@
 #include <linux/acpi.h>
 #include <linux/kallsyms.h>
 #include <linux/edd.h>
+#include <linux/iscsi_ibft.h>
 #include <linux/mmzone.h>
 #include <linux/kexec.h>
 #include <linux/cpufreq.h>
@@ -419,6 +420,9 @@ void __init setup_arch(char **cmdline_p)
 	}
 #endif
 	reserve_crashkernel();
+
+	reserve_ibft_region();
+
 	paging_init();
 	map_vsyscall();
 
diff --git a/drivers/firmware/Kconfig b/drivers/firmware/Kconfig
index 05f02a3..81a3e59 100644
--- a/drivers/firmware/Kconfig
+++ b/drivers/firmware/Kconfig
@@ -93,4 +93,24 @@ config DMIID
 	  information from userspace through /sys/class/dmi/id/ or if you want
 	  DMI-based module auto-loading.
 
+config ISCSI_IBFT_FIND
+	bool "iSCSI Boot Firmware Table Attributes"
+	depends on X86 || IA64
+	default n
+	help
+	  This option enables the kernel to find the region of memory
+	  in which the ISCSI Boot Firmware Table (iBFT) resides. This
+	  is necessary for iSCSI Boot Firmware Table Attributes module to work
+	  properly.
+
+config ISCSI_IBFT
+	tristate "iSCSI Boot Firmware Table Attributes module"
+	depends on ISCSI_IBFT_FIND
+	default	n
+	help
+	  This option enables support for detection and exposing of iSCSI
+	  Boot Firmware Table (iBFT) via sysfs to userspace. If you wish to
+	  detect iSCSI boot parameters dynamically during system boot, say Y.
+	  Otherwise, say N.
+
 endmenu
diff --git a/drivers/firmware/Makefile b/drivers/firmware/Makefile
index 8d4ebc8..4c91471 100644
--- a/drivers/firmware/Makefile
+++ b/drivers/firmware/Makefile
@@ -8,3 +8,5 @@ obj-$(CONFIG_EFI_PCDP)		+= pcdp.o
 obj-$(CONFIG_DELL_RBU)          += dell_rbu.o
 obj-$(CONFIG_DCDBAS)		+= dcdbas.o
 obj-$(CONFIG_DMIID)		+= dmi-id.o
+obj-$(CONFIG_ISCSI_IBFT_FIND)	+= iscsi_ibft_find.o
+obj-$(CONFIG_ISCSI_IBFT)	+= iscsi_ibft.o
diff --git a/drivers/firmware/iscsi_ibft.c b/drivers/firmware/iscsi_ibft.c
new file mode 100644
index 0000000..c2e5738
--- /dev/null
+++ b/drivers/firmware/iscsi_ibft.c
@@ -0,0 +1,985 @@
+/*
+ *  Copyright 2007 Red Hat, Inc.
+ *  by Peter Jones <pjones@redhat.com>
+ *  Copyright 2008 IBM, Inc.
+ *  by Konrad Rzeszutek <konradr@linux.vnet.ibm.com>
+ *
+ * This code exposes the iSCSI Boot Format Table to userland via sysfs.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License v2.0 as published by
+ * the Free Software Foundation
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * Changelog:
+ *
+ *   8 Feb 2008 - Konrad Rzeszutek <konradr@linux.vnet.ibm.com>
+ *    Combined two functions in one: reserve_ibft_region. (v0.4.7)
+ *
+ *  30 Jan 2008 - Konrad Rzeszutek <konradr@linux.vnet.ibm.com>
+ *   Added logic to handle IPv6 addresses. (v0.4.6)
+ *
+ *  25 Jan 2008 - Konrad Rzeszutek <konradr@linux.vnet.ibm.com>
+ *   Added logic to handle badly not-to-spec iBFT. (v0.4.5)
+ *
+ *   4 Jan 2008 - Konrad Rzeszutek <konradr@linux.vnet.ibm.com>
+ *   Added __init to function declarations. (v0.4.4)
+ *
+ *  21 Dec 2007 - Konrad Rzeszutek <konradr@linux.vnet.ibm.com>
+ *   Updated kobject registration, combined unregister functions in one
+ *   and code and style cleanup. (v0.4.3)
+ *
+ *   5 Dec 2007 - Konrad Rzeszutek <konradr@linux.vnet.ibm.com>
+ *   Added end-markers to enums and re-organized kobject registration. (v0.4.2)
+ *
+ *   4 Dec 2007 - Konrad Rzeszutek <konradr@linux.vnet.ibm.com>
+ *   Created 'device' sysfs link to the NIC and style cleanup. (v0.4.1)
+ *
+ *  28 Nov 2007 - Konrad Rzeszutek <konradr@linux.vnet.ibm.com>
+ *   Added sysfs-ibft documentation, moved 'find_ibft' function to
+ *   in its own file and added text attributes for every struct field.  (v0.4)
+ *
+ *  21 Nov 2007 - Konrad Rzeszutek <konradr@linux.vnet.ibm.com>
+ *   Added text attributes emulating OpenFirmware /proc/device-tree naming.
+ *   Removed binary /sysfs interface (v0.3)
+ *
+ *  29 Aug 2007 - Konrad Rzeszutek <konradr@linux.vnet.ibm.com>
+ *   Added functionality in setup.c to reserve iBFT region. (v0.2)
+ *
+ *  27 Aug 2007 - Konrad Rzeszutek <konradr@linux.vnet.ibm.com>
+ *   First version exposing iBFT data via a binary /sysfs. (v0.1)
+ *
+ */
+
+
+#include <linux/blkdev.h>
+#include <linux/ctype.h>
+#include <linux/device.h>
+#include <linux/err.h>
+#include <linux/init.h>
+#include <linux/iscsi_ibft.h>
+#include <linux/limits.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/slab.h>
+#include <linux/stat.h>
+#include <linux/string.h>
+#include <linux/types.h>
+
+#define IBFT_ISCSI_VERSION "0.4.7"
+#define IBFT_ISCSI_DATE "2008-Feb-08"
+
+MODULE_AUTHOR("Peter Jones <pjones@redhat.com> and \
+Konrad Rzeszutek <konradr@us.ibm.com>");
+MODULE_DESCRIPTION("sysfs interface to BIOS iBFT information");
+MODULE_LICENSE("GPL");
+MODULE_VERSION(IBFT_ISCSI_VERSION);
+
+
+struct ibft_table_header {
+	char signature[4];
+	u32 length;
+	u8 revision;
+	u8 checksum;
+	char oem_id[6];
+	char oem_table_id[8];
+	char reserved[24];
+} __attribute__((__packed__));
+
+struct ibft_hdr {
+	u8 id;
+	u8 version;
+	u16 length;
+	u8 index;
+	u8 flags;
+} __attribute__((__packed__));
+
+struct ibft_control {
+	struct ibft_hdr hdr;
+	u16 extensions;
+	u16 initiator_off;
+	u16 nic0_off;
+	u16 tgt0_off;
+	u16 nic1_off;
+	u16 tgt1_off;
+} __attribute__((__packed__));
+
+struct ibft_initiator {
+	struct ibft_hdr hdr;
+	char isns_server[16];
+	char slp_server[16];
+	char pri_radius_server[16];
+	char sec_radius_server[16];
+	u16 initiator_name_len;
+	u16 initiator_name_off;
+} __attribute__((__packed__));
+
+struct ibft_nic {
+	struct ibft_hdr hdr;
+	char ip_addr[16];
+	u8 subnet_mask_prefix;
+	u8 origin;
+	char gateway[16];
+	char primary_dns[16];
+	char secondary_dns[16];
+	char dhcp[16];
+	u16 vlan;
+	char mac[6];
+	u16 pci_bdf;
+	u16 hostname_len;
+	u16 hostname_off;
+} __attribute__((__packed__));
+
+struct ibft_tgt {
+	struct ibft_hdr hdr;
+	char ip_addr[16];
+	u16 port;
+	char lun[8];
+	u8 chap_type;
+	u8 nic_assoc;
+	u16 tgt_name_len;
+	u16 tgt_name_off;
+	u16 chap_name_len;
+	u16 chap_name_off;
+	u16 chap_secret_len;
+	u16 chap_secret_off;
+	u16 rev_chap_name_len;
+	u16 rev_chap_name_off;
+	u16 rev_chap_secret_len;
+	u16 rev_chap_secret_off;
+} __attribute__((__packed__));
+
+/*
+ * The kobject different types and its names.
+ *
+*/
+enum ibft_id {
+	id_reserved = 0, /* We don't support. */
+	id_control = 1, /* Should show up only once and is not exported. */
+	id_initiator = 2,
+	id_nic = 3,
+	id_target = 4,
+	id_extensions = 5, /* We don't support. */
+	id_end_marker,
+};
+
+/*
+ * We do not support the other types, hence the NULL.
+ */
+static const char *ibft_id_names[] =
+	{NULL, NULL, "initiator", "ethernet%d", "target%d", NULL, NULL};
+
+/*
+ * The text attributes names for each of the kobjects.
+*/
+enum ibft_eth_properties_enum {
+	ibft_eth_index,
+	ibft_eth_flags,
+	ibft_eth_ip_addr,
+	ibft_eth_subnet_mask,
+	ibft_eth_origin,
+	ibft_eth_gateway,
+	ibft_eth_primary_dns,
+	ibft_eth_secondary_dns,
+	ibft_eth_dhcp,
+	ibft_eth_vlan,
+	ibft_eth_mac,
+	/* ibft_eth_pci_bdf - this is replaced by link to the device itself. */
+	ibft_eth_hostname,
+	ibft_eth_end_marker,
+};
+
+static const char *ibft_eth_properties[] =
+	{"index", "flags", "ip-addr", "subnet-mask", "origin", "gateway",
+	"primary-dns", "secondary-dns", "dhcp", "vlan", "mac", "hostname",
+	NULL};
+
+enum ibft_tgt_properties_enum {
+	ibft_tgt_index,
+	ibft_tgt_flags,
+	ibft_tgt_ip_addr,
+	ibft_tgt_port,
+	ibft_tgt_lun,
+	ibft_tgt_chap_type,
+	ibft_tgt_nic_assoc,
+	ibft_tgt_name,
+	ibft_tgt_chap_name,
+	ibft_tgt_chap_secret,
+	ibft_tgt_rev_chap_name,
+	ibft_tgt_rev_chap_secret,
+	ibft_tgt_end_marker,
+};
+
+static const char *ibft_tgt_properties[] =
+	{"index", "flags", "ip-addr", "port", "lun", "chap-type", "nic-assoc",
+	"target-name", "chap-name", "chap-secret", "rev-chap-name",
+	"rev-chap-name-secret", NULL};
+
+enum ibft_initiator_properties_enum {
+	ibft_init_index,
+	ibft_init_flags,
+	ibft_init_isns_server,
+	ibft_init_slp_server,
+	ibft_init_pri_radius_server,
+	ibft_init_sec_radius_server,
+	ibft_init_initiator_name,
+	ibft_init_end_marker,
+};
+
+static const char *ibft_initiator_properties[] =
+	{"index", "flags", "isns-server", "slp-server", "pri-radius-server",
+	"sec-radius-server", "initiator-name", NULL};
+
+/*
+ * The kobject and attribute structures.
+ */
+
+struct ibft_kobject {
+	struct ibft_table_header *header;
+	union {
+		struct ibft_initiator *initiator;
+		struct ibft_nic *nic;
+		struct ibft_tgt *tgt;
+		struct ibft_hdr *hdr;
+	};
+	struct kobject kobj;
+	struct list_head node;
+};
+
+struct ibft_attribute {
+	struct attribute attr;
+	ssize_t (*show) (struct  ibft_kobject *entry,
+			 struct ibft_attribute *attr, char *buf);
+	union {
+		struct ibft_initiator *initiator;
+		struct ibft_nic *nic;
+		struct ibft_tgt *tgt;
+		struct ibft_hdr *hdr;
+	};
+	struct kobject *kobj;
+	int type; /* The enum of the type. This can be any value off:
+		ibft_eth_properties_enum, ibft_tgt_properties_enum,
+		or ibft_initiator_properties_enum. */
+	struct list_head node;
+};
+
+static LIST_HEAD(ibft_attr_list);
+static LIST_HEAD(ibft_kobject_list);
+
+static const char nulls[16];
+static struct ibft_table_header *ibft_device;
+
+/*
+ * Helper functions to parse data properly.
+ */
+static ssize_t sprintf_ipaddr(char *buf, u8 *ip)
+{
+	char *str = buf;
+
+	if (ip[0] == 0 && ip[1] == 0 && ip[2] == 0 && ip[3] == 0 &&
+	    ip[4] == 0 && ip[5] == 0 && ip[6] == 0 && ip[7] == 0 &&
+	    ip[8] == 0 && ip[9] == 0 && ip[10] == 0xff && ip[11] == 0xff) {
+		/*
+		 * IPV4
+		 */
+		str += sprintf(buf, NIPQUAD_FMT, ip[12],
+			       ip[13], ip[14], ip[15]);
+	} else {
+		/*
+		 * IPv6
+		 */
+		str += sprintf(str, NIP6_FMT, ntohs(ip[0]), ntohs(ip[1]),
+			       ntohs(ip[2]), ntohs(ip[3]), ntohs(ip[4]),
+			       ntohs(ip[5]), ntohs(ip[6]), ntohs(ip[7]));
+	}
