linux/drivers/scsi/qla2xxx/qla_attr.c

1985 lines
51 KiB
C
Raw Normal View History

/*
* QLogic Fibre Channel HBA Driver
* Copyright (c) 2003-2008 QLogic Corporation
*
* See LICENSE.qla2xxx for copyright and licensing details.
*/
#include "qla_def.h"
#include <linux/kthread.h>
#include <linux/vmalloc.h>
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo <tj@kernel.org> Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 16:04:11 +08:00
#include <linux/slab.h>
#include <linux/delay.h>
static int qla24xx_vport_disable(struct fc_vport *, bool);
/* SYSFS attributes --------------------------------------------------------- */
static ssize_t
sysfs: add parameter "struct bin_attribute *" in .read/.write methods for sysfs binary attributes Well, first of all, I don't want to change so many files either. What I do: Adding a new parameter "struct bin_attribute *" in the .read/.write methods for the sysfs binary attributes. In fact, only the four lines change in fs/sysfs/bin.c and include/linux/sysfs.h do the real work. But I have to update all the files that use binary attributes to make them compatible with the new .read and .write methods. I'm not sure if I missed any. :( Why I do this: For a sysfs attribute, we can get a pointer pointing to the struct attribute in the .show/.store method, while we can't do this for the binary attributes. I don't know why this is different, but this does make it not so handy to use the binary attributes as the regular ones. So I think this patch is reasonable. :) Who benefits from it: The patch that exposes ACPI tables in sysfs requires such an improvement. All the table binary attributes share the same .read method. Parameter "struct bin_attribute *" is used to get the table signature and instance number which are used to distinguish different ACPI table binary attributes. Without this parameter, we need to offer different .read methods for different ACPI table binary attributes. This is impossible as there are various ACPI tables on different platforms, and we don't know what they are until they are loaded. Signed-off-by: Zhang Rui <rui.zhang@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2007-06-09 13:57:22 +08:00
qla2x00_sysfs_read_fw_dump(struct kobject *kobj,
struct bin_attribute *bin_attr,
char *buf, loff_t off, size_t count)
{
struct scsi_qla_host *vha = shost_priv(dev_to_shost(container_of(kobj,
struct device, kobj)));
struct qla_hw_data *ha = vha->hw;
if (ha->fw_dump_reading == 0)
return 0;
return memory_read_from_buffer(buf, count, &off, ha->fw_dump,
ha->fw_dump_len);
}
static ssize_t
sysfs: add parameter "struct bin_attribute *" in .read/.write methods for sysfs binary attributes Well, first of all, I don't want to change so many files either. What I do: Adding a new parameter "struct bin_attribute *" in the .read/.write methods for the sysfs binary attributes. In fact, only the four lines change in fs/sysfs/bin.c and include/linux/sysfs.h do the real work. But I have to update all the files that use binary attributes to make them compatible with the new .read and .write methods. I'm not sure if I missed any. :( Why I do this: For a sysfs attribute, we can get a pointer pointing to the struct attribute in the .show/.store method, while we can't do this for the binary attributes. I don't know why this is different, but this does make it not so handy to use the binary attributes as the regular ones. So I think this patch is reasonable. :) Who benefits from it: The patch that exposes ACPI tables in sysfs requires such an improvement. All the table binary attributes share the same .read method. Parameter "struct bin_attribute *" is used to get the table signature and instance number which are used to distinguish different ACPI table binary attributes. Without this parameter, we need to offer different .read methods for different ACPI table binary attributes. This is impossible as there are various ACPI tables on different platforms, and we don't know what they are until they are loaded. Signed-off-by: Zhang Rui <rui.zhang@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2007-06-09 13:57:22 +08:00
qla2x00_sysfs_write_fw_dump(struct kobject *kobj,
struct bin_attribute *bin_attr,
char *buf, loff_t off, size_t count)
{
struct scsi_qla_host *vha = shost_priv(dev_to_shost(container_of(kobj,
struct device, kobj)));
struct qla_hw_data *ha = vha->hw;
int reading;
if (IS_QLA82XX(ha)) {
DEBUG2(qla_printk(KERN_INFO, ha,
"Firmware dump not supported for ISP82xx\n"));
return count;
}
if (off != 0)
return (0);
reading = simple_strtol(buf, NULL, 10);
switch (reading) {
case 0:
if (!ha->fw_dump_reading)
break;
qla_printk(KERN_INFO, ha,
"Firmware dump cleared on (%ld).\n", vha->host_no);
ha->fw_dump_reading = 0;
ha->fw_dumped = 0;
break;
case 1:
if (ha->fw_dumped && !ha->fw_dump_reading) {
ha->fw_dump_reading = 1;
qla_printk(KERN_INFO, ha,
"Raw firmware dump ready for read on (%ld).\n",
vha->host_no);
}
break;
case 2:
qla2x00_alloc_fw_dump(vha);
break;
case 3:
qla2x00_system_error(vha);
break;
}
return (count);
}
static struct bin_attribute sysfs_fw_dump_attr = {
.attr = {
.name = "fw_dump",
.mode = S_IRUSR | S_IWUSR,
},
.size = 0,
.read = qla2x00_sysfs_read_fw_dump,
.write = qla2x00_sysfs_write_fw_dump,
};
static ssize_t
sysfs: add parameter "struct bin_attribute *" in .read/.write methods for sysfs binary attributes Well, first of all, I don't want to change so many files either. What I do: Adding a new parameter "struct bin_attribute *" in the .read/.write methods for the sysfs binary attributes. In fact, only the four lines change in fs/sysfs/bin.c and include/linux/sysfs.h do the real work. But I have to update all the files that use binary attributes to make them compatible with the new .read and .write methods. I'm not sure if I missed any. :( Why I do this: For a sysfs attribute, we can get a pointer pointing to the struct attribute in the .show/.store method, while we can't do this for the binary attributes. I don't know why this is different, but this does make it not so handy to use the binary attributes as the regular ones. So I think this patch is reasonable. :) Who benefits from it: The patch that exposes ACPI tables in sysfs requires such an improvement. All the table binary attributes share the same .read method. Parameter "struct bin_attribute *" is used to get the table signature and instance number which are used to distinguish different ACPI table binary attributes. Without this parameter, we need to offer different .read methods for different ACPI table binary attributes. This is impossible as there are various ACPI tables on different platforms, and we don't know what they are until they are loaded. Signed-off-by: Zhang Rui <rui.zhang@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2007-06-09 13:57:22 +08:00
qla2x00_sysfs_read_nvram(struct kobject *kobj,
struct bin_attribute *bin_attr,
char *buf, loff_t off, size_t count)
{
struct scsi_qla_host *vha = shost_priv(dev_to_shost(container_of(kobj,
struct device, kobj)));
struct qla_hw_data *ha = vha->hw;
if (!capable(CAP_SYS_ADMIN))
return 0;
if (IS_NOCACHE_VPD_TYPE(ha))
ha->isp_ops->read_optrom(vha, ha->nvram, ha->flt_region_nvram << 2,
ha->nvram_size);
return memory_read_from_buffer(buf, count, &off, ha->nvram,
ha->nvram_size);
}
static ssize_t
sysfs: add parameter "struct bin_attribute *" in .read/.write methods for sysfs binary attributes Well, first of all, I don't want to change so many files either. What I do: Adding a new parameter "struct bin_attribute *" in the .read/.write methods for the sysfs binary attributes. In fact, only the four lines change in fs/sysfs/bin.c and include/linux/sysfs.h do the real work. But I have to update all the files that use binary attributes to make them compatible with the new .read and .write methods. I'm not sure if I missed any. :( Why I do this: For a sysfs attribute, we can get a pointer pointing to the struct attribute in the .show/.store method, while we can't do this for the binary attributes. I don't know why this is different, but this does make it not so handy to use the binary attributes as the regular ones. So I think this patch is reasonable. :) Who benefits from it: The patch that exposes ACPI tables in sysfs requires such an improvement. All the table binary attributes share the same .read method. Parameter "struct bin_attribute *" is used to get the table signature and instance number which are used to distinguish different ACPI table binary attributes. Without this parameter, we need to offer different .read methods for different ACPI table binary attributes. This is impossible as there are various ACPI tables on different platforms, and we don't know what they are until they are loaded. Signed-off-by: Zhang Rui <rui.zhang@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2007-06-09 13:57:22 +08:00
qla2x00_sysfs_write_nvram(struct kobject *kobj,
struct bin_attribute *bin_attr,
char *buf, loff_t off, size_t count)
{
struct scsi_qla_host *vha = shost_priv(dev_to_shost(container_of(kobj,
struct device, kobj)));
struct qla_hw_data *ha = vha->hw;
uint16_t cnt;
if (!capable(CAP_SYS_ADMIN) || off != 0 || count != ha->nvram_size ||
!ha->isp_ops->write_nvram)
return 0;
/* Checksum NVRAM. */
if (IS_FWI2_CAPABLE(ha)) {
uint32_t *iter;
uint32_t chksum;
iter = (uint32_t *)buf;
chksum = 0;
for (cnt = 0; cnt < ((count >> 2) - 1); cnt++)
