leandrof/linux
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
e5d6468fe9
linux/drivers/s390/cio/qdio_debug.c
Julian Wiedmann 0a6e634535 s390/qdio: extend polling support to multiple queues 
When the support for polling drivers was initially added, it only
considered Input Queue 0. But as QDIO interrupts are actually for the
full device and not a single queue, this doesn't really fit for
configurations where multiple Input Queues are used.

Rework the qdio code so that interrupts for a polling driver are not
split up into actions for each queue. Instead deliver the interrupt as
a single event, and let the driver decide which queue needs what action.

When re-enabling the QDIO interrupt via qdio_start_irq(), this means
that the qdio code needs to
(1) put _all_ eligible queues back into a state where they raise IRQs,
(2) and afterwards check _all_ eligible queues for new work to bridge
    the race window.

On the qeth side of things (as the only qdio polling driver), we can now
add CQ polling support to the main NAPI poll routine. It doesn't consume
NAPI budget, and to avoid hogging the CPU we yield control after
completing one full queue worth of buffers.
The subsequent qdio_start_irq() will check for any additional work, and
have us re-schedule the NAPI instance accordingly.

Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
Acked-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2020-03-25 12:07:15 -07:00

359 lines
9.1 KiB
C
Raw Blame History

// SPDX-License-Identifier: GPL-2.0
/*
* Copyright IBM Corp. 2008, 2009
*
* Author: Jan Glauber (jang@linux.vnet.ibm.com)
*/
#include <linux/seq_file.h>
#include <linux/debugfs.h>
#include <linux/uaccess.h>
#include <linux/export.h>
#include <linux/slab.h>
#include <asm/debug.h>
#include "qdio_debug.h"
#include "qdio.h"
debug_info_t *qdio_dbf_setup;
debug_info_t *qdio_dbf_error;
static struct dentry *debugfs_root;
#define QDIO_DEBUGFS_NAME_LEN 10
#define QDIO_DBF_NAME_LEN 20
struct qdio_dbf_entry {
char dbf_name[QDIO_DBF_NAME_LEN];
debug_info_t *dbf_info;
struct list_head dbf_list;
};
static LIST_HEAD(qdio_dbf_list);
static DEFINE_MUTEX(qdio_dbf_list_mutex);
static debug_info_t *qdio_get_dbf_entry(char *name)
{
struct qdio_dbf_entry *entry;
debug_info_t *rc = NULL;
mutex_lock(&qdio_dbf_list_mutex);
list_for_each_entry(entry, &qdio_dbf_list, dbf_list) {
if (strcmp(entry->dbf_name, name) == 0) {
rc = entry->dbf_info;
break;
}
}
mutex_unlock(&qdio_dbf_list_mutex);
return rc;
}
static void qdio_clear_dbf_list(void)
{
struct qdio_dbf_entry *entry, *tmp;
mutex_lock(&qdio_dbf_list_mutex);
list_for_each_entry_safe(entry, tmp, &qdio_dbf_list, dbf_list) {
list_del(&entry->dbf_list);
debug_unregister(entry->dbf_info);
kfree(entry);
}
mutex_unlock(&qdio_dbf_list_mutex);
}
int qdio_allocate_dbf(struct qdio_initialize *init_data,
struct qdio_irq *irq_ptr)
{
char text[QDIO_DBF_NAME_LEN];
struct qdio_dbf_entry *new_entry;
DBF_EVENT("qfmt:%1d", init_data->q_format);
DBF_HEX(init_data->adapter_name, 8);
DBF_EVENT("qpff%4x", init_data->qib_param_field_format);
DBF_HEX(&init_data->qib_param_field, sizeof(void *));
DBF_HEX(&init_data->input_slib_elements, sizeof(void *));
DBF_HEX(&init_data->output_slib_elements, sizeof(void *));
DBF_EVENT("niq:%1d noq:%1d", init_data->no_input_qs,
init_data->no_output_qs);
DBF_HEX(&init_data->input_handler, sizeof(void *));
DBF_HEX(&init_data->output_handler, sizeof(void *));