+	str += sprintf(str, "\n");
+	return str - buf;
+}
+
+static ssize_t sprintf_string(char *str, int len, char *buf)
+{
+	return sprintf(str, "%.*s\n", len, buf);
+}
+
+/*
+ * Helper function to verify the IBFT header.
+ */
+static int ibft_verify_hdr(char *t, struct ibft_hdr *hdr, int id, int length)
+{
+	if (hdr->id != id) {
+		printk(KERN_ERR "iBFT error: The field header.id is " \
+				"expected %d but found %d!\n", \
+				id, hdr->id);
+		return -ENODEV;
+	}
+	if (hdr->length != length) {
+		printk(KERN_ERR "iBFT error: The field header.length is " \
+				"expected %d but found %d!\n", \
+				length, hdr->length);
+		return -ENODEV;
+	}
+
+	return 0;
+}
+
+static void ibft_release(struct kobject *kobj)
+{
+	struct ibft_kobject *ibft =
+		container_of(kobj, struct ibft_kobject, kobj);
+	kfree(ibft);
+}
+
+/*
+ *  Routines for reading of the iBFT data in a human readable fashion.
+ */
+ssize_t ibft_attr_show_initiator(struct ibft_kobject *entry,
+				  struct ibft_attribute *attr,
+				  char *buf)
+{
+	struct ibft_initiator *initiator = entry->initiator;
+	void *ibft_loc = entry->header;
+	char *str = buf;
+
+	if (!initiator)
+		return 0;
+
+	switch (attr->type) {
+	case ibft_init_index:
+		str += sprintf(str, "%d\n", initiator->hdr.index);
+		break;
+	case ibft_init_flags:
+		str += sprintf(str, "%d\n", initiator->hdr.flags);
+		break;
+	case ibft_init_isns_server:
+		str += sprintf_ipaddr(str, initiator->isns_server);
+		break;
+	case ibft_init_slp_server:
+		str += sprintf_ipaddr(str, initiator->slp_server);
+		break;
+	case ibft_init_pri_radius_server:
+		str += sprintf_ipaddr(str, initiator->pri_radius_server);
+		break;
+	case ibft_init_sec_radius_server:
+		str += sprintf_ipaddr(str, initiator->sec_radius_server);
+		break;
+	case ibft_init_initiator_name:
+		str += sprintf_string(str, initiator->initiator_name_len,
+				      (char *)ibft_loc +
+				      initiator->initiator_name_off);
+		break;
+	default:
+		break;
+	}
+
+	return str - buf;
+}
+
+ssize_t ibft_attr_show_nic(struct ibft_kobject *entry,
+			    struct ibft_attribute *attr,
+			    char *buf)
+{
+	struct ibft_nic *nic = entry->nic;
+	void *ibft_loc = entry->header;
+	char *str = buf;
+	char *mac;
+	int val;
+
+	if (!nic)
+		return 0;
+
+	switch (attr->type) {
+	case ibft_eth_index:
+		str += sprintf(str, "%d\n", nic->hdr.index);
+		break;
+	case ibft_eth_flags:
+		str += sprintf(str, "%d\n", nic->hdr.flags);
+		break;
+	case ibft_eth_ip_addr:
+		str += sprintf_ipaddr(str, nic->ip_addr);
+		break;
+	case ibft_eth_subnet_mask:
+		val = ~((1 << (32-nic->subnet_mask_prefix))-1);
+		str += sprintf(str, "%d.%d.%d.%d\n",
+			       (u8)(val >> 24), (u8)(val >> 16),
+			       (u8)(val >> 8), (u8)(val));
+		break;
+	case ibft_eth_origin:
+		str += sprintf(str, "%d\n", nic->origin);
+		break;
+	case ibft_eth_gateway:
+		str += sprintf_ipaddr(str, nic->gateway);
+		break;
+	case ibft_eth_primary_dns:
+		str += sprintf_ipaddr(str, nic->primary_dns);
+		break;
+	case ibft_eth_secondary_dns:
+		str += sprintf_ipaddr(str, nic->secondary_dns);
+		break;
+	case ibft_eth_dhcp:
+		str += sprintf_ipaddr(str, nic->dhcp);
+		break;
+	case ibft_eth_vlan:
+		str += sprintf(str, "%d\n", nic->vlan);
+		break;
+	case ibft_eth_mac:
+		mac = nic->mac;
+		str += sprintf(str, "%02x:%02x:%02x:%02x:%02x:%02x\n",
+			       (u8)mac[0], (u8)mac[1], (u8)mac[2],
+			       (u8)mac[3], (u8)mac[4], (u8)mac[5]);
+		break;
+	case ibft_eth_hostname:
+		str += sprintf_string(str, nic->hostname_len,
+				      (char *)ibft_loc + nic->hostname_off);
+		break;
+	default:
+		break;
+	}
+
+	return str - buf;
+};
+
+ssize_t ibft_attr_show_target(struct ibft_kobject *entry,
+			       struct ibft_attribute *attr,
+			       char *buf)
+{
+	struct ibft_tgt *tgt = entry->tgt;
+	void *ibft_loc = entry->header;
+	char *str = buf;
+	int i;
+
+	if (!tgt)
+		return 0;
+
+	switch (attr->type) {
+	case ibft_tgt_index:
+		str += sprintf(str, "%d\n", tgt->hdr.index);
+		break;
+	case ibft_tgt_flags:
+		str += sprintf(str, "%d\n", tgt->hdr.flags);
+		break;
+	case ibft_tgt_ip_addr:
+		str += sprintf_ipaddr(str, tgt->ip_addr);
+		break;
+	case ibft_tgt_port:
+		str += sprintf(str, "%d\n", tgt->port);
+		break;
+	case ibft_tgt_lun:
+		for (i = 0; i < 8; i++)
+			str += sprintf(str, "%x", (u8)tgt->lun[i]);
+		str += sprintf(str, "\n");
+		break;
+	case ibft_tgt_nic_assoc:
+		str += sprintf(str, "%d\n", tgt->nic_assoc);
+		break;
+	case ibft_tgt_chap_type:
+		str += sprintf(str, "%d\n", tgt->chap_type);
+		break;
+	case ibft_tgt_name:
+		str += sprintf_string(str, tgt->tgt_name_len,
+				      (char *)ibft_loc + tgt->tgt_name_off);
+		break;
+	case ibft_tgt_chap_name:
+		str += sprintf_string(str, tgt->chap_name_len,
+				      (char *)ibft_loc + tgt->chap_name_off);
+		break;
+	case ibft_tgt_chap_secret:
+		str += sprintf_string(str, tgt->chap_secret_len,
+				      (char *)ibft_loc + tgt->chap_secret_off);
+		break;
+	case ibft_tgt_rev_chap_name:
+		str += sprintf_string(str, tgt->rev_chap_name_len,
+				      (char *)ibft_loc +
+				      tgt->rev_chap_name_off);
+		break;
+	case ibft_tgt_rev_chap_secret:
+		str += sprintf_string(str, tgt->rev_chap_secret_len,
+				      (char *)ibft_loc +
+				      tgt->rev_chap_secret_off);
+		break;
+	default:
+		break;
+	}
+
+	return str - buf;
+}
+
+/*
+ * The main routine which allows the user to read the IBFT data.
+ */
+static ssize_t ibft_show_attribute(struct kobject *kobj,
+				    struct attribute *attr,
+				    char *buf)
+{
+	struct ibft_kobject *dev =
+		container_of(kobj, struct ibft_kobject, kobj);
+	struct ibft_attribute *ibft_attr =
+		container_of(attr, struct ibft_attribute, attr);
+	ssize_t ret = -EIO;
+	char *str = buf;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EACCES;
+
+	if (ibft_attr->show)
+		ret = ibft_attr->show(dev, ibft_attr, str);
+
+	return ret;
+}
+
+static struct sysfs_ops ibft_attr_ops = {
+	.show = ibft_show_attribute,
+};
+
+static struct kobj_type ibft_ktype = {
+	.release = ibft_release,
+	.sysfs_ops = &ibft_attr_ops,
+};
+
+static struct kset *ibft_kset;
+
+static int __init ibft_check_and_alloc_device(void)
+{
+	int len;
+	u8 *pos;
+	u8 csum = 0;
+	struct ibft_table_header *hdr;
+
+	hdr = (struct ibft_table_header *)ibft_addr;
+	len = hdr->length;
+
+	/* Sanity checking of iBFT. */
+	if (hdr->revision != 1) {
+		printk(KERN_ERR "iBFT module supports only revision 1, " \
+				"while this is %d.\n", hdr->revision);
+		return 1;
+	}
+	for (pos = (u8 *)hdr; pos < (u8 *)hdr + len; pos++)
+		csum += *pos;
+
+	if (csum) {
+		printk(KERN_ERR "iBFT has incorrect checksum (0x%x)!\n", csum);
+		return 1;
+	}
+
+	ibft_device = kmalloc(len, GFP_KERNEL);
+	if (!ibft_device)
+		return -ENOMEM;
+
+	memcpy(ibft_device, hdr, len);
+
+	return 0;
+}
+
+static void ibft_free_device(struct ibft_table_header *hdr)
+{
+	kfree(hdr);
+}
+
+/*
+ * Helper function for ibft_register_kobjects.
+ */
+static int __init ibft_create_kobject(struct ibft_table_header *header,
+				       struct ibft_hdr *hdr,
+				       struct list_head *list)
+{
+	struct ibft_kobject *ibft_kobj = NULL;
+	struct ibft_nic *nic = (struct ibft_nic *)hdr;
+	struct pci_dev *pci_dev;
+	int rc = 0;
+
+	ibft_kobj = kzalloc(sizeof(*ibft_kobj), GFP_KERNEL);
+	if (!ibft_kobj)
+		return -ENOMEM;
+
+	ibft_kobj->header = header;
+	ibft_kobj->hdr = hdr;
+
+	switch (hdr->id) {
+	case id_initiator:
+		rc = ibft_verify_hdr("initiator", hdr, id_initiator,
+				     sizeof(*ibft_kobj->initiator));
+		break;
+	case id_nic:
+		rc = ibft_verify_hdr("ethernet", hdr, id_nic,
+				     sizeof(*ibft_kobj->nic));
+		break;
+	case id_target:
+		rc = ibft_verify_hdr("target", hdr, id_target,
+				     sizeof(*ibft_kobj->tgt));
+		break;
+	case id_reserved:
+	case id_control:
+	case id_extensions:
+		/* Fields which we don't support. Ignore them */
+		rc = 1;
+		break;
+	default:
+		printk(KERN_ERR "iBFT has unknown structure type (%d). " \
+				"Report this bug to your vendor!\n", hdr->id);
+		rc = 1;
+		break;
+	}
+
+	if (rc) {
+		kfree(ibft_kobj);
+		goto out_invalid_struct;
+	}
+
+	/* kobject fief. We will let the reference counter do the job
+	 of deleting its name if we fail here. */
+	ibft_kobj->kobj.kset = ibft_kset;
+
+	rc = kobject_init_and_add(&ibft_kobj->kobj, &ibft_ktype,
+				  NULL, ibft_id_names[hdr->id], hdr->index);
+
+	if (rc) {
+		kfree(ibft_kobj);
+		goto out;
+	}
+
+	kobject_uevent(&ibft_kobj->kobj, KOBJ_ADD);
+
+	if (hdr->id == id_nic) {
+		pci_dev = pci_get_bus_and_slot((nic->pci_bdf & 0xff00) >> 8,
+					       (nic->pci_bdf & 0xff));
+		if (pci_dev) {
+			rc = sysfs_create_link(&ibft_kobj->kobj,
+					       &pci_dev->dev.kobj, "device");
+			pci_dev_put(pci_dev);
+		}
+	}
+
+	/* Nothing broke so lets add it to the list. */
+	list_add_tail(&ibft_kobj->node, list);
+out:
+	return rc;
+out_invalid_struct:
+	/* Unsupported structs are skipped. */
+	return 0;
+}
+
+/*
+ * Scan the IBFT table structure for the NIC and Target fields. When
+ * found add them on the passed-in list.
+ */
+static int __init ibft_register_kobjects(struct ibft_table_header *header,
+					  struct list_head *list)
+{
+	struct ibft_control *control = NULL;
+	void *ptr, *end;
+	int rc = 0;
+	u16 offset;
+	u16 eot_offset;
+
+	control = (void *)header + sizeof(*header);
+	end = (void *)control + control->hdr.length;
+	eot_offset = (void *)header + header->length -
+		     (void *)control - sizeof(*header);
+	rc = ibft_verify_hdr("control", (struct ibft_hdr *)control, id_control,
+			     sizeof(*control));
+	rc |= control->hdr.index != 0;
+	if (rc)
+		return rc;
+
+	for (ptr = &control->initiator_off; ptr < end; ptr += sizeof(u16)) {
+		offset = *(u16 *)ptr;
+		if (offset && offset < header->length && offset < eot_offset) {