chksum += le32_to_cpu(*iter++);
chksum = ~chksum + 1;
*iter = cpu_to_le32(chksum);
} else {
uint8_t *iter;
uint8_t chksum;
iter = (uint8_t *)buf;
chksum = 0;
for (cnt = 0; cnt < count - 1; cnt++)
chksum += *iter++;
chksum = ~chksum + 1;
*iter = chksum;
}
if (qla2x00_wait_for_hba_online(vha) != QLA_SUCCESS) {
qla_printk(KERN_WARNING, ha,
"HBA not online, failing NVRAM update.\n");
return -EAGAIN;
}
/* Write NVRAM. */
ha->isp_ops->write_nvram(vha, (uint8_t *)buf, ha->nvram_base, count);
ha->isp_ops->read_nvram(vha, (uint8_t *)ha->nvram, ha->nvram_base,
count);
/* NVRAM settings take effect immediately. */
set_bit(ISP_ABORT_NEEDED, &vha->dpc_flags);
qla2xxx_wake_dpc(vha);
qla2x00_wait_for_chip_reset(vha);
return (count);
}
static struct bin_attribute sysfs_nvram_attr = {
.attr = {
.name = "nvram",
.mode = S_IRUSR | S_IWUSR,
},
.size = 512,
.read = qla2x00_sysfs_read_nvram,
.write = qla2x00_sysfs_write_nvram,
};
static ssize_t
sysfs: add parameter "struct bin_attribute *" in .read/.write methods for sysfs binary attributes Well, first of all, I don't want to change so many files either. What I do: Adding a new parameter "struct bin_attribute *" in the .read/.write methods for the sysfs binary attributes. In fact, only the four lines change in fs/sysfs/bin.c and include/linux/sysfs.h do the real work. But I have to update all the files that use binary attributes to make them compatible with the new .read and .write methods. I'm not sure if I missed any. :( Why I do this: For a sysfs attribute, we can get a pointer pointing to the struct attribute in the .show/.store method, while we can't do this for the binary attributes. I don't know why this is different, but this does make it not so handy to use the binary attributes as the regular ones. So I think this patch is reasonable. :) Who benefits from it: The patch that exposes ACPI tables in sysfs requires such an improvement. All the table binary attributes share the same .read method. Parameter "struct bin_attribute *" is used to get the table signature and instance number which are used to distinguish different ACPI table binary attributes. Without this parameter, we need to offer different .read methods for different ACPI table binary attributes. This is impossible as there are various ACPI tables on different platforms, and we don't know what they are until they are loaded. Signed-off-by: Zhang Rui <rui.zhang@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2007-06-09 13:57:22 +08:00
qla2x00_sysfs_read_optrom(struct kobject *kobj,
struct bin_attribute *bin_attr,
char *buf, loff_t off, size_t count)
{
struct scsi_qla_host *vha = shost_priv(dev_to_shost(container_of(kobj,
struct device, kobj)));
struct qla_hw_data *ha = vha->hw;
if (ha->optrom_state != QLA_SREADING)
return 0;
return memory_read_from_buffer(buf, count, &off, ha->optrom_buffer,
ha->optrom_region_size);
}
static ssize_t
sysfs: add parameter "struct bin_attribute *" in .read/.write methods for sysfs binary attributes Well, first of all, I don't want to change so many files either. What I do: Adding a new parameter "struct bin_attribute *" in the .read/.write methods for the sysfs binary attributes. In fact, only the four lines change in fs/sysfs/bin.c and include/linux/sysfs.h do the real work. But I have to update all the files that use binary attributes to make them compatible with the new .read and .write methods. I'm not sure if I missed any. :( Why I do this: For a sysfs attribute, we can get a pointer pointing to the struct attribute in the .show/.store method, while we can't do this for the binary attributes. I don't know why this is different, but this does make it not so handy to use the binary attributes as the regular ones. So I think this patch is reasonable. :) Who benefits from it: The patch that exposes ACPI tables in sysfs requires such an improvement. All the table binary attributes share the same .read method. Parameter "struct bin_attribute *" is used to get the table signature and instance number which are used to distinguish different ACPI table binary attributes. Without this parameter, we need to offer different .read methods for different ACPI table binary attributes. This is impossible as there are various ACPI tables on different platforms, and we don't know what they are until they are loaded. Signed-off-by: Zhang Rui <rui.zhang@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2007-06-09 13:57:22 +08:00
qla2x00_sysfs_write_optrom(struct kobject *kobj,
struct bin_attribute *bin_attr,
char *buf, loff_t off, size_t count)
{
struct scsi_qla_host *vha = shost_priv(dev_to_shost(container_of(kobj,
struct device, kobj)));
struct qla_hw_data *ha = vha->hw;
if (ha->optrom_state != QLA_SWRITING)
return -EINVAL;
if (off > ha->optrom_region_size)
return -ERANGE;
if (off + count > ha->optrom_region_size)
count = ha->optrom_region_size - off;
memcpy(&ha->optrom_buffer[off], buf, count);
return count;
}
static struct bin_attribute sysfs_optrom_attr = {
.attr = {
.name = "optrom",
.mode = S_IRUSR | S_IWUSR,
},
.size = 0,
.read = qla2x00_sysfs_read_optrom,
.write = qla2x00_sysfs_write_optrom,
};
static ssize_t
sysfs: add parameter "struct bin_attribute *" in .read/.write methods for sysfs binary attributes Well, first of all, I don't want to change so many files either. What I do: Adding a new parameter "struct bin_attribute *" in the .read/.write methods for the sysfs binary attributes. In fact, only the four lines change in fs/sysfs/bin.c and include/linux/sysfs.h do the real work. But I have to update all the files that use binary attributes to make them compatible with the new .read and .write methods. I'm not sure if I missed any. :( Why I do this: For a sysfs attribute, we can get a pointer pointing to the struct attribute in the .show/.store method, while we can't do this for the binary attributes. I don't know why this is different, but this does make it not so handy to use the binary attributes as the regular ones. So I think this patch is reasonable. :) Who benefits from it: The patch that exposes ACPI tables in sysfs requires such an improvement. All the table binary attributes share the same .read method. Parameter "struct bin_attribute *" is used to get the table signature and instance number which are used to distinguish different ACPI table binary attributes. Without this parameter, we need to offer different .read methods for different ACPI table binary attributes. This is impossible as there are various ACPI tables on different platforms, and we don't know what they are until they are loaded. Signed-off-by: Zhang Rui <rui.zhang@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2007-06-09 13:57:22 +08:00
qla2x00_sysfs_write_optrom_ctl(struct kobject *kobj,
struct bin_attribute *bin_attr,
char *buf, loff_t off, size_t count)
{
struct scsi_qla_host *vha = shost_priv(dev_to_shost(container_of(kobj,
struct device, kobj)));
struct qla_hw_data *ha = vha->hw;
uint32_t start = 0;
uint32_t size = ha->optrom_size;
int val, valid;
if (off)
return 0;
if (unlikely(pci_channel_offline(ha->pdev)))
return 0;
if (sscanf(buf, "%d:%x:%x", &val, &start, &size) < 1)
return -EINVAL;
if (start > ha->optrom_size)
return -EINVAL;
switch (val) {
case 0:
if (ha->optrom_state != QLA_SREADING &&
ha->optrom_state != QLA_SWRITING)
break;
ha->optrom_state = QLA_SWAITING;
DEBUG2(qla_printk(KERN_INFO, ha,
"Freeing flash region allocation -- 0x%x bytes.\n",
ha->optrom_region_size));
vfree(ha->optrom_buffer);
ha->optrom_buffer = NULL;
break;
case 1:
if (ha->optrom_state != QLA_SWAITING)
break;
ha->optrom_region_start = start;
ha->optrom_region_size = start + size > ha->optrom_size ?
ha->optrom_size - start : size;
ha->optrom_state = QLA_SREADING;
ha->optrom_buffer = vmalloc(ha->optrom_region_size);
if (ha->optrom_buffer == NULL) {
qla_printk(KERN_WARNING, ha,
"Unable to allocate memory for optrom retrieval "
"(%x).\n", ha->optrom_region_size);
ha->optrom_state = QLA_SWAITING;
return count;
}
if (qla2x00_wait_for_hba_online(vha) != QLA_SUCCESS) {
qla_printk(KERN_WARNING, ha,
"HBA not online, failing NVRAM update.\n");
return -EAGAIN;
}
DEBUG2(qla_printk(KERN_INFO, ha,
"Reading flash region -- 0x%x/0x%x.\n",
ha->optrom_region_start, ha->optrom_region_size));
memset(ha->optrom_buffer, 0, ha->optrom_region_size);
ha->isp_ops->read_optrom(vha, ha->optrom_buffer,
ha->optrom_region_start, ha->optrom_region_size);
break;
case 2:
if (ha->optrom_state != QLA_SWAITING)
break;
/*
* We need to be more restrictive on which FLASH regions are
* allowed to be updated via user-space. Regions accessible
* via this method include:
*
* ISP21xx/ISP22xx/ISP23xx type boards:
*
* 0x000000 -> 0x020000 -- Boot code.
*
* ISP2322/ISP24xx type boards:
*
* 0x000000 -> 0x07ffff -- Boot code.
* 0x080000 -> 0x0fffff -- Firmware.
*
* ISP25xx type boards:
*
* 0x000000 -> 0x07ffff -- Boot code.
* 0x080000 -> 0x0fffff -- Firmware.
* 0x120000 -> 0x12ffff -- VPD and HBA parameters.
*/
valid = 0;
if (ha->optrom_size == OPTROM_SIZE_2300 && start == 0)
valid = 1;
else if (start == (ha->flt_region_boot * 4) ||
start == (ha->flt_region_fw * 4))
valid = 1;
else if (IS_QLA25XX(ha) || IS_QLA8XXX_TYPE(ha))
valid = 1;
if (!valid) {
qla_printk(KERN_WARNING, ha,
"Invalid start region 0x%x/0x%x.\n", start, size);
return -EINVAL;
}
ha->optrom_region_start = start;
ha->optrom_region_size = start + size > ha->optrom_size ?