DBF_HEX(&init_data->int_parm, sizeof(long));
DBF_HEX(&init_data->input_sbal_addr_array, sizeof(void *));
DBF_HEX(&init_data->output_sbal_addr_array, sizeof(void *));
DBF_EVENT("irq:%8lx", (unsigned long)irq_ptr);
/* allocate trace view for the interface */
snprintf(text, QDIO_DBF_NAME_LEN, "qdio_%s",
dev_name(&init_data->cdev->dev));
irq_ptr->debug_area = qdio_get_dbf_entry(text);
if (irq_ptr->debug_area)
DBF_DEV_EVENT(DBF_ERR, irq_ptr, "dbf reused");
else {
irq_ptr->debug_area = debug_register(text, 2, 1, 16);
if (!irq_ptr->debug_area)
return -ENOMEM;
if (debug_register_view(irq_ptr->debug_area,
&debug_hex_ascii_view)) {
debug_unregister(irq_ptr->debug_area);
return -ENOMEM;
}
debug_set_level(irq_ptr->debug_area, DBF_WARN);
DBF_DEV_EVENT(DBF_ERR, irq_ptr, "dbf created");
new_entry = kzalloc(sizeof(struct qdio_dbf_entry), GFP_KERNEL);
if (!new_entry) {
debug_unregister(irq_ptr->debug_area);
return -ENOMEM;
}
strlcpy(new_entry->dbf_name, text, QDIO_DBF_NAME_LEN);
new_entry->dbf_info = irq_ptr->debug_area;
mutex_lock(&qdio_dbf_list_mutex);
list_add(&new_entry->dbf_list, &qdio_dbf_list);
mutex_unlock(&qdio_dbf_list_mutex);
}
return 0;
}
static int qstat_show(struct seq_file *m, void *v)
{
unsigned char state;
struct qdio_q *q = m->private;
int i;
if (!q)
return 0;
seq_printf(m, "Timestamp: %Lx Last AI: %Lx\n",
q->timestamp, last_ai_time);
seq_printf(m, "nr_used: %d ftc: %d\n",
atomic_read(&q->nr_buf_used), q->first_to_check);
if (q->is_input_q) {
seq_printf(m, "ack start: %d ack count: %d\n",
q->u.in.ack_start, q->u.in.ack_count);
seq_printf(m, "DSCI: %x IRQs disabled: %u\n",
*(u8 *)q->irq_ptr->dsci,
test_bit(QDIO_IRQ_DISABLED,
&q->irq_ptr->poll_state));
}
seq_printf(m, "SBAL states:\n");
seq_printf(m, "|0 |8 |16 |24 |32 |40 |48 |56 63|\n");
for (i = 0; i < QDIO_MAX_BUFFERS_PER_Q; i++) {
debug_get_buf_state(q, i, &state);
switch (state) {
case SLSB_P_INPUT_NOT_INIT:
case SLSB_P_OUTPUT_NOT_INIT:
seq_printf(m, "N");
break;
case SLSB_P_OUTPUT_PENDING:
seq_printf(m, "P");
break;
case SLSB_P_INPUT_PRIMED:
case SLSB_CU_OUTPUT_PRIMED:
seq_printf(m, "+");
break;
case SLSB_P_INPUT_ACK:
seq_printf(m, "A");
break;
case SLSB_P_INPUT_ERROR:
case SLSB_P_OUTPUT_ERROR:
seq_printf(m, "x");
break;
case SLSB_CU_INPUT_EMPTY:
case SLSB_P_OUTPUT_EMPTY:
seq_printf(m, "-");
break;
case SLSB_P_INPUT_HALTED:
case SLSB_P_OUTPUT_HALTED:
seq_printf(m, ".");
break;
default:
seq_printf(m, "?");
}
if (i == 63)
seq_printf(m, "\n");
}
seq_printf(m, "\n");
seq_printf(m, "|64 |72 |80 |88 |96 |104 |112 | 127|\n");
seq_printf(m, "\nSBAL statistics:");
if (!q->irq_ptr->perf_stat_enabled) {
seq_printf(m, " disabled\n");
return 0;
}
seq_printf(m, "\n1 2.. 4.. 8.. "
"16.. 32.. 64.. 127\n");
for (i = 0; i < ARRAY_SIZE(q->q_stats.nr_sbals); i++)
seq_printf(m, "%-10u ", q->q_stats.nr_sbals[i]);
seq_printf(m, "\nError NOP Total\n%-10u %-10u %-10u\n\n",
q->q_stats.nr_sbal_error, q->q_stats.nr_sbal_nop,
q->q_stats.nr_sbal_total);
return 0;
}
DEFINE_SHOW_ATTRIBUTE(qstat);
static char *qperf_names[] = {
"Assumed adapter interrupts",
"QDIO interrupts",
"Requested PCIs",
"Inbound tasklet runs",
"Inbound tasklet resched",
"Inbound tasklet resched2",
"Outbound tasklet runs",
"SIGA read",
"SIGA write",
"SIGA sync",
"Inbound calls",
"Inbound handler",
"Inbound stop_polling",
"Inbound queue full",
"Outbound calls",
"Outbound handler",
"Outbound queue full",
"Outbound fast_requeue",