+			rc = ibft_create_kobject(header,
+						 (void *)header + offset,
+						 list);
+			if (rc)
+				break;
+		}
+	}
+
+	return rc;
+}
+
+static void ibft_unregister(struct list_head *attr_list,
+			     struct list_head *kobj_list)
+{
+	struct ibft_kobject *data = NULL, *n;
+	struct ibft_attribute *attr = NULL, *m;
+
+	list_for_each_entry_safe(attr, m, attr_list, node) {
+		sysfs_remove_file(attr->kobj, &attr->attr);
+		list_del(&attr->node);
+		kfree(attr);
+	};
+	list_del_init(attr_list);
+
+	list_for_each_entry_safe(data, n, kobj_list, node) {
+		list_del(&data->node);
+		if (data->hdr->id == id_nic)
+			sysfs_remove_link(&data->kobj, "device");
+		kobject_put(&data->kobj);
+	};
+	list_del_init(kobj_list);
+}
+
+static int __init ibft_create_attribute(struct ibft_kobject *kobj_data,
+					 int type,
+					 const char *name,
+					 ssize_t (*show)(struct ibft_kobject *,
+							 struct ibft_attribute*,
+							 char *buf),
+					 struct list_head *list)
+{
+	struct ibft_attribute *attr = NULL;
+	struct ibft_hdr *hdr = kobj_data->hdr;
+
+	attr = kmalloc(sizeof(*attr), GFP_KERNEL);
+	if (!attr)
+		return -ENOMEM;
+
+	attr->attr.name = name;
+	attr->attr.mode = S_IRUSR;
+	attr->attr.owner = THIS_MODULE;
+
+	attr->hdr = hdr;
+	attr->show = show;
+	attr->kobj = &kobj_data->kobj;
+	attr->type = type;
+
+	list_add_tail(&attr->node, list);
+
+	return 0;
+}
+
+static int __init ibft_check_nic_for(struct ibft_nic *nic, int entry)
+{
+	int rc = 0;
+
+	switch (entry) {
+	case ibft_eth_index:
+	case ibft_eth_flags:
+		rc = 1;
+		break;
+	case ibft_eth_ip_addr:
+		if (!memcmp(nic->dhcp, nulls, sizeof(nic->dhcp)))
+			rc = 1;
+		break;
+	case ibft_eth_subnet_mask:
+		if (!memcmp(nic->dhcp, nulls, sizeof(nic->dhcp)))
+			rc = 1;
+		break;
+	case ibft_eth_origin:
+		rc = 1;
+		break;
+	case ibft_eth_gateway:
+		if (memcmp(nic->gateway, nulls, sizeof(nic->gateway)))
+			rc = 1;
+		break;
+	case ibft_eth_primary_dns:
+		if (memcmp(nic->primary_dns, nulls,
+			   sizeof(nic->primary_dns)))
+			rc = 1;
+		break;
+	case ibft_eth_secondary_dns:
+		if (memcmp(nic->secondary_dns, nulls,
+			   sizeof(nic->secondary_dns)))
+			rc = 1;
+		break;
+	case ibft_eth_dhcp:
+		if (memcmp(nic->dhcp, nulls, sizeof(nic->dhcp)))
+			rc = 1;
+		break;
+	case ibft_eth_vlan:
+	case ibft_eth_mac:
+		rc = 1;
+		break;
+	case ibft_eth_hostname:
+		if (nic->hostname_off)
+			rc = 1;
+		break;
+	default:
+		break;
+	}
+
+	return rc;
+}
+
+static int __init ibft_check_tgt_for(struct ibft_tgt *tgt, int entry)
+{
+	int rc = 0;
+
+	switch (entry) {
+	case ibft_tgt_index:
+	case ibft_tgt_flags:
+	case ibft_tgt_ip_addr:
+	case ibft_tgt_port:
+	case ibft_tgt_lun:
+	case ibft_tgt_nic_assoc:
+	case ibft_tgt_chap_type:
+		rc = 1;
+	case ibft_tgt_name:
+		if (tgt->tgt_name_len)
+			rc = 1;
+		break;
+	case ibft_tgt_chap_name:
+	case ibft_tgt_chap_secret:
+		if (tgt->chap_name_len)
+			rc = 1;
+		break;
+	case ibft_tgt_rev_chap_name:
+	case ibft_tgt_rev_chap_secret:
+		if (tgt->rev_chap_name_len)
+			rc = 1;
+		break;
+	default:
+		break;
+	}
+
+	return rc;
+}
+
+static int __init ibft_check_initiator_for(struct ibft_initiator *init,
+					    int entry)
+{
+	int rc = 0;
+
+	switch (entry) {
+	case ibft_init_index:
+	case ibft_init_flags:
+		rc = 1;
+		break;
+	case ibft_init_isns_server:
+		if (memcmp(init->isns_server, nulls,
+			   sizeof(init->isns_server)))
+			rc = 1;
+		break;
+	case ibft_init_slp_server:
+		if (memcmp(init->slp_server, nulls,
+			   sizeof(init->slp_server)))
+			rc = 1;
+		break;
+	case ibft_init_pri_radius_server:
+		if (memcmp(init->pri_radius_server, nulls,
+			   sizeof(init->pri_radius_server)))
+			rc = 1;
+		break;
+	case ibft_init_sec_radius_server:
+		if (memcmp(init->sec_radius_server, nulls,
+			   sizeof(init->sec_radius_server)))
+			rc = 1;
+		break;
+	case ibft_init_initiator_name:
+		if (init->initiator_name_len)
+			rc = 1;
+		break;
+	default:
+		break;
+	}
+
+	return rc;
+}
+
+/*
+ *  Register the attributes for all of the kobjects.
+ */
+static int __init ibft_register_attributes(struct list_head *kobject_list,
+					    struct list_head *attr_list)
+{
+	int rc = 0, i = 0;
+	struct ibft_kobject *data = NULL;
+	struct ibft_attribute *attr = NULL, *m;
+
+	list_for_each_entry(data, kobject_list, node) {
+		switch (data->hdr->id) {
+		case id_nic:
+			for (i = 0; i < ibft_eth_end_marker && !rc; i++)
+				if (ibft_check_nic_for(data->nic, i))
+					rc = ibft_create_attribute(data, i,
+						ibft_eth_properties[i],
+						ibft_attr_show_nic, attr_list);
+			break;
+		case id_target:
+			for (i = 0; i < ibft_tgt_end_marker && !rc; i++)
+				if (ibft_check_tgt_for(data->tgt, i))
+					rc = ibft_create_attribute(data, i,
+						ibft_tgt_properties[i],
+						ibft_attr_show_target,
+						attr_list);
+			break;
+		case id_initiator:
+			for (i = 0; i < ibft_init_end_marker && !rc; i++)
+				if (ibft_check_initiator_for(
+					data->initiator, i))
+					rc = ibft_create_attribute(data, i,
+						ibft_initiator_properties[i],
+						ibft_attr_show_initiator,
+						attr_list);
+			break;
+		default:
+			break;
+		}
+		if (rc)
+			break;
+	}
+	list_for_each_entry_safe(attr, m, attr_list, node) {
+		rc = sysfs_create_file(attr->kobj, &attr->attr);
+		if (rc) {
+			list_del(&attr->node);
+			kfree(attr);
+			break;
+		}
+	}
+
+	return rc;
+}
+
+/*
+ * ibft_init() - creates sysfs tree entries for iBFT data.
+ */
+static int __init ibft_init(void)
+{
+	int rc = 0;
+
+	ibft_kset = kset_create_and_add("ibft", NULL, firmware_kobj);
+	if (!ibft_kset)
+		return -ENOMEM;
+
+	if (ibft_addr) {
+		printk(KERN_INFO "iBFT detected at 0x%lx.\n",
+		       virt_to_phys((void *)ibft_addr));
+
+		rc = ibft_check_and_alloc_device();
+		if (rc)
+			goto out_firmware_unregister;
+
+		/* Scan the IBFT for data and register kobjects. */
+		rc = ibft_register_kobjects(ibft_device, &ibft_kobject_list);
+		if (rc)
+			goto out_free;
+
+		/* Register the attributes */
+		rc = ibft_register_attributes(&ibft_kobject_list,
+					      &ibft_attr_list);
+		if (rc)
+			goto out_free;
+	} else
+		printk(KERN_INFO "No iBFT detected.\n");
+
+	return 0;
+
+out_free:
+	ibft_unregister(&ibft_attr_list, &ibft_kobject_list);
+	ibft_free_device(ibft_device);
+out_firmware_unregister:
+	kset_unregister(ibft_kset);
+	return rc;
+}
+
+static void __exit ibft_exit(void)
+{
+	ibft_unregister(&ibft_attr_list, &ibft_kobject_list);
+	ibft_free_device(ibft_device);
+	kset_unregister(ibft_kset);
+}
+
+module_init(ibft_init);
+module_exit(ibft_exit);
diff --git a/drivers/firmware/iscsi_ibft_find.c b/drivers/firmware/iscsi_ibft_find.c
new file mode 100644
index 0000000..cb97513
--- /dev/null
+++ b/drivers/firmware/iscsi_ibft_find.c
@@ -0,0 +1,83 @@
+/*
+ *  Copyright 2007 Red Hat, Inc.
+ *  by Peter Jones <pjones@redhat.com>
+ *  Copyright 2007 IBM, Inc.
+ *  by Konrad Rzeszutek <konradr@linux.vnet.ibm.com>
+ *
+ * This code finds the iSCSI Boot Format Table.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License v2.0 as published by
+ * the Free Software Foundation
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#include <linux/bootmem.h>
+#include <linux/blkdev.h>
+#include <linux/ctype.h>
+#include <linux/device.h>
+#include <linux/err.h>
+#include <linux/init.h>
+#include <linux/limits.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/slab.h>
+#include <linux/stat.h>
+#include <linux/string.h>
+#include <linux/types.h>
+
+#include <asm/mmzone.h>
+
+/*
+ * Physical location of iSCSI Boot Format Table.
+ */
+unsigned long ibft_addr;
+EXPORT_SYMBOL(ibft_addr);
+
+#define IBFT_SIGN "iBFT"
+#define IBFT_SIGN_LEN 4
+#define IBFT_START 0x80000 /* 512kB */
+#define IBFT_END 0x100000 /* 1MB */
+#define VGA_MEM 0xA0000 /* VGA buffer */
+#define VGA_SIZE 0x20000 /* 128kB */
+
+
+/*
+ * Routine used to find the iSCSI Boot Format Table. the logical
+ * kernel address is set in the ibft_addr variable.
+ */
+void __init reserve_ibft_region(void)
+{
+	unsigned long pos;
+	unsigned int len = 0;
+	void *virt;
+
+	ibft_addr = 0;
+
+	for (pos = IBFT_START; pos < IBFT_END; pos += 16) {
+		/* The table can't be inside the VGA BIOS reserved space,
+		 * so skip that area */
+		if (pos == VGA_MEM)
+			pos += VGA_SIZE;
+		virt = phys_to_virt(pos);
+		if (memcmp(virt, IBFT_SIGN, IBFT_SIGN_LEN) == 0) {
+			unsigned long *addr =
+			    (unsigned long *)phys_to_virt(pos + 4);
+			len = *addr;
+			/* if the length of the table extends past 1M,
+			 * the table cannot be valid. */
+			if (pos + len <= (IBFT_END-1)) {
+				ibft_addr = (unsigned long)virt;
+				break;
+			}
+		}
+	}
+	if (ibft_addr)
+		reserve_bootmem(pos,
+				PAGE_ALIGN(len));
+}
+EXPORT_SYMBOL(reserve_ibft_region);
diff --git a/include/linux/iscsi_ibft.h b/include/linux/iscsi_ibft.h
new file mode 100644
index 0000000..d58857d
--- /dev/null
+++ b/include/linux/iscsi_ibft.h
@@ -0,0 +1,38 @@
+/*
+ *  Copyright 2007 Red Hat, Inc.
+ *  by Peter Jones <pjones@redhat.com>
+ *  Copyright 2007 IBM, Inc.
+ *  by Konrad Rzeszutek <konradr@linux.vnet.ibm.com>
+ *
+ * This code exposes the iSCSI Boot Format Table to userland via sysfs.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License v2.0 as published by
+ * the Free Software Foundation
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#ifndef ISCSI_IBFT_H
+#define ISCSI_IBFT_H
+
+/*
+ * Physical location of iSCSI Boot Format Table.
+ * If the value is NULL there is no iBFT on the machine.
+ */
+extern unsigned long ibft_addr;
+
+/*
+ * Routine used to find and reserve the iSCSI Boot Format Table. The
+ * mapped address is set in the ibft_addr variable.
+ */
+#ifdef CONFIG_ISCSI_IBFT_FIND
+extern void __init reserve_ibft_region(void);
+#else
+static inline void reserve_ibft_region(void) { }
+#endif
+
+#endif /* ISCSI_IBFT_H */