ha->optrom_size - start : size;
ha->optrom_state = QLA_SWRITING;
ha->optrom_buffer = vmalloc(ha->optrom_region_size);
if (ha->optrom_buffer == NULL) {
qla_printk(KERN_WARNING, ha,
"Unable to allocate memory for optrom update "
"(%x).\n", ha->optrom_region_size);
ha->optrom_state = QLA_SWAITING;
return count;
}
DEBUG2(qla_printk(KERN_INFO, ha,
"Staging flash region write -- 0x%x/0x%x.\n",
ha->optrom_region_start, ha->optrom_region_size));
memset(ha->optrom_buffer, 0, ha->optrom_region_size);
break;
case 3:
if (ha->optrom_state != QLA_SWRITING)
break;
if (qla2x00_wait_for_hba_online(vha) != QLA_SUCCESS) {
qla_printk(KERN_WARNING, ha,
"HBA not online, failing flash update.\n");
return -EAGAIN;
}
DEBUG2(qla_printk(KERN_INFO, ha,
"Writing flash region -- 0x%x/0x%x.\n",
ha->optrom_region_start, ha->optrom_region_size));
ha->isp_ops->write_optrom(vha, ha->optrom_buffer,
ha->optrom_region_start, ha->optrom_region_size);
break;
default:
count = -EINVAL;
}
return count;
}
static struct bin_attribute sysfs_optrom_ctl_attr = {
.attr = {
.name = "optrom_ctl",
.mode = S_IWUSR,
},
.size = 0,
.write = qla2x00_sysfs_write_optrom_ctl,
};
static ssize_t
sysfs: add parameter "struct bin_attribute *" in .read/.write methods for sysfs binary attributes Well, first of all, I don't want to change so many files either. What I do: Adding a new parameter "struct bin_attribute *" in the .read/.write methods for the sysfs binary attributes. In fact, only the four lines change in fs/sysfs/bin.c and include/linux/sysfs.h do the real work. But I have to update all the files that use binary attributes to make them compatible with the new .read and .write methods. I'm not sure if I missed any. :( Why I do this: For a sysfs attribute, we can get a pointer pointing to the struct attribute in the .show/.store method, while we can't do this for the binary attributes. I don't know why this is different, but this does make it not so handy to use the binary attributes as the regular ones. So I think this patch is reasonable. :) Who benefits from it: The patch that exposes ACPI tables in sysfs requires such an improvement. All the table binary attributes share the same .read method. Parameter "struct bin_attribute *" is used to get the table signature and instance number which are used to distinguish different ACPI table binary attributes. Without this parameter, we need to offer different .read methods for different ACPI table binary attributes. This is impossible as there are various ACPI tables on different platforms, and we don't know what they are until they are loaded. Signed-off-by: Zhang Rui <rui.zhang@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2007-06-09 13:57:22 +08:00
qla2x00_sysfs_read_vpd(struct kobject *kobj,
struct bin_attribute *bin_attr,
char *buf, loff_t off, size_t count)
{
struct scsi_qla_host *vha = shost_priv(dev_to_shost(container_of(kobj,
struct device, kobj)));
struct qla_hw_data *ha = vha->hw;
if (unlikely(pci_channel_offline(ha->pdev)))
return 0;
if (!capable(CAP_SYS_ADMIN))
return 0;
if (IS_NOCACHE_VPD_TYPE(ha))
ha->isp_ops->read_optrom(vha, ha->vpd, ha->flt_region_vpd << 2,
ha->vpd_size);
return memory_read_from_buffer(buf, count, &off, ha->vpd, ha->vpd_size);
}
static ssize_t
sysfs: add parameter "struct bin_attribute *" in .read/.write methods for sysfs binary attributes Well, first of all, I don't want to change so many files either. What I do: Adding a new parameter "struct bin_attribute *" in the .read/.write methods for the sysfs binary attributes. In fact, only the four lines change in fs/sysfs/bin.c and include/linux/sysfs.h do the real work. But I have to update all the files that use binary attributes to make them compatible with the new .read and .write methods. I'm not sure if I missed any. :( Why I do this: For a sysfs attribute, we can get a pointer pointing to the struct attribute in the .show/.store method, while we can't do this for the binary attributes. I don't know why this is different, but this does make it not so handy to use the binary attributes as the regular ones. So I think this patch is reasonable. :) Who benefits from it: The patch that exposes ACPI tables in sysfs requires such an improvement. All the table binary attributes share the same .read method. Parameter "struct bin_attribute *" is used to get the table signature and instance number which are used to distinguish different ACPI table binary attributes. Without this parameter, we need to offer different .read methods for different ACPI table binary attributes. This is impossible as there are various ACPI tables on different platforms, and we don't know what they are until they are loaded. Signed-off-by: Zhang Rui <rui.zhang@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2007-06-09 13:57:22 +08:00
qla2x00_sysfs_write_vpd(struct kobject *kobj,
struct bin_attribute *bin_attr,
char *buf, loff_t off, size_t count)
{
struct scsi_qla_host *vha = shost_priv(dev_to_shost(container_of(kobj,
struct device, kobj)));
struct qla_hw_data *ha = vha->hw;
uint8_t *tmp_data;
if (unlikely(pci_channel_offline(ha->pdev)))
return 0;
if (!capable(CAP_SYS_ADMIN) || off != 0 || count != ha->vpd_size ||
!ha->isp_ops->write_nvram)
return 0;
if (qla2x00_wait_for_hba_online(vha) != QLA_SUCCESS) {
qla_printk(KERN_WARNING, ha,
"HBA not online, failing VPD update.\n");
return -EAGAIN;
}
/* Write NVRAM. */
ha->isp_ops->write_nvram(vha, (uint8_t *)buf, ha->vpd_base, count);
ha->isp_ops->read_nvram(vha, (uint8_t *)ha->vpd, ha->vpd_base, count);
/* Update flash version information for 4Gb & above. */
if (!IS_FWI2_CAPABLE(ha))
goto done;
tmp_data = vmalloc(256);
if (!tmp_data) {
qla_printk(KERN_WARNING, ha,
"Unable to allocate memory for VPD information update.\n");
goto done;
}
ha->isp_ops->get_flash_version(vha, tmp_data);
vfree(tmp_data);
done:
return count;
}
static struct bin_attribute sysfs_vpd_attr = {
.attr = {
.name = "vpd",
.mode = S_IRUSR | S_IWUSR,
},
.size = 0,
.read = qla2x00_sysfs_read_vpd,
.write = qla2x00_sysfs_write_vpd,
};
static ssize_t
sysfs: add parameter "struct bin_attribute *" in .read/.write methods for sysfs binary attributes Well, first of all, I don't want to change so many files either. What I do: Adding a new parameter "struct bin_attribute *" in the .read/.write methods for the sysfs binary attributes. In fact, only the four lines change in fs/sysfs/bin.c and include/linux/sysfs.h do the real work. But I have to update all the files that use binary attributes to make them compatible with the new .read and .write methods. I'm not sure if I missed any. :( Why I do this: For a sysfs attribute, we can get a pointer pointing to the struct attribute in the .show/.store method, while we can't do this for the binary attributes. I don't know why this is different, but this does make it not so handy to use the binary attributes as the regular ones. So I think this patch is reasonable. :) Who benefits from it: The patch that exposes ACPI tables in sysfs requires such an improvement. All the table binary attributes share the same .read method. Parameter "struct bin_attribute *" is used to get the table signature and instance number which are used to distinguish different ACPI table binary attributes. Without this parameter, we need to offer different .read methods for different ACPI table binary attributes. This is impossible as there are various ACPI tables on different platforms, and we don't know what they are until they are loaded. Signed-off-by: Zhang Rui <rui.zhang@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2007-06-09 13:57:22 +08:00
qla2x00_sysfs_read_sfp(struct kobject *kobj,
struct bin_attribute *bin_attr,
char *buf, loff_t off, size_t count)
{
struct scsi_qla_host *vha = shost_priv(dev_to_shost(container_of(kobj,