"Outbound target_full",
"QEBSM eqbs",
"QEBSM eqbs partial",
"QEBSM sqbs",
"QEBSM sqbs partial",
"Discarded interrupts"
};
static int qperf_show(struct seq_file *m, void *v)
{
struct qdio_irq *irq_ptr = m->private;
unsigned int *stat;
int i;
if (!irq_ptr)
return 0;
if (!irq_ptr->perf_stat_enabled) {
seq_printf(m, "disabled\n");
return 0;
}
stat = (unsigned int *)&irq_ptr->perf_stat;
for (i = 0; i < ARRAY_SIZE(qperf_names); i++)
seq_printf(m, "%26s:\t%u\n",
qperf_names[i], *(stat + i));
return 0;
}
static ssize_t qperf_seq_write(struct file *file, const char __user *ubuf,
size_t count, loff_t *off)
{
struct seq_file *seq = file->private_data;
struct qdio_irq *irq_ptr = seq->private;
struct qdio_q *q;
unsigned long val;
int ret, i;
if (!irq_ptr)
return 0;
ret = kstrtoul_from_user(ubuf, count, 10, &val);
if (ret)
return ret;
switch (val) {
case 0:
irq_ptr->perf_stat_enabled = 0;
memset(&irq_ptr->perf_stat, 0, sizeof(irq_ptr->perf_stat));
for_each_input_queue(irq_ptr, q, i)
memset(&q->q_stats, 0, sizeof(q->q_stats));
for_each_output_queue(irq_ptr, q, i)
memset(&q->q_stats, 0, sizeof(q->q_stats));
break;
case 1:
irq_ptr->perf_stat_enabled = 1;
break;
}
return count;
}
static int qperf_seq_open(struct inode *inode, struct file *filp)
{
return single_open(filp, qperf_show,
file_inode(filp)->i_private);
}
static const struct file_operations debugfs_perf_fops = {
.owner = THIS_MODULE,
.open = qperf_seq_open,
.read = seq_read,
.write = qperf_seq_write,
.llseek = seq_lseek,
.release = single_release,
};
static void setup_debugfs_entry(struct qdio_q *q)
{
char name[QDIO_DEBUGFS_NAME_LEN];
snprintf(name, QDIO_DEBUGFS_NAME_LEN, "%s_%d",
q->is_input_q ? "input" : "output",
q->nr);
q->debugfs_q = debugfs_create_file(name, 0444,
q->irq_ptr->debugfs_dev, q, &qstat_fops);
if (IS_ERR(q->debugfs_q))
q->debugfs_q = NULL;
}
void qdio_setup_debug_entries(struct qdio_irq *irq_ptr, struct ccw_device *cdev)
{
struct qdio_q *q;
int i;
irq_ptr->debugfs_dev = debugfs_create_dir(dev_name(&cdev->dev),
debugfs_root);
if (IS_ERR(irq_ptr->debugfs_dev))
irq_ptr->debugfs_dev = NULL;
irq_ptr->debugfs_perf = debugfs_create_file("statistics",
S_IFREG | S_IRUGO | S_IWUSR,
irq_ptr->debugfs_dev, irq_ptr,
&debugfs_perf_fops);
if (IS_ERR(irq_ptr->debugfs_perf))
irq_ptr->debugfs_perf = NULL;
for_each_input_queue(irq_ptr, q, i)
setup_debugfs_entry(q);
for_each_output_queue(irq_ptr, q, i)
setup_debugfs_entry(q);
}
void qdio_shutdown_debug_entries(struct qdio_irq *irq_ptr)
{
struct qdio_q *q;
int i;
for_each_input_queue(irq_ptr, q, i)
debugfs_remove(q->debugfs_q);
for_each_output_queue(irq_ptr, q, i)
debugfs_remove(q->debugfs_q);
debugfs_remove(irq_ptr->debugfs_perf);
debugfs_remove(irq_ptr->debugfs_dev);
}
int __init qdio_debug_init(void)
{
debugfs_root = debugfs_create_dir("qdio", NULL);
qdio_dbf_setup = debug_register("qdio_setup", 16, 1, 16);
debug_register_view(qdio_dbf_setup, &debug_hex_ascii_view);
debug_set_level(qdio_dbf_setup, DBF_INFO);
DBF_EVENT("dbf created\n");
qdio_dbf_error = debug_register("qdio_error", 4, 1, 16);
debug_register_view(qdio_dbf_error, &debug_hex_ascii_view);
debug_set_level(qdio_dbf_error, DBF_INFO);
DBF_ERROR("dbf created\n");
return 0;
}
void qdio_debug_exit(void)
{
qdio_clear_dbf_list();
debugfs_remove(debugfs_root);
debug_unregister(qdio_dbf_setup);
debug_unregister(qdio_dbf_error);
}
Reference in New Issue View Git Blame Copy Permalink