^ permalink raw reply	[flat|nested] 8+ messages in thread

* REPOST: [PATCH] Add iSCSI iBFT support (v0.4.7)
  2008-02-08 19:47 [PATCH] Add iSCSI iBFT support (v0.4.7) Konrad Rzeszutek
@ 2008-02-09  3:10 ` Konrad Rzeszutek
  2008-02-09  5:17   ` James Bottomley
                     ` (2 more replies)
  0 siblings, 3 replies; 8+ messages in thread
From: Konrad Rzeszutek @ 2008-02-09  3:10 UTC (permalink / raw)
  To: Andrew Morton, linux-kernel, Greg KH, michaelc, James Bottomley
  Cc: dwm, darnok, pjones, konradr, konradr, randy.dunlap, hpa, lenb,
	mike.anderson, dwm, arjan, Andy Whitcroft

Whoops. I've attached an incorrect patch in the previous
e-mail (http://lkml.org/lkml/2008/2/8/350) that didn't take
in to account the 'reserve_bootmem' parameters changes.

Here is fresher copy which has been tested on 2.6.24-git19
on a machine with iBFT and without.

This patch (v0.4.7) adds /sysfs/firmware/ibft/[initiator|targetX|ethernetX]
directories along with text properties which export the the iSCSI
Boot Firmware Table (iBFT) structure.

What is iSCSI Boot Firmware Table? It is a mechanism for the iSCSI
tools to extract from the machine NICs the iSCSI connection information
so that they can automagically mount the iSCSI share/target. Currently
the iSCSI information is hard-coded in the initrd. The /sysfs entries
are read-only one-name-and-value fields.

The usual set of data exposed is:

# for a in `find /sys/firmware/ibft/ -type f -print`; do  echo -n "$a: ";  cat $a; done
/sys/firmware/ibft/target0/target-name: iqn.2007.com.intel-sbx44:storage-10gb
/sys/firmware/ibft/target0/nic-assoc: 0
/sys/firmware/ibft/target0/chap-type: 0
/sys/firmware/ibft/target0/lun: 00000000
/sys/firmware/ibft/target0/port: 3260
/sys/firmware/ibft/target0/ip-addr: 192.168.79.116
/sys/firmware/ibft/target0/flags: 3
/sys/firmware/ibft/target0/index: 0
/sys/firmware/ibft/ethernet0/mac: 00:11:25:9d:8b:01
/sys/firmware/ibft/ethernet0/vlan: 0
/sys/firmware/ibft/ethernet0/gateway: 192.168.79.254
/sys/firmware/ibft/ethernet0/origin: 0
/sys/firmware/ibft/ethernet0/subnet-mask: 255.255.252.0
/sys/firmware/ibft/ethernet0/ip-addr: 192.168.77.41
/sys/firmware/ibft/ethernet0/flags: 7
/sys/firmware/ibft/ethernet0/index: 0
/sys/firmware/ibft/initiator/initiator-name: iqn.2007-07.com:konrad.initiator
/sys/firmware/ibft/initiator/flags: 3
/sys/firmware/ibft/initiator/index: 0

For full details of the IBFT structure please take a look at:
ftp://ftp.software.ibm.com/systems/support/system_x_pdf/ibm_iscsi_boot_firmware_table_v1.02.pdf

Signed-off-by: Konrad Rzeszutek <konradr@linux.vnet.ibm.com>
Cc: Mike Christie <michaelc@cs.wisc.edu>
Cc: Peter Jones <pjones@redhat.com>
Cc: James Bottomley <James.Bottomley@hansenpartnership.com>
Cc: Greg KH <greg@kroah.com>


 Documentation/ABI/testing/sysfs-ibft |   23 
 arch/x86/kernel/setup_32.c           |    3 
 arch/x86/kernel/setup_64.c           |    4 
 drivers/firmware/Kconfig             |   20 
 drivers/firmware/Makefile            |    2 
 drivers/firmware/iscsi_ibft.c        |  985 +++++++++++++++++++++++++++++++++++
 drivers/firmware/iscsi_ibft_find.c   |   82 ++
 include/linux/iscsi_ibft.h           |   38 +
 8 files changed, 1157 insertions(+)


diff --git a/Documentation/ABI/testing/sysfs-ibft b/Documentation/ABI/testing/sysfs-ibft
new file mode 100644
index 0000000..062c009
--- /dev/null
+++ b/Documentation/ABI/testing/sysfs-ibft
@@ -0,0 +1,23 @@
+What:		/sys/firmware/ibft/initiator
+Date:		November 2007
+Contact:	Konrad Rzeszutek <konradr@linux.vnet.ibm.com>
+Description:	The /sys/firmware/ibft/initiator directory will contain
+		files that expose the iSCSI Boot Firmware Table initiator data.
+		Usually this contains the Initiator name.
+
+What:		/sys/firmware/ibft/targetX
+Date:		November 2007
+Contact:	Konrad Rzeszutek <konradr@linux.vnet.ibm.com>
+Description:	The /sys/firmware/ibft/targetX directory will contain
+		files that expose the iSCSI Boot Firmware Table target data.
+		Usually this contains the target's IP address, boot LUN,
+		target name, and what NIC it is associated with. It can also
+		contain the CHAP name (and password), the reverse CHAP
+		name (and password)
+
+What:		/sys/firmware/ibft/ethernetX
+Date:		November 2007
+Contact:	Konrad Rzeszutek <konradr@linux.vnet.ibm.com>
+Description:	The /sys/firmware/ibft/ethernetX directory will contain
+		files that expose the iSCSI Boot Firmware Table NIC data.
+		This can this can the IP address, MAC, and gateway of the NIC.
diff --git a/arch/x86/kernel/setup_32.c b/arch/x86/kernel/setup_32.c
index d1d8c34..9bd9482 100644
--- a/arch/x86/kernel/setup_32.c
+++ b/arch/x86/kernel/setup_32.c
@@ -39,6 +39,7 @@
 #include <linux/efi.h>
 #include <linux/init.h>
 #include <linux/edd.h>
+#include <linux/iscsi_ibft.h>
 #include <linux/nodemask.h>
 #include <linux/kexec.h>
 #include <linux/crash_dump.h>
@@ -651,6 +652,8 @@ void __init setup_bootmem_allocator(void)
 #endif
 	numa_kva_reserve();
 	reserve_crashkernel();
+
+	reserve_ibft_region();
 }
 
 /*
diff --git a/arch/x86/kernel/setup_64.c b/arch/x86/kernel/setup_64.c
index c0d8208..8efab8f 100644
--- a/arch/x86/kernel/setup_64.c
+++ b/arch/x86/kernel/setup_64.c
@@ -33,6 +33,7 @@
 #include <linux/acpi.h>
 #include <linux/kallsyms.h>
 #include <linux/edd.h>
+#include <linux/iscsi_ibft.h>
 #include <linux/mmzone.h>
 #include <linux/kexec.h>
 #include <linux/cpufreq.h>
@@ -425,6 +426,9 @@ void __init setup_arch(char **cmdline_p)
 	}
 #endif
 	reserve_crashkernel();
+
+	reserve_ibft_region();
+
 	paging_init();
 	map_vsyscall();
 
diff --git a/drivers/firmware/Kconfig b/drivers/firmware/Kconfig
index 05f02a3..40ffd76 100644
--- a/drivers/firmware/Kconfig
+++ b/drivers/firmware/Kconfig
@@ -93,4 +93,24 @@ config DMIID
 	  information from userspace through /sys/class/dmi/id/ or if you want
 	  DMI-based module auto-loading.
 