struct device, kobj)));
struct qla_hw_data *ha = vha->hw;
uint16_t iter, addr, offset;
int rval;
if (!capable(CAP_SYS_ADMIN) || off != 0 || count != SFP_DEV_SIZE * 2)
return 0;
if (ha->sfp_data)
goto do_read;
ha->sfp_data = dma_pool_alloc(ha->s_dma_pool, GFP_KERNEL,
&ha->sfp_data_dma);
if (!ha->sfp_data) {
qla_printk(KERN_WARNING, ha,
"Unable to allocate memory for SFP read-data.\n");
return 0;
}
do_read:
memset(ha->sfp_data, 0, SFP_BLOCK_SIZE);
addr = 0xa0;
for (iter = 0, offset = 0; iter < (SFP_DEV_SIZE * 2) / SFP_BLOCK_SIZE;
iter++, offset += SFP_BLOCK_SIZE) {
if (iter == 4) {
/* Skip to next device address. */
addr = 0xa2;
offset = 0;
}
rval = qla2x00_read_sfp(vha, ha->sfp_data_dma, addr, offset,
SFP_BLOCK_SIZE);
if (rval != QLA_SUCCESS) {
qla_printk(KERN_WARNING, ha,
"Unable to read SFP data (%x/%x/%x).\n", rval,
addr, offset);
count = 0;
break;
}
memcpy(buf, ha->sfp_data, SFP_BLOCK_SIZE);
buf += SFP_BLOCK_SIZE;
}
return count;
}
static struct bin_attribute sysfs_sfp_attr = {
.attr = {
.name = "sfp",
.mode = S_IRUSR | S_IWUSR,
},
.size = SFP_DEV_SIZE * 2,
.read = qla2x00_sysfs_read_sfp,
};
static ssize_t
qla2x00_sysfs_write_reset(struct kobject *kobj,
struct bin_attribute *bin_attr,
char *buf, loff_t off, size_t count)
{
struct scsi_qla_host *vha = shost_priv(dev_to_shost(container_of(kobj,
struct device, kobj)));
struct qla_hw_data *ha = vha->hw;
struct scsi_qla_host *base_vha = pci_get_drvdata(ha->pdev);
int type;
if (off != 0)
return 0;
type = simple_strtol(buf, NULL, 10);
switch (type) {
case 0x2025c:
qla_printk(KERN_INFO, ha,
"Issuing ISP reset on (%ld).\n", vha->host_no);
scsi_block_requests(vha->host);
set_bit(ISP_ABORT_NEEDED, &vha->dpc_flags);
qla2xxx_wake_dpc(vha);
qla2x00_wait_for_chip_reset(vha);
scsi_unblock_requests(vha->host);
break;
case 0x2025d:
if (!IS_QLA81XX(ha))
break;
qla_printk(KERN_INFO, ha,
"Issuing MPI reset on (%ld).\n", vha->host_no);
/* Make sure FC side is not in reset */
qla2x00_wait_for_hba_online(vha);
/* Issue MPI reset */
scsi_block_requests(vha->host);
if (qla81xx_restart_mpi_firmware(vha) != QLA_SUCCESS)
qla_printk(KERN_WARNING, ha,
"MPI reset failed on (%ld).\n", vha->host_no);
scsi_unblock_requests(vha->host);
break;
case 0x2025e:
if (!IS_QLA82XX(ha) || vha != base_vha) {
qla_printk(KERN_INFO, ha,
"FCoE ctx reset not supported for host%ld.\n",
vha->host_no);
return count;
}
qla_printk(KERN_INFO, ha,
"Issuing FCoE CTX reset on host%ld.\n", vha->host_no);
set_bit(FCOE_CTX_RESET_NEEDED, &vha->dpc_flags);
qla2xxx_wake_dpc(vha);
qla2x00_wait_for_fcoe_ctx_reset(vha);
break;
}
return count;
}
static struct bin_attribute sysfs_reset_attr = {
.attr = {
.name = "reset",
.mode = S_IWUSR,
},
.size = 0,
.write = qla2x00_sysfs_write_reset,
};
static ssize_t
qla2x00_sysfs_write_edc(struct kobject *kobj,
struct bin_attribute *bin_attr,
char *buf, loff_t off, size_t count)
{
struct scsi_qla_host *vha = shost_priv(dev_to_shost(container_of(kobj,
struct device, kobj)));
struct qla_hw_data *ha = vha->hw;
uint16_t dev, adr, opt, len;
int rval;
ha->edc_data_len = 0;
if (!capable(CAP_SYS_ADMIN) || off != 0 || count < 8)
return 0;
if (!ha->edc_data) {
ha->edc_data = dma_pool_alloc(ha->s_dma_pool, GFP_KERNEL,
&ha->edc_data_dma);
if (!ha->edc_data) {
DEBUG2(qla_printk(KERN_INFO, ha,
"Unable to allocate memory for EDC write.\n"));
return 0;
}
}
dev = le16_to_cpup((void *)&buf[0]);
adr = le16_to_cpup((void *)&buf[2]);
opt = le16_to_cpup((void *)&buf[4]);
len = le16_to_cpup((void *)&buf[6]);
if (!(opt & BIT_0))
if (len == 0 || len > DMA_POOL_SIZE || len > count - 8)
return -EINVAL;
memcpy(ha->edc_data, &buf[8], len);
rval = qla2x00_write_edc(vha, dev, adr, ha->edc_data_dma,
ha->edc_data, len, opt);
if (rval != QLA_SUCCESS) {
DEBUG2(qla_printk(KERN_INFO, ha,
"Unable to write EDC (%x) %02x:%02x:%04x:%02x:%02x.\n",
rval, dev, adr, opt, len, *buf));
return 0;
}
return count;
}
static struct bin_attribute sysfs_edc_attr = {
.attr = {
.name = "edc",
.mode = S_IWUSR,
},
.size = 0,
.write = qla2x00_sysfs_write_edc,
};
static ssize_t
qla2x00_sysfs_write_edc_status(struct kobject *kobj,
struct bin_attribute *bin_attr,
char *buf, loff_t off, size_t count)
{
struct scsi_qla_host *vha = shost_priv(dev_to_shost(container_of(kobj,
struct device, kobj)));
struct qla_hw_data *ha = vha->hw;
uint16_t dev, adr, opt, len;
int rval;
ha->edc_data_len = 0;
if (!capable(CAP_SYS_ADMIN) || off != 0 || count < 8)
return 0;
if (!ha->edc_data) {
ha->edc_data = dma_pool_alloc(ha->s_dma_pool, GFP_KERNEL,
&ha->edc_data_dma);
if (!ha->edc_data) {
DEBUG2(qla_printk(KERN_INFO, ha,
"Unable to allocate memory for EDC status.\n"));
return 0;
}
}
dev = le16_to_cpup((void *)&buf[0]);
adr = le16_to_cpup((void *)&buf[2]);
opt = le16_to_cpup((void *)&buf[4]);
len = le16_to_cpup((void *)&buf[6]);
if (!(opt & BIT_0))
if (len == 0 || len > DMA_POOL_SIZE)
return -EINVAL;
memset(ha->edc_data, 0, len);
rval = qla2x00_read_edc(vha, dev, adr, ha->edc_data_dma,
ha->edc_data, len, opt);
if (rval != QLA_SUCCESS) {
DEBUG2(qla_printk(KERN_INFO, ha,
"Unable to write EDC status (%x) %02x:%02x:%04x:%02x.\n",
rval, dev, adr, opt, len));
return 0;
}
ha->edc_data_len = len;
return count;
}
static ssize_t
qla2x00_sysfs_read_edc_status(struct kobject *kobj,
struct bin_attribute *bin_attr,
char *buf, loff_t off, size_t count)
{
struct scsi_qla_host *vha = shost_priv(dev_to_shost(container_of(kobj,
struct device, kobj)));
struct qla_hw_data *ha = vha->hw;
if (!capable(CAP_SYS_ADMIN) || off != 0 || count == 0)
return 0;
if (!ha->edc_data || ha->edc_data_len == 0 || ha->edc_data_len > count)
return -EINVAL;
memcpy(buf, ha->edc_data, ha->edc_data_len);
return ha->edc_data_len;
}
static struct bin_attribute sysfs_edc_status_attr = {
.attr = {
.name = "edc_status",
.mode = S_IRUSR | S_IWUSR,
},
.size = 0,
.write = qla2x00_sysfs_write_edc_status,
.read = qla2x00_sysfs_read_edc_status,
};
static ssize_t
qla2x00_sysfs_read_xgmac_stats(struct kobject *kobj,
struct bin_attribute *bin_attr,
char *buf, loff_t off, size_t count)
{
struct scsi_qla_host *vha = shost_priv(dev_to_shost(container_of(kobj,
struct device, kobj)));
struct qla_hw_data *ha = vha->hw;
int rval;
uint16_t actual_size;
if (!capable(CAP_SYS_ADMIN) || off != 0 || count > XGMAC_DATA_SIZE)
return 0;
if (ha->xgmac_data)
goto do_read;
ha->xgmac_data = dma_alloc_coherent(&ha->pdev->dev, XGMAC_DATA_SIZE,
&ha->xgmac_data_dma, GFP_KERNEL);
if (!ha->xgmac_data) {
qla_printk(KERN_WARNING, ha,
"Unable to allocate memory for XGMAC read-data.\n");
return 0;
}
do_read:
actual_size = 0;
memset(ha->xgmac_data, 0, XGMAC_DATA_SIZE);
rval = qla2x00_get_xgmac_stats(vha, ha->xgmac_data_dma,
XGMAC_DATA_SIZE, &actual_size);
if (rval != QLA_SUCCESS) {
qla_printk(KERN_WARNING, ha,
"Unable to read XGMAC data (%x).\n", rval);
count = 0;
}
count = actual_size > count ? count: actual_size;
memcpy(buf, ha->xgmac_data, count);
return count;
}
static struct bin_attribute sysfs_xgmac_stats_attr = {
.attr = {
.name = "xgmac_stats",
.mode = S_IRUSR,
},
.size = 0,
.read = qla2x00_sysfs_read_xgmac_stats,
};
static ssize_t
qla2x00_sysfs_read_dcbx_tlv(struct kobject *kobj,
struct bin_attribute *bin_attr,
char *buf, loff_t off, size_t count)
{
struct scsi_qla_host *vha = shost_priv(dev_to_shost(container_of(kobj,
struct device, kobj)));
struct qla_hw_data *ha = vha->hw;
int rval;
uint16_t actual_size;
if (!capable(CAP_SYS_ADMIN) || off != 0 || count > DCBX_TLV_DATA_SIZE)
return 0;
if (ha->dcbx_tlv)
goto do_read;
ha->dcbx_tlv = dma_alloc_coherent(&ha->pdev->dev, DCBX_TLV_DATA_SIZE,
&ha->dcbx_tlv_dma, GFP_KERNEL);
if (!ha->dcbx_tlv) {
qla_printk(KERN_WARNING, ha,
"Unable to allocate memory for DCBX TLV read-data.\n");
return 0;
}
do_read:
actual_size = 0;
memset(ha->dcbx_tlv, 0, DCBX_TLV_DATA_SIZE);
rval = qla2x00_get_dcbx_params(vha, ha->dcbx_tlv_dma,
DCBX_TLV_DATA_SIZE);
if (rval != QLA_SUCCESS) {
qla_printk(KERN_WARNING, ha,
"Unable to read DCBX TLV data (%x).\n", rval);
count = 0;
}
memcpy(buf, ha->dcbx_tlv, count);
return count;
}
static struct bin_attribute sysfs_dcbx_tlv_attr = {
.attr = {
.name = "dcbx_tlv",
.mode = S_IRUSR,
},
.size = 0,
.read = qla2x00_sysfs_read_dcbx_tlv,
};
static struct sysfs_entry {
char *name;
struct bin_attribute *attr;
int is4GBp_only;
} bin_file_entries[] = {
{ "fw_dump", &sysfs_fw_dump_attr, },
{ "nvram", &sysfs_nvram_attr, },
{ "optrom", &sysfs_optrom_attr, },
{ "optrom_ctl", &sysfs_optrom_ctl_attr, },
{ "vpd", &sysfs_vpd_attr, 1 },
{ "sfp", &sysfs_sfp_attr, 1 },
{ "reset", &sysfs_reset_attr, },
{ "edc", &sysfs_edc_attr, 2 },
{ "edc_status", &sysfs_edc_status_attr, 2 },
{ "xgmac_stats", &sysfs_xgmac_stats_attr, 3 },
{ "dcbx_tlv", &sysfs_dcbx_tlv_attr, 3 },
{ NULL },
};
void
qla2x00_alloc_sysfs_attr(scsi_qla_host_t *vha)
{
struct Scsi_Host *host = vha->host;
struct sysfs_entry *iter;
int ret;
for (iter = bin_file_entries; iter->name; iter++) {
if (iter->is4GBp_only && !IS_FWI2_CAPABLE(vha->hw))
continue;
if (iter->is4GBp_only == 2 && !IS_QLA25XX(vha->hw))
continue;
if (iter->is4GBp_only == 3 && !(IS_QLA8XXX_TYPE(vha->hw)))
continue;
ret = sysfs_create_bin_file(&host->shost_gendev.kobj,
iter->attr);
if (ret)
qla_printk(KERN_INFO, vha->hw,
"Unable to create sysfs %s binary attribute "
"(%d).\n", iter->name, ret);
}
}
void
qla2x00_free_sysfs_attr(scsi_qla_host_t *vha)
{
struct Scsi_Host *host = vha->host;
struct sysfs_entry *iter;
struct qla_hw_data *ha = vha->hw;
for (iter = bin_file_entries; iter->name; iter++) {
if (iter->is4GBp_only && !IS_FWI2_CAPABLE(ha))
continue;
if (iter->is4GBp_only == 2 && !IS_QLA25XX(ha))
continue;
if (iter->is4GBp_only == 3 && !!(IS_QLA8XXX_TYPE(vha->hw)))
continue;
sysfs_remove_bin_file(&host->shost_gendev.kobj,
iter->attr);
}
if (ha->beacon_blink_led == 1)
ha->isp_ops->beacon_off(vha);
}
/* Scsi_Host attributes. */
static ssize_t
qla2x00_drvr_version_show(struct device *dev,
struct device_attribute *attr, char *buf)
{
return snprintf(buf, PAGE_SIZE, "%s\n", qla2x00_version_str);
}
static ssize_t
qla2x00_fw_version_show(struct device *dev,
struct device_attribute *attr, char *buf)
{
scsi_qla_host_t *vha = shost_priv(class_to_shost(dev));
struct qla_hw_data *ha = vha->hw;
char fw_str[128];
return snprintf(buf, PAGE_SIZE, "%s\n",
ha->isp_ops->fw_version_str(vha, fw_str));
}
static ssize_t
qla2x00_serial_num_show(struct device *dev, struct device_attribute *attr,
char *buf)
{
scsi_qla_host_t *vha = shost_priv(class_to_shost(dev));
struct qla_hw_data *ha = vha->hw;
uint32_t sn;
if (IS_FWI2_CAPABLE(ha)) {
qla2xxx_get_vpd_field(vha, "SN", buf, PAGE_SIZE);
return snprintf(buf, PAGE_SIZE, "%s\n", buf);
}
sn = ((ha->serial0 & 0x1f) << 16) | (ha->serial2 << 8) | ha->serial1;
return snprintf(buf, PAGE_SIZE, "%c%05d\n", 'A' + sn / 100000,
sn % 100000);
}
static ssize_t
qla2x00_isp_name_show(struct device *dev, struct device_attribute *attr,
char *buf)
{
scsi_qla_host_t *vha = shost_priv(class_to_shost(dev));
return snprintf(buf, PAGE_SIZE, "ISP%04X\n", vha->hw->pdev->device);
}
static ssize_t
qla2x00_isp_id_show(struct device *dev, struct device_attribute *attr,
char *buf)
{
scsi_qla_host_t *vha = shost_priv(class_to_shost(dev));
struct qla_hw_data *ha = vha->hw;
return snprintf(buf, PAGE_SIZE, "%04x %04x %04x %04x\n",
ha->product_id[0], ha->product_id[1], ha->product_id[2],
ha->product_id[3]);
}
static ssize_t
qla2x00_model_name_show(struct device *dev, struct device_attribute *attr,
char *buf)
{
scsi_qla_host_t *vha = shost_priv(class_to_shost(dev));
return snprintf(buf, PAGE_SIZE, "%s\n", vha->hw->model_number);
}
static ssize_t
qla2x00_model_desc_show(struct device *dev, struct device_attribute *attr,
char *buf)
{
scsi_qla_host_t *vha = shost_priv(class_to_shost(dev));
return snprintf(buf, PAGE_SIZE, "%s\n",
vha->hw->model_desc ? vha->hw->model_desc : "");
}
static ssize_t
qla2x00_pci_info_show(struct device *dev, struct device_attribute *attr,
char *buf)
{
scsi_qla_host_t *vha = shost_priv(class_to_shost(dev));
char pci_info[30];
return snprintf(buf, PAGE_SIZE, "%s\n",
vha->hw->isp_ops->pci_info_str(vha, pci_info));
}
static ssize_t
qla2x00_link_state_show(struct device *dev, struct device_attribute *attr,
char *buf)
{
scsi_qla_host_t *vha = shost_priv(class_to_shost(dev));
struct qla_hw_data *ha = vha->hw;
int len = 0;
if (atomic_read(&vha->loop_state) == LOOP_DOWN ||
atomic_read(&vha->loop_state) == LOOP_DEAD ||
vha->device_flags & DFLG_NO_CABLE)
len = snprintf(buf, PAGE_SIZE, "Link Down\n");
else if (atomic_read(&vha->loop_state) != LOOP_READY ||
test_bit(ABORT_ISP_ACTIVE, &vha->dpc_flags) ||
test_bit(ISP_ABORT_NEEDED, &vha->dpc_flags))
len = snprintf(buf, PAGE_SIZE, "Unknown Link State\n");
else {
len = snprintf(buf, PAGE_SIZE, "Link Up - ");
switch (ha->current_topology) {
case ISP_CFG_NL:
len += snprintf(buf + len, PAGE_SIZE-len, "Loop\n");
break;
case ISP_CFG_FL:
len += snprintf(buf + len, PAGE_SIZE-len, "FL_Port\n");
break;
case ISP_CFG_N:
len += snprintf(buf + len, PAGE_SIZE-len,
"N_Port to N_Port\n");
break;
case ISP_CFG_F:
len += snprintf(buf + len, PAGE_SIZE-len, "F_Port\n");
break;
default:
len += snprintf(buf + len, PAGE_SIZE-len, "Loop\n");
break;
}
}
return len;
}
static ssize_t
qla2x00_zio_show(struct device *dev, struct device_attribute *attr,
char *buf)
{
scsi_qla_host_t *vha = shost_priv(class_to_shost(dev));
int len = 0;
switch (vha->hw->zio_mode) {
case QLA_ZIO_MODE_6:
len += snprintf(buf + len, PAGE_SIZE-len, "Mode 6\n");
break;
case QLA_ZIO_DISABLED:
len += snprintf(buf + len, PAGE_SIZE-len, "Disabled\n");
break;
}
return len;
}
static ssize_t
qla2x00_zio_store(struct device *dev, struct device_attribute *attr,
const char *buf, size_t count)
{
scsi_qla_host_t *vha = shost_priv(class_to_shost(dev));
struct qla_hw_data *ha = vha->hw;
int val = 0;
uint16_t zio_mode;
if (!IS_ZIO_SUPPORTED(ha))
return -ENOTSUPP;
if (sscanf(buf, "%d", &val) != 1)
return -EINVAL;
if (val)
zio_mode = QLA_ZIO_MODE_6;
else
zio_mode = QLA_ZIO_DISABLED;
/* Update per-hba values and queue a reset. */
if (zio_mode != QLA_ZIO_DISABLED || ha->zio_mode != QLA_ZIO_DISABLED) {
ha->zio_mode = zio_mode;
set_bit(ISP_ABORT_NEEDED, &vha->dpc_flags);
}
return strlen(buf);
}
static ssize_t
qla2x00_zio_timer_show(struct device *dev, struct device_attribute *attr,
char *buf)
{
scsi_qla_host_t *vha = shost_priv(class_to_shost(dev));
return snprintf(buf, PAGE_SIZE, "%d us\n", vha->hw->zio_timer * 100);
}
static ssize_t
qla2x00_zio_timer_store(struct device *dev, struct device_attribute *attr,
const char *buf, size_t count)
{
scsi_qla_host_t *vha = shost_priv(class_to_shost(dev));
int val = 0;
uint16_t zio_timer;
if (sscanf(buf, "%d", &val) != 1)
return -EINVAL;
if (val > 25500 || val < 100)
return -ERANGE;
zio_timer = (uint16_t)(val / 100);
vha->hw->zio_timer = zio_timer;
return strlen(buf);
}
static ssize_t
qla2x00_beacon_show(struct device *dev, struct device_attribute *attr,
char *buf)
{
scsi_qla_host_t *vha = shost_priv(class_to_shost(dev));
int len = 0;
if (vha->hw->beacon_blink_led)
len += snprintf(buf + len, PAGE_SIZE-len, "Enabled\n");
else
len += snprintf(buf + len, PAGE_SIZE-len, "Disabled\n");
return len;
}
static ssize_t
qla2x00_beacon_store(struct device *dev, struct device_attribute *attr,
const char *buf, size_t count)
{
scsi_qla_host_t *vha = shost_priv(class_to_shost(dev));
struct qla_hw_data *ha = vha->hw;
int val = 0;
int rval;
if (IS_QLA2100(ha) || IS_QLA2200(ha))
return -EPERM;
if (test_bit(ABORT_ISP_ACTIVE, &vha->dpc_flags)) {
qla_printk(KERN_WARNING, ha,
"Abort ISP active -- ignoring beacon request.\n");
return -EBUSY;
}