+config ISCSI_IBFT_FIND
+	bool "iSCSI Boot Firmware Table Attributes"
+	depends on X86
+	default n
+	help
+	  This option enables the kernel to find the region of memory
+	  in which the ISCSI Boot Firmware Table (iBFT) resides. This
+	  is necessary for iSCSI Boot Firmware Table Attributes module to work
+	  properly.
+
+config ISCSI_IBFT
+	tristate "iSCSI Boot Firmware Table Attributes module"
+	depends on ISCSI_IBFT_FIND
+	default	n
+	help
+	  This option enables support for detection and exposing of iSCSI
+	  Boot Firmware Table (iBFT) via sysfs to userspace. If you wish to
+	  detect iSCSI boot parameters dynamically during system boot, say Y.
+	  Otherwise, say N.
+
 endmenu
diff --git a/drivers/firmware/Makefile b/drivers/firmware/Makefile
index 8d4ebc8..4c91471 100644
--- a/drivers/firmware/Makefile
+++ b/drivers/firmware/Makefile
@@ -8,3 +8,5 @@ obj-$(CONFIG_EFI_PCDP)		+= pcdp.o
 obj-$(CONFIG_DELL_RBU)          += dell_rbu.o
 obj-$(CONFIG_DCDBAS)		+= dcdbas.o
 obj-$(CONFIG_DMIID)		+= dmi-id.o
+obj-$(CONFIG_ISCSI_IBFT_FIND)	+= iscsi_ibft_find.o
+obj-$(CONFIG_ISCSI_IBFT)	+= iscsi_ibft.o
diff --git a/drivers/firmware/iscsi_ibft.c b/drivers/firmware/iscsi_ibft.c
new file mode 100644
index 0000000..c2e5738
--- /dev/null
+++ b/drivers/firmware/iscsi_ibft.c
@@ -0,0 +1,985 @@
+/*
+ *  Copyright 2007 Red Hat, Inc.
+ *  by Peter Jones <pjones@redhat.com>
+ *  Copyright 2008 IBM, Inc.
+ *  by Konrad Rzeszutek <konradr@linux.vnet.ibm.com>
+ *
+ * This code exposes the iSCSI Boot Format Table to userland via sysfs.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License v2.0 as published by
+ * the Free Software Foundation
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * Changelog:
+ *
+ *   8 Feb 2008 - Konrad Rzeszutek <konradr@linux.vnet.ibm.com>
+ *    Combined two functions in one: reserve_ibft_region. (v0.4.7)
+ *
+ *  30 Jan 2008 - Konrad Rzeszutek <konradr@linux.vnet.ibm.com>
+ *   Added logic to handle IPv6 addresses. (v0.4.6)
+ *
+ *  25 Jan 2008 - Konrad Rzeszutek <konradr@linux.vnet.ibm.com>
+ *   Added logic to handle badly not-to-spec iBFT. (v0.4.5)
+ *
+ *   4 Jan 2008 - Konrad Rzeszutek <konradr@linux.vnet.ibm.com>
+ *   Added __init to function declarations. (v0.4.4)
+ *
+ *  21 Dec 2007 - Konrad Rzeszutek <konradr@linux.vnet.ibm.com>
+ *   Updated kobject registration, combined unregister functions in one
+ *   and code and style cleanup. (v0.4.3)
+ *
+ *   5 Dec 2007 - Konrad Rzeszutek <konradr@linux.vnet.ibm.com>
+ *   Added end-markers to enums and re-organized kobject registration. (v0.4.2)
+ *
+ *   4 Dec 2007 - Konrad Rzeszutek <konradr@linux.vnet.ibm.com>
+ *   Created 'device' sysfs link to the NIC and style cleanup. (v0.4.1)
+ *
+ *  28 Nov 2007 - Konrad Rzeszutek <konradr@linux.vnet.ibm.com>
+ *   Added sysfs-ibft documentation, moved 'find_ibft' function to
+ *   in its own file and added text attributes for every struct field.  (v0.4)
+ *
+ *  21 Nov 2007 - Konrad Rzeszutek <konradr@linux.vnet.ibm.com>
+ *   Added text attributes emulating OpenFirmware /proc/device-tree naming.
+ *   Removed binary /sysfs interface (v0.3)
+ *
+ *  29 Aug 2007 - Konrad Rzeszutek <konradr@linux.vnet.ibm.com>
+ *   Added functionality in setup.c to reserve iBFT region. (v0.2)
+ *
+ *  27 Aug 2007 - Konrad Rzeszutek <konradr@linux.vnet.ibm.com>
+ *   First version exposing iBFT data via a binary /sysfs. (v0.1)
+ *
+ */
+
+
+#include <linux/blkdev.h>
+#include <linux/ctype.h>
+#include <linux/device.h>
+#include <linux/err.h>
+#include <linux/init.h>
+#include <linux/iscsi_ibft.h>
+#include <linux/limits.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/slab.h>
+#include <linux/stat.h>
+#include <linux/string.h>
+#include <linux/types.h>
+
+#define IBFT_ISCSI_VERSION "0.4.7"
+#define IBFT_ISCSI_DATE "2008-Feb-08"
+
+MODULE_AUTHOR("Peter Jones <pjones@redhat.com> and \
+Konrad Rzeszutek <konradr@us.ibm.com>");
+MODULE_DESCRIPTION("sysfs interface to BIOS iBFT information");
+MODULE_LICENSE("GPL");
+MODULE_VERSION(IBFT_ISCSI_VERSION);
+
+
+struct ibft_table_header {
+	char signature[4];
+	u32 length;
+	u8 revision;
+	u8 checksum;
+	char oem_id[6];
+	char oem_table_id[8];
+	char reserved[24];
+} __attribute__((__packed__));
+
+struct ibft_hdr {
+	u8 id;
+	u8 version;
+	u16 length;
+	u8 index;
+	u8 flags;
+} __attribute__((__packed__));
+
+struct ibft_control {
+	struct ibft_hdr hdr;
+	u16 extensions;
+	u16 initiator_off;
+	u16 nic0_off;
+	u16 tgt0_off;
+	u16 nic1_off;
+	u16 tgt1_off;
+} __attribute__((__packed__));
+
+struct ibft_initiator {
+	struct ibft_hdr hdr;
+	char isns_server[16];
+	char slp_server[16];
+	char pri_radius_server[16];
+	char sec_radius_server[16];
+	u16 initiator_name_len;
+	u16 initiator_name_off;
+} __attribute__((__packed__));
+
+struct ibft_nic {
+	struct ibft_hdr hdr;
+	char ip_addr[16];
+	u8 subnet_mask_prefix;
+	u8 origin;
+	char gateway[16];
+	char primary_dns[16];
+	char secondary_dns[16];
+	char dhcp[16];
+	u16 vlan;
+	char mac[6];
+	u16 pci_bdf;
+	u16 hostname_len;
+	u16 hostname_off;
+} __attribute__((__packed__));
+
+struct ibft_tgt {
+	struct ibft_hdr hdr;
+	char ip_addr[16];
+	u16 port;
+	char lun[8];
+	u8 chap_type;
+	u8 nic_assoc;
+	u16 tgt_name_len;
+	u16 tgt_name_off;
+	u16 chap_name_len;
+	u16 chap_name_off;
+	u16 chap_secret_len;
+	u16 chap_secret_off;
+	u16 rev_chap_name_len;
+	u16 rev_chap_name_off;
+	u16 rev_chap_secret_len;
+	u16 rev_chap_secret_off;
+} __attribute__((__packed__));
+
+/*
+ * The kobject different types and its names.
+ *
+*/
+enum ibft_id {
+	id_reserved = 0, /* We don't support. */
+	id_control = 1, /* Should show up only once and is not exported. */
+	id_initiator = 2,
+	id_nic = 3,
+	id_target = 4,
+	id_extensions = 5, /* We don't support. */
+	id_end_marker,
+};
+
+/*
+ * We do not support the other types, hence the NULL.
+ */
+static const char *ibft_id_names[] =
+	{NULL, NULL, "initiator", "ethernet%d", "target%d", NULL, NULL};
+
+/*
+ * The text attributes names for each of the kobjects.
+*/
+enum ibft_eth_properties_enum {
+	ibft_eth_index,
+	ibft_eth_flags,
+	ibft_eth_ip_addr,
+	ibft_eth_subnet_mask,
+	ibft_eth_origin,
+	ibft_eth_gateway,
+	ibft_eth_primary_dns,
+	ibft_eth_secondary_dns,
+	ibft_eth_dhcp,
+	ibft_eth_vlan,
+	ibft_eth_mac,
+	/* ibft_eth_pci_bdf - this is replaced by link to the device itself. */
+	ibft_eth_hostname,
+	ibft_eth_end_marker,
+};
+
+static const char *ibft_eth_properties[] =
+	{"index", "flags", "ip-addr", "subnet-mask", "origin", "gateway",
+	"primary-dns", "secondary-dns", "dhcp", "vlan", "mac", "hostname",
+	NULL};
+
+enum ibft_tgt_properties_enum {
+	ibft_tgt_index,
+	ibft_tgt_flags,
+	ibft_tgt_ip_addr,
+	ibft_tgt_port,
+	ibft_tgt_lun,
+	ibft_tgt_chap_type,
+	ibft_tgt_nic_assoc,
+	ibft_tgt_name,
+	ibft_tgt_chap_name,
+	ibft_tgt_chap_secret,
+	ibft_tgt_rev_chap_name,
+	ibft_tgt_rev_chap_secret,
+	ibft_tgt_end_marker,
+};
+
+static const char *ibft_tgt_properties[] =
+	{"index", "flags", "ip-addr", "port", "lun", "chap-type", "nic-assoc",
+	"target-name", "chap-name", "chap-secret", "rev-chap-name",
+	"rev-chap-name-secret", NULL};
+
+enum ibft_initiator_properties_enum {
+	ibft_init_index,
+	ibft_init_flags,
+	ibft_init_isns_server,
+	ibft_init_slp_server,
+	ibft_init_pri_radius_server,
+	ibft_init_sec_radius_server,
+	ibft_init_initiator_name,
+	ibft_init_end_marker,
+};
+
+static const char *ibft_initiator_properties[] =
+	{"index", "flags", "isns-server", "slp-server", "pri-radius-server",
+	"sec-radius-server", "initiator-name", NULL};
+
+/*
+ * The kobject and attribute structures.
+ */
+
+struct ibft_kobject {
+	struct ibft_table_header *header;
+	union {
+		struct ibft_initiator *initiator;
+		struct ibft_nic *nic;
+		struct ibft_tgt *tgt;
+		struct ibft_hdr *hdr;
+	};
+	struct kobject kobj;
+	struct list_head node;
+};
+
+struct ibft_attribute {
+	struct attribute attr;
+	ssize_t (*show) (struct  ibft_kobject *entry,
+			 struct ibft_attribute *attr, char *buf);
+	union {
+		struct ibft_initiator *initiator;
+		struct ibft_nic *nic;
+		struct ibft_tgt *tgt;
+		struct ibft_hdr *hdr;
+	};
+	struct kobject *kobj;
+	int type; /* The enum of the type. This can be any value off:
+		ibft_eth_properties_enum, ibft_tgt_properties_enum,
+		or ibft_initiator_properties_enum. */
+	struct list_head node;
+};
+
+static LIST_HEAD(ibft_attr_list);
+static LIST_HEAD(ibft_kobject_list);
+
+static const char nulls[16];
+static struct ibft_table_header *ibft_device;
+
+/*
+ * Helper functions to parse data properly.
+ */
+static ssize_t sprintf_ipaddr(char *buf, u8 *ip)
+{
+	char *str = buf;
+
+	if (ip[0] == 0 && ip[1] == 0 && ip[2] == 0 && ip[3] == 0 &&
+	    ip[4] == 0 && ip[5] == 0 && ip[6] == 0 && ip[7] == 0 &&
+	    ip[8] == 0 && ip[9] == 0 && ip[10] == 0xff && ip[11] == 0xff) {
+		/*
+		 * IPV4
+		 */
+		str += sprintf(buf, NIPQUAD_FMT, ip[12],
+			       ip[13], ip[14], ip[15]);
+	} else {
+		/*
+		 * IPv6
+		 */
+		str += sprintf(str, NIP6_FMT, ntohs(ip[0]), ntohs(ip[1]),
+			       ntohs(ip[2]), ntohs(ip[3]), ntohs(ip[4]),
+			       ntohs(ip[5]), ntohs(ip[6]), ntohs(ip[7]));
+	}
+	str += sprintf(str, "\n");
+	return str - buf;
+}
+
+static ssize_t sprintf_string(char *str, int len, char *buf)
+{
+	return sprintf(str, "%.*s\n", len, buf);
+}
+
+/*
+ * Helper function to verify the IBFT header.
+ */
+static int ibft_verify_hdr(char *t, struct ibft_hdr *hdr, int id, int length)
+{
+	if (hdr->id != id) {
+		printk(KERN_ERR "iBFT error: The field header.id is " \
+				"expected %d but found %d!\n", \
+				id, hdr->id);