if (sscanf(buf, "%d", &val) != 1)
return -EINVAL;
if (val)
rval = ha->isp_ops->beacon_on(vha);
else
rval = ha->isp_ops->beacon_off(vha);
if (rval != QLA_SUCCESS)
count = 0;
return count;
}
static ssize_t
qla2x00_optrom_bios_version_show(struct device *dev,
struct device_attribute *attr, char *buf)
{
scsi_qla_host_t *vha = shost_priv(class_to_shost(dev));
struct qla_hw_data *ha = vha->hw;
return snprintf(buf, PAGE_SIZE, "%d.%02d\n", ha->bios_revision[1],
ha->bios_revision[0]);
}
static ssize_t
qla2x00_optrom_efi_version_show(struct device *dev,
struct device_attribute *attr, char *buf)
{
scsi_qla_host_t *vha = shost_priv(class_to_shost(dev));
struct qla_hw_data *ha = vha->hw;
return snprintf(buf, PAGE_SIZE, "%d.%02d\n", ha->efi_revision[1],
ha->efi_revision[0]);
}
static ssize_t
qla2x00_optrom_fcode_version_show(struct device *dev,
struct device_attribute *attr, char *buf)
{
scsi_qla_host_t *vha = shost_priv(class_to_shost(dev));
struct qla_hw_data *ha = vha->hw;
return snprintf(buf, PAGE_SIZE, "%d.%02d\n", ha->fcode_revision[1],
ha->fcode_revision[0]);
}
static ssize_t
qla2x00_optrom_fw_version_show(struct device *dev,
struct device_attribute *attr, char *buf)
{
scsi_qla_host_t *vha = shost_priv(class_to_shost(dev));
struct qla_hw_data *ha = vha->hw;
return snprintf(buf, PAGE_SIZE, "%d.%02d.%02d %d\n",
ha->fw_revision[0], ha->fw_revision[1], ha->fw_revision[2],
ha->fw_revision[3]);
}
static ssize_t
qla2x00_total_isp_aborts_show(struct device *dev,
struct device_attribute *attr, char *buf)
{
scsi_qla_host_t *vha = shost_priv(class_to_shost(dev));
struct qla_hw_data *ha = vha->hw;
return snprintf(buf, PAGE_SIZE, "%d\n",
ha->qla_stats.total_isp_aborts);
}
static ssize_t
qla24xx_84xx_fw_version_show(struct device *dev,
struct device_attribute *attr, char *buf)
{
int rval = QLA_SUCCESS;
uint16_t status[2] = {0, 0};
scsi_qla_host_t *vha = shost_priv(class_to_shost(dev));
struct qla_hw_data *ha = vha->hw;
if (!IS_QLA84XX(ha))
return snprintf(buf, PAGE_SIZE, "\n");
if (ha->cs84xx && ha->cs84xx->op_fw_version == 0)
rval = qla84xx_verify_chip(vha, status);
if ((rval == QLA_SUCCESS) && (status[0] == 0))
return snprintf(buf, PAGE_SIZE, "%u\n",
(uint32_t)ha->cs84xx->op_fw_version);
return snprintf(buf, PAGE_SIZE, "\n");
}
static ssize_t
qla2x00_mpi_version_show(struct device *dev, struct device_attribute *attr,
char *buf)
{
scsi_qla_host_t *vha = shost_priv(class_to_shost(dev));
struct qla_hw_data *ha = vha->hw;
if (!IS_QLA81XX(ha))
return snprintf(buf, PAGE_SIZE, "\n");
return snprintf(buf, PAGE_SIZE, "%d.%02d.%02d (%x)\n",
ha->mpi_version[0], ha->mpi_version[1], ha->mpi_version[2],
ha->mpi_capabilities);
}
static ssize_t
qla2x00_phy_version_show(struct device *dev, struct device_attribute *attr,
char *buf)
{
scsi_qla_host_t *vha = shost_priv(class_to_shost(dev));
struct qla_hw_data *ha = vha->hw;
if (!IS_QLA81XX(ha))
return snprintf(buf, PAGE_SIZE, "\n");
return snprintf(buf, PAGE_SIZE, "%d.%02d.%02d\n",
ha->phy_version[0], ha->phy_version[1], ha->phy_version[2]);
}
static ssize_t
qla2x00_flash_block_size_show(struct device *dev,
struct device_attribute *attr, char *buf)
{
scsi_qla_host_t *vha = shost_priv(class_to_shost(dev));
struct qla_hw_data *ha = vha->hw;
return snprintf(buf, PAGE_SIZE, "0x%x\n", ha->fdt_block_size);
}
static ssize_t
qla2x00_vlan_id_show(struct device *dev, struct device_attribute *attr,
char *buf)
{
scsi_qla_host_t *vha = shost_priv(class_to_shost(dev));
if (!IS_QLA8XXX_TYPE(vha->hw))
return snprintf(buf, PAGE_SIZE, "\n");
return snprintf(buf, PAGE_SIZE, "%d\n", vha->fcoe_vlan_id);
}
static ssize_t
qla2x00_vn_port_mac_address_show(struct device *dev,
struct device_attribute *attr, char *buf)
{
scsi_qla_host_t *vha = shost_priv(class_to_shost(dev));
if (!IS_QLA8XXX_TYPE(vha->hw))
return snprintf(buf, PAGE_SIZE, "\n");
return snprintf(buf, PAGE_SIZE, "%02x:%02x:%02x:%02x:%02x:%02x\n",
vha->fcoe_vn_port_mac[5], vha->fcoe_vn_port_mac[4],
vha->fcoe_vn_port_mac[3], vha->fcoe_vn_port_mac[2],
vha->fcoe_vn_port_mac[1], vha->fcoe_vn_port_mac[0]);
}
static ssize_t
qla2x00_fabric_param_show(struct device *dev, struct device_attribute *attr,
char *buf)
{
scsi_qla_host_t *vha = shost_priv(class_to_shost(dev));
return snprintf(buf, PAGE_SIZE, "%d\n", vha->hw->switch_cap);
}
static ssize_t
qla2x00_fw_state_show(struct device *dev, struct device_attribute *attr,
char *buf)
{
scsi_qla_host_t *vha = shost_priv(class_to_shost(dev));
int rval = QLA_FUNCTION_FAILED;
uint16_t state[5];
if (test_bit(ABORT_ISP_ACTIVE, &vha->dpc_flags) ||
test_bit(ISP_ABORT_NEEDED, &vha->dpc_flags))
DEBUG2_3_11(printk("%s(%ld): isp reset in progress.\n",
__func__, vha->host_no));
else if (!vha->hw->flags.eeh_busy)
rval = qla2x00_get_firmware_state(vha, state);
if (rval != QLA_SUCCESS)
memset(state, -1, sizeof(state));
return snprintf(buf, PAGE_SIZE, "0x%x 0x%x 0x%x 0x%x 0x%x\n", state[0],
state[1], state[2], state[3], state[4]);
}
static DEVICE_ATTR(driver_version, S_IRUGO, qla2x00_drvr_version_show, NULL);
static DEVICE_ATTR(fw_version, S_IRUGO, qla2x00_fw_version_show, NULL);
static DEVICE_ATTR(serial_num, S_IRUGO, qla2x00_serial_num_show, NULL);
static DEVICE_ATTR(isp_name, S_IRUGO, qla2x00_isp_name_show, NULL);
static DEVICE_ATTR(isp_id, S_IRUGO, qla2x00_isp_id_show, NULL);
static DEVICE_ATTR(model_name, S_IRUGO, qla2x00_model_name_show, NULL);
static DEVICE_ATTR(model_desc, S_IRUGO, qla2x00_model_desc_show, NULL);
static DEVICE_ATTR(pci_info, S_IRUGO, qla2x00_pci_info_show, NULL);
static DEVICE_ATTR(link_state, S_IRUGO, qla2x00_link_state_show, NULL);
static DEVICE_ATTR(zio, S_IRUGO | S_IWUSR, qla2x00_zio_show, qla2x00_zio_store);
static DEVICE_ATTR(zio_timer, S_IRUGO | S_IWUSR, qla2x00_zio_timer_show,
qla2x00_zio_timer_store);
static DEVICE_ATTR(beacon, S_IRUGO | S_IWUSR, qla2x00_beacon_show,
qla2x00_beacon_store);
static DEVICE_ATTR(optrom_bios_version, S_IRUGO,
qla2x00_optrom_bios_version_show, NULL);
static DEVICE_ATTR(optrom_efi_version, S_IRUGO,
qla2x00_optrom_efi_version_show, NULL);
static DEVICE_ATTR(optrom_fcode_version, S_IRUGO,
qla2x00_optrom_fcode_version_show, NULL);
static DEVICE_ATTR(optrom_fw_version, S_IRUGO, qla2x00_optrom_fw_version_show,
NULL);
static DEVICE_ATTR(84xx_fw_version, S_IRUGO, qla24xx_84xx_fw_version_show,
NULL);
static DEVICE_ATTR(total_isp_aborts, S_IRUGO, qla2x00_total_isp_aborts_show,
NULL);
static DEVICE_ATTR(mpi_version, S_IRUGO, qla2x00_mpi_version_show, NULL);
static DEVICE_ATTR(phy_version, S_IRUGO, qla2x00_phy_version_show, NULL);
static DEVICE_ATTR(flash_block_size, S_IRUGO, qla2x00_flash_block_size_show,
NULL);
static DEVICE_ATTR(vlan_id, S_IRUGO, qla2x00_vlan_id_show, NULL);
static DEVICE_ATTR(vn_port_mac_address, S_IRUGO,
qla2x00_vn_port_mac_address_show, NULL);
static DEVICE_ATTR(fabric_param, S_IRUGO, qla2x00_fabric_param_show, NULL);
static DEVICE_ATTR(fw_state, S_IRUGO, qla2x00_fw_state_show, NULL);
struct device_attribute *qla2x00_host_attrs[] = {
&dev_attr_driver_version,
&dev_attr_fw_version,
&dev_attr_serial_num,
&dev_attr_isp_name,
&dev_attr_isp_id,
&dev_attr_model_name,
&dev_attr_model_desc,
&dev_attr_pci_info,
&dev_attr_link_state,
&dev_attr_zio,
&dev_attr_zio_timer,
&dev_attr_beacon,
&dev_attr_optrom_bios_version,
&dev_attr_optrom_efi_version,
&dev_attr_optrom_fcode_version,
&dev_attr_optrom_fw_version,
&dev_attr_84xx_fw_version,
&dev_attr_total_isp_aborts,
&dev_attr_mpi_version,
&dev_attr_phy_version,
&dev_attr_flash_block_size,
&dev_attr_vlan_id,
&dev_attr_vn_port_mac_address,
&dev_attr_fabric_param,
&dev_attr_fw_state,
NULL,
};
/* Host attributes. */
static void
qla2x00_get_host_port_id(struct Scsi_Host *shost)
{
scsi_qla_host_t *vha = shost_priv(shost);
fc_host_port_id(shost) = vha->d_id.b.domain << 16 |
vha->d_id.b.area << 8 | vha->d_id.b.al_pa;
}
static void
qla2x00_get_host_speed(struct Scsi_Host *shost)
{
struct qla_hw_data *ha = ((struct scsi_qla_host *)
(shost_priv(shost)))->hw;
u32 speed = FC_PORTSPEED_UNKNOWN;
switch (ha->link_data_rate) {
case PORT_SPEED_1GB:
speed = FC_PORTSPEED_1GBIT;
break;
case PORT_SPEED_2GB:
speed = FC_PORTSPEED_2GBIT;
break;
case PORT_SPEED_4GB:
speed = FC_PORTSPEED_4GBIT;
break;
case PORT_SPEED_8GB:
speed = FC_PORTSPEED_8GBIT;
break;
case PORT_SPEED_10GB:
speed = FC_PORTSPEED_10GBIT;
break;
}
fc_host_speed(shost) = speed;
}
static void
qla2x00_get_host_port_type(struct Scsi_Host *shost)
{
scsi_qla_host_t *vha = shost_priv(shost);
uint32_t port_type = FC_PORTTYPE_UNKNOWN;
if (vha->vp_idx) {
fc_host_port_type(shost) = FC_PORTTYPE_NPIV;
return;
}
switch (vha->hw->current_topology) {
case ISP_CFG_NL:
port_type = FC_PORTTYPE_LPORT;
break;
case ISP_CFG_FL:
port_type = FC_PORTTYPE_NLPORT;
break;
case ISP_CFG_N:
port_type = FC_PORTTYPE_PTP;
break;
case ISP_CFG_F:
port_type = FC_PORTTYPE_NPORT;
break;
}
fc_host_port_type(shost) = port_type;
}
static void
qla2x00_get_starget_node_name(struct scsi_target *starget)
{
struct Scsi_Host *host = dev_to_shost(starget->dev.parent);
scsi_qla_host_t *vha = shost_priv(host);
fc_port_t *fcport;
u64 node_name = 0;
list_for_each_entry(fcport, &vha->vp_fcports, list) {
if (fcport->rport &&
starget->id == fcport->rport->scsi_target_id) {
node_name = wwn_to_u64(fcport->node_name);
break;
}
}
fc_starget_node_name(starget) = node_name;
}
static void
qla2x00_get_starget_port_name(struct scsi_target *starget)
{
struct Scsi_Host *host = dev_to_shost(starget->dev.parent);
scsi_qla_host_t *vha = shost_priv(host);
fc_port_t *fcport;
u64 port_name = 0;
list_for_each_entry(fcport, &vha->vp_fcports, list) {
if (fcport->rport &&
starget->id == fcport->rport->scsi_target_id) {
port_name = wwn_to_u64(fcport->port_name);
break;
}
}
fc_starget_port_name(starget) = port_name;
}
static void
qla2x00_get_starget_port_id(struct scsi_target *starget)
{
struct Scsi_Host *host = dev_to_shost(starget->dev.parent);
scsi_qla_host_t *vha = shost_priv(host);
fc_port_t *fcport;
uint32_t port_id = ~0U;
list_for_each_entry(fcport, &vha->vp_fcports, list) {
if (fcport->rport &&
starget->id == fcport->rport->scsi_target_id) {
port_id = fcport->d_id.b.domain << 16 |
fcport->d_id.b.area << 8 | fcport->d_id.b.al_pa;
break;
}
}
fc_starget_port_id(starget) = port_id;
}
static void
qla2x00_set_rport_loss_tmo(struct fc_rport *rport, uint32_t timeout)
{
if (timeout)
rport->dev_loss_tmo = timeout;
else
rport->dev_loss_tmo = 1;
}
static void
qla2x00_dev_loss_tmo_callbk(struct fc_rport *rport)
{
struct Scsi_Host *host = rport_to_shost(rport);
fc_port_t *fcport = *(fc_port_t **)rport->dd_data;
if (!fcport)
return;
if (test_bit(ABORT_ISP_ACTIVE, &fcport->vha->dpc_flags))
return;
if (unlikely(pci_channel_offline(fcport->vha->hw->pdev))) {
qla2x00_abort_all_cmds(fcport->vha, DID_NO_CONNECT << 16);
return;
}
/*
* Transport has effectively 'deleted' the rport, clear
* all local references.
*/
spin_lock_irq(host->host_lock);
fcport->rport = NULL;
*((fc_port_t **)rport->dd_data) = NULL;
spin_unlock_irq(host->host_lock);
}
static void
qla2x00_terminate_rport_io(struct fc_rport *rport)
{
fc_port_t *fcport = *(fc_port_t **)rport->dd_data;
if (!fcport)
return;
if (test_bit(ABORT_ISP_ACTIVE, &fcport->vha->dpc_flags))
return;
if (unlikely(pci_channel_offline(fcport->vha->hw->pdev))) {
qla2x00_abort_all_cmds(fcport->vha, DID_NO_CONNECT << 16);
return;
}
/*
* At this point all fcport's software-states are cleared. Perform any
* final cleanup of firmware resources (PCBs and XCBs).
*/
if (fcport->loop_id != FC_NO_LOOP_ID &&
!test_bit(UNLOADING, &fcport->vha->dpc_flags))
fcport->vha->hw->isp_ops->fabric_logout(fcport->vha,
fcport->loop_id, fcport->d_id.b.domain,
fcport->d_id.b.area, fcport->d_id.b.al_pa);
}
static int
qla2x00_issue_lip(struct Scsi_Host *shost)
{
scsi_qla_host_t *vha = shost_priv(shost);
qla2x00_loop_reset(vha);
return 0;
}
static struct fc_host_statistics *
qla2x00_get_fc_host_stats(struct Scsi_Host *shost)
{
scsi_qla_host_t *vha = shost_priv(shost);
struct qla_hw_data *ha = vha->hw;
struct scsi_qla_host *base_vha = pci_get_drvdata(ha->pdev);
int rval;
struct link_statistics *stats;
dma_addr_t stats_dma;
struct fc_host_statistics *pfc_host_stat;
pfc_host_stat = &ha->fc_host_stat;
memset(pfc_host_stat, -1, sizeof(struct fc_host_statistics));
if (test_bit(UNLOADING, &vha->dpc_flags))
goto done;
if (unlikely(pci_channel_offline(ha->pdev)))
goto done;
stats = dma_pool_alloc(ha->s_dma_pool, GFP_KERNEL, &stats_dma);
if (stats == NULL) {
DEBUG2_3_11(printk("%s(%ld): Failed to allocate memory.\n",
__func__, base_vha->host_no));
goto done;
}
memset(stats, 0, DMA_POOL_SIZE);
rval = QLA_FUNCTION_FAILED;
if (IS_FWI2_CAPABLE(ha)) {
rval = qla24xx_get_isp_stats(base_vha, stats, stats_dma);
} else if (atomic_read(&base_vha->loop_state) == LOOP_READY &&
!test_bit(ABORT_ISP_ACTIVE, &base_vha->dpc_flags) &&
!test_bit(ISP_ABORT_NEEDED, &base_vha->dpc_flags) &&
!ha->dpc_active) {
/* Must be in a 'READY' state for statistics retrieval. */
rval = qla2x00_get_link_status(base_vha, base_vha->loop_id,
stats, stats_dma);
}
if (rval != QLA_SUCCESS)
goto done_free;
pfc_host_stat->link_failure_count = stats->link_fail_cnt;
pfc_host_stat->loss_of_sync_count = stats->loss_sync_cnt;
pfc_host_stat->loss_of_signal_count = stats->loss_sig_cnt;
pfc_host_stat->prim_seq_protocol_err_count = stats->prim_seq_err_cnt;
pfc_host_stat->invalid_tx_word_count = stats->inval_xmit_word_cnt;
pfc_host_stat->invalid_crc_count = stats->inval_crc_cnt;
if (IS_FWI2_CAPABLE(ha)) {
pfc_host_stat->lip_count = stats->lip_cnt;
pfc_host_stat->tx_frames = stats->tx_frames;
pfc_host_stat->rx_frames = stats->rx_frames;
pfc_host_stat->dumped_frames = stats->dumped_frames;
pfc_host_stat->nos_count = stats->nos_rcvd;
}
pfc_host_stat->fcp_input_megabytes = ha->qla_stats.input_bytes >> 20;
pfc_host_stat->fcp_output_megabytes = ha->qla_stats.output_bytes >> 20;
done_free:
dma_pool_free(ha->s_dma_pool, stats, stats_dma);
done:
return pfc_host_stat;
}
static void
qla2x00_get_host_symbolic_name(struct Scsi_Host *shost)
{
scsi_qla_host_t *vha = shost_priv(shost);
qla2x00_get_sym_node_name(vha, fc_host_symbolic_name(shost));
}
static void
qla2x00_set_host_system_hostname(struct Scsi_Host *shost)
{
scsi_qla_host_t *vha = shost_priv(shost);
set_bit(REGISTER_FDMI_NEEDED, &vha->dpc_flags);
}
static void
qla2x00_get_host_fabric_name(struct Scsi_Host *shost)
{
scsi_qla_host_t *vha = shost_priv(shost);
u64 node_name;
if (vha->device_flags & SWITCH_FOUND)
node_name = wwn_to_u64(vha->fabric_node_name);
else
node_name = wwn_to_u64(vha->node_name);
fc_host_fabric_name(shost) = node_name;
}
static void
qla2x00_get_host_port_state(struct Scsi_Host *shost)
{
scsi_qla_host_t *vha = shost_priv(shost);
struct scsi_qla_host *base_vha = pci_get_drvdata(vha->hw->pdev);
if (!base_vha->flags.online)
fc_host_port_state(shost) = FC_PORTSTATE_OFFLINE;
else if (atomic_read(&base_vha->loop_state) == LOOP_TIMEOUT)
fc_host_port_state(shost) = FC_PORTSTATE_UNKNOWN;
else
fc_host_port_state(shost) = FC_PORTSTATE_ONLINE;
}
static int
qla24xx_vport_create(struct fc_vport *fc_vport, bool disable)
{
int ret = 0;
uint8_t qos = 0;
scsi_qla_host_t *base_vha = shost_priv(fc_vport->shost);
scsi_qla_host_t *vha = NULL;
struct qla_hw_data *ha = base_vha->hw;
uint16_t options = 0;
int cnt;
struct req_que *req = ha->req_q_map[0];
ret = qla24xx_vport_create_req_sanity_check(fc_vport);
if (ret) {
DEBUG15(printk("qla24xx_vport_create_req_sanity_check failed, "
"status %x\n", ret));
return (ret);
}
vha = qla24xx_create_vhost(fc_vport);
if (vha == NULL) {
DEBUG15(printk ("qla24xx_create_vhost failed, vha = %p\n",
vha));
return FC_VPORT_FAILED;