+		return -ENODEV;
+	}
+	if (hdr->length != length) {
+		printk(KERN_ERR "iBFT error: The field header.length is " \
+				"expected %d but found %d!\n", \
+				length, hdr->length);
+		return -ENODEV;
+	}
+
+	return 0;
+}
+
+static void ibft_release(struct kobject *kobj)
+{
+	struct ibft_kobject *ibft =
+		container_of(kobj, struct ibft_kobject, kobj);
+	kfree(ibft);
+}
+
+/*
+ *  Routines for reading of the iBFT data in a human readable fashion.
+ */
+ssize_t ibft_attr_show_initiator(struct ibft_kobject *entry,
+				  struct ibft_attribute *attr,
+				  char *buf)
+{
+	struct ibft_initiator *initiator = entry->initiator;
+	void *ibft_loc = entry->header;
+	char *str = buf;
+
+	if (!initiator)
+		return 0;
+
+	switch (attr->type) {
+	case ibft_init_index:
+		str += sprintf(str, "%d\n", initiator->hdr.index);
+		break;
+	case ibft_init_flags:
+		str += sprintf(str, "%d\n", initiator->hdr.flags);
+		break;
+	case ibft_init_isns_server:
+		str += sprintf_ipaddr(str, initiator->isns_server);
+		break;
+	case ibft_init_slp_server:
+		str += sprintf_ipaddr(str, initiator->slp_server);
+		break;
+	case ibft_init_pri_radius_server:
+		str += sprintf_ipaddr(str, initiator->pri_radius_server);
+		break;
+	case ibft_init_sec_radius_server:
+		str += sprintf_ipaddr(str, initiator->sec_radius_server);
+		break;
+	case ibft_init_initiator_name:
+		str += sprintf_string(str, initiator->initiator_name_len,
+				      (char *)ibft_loc +
+				      initiator->initiator_name_off);
+		break;
+	default:
+		break;
+	}
+
+	return str - buf;
+}
+
+ssize_t ibft_attr_show_nic(struct ibft_kobject *entry,
+			    struct ibft_attribute *attr,
+			    char *buf)
+{
+	struct ibft_nic *nic = entry->nic;
+	void *ibft_loc = entry->header;
+	char *str = buf;
+	char *mac;
+	int val;
+
+	if (!nic)
+		return 0;
+
+	switch (attr->type) {
+	case ibft_eth_index:
+		str += sprintf(str, "%d\n", nic->hdr.index);
+		break;
+	case ibft_eth_flags:
+		str += sprintf(str, "%d\n", nic->hdr.flags);
+		break;
+	case ibft_eth_ip_addr:
+		str += sprintf_ipaddr(str, nic->ip_addr);
+		break;
+	case ibft_eth_subnet_mask:
+		val = ~((1 << (32-nic->subnet_mask_prefix))-1);
+		str += sprintf(str, "%d.%d.%d.%d\n",
+			       (u8)(val >> 24), (u8)(val >> 16),
+			       (u8)(val >> 8), (u8)(val));
+		break;
+	case ibft_eth_origin:
+		str += sprintf(str, "%d\n", nic->origin);
+		break;
+	case ibft_eth_gateway:
+		str += sprintf_ipaddr(str, nic->gateway);
+		break;
+	case ibft_eth_primary_dns:
+		str += sprintf_ipaddr(str, nic->primary_dns);
+		break;
+	case ibft_eth_secondary_dns:
+		str += sprintf_ipaddr(str, nic->secondary_dns);
+		break;
+	case ibft_eth_dhcp:
+		str += sprintf_ipaddr(str, nic->dhcp);
+		break;
+	case ibft_eth_vlan:
+		str += sprintf(str, "%d\n", nic->vlan);
+		break;
+	case ibft_eth_mac:
+		mac = nic->mac;
+		str += sprintf(str, "%02x:%02x:%02x:%02x:%02x:%02x\n",
+			       (u8)mac[0], (u8)mac[1], (u8)mac[2],
+			       (u8)mac[3], (u8)mac[4], (u8)mac[5]);
+		break;
+	case ibft_eth_hostname:
+		str += sprintf_string(str, nic->hostname_len,
+				      (char *)ibft_loc + nic->hostname_off);
+		break;
+	default:
+		break;
+	}
+
+	return str - buf;
+};
+
+ssize_t ibft_attr_show_target(struct ibft_kobject *entry,
+			       struct ibft_attribute *attr,
+			       char *buf)
+{
+	struct ibft_tgt *tgt = entry->tgt;
+	void *ibft_loc = entry->header;
+	char *str = buf;
+	int i;
+
+	if (!tgt)
+		return 0;
+
+	switch (attr->type) {
+	case ibft_tgt_index:
+		str += sprintf(str, "%d\n", tgt->hdr.index);
+		break;
+	case ibft_tgt_flags:
+		str += sprintf(str, "%d\n", tgt->hdr.flags);
+		break;
+	case ibft_tgt_ip_addr:
+		str += sprintf_ipaddr(str, tgt->ip_addr);
+		break;
+	case ibft_tgt_port:
+		str += sprintf(str, "%d\n", tgt->port);
+		break;
+	case ibft_tgt_lun:
+		for (i = 0; i < 8; i++)
+			str += sprintf(str, "%x", (u8)tgt->lun[i]);
+		str += sprintf(str, "\n");
+		break;
+	case ibft_tgt_nic_assoc:
+		str += sprintf(str, "%d\n", tgt->nic_assoc);
+		break;
+	case ibft_tgt_chap_type:
+		str += sprintf(str, "%d\n", tgt->chap_type);
+		break;
+	case ibft_tgt_name:
+		str += sprintf_string(str, tgt->tgt_name_len,
+				      (char *)ibft_loc + tgt->tgt_name_off);
+		break;
+	case ibft_tgt_chap_name:
+		str += sprintf_string(str, tgt->chap_name_len,
+				      (char *)ibft_loc + tgt->chap_name_off);
+		break;
+	case ibft_tgt_chap_secret:
+		str += sprintf_string(str, tgt->chap_secret_len,
+				      (char *)ibft_loc + tgt->chap_secret_off);
+		break;
+	case ibft_tgt_rev_chap_name:
+		str += sprintf_string(str, tgt->rev_chap_name_len,
+				      (char *)ibft_loc +
+				      tgt->rev_chap_name_off);
+		break;
+	case ibft_tgt_rev_chap_secret:
+		str += sprintf_string(str, tgt->rev_chap_secret_len,
+				      (char *)ibft_loc +
+				      tgt->rev_chap_secret_off);
+		break;
+	default:
+		break;
+	}
+
+	return str - buf;
+}
+
+/*
+ * The main routine which allows the user to read the IBFT data.
+ */
+static ssize_t ibft_show_attribute(struct kobject *kobj,
+				    struct attribute *attr,
+				    char *buf)
+{
+	struct ibft_kobject *dev =
+		container_of(kobj, struct ibft_kobject, kobj);
+	struct ibft_attribute *ibft_attr =
+		container_of(attr, struct ibft_attribute, attr);
+	ssize_t ret = -EIO;
+	char *str = buf;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EACCES;
+
+	if (ibft_attr->show)
+		ret = ibft_attr->show(dev, ibft_attr, str);
+
+	return ret;
+}
+
+static struct sysfs_ops ibft_attr_ops = {
+	.show = ibft_show_attribute,
+};
+
+static struct kobj_type ibft_ktype = {
+	.release = ibft_release,
+	.sysfs_ops = &ibft_attr_ops,
+};
+
+static struct kset *ibft_kset;
+
+static int __init ibft_check_and_alloc_device(void)
+{
+	int len;
+	u8 *pos;
+	u8 csum = 0;
+	struct ibft_table_header *hdr;
+
+	hdr = (struct ibft_table_header *)ibft_addr;
+	len = hdr->length;
+
+	/* Sanity checking of iBFT. */
+	if (hdr->revision != 1) {
+		printk(KERN_ERR "iBFT module supports only revision 1, " \
+				"while this is %d.\n", hdr->revision);
+		return 1;
+	}
+	for (pos = (u8 *)hdr; pos < (u8 *)hdr + len; pos++)
+		csum += *pos;
+
+	if (csum) {
+		printk(KERN_ERR "iBFT has incorrect checksum (0x%x)!\n", csum);
+		return 1;
+	}
+
+	ibft_device = kmalloc(len, GFP_KERNEL);
+	if (!ibft_device)
+		return -ENOMEM;
+
+	memcpy(ibft_device, hdr, len);
+
+	return 0;
+}
+
+static void ibft_free_device(struct ibft_table_header *hdr)
+{
+	kfree(hdr);
+}
+
+/*
+ * Helper function for ibft_register_kobjects.
+ */
+static int __init ibft_create_kobject(struct ibft_table_header *header,
+				       struct ibft_hdr *hdr,
+				       struct list_head *list)
+{
+	struct ibft_kobject *ibft_kobj = NULL;
+	struct ibft_nic *nic = (struct ibft_nic *)hdr;
+	struct pci_dev *pci_dev;
+	int rc = 0;
+
+	ibft_kobj = kzalloc(sizeof(*ibft_kobj), GFP_KERNEL);
+	if (!ibft_kobj)
+		return -ENOMEM;
+
+	ibft_kobj->header = header;
+	ibft_kobj->hdr = hdr;
+
+	switch (hdr->id) {
+	case id_initiator:
+		rc = ibft_verify_hdr("initiator", hdr, id_initiator,
+				     sizeof(*ibft_kobj->initiator));
+		break;
+	case id_nic:
+		rc = ibft_verify_hdr("ethernet", hdr, id_nic,
+				     sizeof(*ibft_kobj->nic));
+		break;
+	case id_target:
+		rc = ibft_verify_hdr("target", hdr, id_target,
+				     sizeof(*ibft_kobj->tgt));
+		break;
+	case id_reserved:
+	case id_control:
+	case id_extensions:
+		/* Fields which we don't support. Ignore them */
+		rc = 1;
+		break;
+	default:
+		printk(KERN_ERR "iBFT has unknown structure type (%d). " \
+				"Report this bug to your vendor!\n", hdr->id);
+		rc = 1;
+		break;
+	}
+
+	if (rc) {
+		kfree(ibft_kobj);
+		goto out_invalid_struct;
+	}
+
+	/* kobject fief. We will let the reference counter do the job
+	 of deleting its name if we fail here. */
+	ibft_kobj->kobj.kset = ibft_kset;
+
+	rc = kobject_init_and_add(&ibft_kobj->kobj, &ibft_ktype,
+				  NULL, ibft_id_names[hdr->id], hdr->index);
+
+	if (rc) {
+		kfree(ibft_kobj);
+		goto out;
+	}
+
+	kobject_uevent(&ibft_kobj->kobj, KOBJ_ADD);
+
+	if (hdr->id == id_nic) {
+		pci_dev = pci_get_bus_and_slot((nic->pci_bdf & 0xff00) >> 8,
+					       (nic->pci_bdf & 0xff));
+		if (pci_dev) {
+			rc = sysfs_create_link(&ibft_kobj->kobj,
+					       &pci_dev->dev.kobj, "device");
+			pci_dev_put(pci_dev);
+		}
+	}
+
+	/* Nothing broke so lets add it to the list. */
+	list_add_tail(&ibft_kobj->node, list);
+out:
+	return rc;
+out_invalid_struct:
+	/* Unsupported structs are skipped. */
+	return 0;
+}
+
+/*
+ * Scan the IBFT table structure for the NIC and Target fields. When
+ * found add them on the passed-in list.
+ */
+static int __init ibft_register_kobjects(struct ibft_table_header *header,
+					  struct list_head *list)
+{
+	struct ibft_control *control = NULL;
+	void *ptr, *end;
+	int rc = 0;
+	u16 offset;
+	u16 eot_offset;
+
+	control = (void *)header + sizeof(*header);
+	end = (void *)control + control->hdr.length;
+	eot_offset = (void *)header + header->length -
+		     (void *)control - sizeof(*header);
+	rc = ibft_verify_hdr("control", (struct ibft_hdr *)control, id_control,
+			     sizeof(*control));
+	rc |= control->hdr.index != 0;
+	if (rc)
+		return rc;
+
+	for (ptr = &control->initiator_off; ptr < end; ptr += sizeof(u16)) {
+		offset = *(u16 *)ptr;
+		if (offset && offset < header->length && offset < eot_offset) {
+			rc = ibft_create_kobject(header,
+						 (void *)header + offset,
+						 list);
+			if (rc)
+				break;
+		}
+	}
+
+	return rc;
+}
+
+static void ibft_unregister(struct list_head *attr_list,
+			     struct list_head *kobj_list)
+{
+	struct ibft_kobject *data = NULL, *n;
+	struct ibft_attribute *attr = NULL, *m;
+
+	list_for_each_entry_safe(attr, m, attr_list, node) {
+		sysfs_remove_file(attr->kobj, &attr->attr);
+		list_del(&attr->node);
+		kfree(attr);
+	};
+	list_del_init(attr_list);
+
+	list_for_each_entry_safe(data, n, kobj_list, node) {
+		list_del(&data->node);
+		if (data->hdr->id == id_nic)
+			sysfs_remove_link(&data->kobj, "device");
+		kobject_put(&data->kobj);
+	};
+	list_del_init(kobj_list);
+}
+
+static int __init ibft_create_attribute(struct ibft_kobject *kobj_data,
+					 int type,
+					 const char *name,
+					 ssize_t (*show)(struct ibft_kobject *,
+							 struct ibft_attribute*,
+							 char *buf),
+					 struct list_head *list)
+{
+	struct ibft_attribute *attr = NULL;
+	struct ibft_hdr *hdr = kobj_data->hdr;
+
+	attr = kmalloc(sizeof(*attr), GFP_KERNEL);
+	if (!attr)
+		return -ENOMEM;
+
+	attr->attr.name = name;
+	attr->attr.mode = S_IRUSR;
+	attr->attr.owner = THIS_MODULE;
+
+	attr->hdr = hdr;
+	attr->show = show;
+	attr->kobj = &kobj_data->kobj;
+	attr->type = type;
+
+	list_add_tail(&attr->node, list);
+
+	return 0;
+}
+
+static int __init ibft_check_nic_for(struct ibft_nic *nic, int entry)
+{
+	int rc = 0;
+
+	switch (entry) {
+	case ibft_eth_index:
+	case ibft_eth_flags:
+		rc = 1;
+		break;
+	case ibft_eth_ip_addr:
+		if (!memcmp(nic->dhcp, nulls, sizeof(nic->dhcp)))
+			rc = 1;
+		break;
+	case ibft_eth_subnet_mask:
+		if (!memcmp(nic->dhcp, nulls, sizeof(nic->dhcp)))
+			rc = 1;
+		break;
+	case ibft_eth_origin:
+		rc = 1;
+		break;
+	case ibft_eth_gateway:
+		if (memcmp(nic->gateway, nulls, sizeof(nic->gateway)))
+			rc = 1;
+		break;
+	case ibft_eth_primary_dns:
+		if (memcmp(nic->primary_dns, nulls,
+			   sizeof(nic->primary_dns)))
+			rc = 1;
+		break;
+	case ibft_eth_secondary_dns:
+		if (memcmp(nic->secondary_dns, nulls,
+			   sizeof(nic->secondary_dns)))
+			rc = 1;
+		break;
+	case ibft_eth_dhcp:
+		if (memcmp(nic->dhcp, nulls, sizeof(nic->dhcp)))
+			rc = 1;
+		break;
+	case ibft_eth_vlan:
+	case ibft_eth_mac:
+		rc = 1;
+		break;
+	case ibft_eth_hostname:
+		if (nic->hostname_off)
+			rc = 1;
+		break;
+	default:
+		break;
+	}
+
+	return rc;
+}
+
+static int __init ibft_check_tgt_for(struct ibft_tgt *tgt, int entry)
+{
+	int rc = 0;
+
+	switch (entry) {
+	case ibft_tgt_index:
+	case ibft_tgt_flags:
+	case ibft_tgt_ip_addr:
+	case ibft_tgt_port:
+	case ibft_tgt_lun:
+	case ibft_tgt_nic_assoc:
+	case ibft_tgt_chap_type:
+		rc = 1;
+	case ibft_tgt_name:
+		if (tgt->tgt_name_len)
+			rc = 1;
+		break;
+	case ibft_tgt_chap_name:
+	case ibft_tgt_chap_secret:
+		if (tgt->chap_name_len)
+			rc = 1;
+		break;
+	case ibft_tgt_rev_chap_name:
+	case ibft_tgt_rev_chap_secret:
+		if (tgt->rev_chap_name_len)
+			rc = 1;
+		break;
+	default:
+		break;
+	}
+
+	return rc;
+}
+
+static int __init ibft_check_initiator_for(struct ibft_initiator *init,
+					    int entry)
+{
+	int rc = 0;
+
+	switch (entry) {
+	case ibft_init_index:
+	case ibft_init_flags:
+		rc = 1;
+		break;
+	case ibft_init_isns_server:
+		if (memcmp(init->isns_server, nulls,
+			   sizeof(init->isns_server)))
+			rc = 1;
+		break;
+	case ibft_init_slp_server:
+		if (memcmp(init->slp_server, nulls,
+			   sizeof(init->slp_server)))
+			rc = 1;
+		break;
+	case ibft_init_pri_radius_server:
+		if (memcmp(init->pri_radius_server, nulls,
+			   sizeof(init->pri_radius_server)))
+			rc = 1;
+		break;
+	case ibft_init_sec_radius_server:
+		if (memcmp(init->sec_radius_server, nulls,
+			   sizeof(init->sec_radius_server)))
+			rc = 1;
+		break;
+	case ibft_init_initiator_name:
+		if (init->initiator_name_len)
+			rc = 1;
+		break;
+	default:
+		break;
+	}
+
+	return rc;
+}
+
+/*
+ *  Register the attributes for all of the kobjects.
+ */
+static int __init ibft_register_attributes(struct list_head *kobject_list,
+					    struct list_head *attr_list)
+{
+	int rc = 0, i = 0;
+	struct ibft_kobject *data = NULL;
+	struct ibft_attribute *attr = NULL, *m;
+
+	list_for_each_entry(data, kobject_list, node) {
+		switch (data->hdr->id) {
+		case id_nic:
+			for (i = 0; i < ibft_eth_end_marker && !rc; i++)
+				if (ibft_check_nic_for(data->nic, i))
+					rc = ibft_create_attribute(data, i,
+						ibft_eth_properties[i],
+						ibft_attr_show_nic, attr_list);
+			break;
+		case id_target:
+			for (i = 0; i < ibft_tgt_end_marker && !rc; i++)
+				if (ibft_check_tgt_for(data->tgt, i))
+					rc = ibft_create_attribute(data, i,
+						ibft_tgt_properties[i],
+						ibft_attr_show_target,
+						attr_list);
+			break;
+		case id_initiator:
+			for (i = 0; i < ibft_init_end_marker && !rc; i++)
+				if (ibft_check_initiator_for(
+					data->initiator, i))
+					rc = ibft_create_attribute(data, i,
+						ibft_initiator_properties[i],
+						ibft_attr_show_initiator,
+						attr_list);
+			break;
+		default:
+			break;
+		}
+		if (rc)
+			break;
+	}
+	list_for_each_entry_safe(attr, m, attr_list, node) {
+		rc = sysfs_create_file(attr->kobj, &attr->attr);
+		if (rc) {
+			list_del(&attr->node);
+			kfree(attr);
+			break;
+		}
+	}
+
+	return rc;
+}
+
+/*
+ * ibft_init() - creates sysfs tree entries for iBFT data.
+ */
+static int __init ibft_init(void)
+{
+	int rc = 0;
+
+	ibft_kset = kset_create_and_add("ibft", NULL, firmware_kobj);
+	if (!ibft_kset)
+		return -ENOMEM;
+
+	if (ibft_addr) {
+		printk(KERN_INFO "iBFT detected at 0x%lx.\n",
+		       virt_to_phys((void *)ibft_addr));
+
+		rc = ibft_check_and_alloc_device();
+		if (rc)
+			goto out_firmware_unregister;
+
+		/* Scan the IBFT for data and register kobjects. */
+		rc = ibft_register_kobjects(ibft_device, &ibft_kobject_list);
+		if (rc)
+			goto out_free;
+
+		/* Register the attributes */
+		rc = ibft_register_attributes(&ibft_kobject_list,
+					      &ibft_attr_list);
+		if (rc)
+			goto out_free;
+	} else
+		printk(KERN_INFO "No iBFT detected.\n");
+
+	return 0;
+
+out_free:
+	ibft_unregister(&ibft_attr_list, &ibft_kobject_list);
+	ibft_free_device(ibft_device);
+out_firmware_unregister:
+	kset_unregister(ibft_kset);
+	return rc;
+}
+
+static void __exit ibft_exit(void)
+{
+	ibft_unregister(&ibft_attr_list, &ibft_kobject_list);
+	ibft_free_device(ibft_device);
+	kset_unregister(ibft_kset);
+}
+
+module_init(ibft_init);
+module_exit(ibft_exit);
diff --git a/drivers/firmware/iscsi_ibft_find.c b/drivers/firmware/iscsi_ibft_find.c
new file mode 100644
index 0000000..1ee56b2
--- /dev/null
+++ b/drivers/firmware/iscsi_ibft_find.c
@@ -0,0 +1,82 @@
+/*
+ *  Copyright 2007 Red Hat, Inc.
+ *  by Peter Jones <pjones@redhat.com>
+ *  Copyright 2007 IBM, Inc.
+ *  by Konrad Rzeszutek <konradr@linux.vnet.ibm.com>
+ *
+ * This code finds the iSCSI Boot Format Table.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License v2.0 as published by
+ * the Free Software Foundation
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#include <linux/bootmem.h>
+#include <linux/blkdev.h>
+#include <linux/ctype.h>
+#include <linux/device.h>
+#include <linux/err.h>
+#include <linux/init.h>
+#include <linux/limits.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/slab.h>
+#include <linux/stat.h>
+#include <linux/string.h>
+#include <linux/types.h>
+
+#include <asm/mmzone.h>
+
+/*
+ * Physical location of iSCSI Boot Format Table.
+ */
+unsigned long ibft_addr;
+EXPORT_SYMBOL(ibft_addr);
+
+#define IBFT_SIGN "iBFT"
+#define IBFT_SIGN_LEN 4
+#define IBFT_START 0x80000 /* 512kB */
+#define IBFT_END 0x100000 /* 1MB */
+#define VGA_MEM 0xA0000 /* VGA buffer */
+#define VGA_SIZE 0x20000 /* 128kB */
+
+
+/*
+ * Routine used to find the iSCSI Boot Format Table. the logical
+ * kernel address is set in the ibft_addr variable.
+ */
+void __init reserve_ibft_region(void)
+{
+	unsigned long pos;
+	unsigned int len = 0;
+	void *virt;
+
+	ibft_addr = 0;
+
+	for (pos = IBFT_START; pos < IBFT_END; pos += 16) {
+		/* The table can't be inside the VGA BIOS reserved space,
+		 * so skip that area */
+		if (pos == VGA_MEM)
+			pos += VGA_SIZE;
+		virt = phys_to_virt(pos);
+		if (memcmp(virt, IBFT_SIGN, IBFT_SIGN_LEN) == 0) {
+			unsigned long *addr =
+			    (unsigned long *)phys_to_virt(pos + 4);
+			len = *addr;
+			/* if the length of the table extends past 1M,
+			 * the table cannot be valid. */
+			if (pos + len <= (IBFT_END-1)) {
+				ibft_addr = (unsigned long)virt;
+				break;
+			}
+		}
+	}
+	if (ibft_addr)
+		reserve_bootmem(pos, PAGE_ALIGN(len), BOOTMEM_DEFAULT);
+}
+EXPORT_SYMBOL(reserve_ibft_region);
diff --git a/include/linux/iscsi_ibft.h b/include/linux/iscsi_ibft.h
new file mode 100644
index 0000000..d58857d
--- /dev/null
+++ b/include/linux/iscsi_ibft.h
@@ -0,0 +1,38 @@
+/*
+ *  Copyright 2007 Red Hat, Inc.
+ *  by Peter Jones <pjones@redhat.com>
+ *  Copyright 2007 IBM, Inc.
+ *  by Konrad Rzeszutek <konradr@linux.vnet.ibm.com>
+ *
+ * This code exposes the iSCSI Boot Format Table to userland via sysfs.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License v2.0 as published by
+ * the Free Software Foundation
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#ifndef ISCSI_IBFT_H
+#define ISCSI_IBFT_H
+
+/*
+ * Physical location of iSCSI Boot Format Table.
+ * If the value is NULL there is no iBFT on the machine.
+ */
+extern unsigned long ibft_addr;
+
+/*
+ * Routine used to find and reserve the iSCSI Boot Format Table. The
+ * mapped address is set in the ibft_addr variable.
+ */
+#ifdef CONFIG_ISCSI_IBFT_FIND
+extern void __init reserve_ibft_region(void);
+#else
+static inline void reserve_ibft_region(void) { }
+#endif
+
+#endif /* ISCSI_IBFT_H */