}
if (disable) {
atomic_set(&vha->vp_state, VP_OFFLINE);
fc_vport_set_state(fc_vport, FC_VPORT_DISABLED);
} else
atomic_set(&vha->vp_state, VP_FAILED);
/* ready to create vport */
qla_printk(KERN_INFO, vha->hw, "VP entry id %d assigned.\n",
vha->vp_idx);
/* initialized vport states */
atomic_set(&vha->loop_state, LOOP_DOWN);
vha->vp_err_state= VP_ERR_PORTDWN;
vha->vp_prev_err_state= VP_ERR_UNKWN;
/* Check if physical ha port is Up */
if (atomic_read(&base_vha->loop_state) == LOOP_DOWN ||
atomic_read(&base_vha->loop_state) == LOOP_DEAD) {
/* Don't retry or attempt login of this virtual port */
DEBUG15(printk ("scsi(%ld): pport loop_state is not UP.\n",
base_vha->host_no));
atomic_set(&vha->loop_state, LOOP_DEAD);
if (!disable)
fc_vport_set_state(fc_vport, FC_VPORT_LINKDOWN);
}
if (IS_QLA25XX(ha) && ql2xenabledif) {
if (ha->fw_attributes & BIT_4) {
vha->flags.difdix_supported = 1;
DEBUG18(qla_printk(KERN_INFO, ha,
"Registering for DIF/DIX type 1 and 3"
" protection.\n"));
scsi_host_set_prot(vha->host,
SHOST_DIF_TYPE1_PROTECTION
| SHOST_DIF_TYPE3_PROTECTION
| SHOST_DIX_TYPE1_PROTECTION
| SHOST_DIX_TYPE3_PROTECTION);
scsi_host_set_guard(vha->host, SHOST_DIX_GUARD_CRC);
} else
vha->flags.difdix_supported = 0;
}
if (scsi_add_host_with_dma(vha->host, &fc_vport->dev,
&ha->pdev->dev)) {
DEBUG15(printk("scsi(%ld): scsi_add_host failure for VP[%d].\n",
vha->host_no, vha->vp_idx));
goto vport_create_failed_2;
}
/* initialize attributes */
fc_host_node_name(vha->host) = wwn_to_u64(vha->node_name);
fc_host_port_name(vha->host) = wwn_to_u64(vha->port_name);
fc_host_supported_classes(vha->host) =
fc_host_supported_classes(base_vha->host);
fc_host_supported_speeds(vha->host) =
fc_host_supported_speeds(base_vha->host);
qla24xx_vport_disable(fc_vport, disable);
if (ha->flags.cpu_affinity_enabled) {
req = ha->req_q_map[1];
goto vport_queue;
} else if (ql2xmaxqueues == 1 || !ha->npiv_info)
goto vport_queue;
/* Create a request queue in QoS mode for the vport */
for (cnt = 0; cnt < ha->nvram_npiv_size; cnt++) {
if (memcmp(ha->npiv_info[cnt].port_name, vha->port_name, 8) == 0
&& memcmp(ha->npiv_info[cnt].node_name, vha->node_name,
8) == 0) {
qos = ha->npiv_info[cnt].q_qos;
break;
}
}
if (qos) {
ret = qla25xx_create_req_que(ha, options, vha->vp_idx, 0, 0,
qos);
if (!ret)
qla_printk(KERN_WARNING, ha,
"Can't create request queue for vp_idx:%d\n",
vha->vp_idx);
else {
DEBUG2(qla_printk(KERN_INFO, ha,
"Request Que:%d (QoS: %d) created for vp_idx:%d\n",
ret, qos, vha->vp_idx));
req = ha->req_q_map[ret];
}
}
vport_queue:
vha->req = req;
return 0;
vport_create_failed_2:
qla24xx_disable_vp(vha);
qla24xx_deallocate_vp_id(vha);
scsi_host_put(vha->host);
return FC_VPORT_FAILED;
}
static int
qla24xx_vport_delete(struct fc_vport *fc_vport)
{
scsi_qla_host_t *vha = fc_vport->dd_data;
fc_port_t *fcport, *tfcport;
struct qla_hw_data *ha = vha->hw;
uint16_t id = vha->vp_idx;
while (test_bit(LOOP_RESYNC_ACTIVE, &vha->dpc_flags) ||
test_bit(FCPORT_UPDATE_NEEDED, &vha->dpc_flags))
msleep(1000);
qla24xx_disable_vp(vha);
fc_remove_host(vha->host);
scsi_remove_host(vha->host);
list_for_each_entry_safe(fcport, tfcport, &vha->vp_fcports, list) {
list_del(&fcport->list);
kfree(fcport);
fcport = NULL;
}
qla24xx_deallocate_vp_id(vha);
mutex_lock(&ha->vport_lock);
ha->cur_vport_count--;
clear_bit(vha->vp_idx, ha->vp_idx_map);
mutex_unlock(&ha->vport_lock);
if (vha->timer_active) {
qla2x00_vp_stop_timer(vha);
DEBUG15(printk ("scsi(%ld): timer for the vport[%d] = %p "
"has stopped\n",
vha->host_no, vha->vp_idx, vha));
}
if (vha->req->id && !ha->flags.cpu_affinity_enabled) {
if (qla25xx_delete_req_que(vha, vha->req) != QLA_SUCCESS)
qla_printk(KERN_WARNING, ha,
"Queue delete failed.\n");
}
scsi_host_put(vha->host);
qla_printk(KERN_INFO, ha, "vport %d deleted\n", id);
return 0;
}
static int
qla24xx_vport_disable(struct fc_vport *fc_vport, bool disable)
{
scsi_qla_host_t *vha = fc_vport->dd_data;
if (disable)
qla24xx_disable_vp(vha);
else
qla24xx_enable_vp(vha);
return 0;
}
struct fc_function_template qla2xxx_transport_functions = {
.show_host_node_name = 1,
.show_host_port_name = 1,
.show_host_supported_classes = 1,
.show_host_supported_speeds = 1,
.get_host_port_id = qla2x00_get_host_port_id,
.show_host_port_id = 1,
.get_host_speed = qla2x00_get_host_speed,
.show_host_speed = 1,
.get_host_port_type = qla2x00_get_host_port_type,
.show_host_port_type = 1,
.get_host_symbolic_name = qla2x00_get_host_symbolic_name,
.show_host_symbolic_name = 1,
.set_host_system_hostname = qla2x00_set_host_system_hostname,
.show_host_system_hostname = 1,
.get_host_fabric_name = qla2x00_get_host_fabric_name,
.show_host_fabric_name = 1,
.get_host_port_state = qla2x00_get_host_port_state,
.show_host_port_state = 1,
.dd_fcrport_size = sizeof(struct fc_port *),
.show_rport_supported_classes = 1,
.get_starget_node_name = qla2x00_get_starget_node_name,
.show_starget_node_name = 1,
.get_starget_port_name = qla2x00_get_starget_port_name,
.show_starget_port_name = 1,
.get_starget_port_id = qla2x00_get_starget_port_id,
.show_starget_port_id = 1,
.set_rport_dev_loss_tmo = qla2x00_set_rport_loss_tmo,
.show_rport_dev_loss_tmo = 1,
.issue_fc_host_lip = qla2x00_issue_lip,
.dev_loss_tmo_callbk = qla2x00_dev_loss_tmo_callbk,
.terminate_rport_io = qla2x00_terminate_rport_io,
.get_fc_host_stats = qla2x00_get_fc_host_stats,
.vport_create = qla24xx_vport_create,
.vport_disable = qla24xx_vport_disable,
.vport_delete = qla24xx_vport_delete,
.bsg_request = qla24xx_bsg_request,
.bsg_timeout = qla24xx_bsg_timeout,
};
struct fc_function_template qla2xxx_transport_vport_functions = {
.show_host_node_name = 1,
.show_host_port_name = 1,
.show_host_supported_classes = 1,
.get_host_port_id = qla2x00_get_host_port_id,
.show_host_port_id = 1,
.get_host_speed = qla2x00_get_host_speed,
.show_host_speed = 1,
.get_host_port_type = qla2x00_get_host_port_type,
.show_host_port_type = 1,
.get_host_symbolic_name = qla2x00_get_host_symbolic_name,
.show_host_symbolic_name = 1,
.set_host_system_hostname = qla2x00_set_host_system_hostname,
.show_host_system_hostname = 1,
.get_host_fabric_name = qla2x00_get_host_fabric_name,
.show_host_fabric_name = 1,
.get_host_port_state = qla2x00_get_host_port_state,
.show_host_port_state = 1,
.dd_fcrport_size = sizeof(struct fc_port *),
.show_rport_supported_classes = 1,
.get_starget_node_name = qla2x00_get_starget_node_name,
.show_starget_node_name = 1,
.get_starget_port_name = qla2x00_get_starget_port_name,
.show_starget_port_name = 1,
.get_starget_port_id = qla2x00_get_starget_port_id,
.show_starget_port_id = 1,
.set_rport_dev_loss_tmo = qla2x00_set_rport_loss_tmo,
.show_rport_dev_loss_tmo = 1,
.issue_fc_host_lip = qla2x00_issue_lip,
.dev_loss_tmo_callbk = qla2x00_dev_loss_tmo_callbk,
.terminate_rport_io = qla2x00_terminate_rport_io,
.get_fc_host_stats = qla2x00_get_fc_host_stats,
.bsg_request = qla24xx_bsg_request,
.bsg_timeout = qla24xx_bsg_timeout,
};
void
qla2x00_init_host_attr(scsi_qla_host_t *vha)
{
struct qla_hw_data *ha = vha->hw;
u32 speed = FC_PORTSPEED_UNKNOWN;
fc_host_node_name(vha->host) = wwn_to_u64(vha->node_name);
fc_host_port_name(vha->host) = wwn_to_u64(vha->port_name);
fc_host_supported_classes(vha->host) = FC_COS_CLASS3;
fc_host_max_npiv_vports(vha->host) = ha->max_npiv_vports;
fc_host_npiv_vports_inuse(vha->host) = ha->cur_vport_count;
if (IS_QLA8XXX_TYPE(ha))
speed = FC_PORTSPEED_10GBIT;
else if (IS_QLA25XX(ha))
speed = FC_PORTSPEED_8GBIT | FC_PORTSPEED_4GBIT |
FC_PORTSPEED_2GBIT | FC_PORTSPEED_1GBIT;
else if (IS_QLA24XX_TYPE(ha))
speed = FC_PORTSPEED_4GBIT | FC_PORTSPEED_2GBIT |
FC_PORTSPEED_1GBIT;
else if (IS_QLA23XX(ha))
speed = FC_PORTSPEED_2GBIT | FC_PORTSPEED_1GBIT;
else
speed = FC_PORTSPEED_1GBIT;
fc_host_supported_speeds(vha->host) = speed;
}