^ permalink raw reply	[flat|nested] 8+ messages in thread

* Re: REPOST: [PATCH] Add iSCSI iBFT support (v0.4.7)
  2008-02-09  3:10 ` REPOST: " Konrad Rzeszutek
@ 2008-02-09  5:17   ` James Bottomley
  2008-02-09  5:23   ` James Bottomley
  2008-02-09  5:32   ` James Bottomley
  2 siblings, 0 replies; 8+ messages in thread
From: James Bottomley @ 2008-02-09  5:17 UTC (permalink / raw)
  To: Konrad Rzeszutek
  Cc: Andrew Morton, linux-kernel, Greg KH, michaelc, dwm, darnok,
	pjones, konradr, konradr, randy.dunlap, hpa, lenb, mike.anderson,
	dwm, arjan, Andy Whitcroft


On Fri, 2008-02-08 at 23:10 -0400, Konrad Rzeszutek wrote:
> +/*
> + * Physical location of iSCSI Boot Format Table.

This is now the Virtual address, isn't it? So just drop the Physical.

> + */
> +unsigned long ibft_addr;
> +EXPORT_SYMBOL(ibft_addr);

And since it is the virtual address, there's no point in keeping it
unsigned long and doing all the casting.  It should be struct
ibft_table_hdr *, shouldn't it?

James



^ permalink raw reply	[flat|nested] 8+ messages in thread

* Re: REPOST: [PATCH] Add iSCSI iBFT support (v0.4.7)
  2008-02-09  3:10 ` REPOST: " Konrad Rzeszutek
  2008-02-09  5:17   ` James Bottomley
@ 2008-02-09  5:23   ` James Bottomley
  2008-02-09  5:32   ` James Bottomley
  2 siblings, 0 replies; 8+ messages in thread
From: James Bottomley @ 2008-02-09  5:23 UTC (permalink / raw)
  To: Konrad Rzeszutek
  Cc: Andrew Morton, linux-kernel, Greg KH, michaelc, dwm, darnok,
	pjones, konradr, konradr, randy.dunlap, hpa, lenb, mike.anderson,
	dwm, arjan, Andy Whitcroft

On Fri, 2008-02-08 at 23:10 -0400, Konrad Rzeszutek wrote:
> +       ibft_device = kmalloc(len, GFP_KERNEL);
> +       if (!ibft_device)
> +               return -ENOMEM;
> +
> +       memcpy(ibft_device, hdr, len);

This piece looks a bit odd.  you're making ibft_device an exact
duplicate of ibft_addr (which is reserved in bootmem and lives as long
as the kernel does).  I can't seem to find anywhere you actually modify
ibft_device, so why not just use the original ibft_addr here instead of
making a copy?

James



^ permalink raw reply	[flat|nested] 8+ messages in thread

* Re: REPOST: [PATCH] Add iSCSI iBFT support (v0.4.7)
  2008-02-09  3:10 ` REPOST: " Konrad Rzeszutek
  2008-02-09  5:17   ` James Bottomley
  2008-02-09  5:23   ` James Bottomley
@ 2008-02-09  5:32   ` James Bottomley
  2008-02-11 23:35     ` Konrad Rzeszutek
  2 siblings, 1 reply; 8+ messages in thread
From: James Bottomley @ 2008-02-09  5:32 UTC (permalink / raw)
  To: Konrad Rzeszutek
  Cc: Andrew Morton, linux-kernel, Greg KH, michaelc, dwm, darnok,
	pjones, konradr, konradr, randy.dunlap, hpa, lenb, mike.anderson,
	dwm, arjan, Andy Whitcroft

On Fri, 2008-02-08 at 23:10 -0400, Konrad Rzeszutek wrote:
> +       if (hdr->id == id_nic) {
> +               pci_dev = pci_get_bus_and_slot((nic->pci_bdf & 0xff00) >> 8,
> +                                              (nic->pci_bdf & 0xff));
> +               if (pci_dev) {
> +                       rc = sysfs_create_link(&ibft_kobj->kobj,
> +                                              &pci_dev->dev.kobj, "device");
> +                       pci_dev_put(pci_dev);
> +               }

pci_get_bus_and_slot fails in the presence of PCI domains, which are
getting to be fairly common even in medium sized servers ... what
happens in that case?

James



^ permalink raw reply	[flat|nested] 8+ messages in thread

* Re: REPOST: [PATCH] Add iSCSI iBFT support (v0.4.7)
  2008-02-09  5:32   ` James Bottomley
@ 2008-02-11 23:35     ` Konrad Rzeszutek
  2008-02-12  0:45       ` Greg KH
  0 siblings, 1 reply; 8+ messages in thread
From: Konrad Rzeszutek @ 2008-02-11 23:35 UTC (permalink / raw)
  To: James Bottomley
  Cc: Andrew Morton, linux-kernel, Greg KH, michaelc, dwm, darnok,
	pjones, konradr, konradr, randy.dunlap, hpa, lenb, mike.anderson,
	dwm, arjan, Andy Whitcroft

> > +       if (hdr->id == id_nic) {
> > +               pci_dev = pci_get_bus_and_slot((nic->pci_bdf & 0xff00) >>
> > 8, +                                              (nic->pci_bdf & 0xff));
> pci_get_bus_and_slot fails in the presence of PCI domains, which are
> getting to be fairly common even in medium sized servers ... what
> happens in that case?

The specification did not take that in to account. The Bus/Dev/Func 
information is only present there - no domain information.

Do you think I should implement a search functionality for this so that is 
searches through the domains for the bus/dev/func and picks the 
first found match (which might be the wrong match)? Or wait to get the spec
updated so that the iBFT will have this data?

^ permalink raw reply	[flat|nested] 8+ messages in thread

* Re: REPOST: [PATCH] Add iSCSI iBFT support (v0.4.7)
  2008-02-11 23:35     ` Konrad Rzeszutek
@ 2008-02-12  0:45       ` Greg KH
  2008-02-12  0:56         ` H. Peter Anvin
  0 siblings, 1 reply; 8+ messages in thread
From: Greg KH @ 2008-02-12  0:45 UTC (permalink / raw)
  To: Konrad Rzeszutek
  Cc: James Bottomley, Andrew Morton, linux-kernel, michaelc, dwm,
	darnok, pjones, konradr, konradr, randy.dunlap, hpa, lenb,
	mike.anderson, dwm, arjan, Andy Whitcroft

On Mon, Feb 11, 2008 at 06:35:16PM -0500, Konrad Rzeszutek wrote:
> > > +       if (hdr->id == id_nic) {
> > > +               pci_dev = pci_get_bus_and_slot((nic->pci_bdf & 0xff00) >>
> > > 8, +                                              (nic->pci_bdf & 0xff));
> > pci_get_bus_and_slot fails in the presence of PCI domains, which are
> > getting to be fairly common even in medium sized servers ... what
> > happens in that case?
> 
> The specification did not take that in to account. The Bus/Dev/Func 
> information is only present there - no domain information.

That's really broken then.  Common i386 boxes these days have multiple
PCI domains, it's not all that uncommon at all.

And almost all big 64 bit boxes have them.

> Do you think I should implement a search functionality for this so that is 
> searches through the domains for the bus/dev/func and picks the 
> first found match (which might be the wrong match)?

I'm guessing that bad things would happen if you picked the wrong pci
device, right?

> Or wait to get the spec updated so that the iBFT will have this data?

I think the spec needs to be fixed.  Or at least clarified (is it domain
0 only?)

thanks,

greg k-h

^ permalink raw reply	[flat|nested] 8+ messages in thread

* Re: REPOST: [PATCH] Add iSCSI iBFT support (v0.4.7)
  2008-02-12  0:45       ` Greg KH
@ 2008-02-12  0:56         ` H. Peter Anvin
  0 siblings, 0 replies; 8+ messages in thread
From: H. Peter Anvin @ 2008-02-12  0:56 UTC (permalink / raw)
  To: Greg KH
  Cc: Konrad Rzeszutek, James Bottomley, Andrew Morton, linux-kernel,
	michaelc, dwm, darnok, pjones, konradr, konradr, randy.dunlap,
	lenb, mike.anderson, dwm, arjan, Andy Whitcroft

Greg KH wrote:
> On Mon, Feb 11, 2008 at 06:35:16PM -0500, Konrad Rzeszutek wrote:
>>>> +       if (hdr->id == id_nic) {
>>>> +               pci_dev = pci_get_bus_and_slot((nic->pci_bdf & 0xff00) >>
>>>> 8, +                                              (nic->pci_bdf & 0xff));
>>> pci_get_bus_and_slot fails in the presence of PCI domains, which are
>>> getting to be fairly common even in medium sized servers ... what
>>> happens in that case?
>> The specification did not take that in to account. The Bus/Dev/Func 
>> information is only present there - no domain information.
> 
> That's really broken then.  Common i386 boxes these days have multiple
> PCI domains, it's not all that uncommon at all.
> 
> And almost all big 64 bit boxes have them.

Yes, but common x86 boxes can't *boot* from anything but domain 0 (in 
particular, the domain which the cf8/cfc registers on the boot processor 
activate.)

	-hpa

^ permalink raw reply	[flat|nested] 8+ messages in thread

end of thread, other threads:[~2008-02-12  1:15 UTC | newest]

Thread overview: 8+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2008-02-08 19:47 [PATCH] Add iSCSI iBFT support (v0.4.7) Konrad Rzeszutek
2008-02-09  3:10 ` REPOST: " Konrad Rzeszutek
2008-02-09  5:17   ` James Bottomley
2008-02-09  5:23   ` James Bottomley
2008-02-09  5:32   ` James Bottomley
2008-02-11 23:35     ` Konrad Rzeszutek
2008-02-12  0:45       ` Greg KH
2008-02-12  0:56         ` H. Peter Anvin

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).