[linux-block.git] / io_uring / fdinfo.c

// SPDX-License-Identifier: GPL-2.0
#include <linux/kernel.h>
#include <linux/errno.h>
#include <linux/fs.h>
#include <linux/file.h>
#include <linux/proc_fs.h>
#include <linux/seq_file.h>
#include <linux/io_uring.h>

#include <uapi/linux/io_uring.h>

#include "io_uring_types.h"
#include "io_uring.h"
#include "sqpoll.h"
#include "fdinfo.h"
#include "cancel.h"

#ifdef CONFIG_PROC_FS
static __cold int io_uring_show_cred(struct seq_file *m, unsigned int id,
		const struct cred *cred)
{
	struct user_namespace *uns = seq_user_ns(m);
	struct group_info *gi;
	kernel_cap_t cap;
	unsigned __capi;
	int g;

	seq_printf(m, "%5d\n", id);
	seq_put_decimal_ull(m, "\tUid:\t", from_kuid_munged(uns, cred->uid));
	seq_put_decimal_ull(m, "\t\t", from_kuid_munged(uns, cred->euid));
	seq_put_decimal_ull(m, "\t\t", from_kuid_munged(uns, cred->suid));
	seq_put_decimal_ull(m, "\t\t", from_kuid_munged(uns, cred->fsuid));
	seq_put_decimal_ull(m, "\n\tGid:\t", from_kgid_munged(uns, cred->gid));
	seq_put_decimal_ull(m, "\t\t", from_kgid_munged(uns, cred->egid));
	seq_put_decimal_ull(m, "\t\t", from_kgid_munged(uns, cred->sgid));
	seq_put_decimal_ull(m, "\t\t", from_kgid_munged(uns, cred->fsgid));
	seq_puts(m, "\n\tGroups:\t");
	gi = cred->group_info;
	for (g = 0; g < gi->ngroups; g++) {
		seq_put_decimal_ull(m, g ? " " : "",
					from_kgid_munged(uns, gi->gid[g]));
	}
	seq_puts(m, "\n\tCapEff:\t");
	cap = cred->cap_effective;
	CAP_FOR_EACH_U32(__capi)
		seq_put_hex_ll(m, NULL, cap.cap[CAP_LAST_U32 - __capi], 8);
	seq_putc(m, '\n');
	return 0;
}

static __cold void __io_uring_show_fdinfo(struct io_ring_ctx *ctx,
					  struct seq_file *m)
{
	struct io_sq_data *sq = NULL;
	struct io_overflow_cqe *ocqe;
	struct io_rings *r = ctx->rings;
	unsigned int sq_mask = ctx->sq_entries - 1, cq_mask = ctx->cq_entries - 1;
	unsigned int sq_head = READ_ONCE(r->sq.head);
	unsigned int sq_tail = READ_ONCE(r->sq.tail);
	unsigned int cq_head = READ_ONCE(r->cq.head);
	unsigned int cq_tail = READ_ONCE(r->cq.tail);
	unsigned int cq_shift = 0;
	unsigned int sq_entries, cq_entries;
	bool has_lock;
	bool is_cqe32 = (ctx->flags & IORING_SETUP_CQE32);
	unsigned int i;

	if (is_cqe32)
		cq_shift = 1;

	/*
	 * we may get imprecise sqe and cqe info if uring is actively running
	 * since we get cached_sq_head and cached_cq_tail without uring_lock
	 * and sq_tail and cq_head are changed by userspace. But it's ok since
	 * we usually use these info when it is stuck.
	 */
	seq_printf(m, "SqMask:\t0x%x\n", sq_mask);
	seq_printf(m, "SqHead:\t%u\n", sq_head);
	seq_printf(m, "SqTail:\t%u\n", sq_tail);
	seq_printf(m, "CachedSqHead:\t%u\n", ctx->cached_sq_head);
	seq_printf(m, "CqMask:\t0x%x\n", cq_mask);
	seq_printf(m, "CqHead:\t%u\n", cq_head);
	seq_printf(m, "CqTail:\t%u\n", cq_tail);
	seq_printf(m, "CachedCqTail:\t%u\n", ctx->cached_cq_tail);
	seq_printf(m, "SQEs:\t%u\n", sq_tail - ctx->cached_sq_head);
	sq_entries = min(sq_tail - sq_head, ctx->sq_entries);
	for (i = 0; i < sq_entries; i++) {
		unsigned int entry = i + sq_head;
		unsigned int sq_idx = READ_ONCE(ctx->sq_array[entry & sq_mask]);
		struct io_uring_sqe *sqe;

		if (sq_idx > sq_mask)
			continue;
		sqe = &ctx->sq_sqes[sq_idx];
		seq_printf(m, "%5u: opcode:%d, fd:%d, flags:%x, user_data:%llu\n",
			   sq_idx, sqe->opcode, sqe->fd, sqe->flags,
			   sqe->user_data);
	}
	seq_printf(m, "CQEs:\t%u\n", cq_tail - cq_head);
	cq_entries = min(cq_tail - cq_head, ctx->cq_entries);
	for (i = 0; i < cq_entries; i++) {
		unsigned int entry = i + cq_head;
		struct io_uring_cqe *cqe = &r->cqes[(entry & cq_mask) << cq_shift];

		if (!is_cqe32) {
			seq_printf(m, "%5u: user_data:%llu, res:%d, flag:%x\n",
			   entry & cq_mask, cqe->user_data, cqe->res,
			   cqe->flags);
		} else {
			seq_printf(m, "%5u: user_data:%llu, res:%d, flag:%x, "
				"extra1:%llu, extra2:%llu\n",
				entry & cq_mask, cqe->user_data, cqe->res,
				cqe->flags, cqe->big_cqe[0], cqe->big_cqe[1]);
		}
	}

	/*
	 * Avoid ABBA deadlock between the seq lock and the io_uring mutex,
	 * since fdinfo case grabs it in the opposite direction of normal use
	 * cases. If we fail to get the lock, we just don't iterate any
	 * structures that could be going away outside the io_uring mutex.
	 */
	has_lock = mutex_trylock(&ctx->uring_lock);

	if (has_lock && (ctx->flags & IORING_SETUP_SQPOLL)) {
		sq = ctx->sq_data;
		if (!sq->thread)
			sq = NULL;
	}

	seq_printf(m, "SqThread:\t%d\n", sq ? task_pid_nr(sq->thread) : -1);
	seq_printf(m, "SqThreadCpu:\t%d\n", sq ? task_cpu(sq->thread) : -1);
	seq_printf(m, "UserFiles:\t%u\n", ctx->nr_user_files);
	for (i = 0; has_lock && i < ctx->nr_user_files; i++) {
		struct file *f = io_file_from_index(&ctx->file_table, i);

		if (f)
			seq_printf(m, "%5u: %s\n", i, file_dentry(f)->d_iname);
		else
			seq_printf(m, "%5u: <none>\n", i);
	}
	seq_printf(m, "UserBufs:\t%u\n", ctx->nr_user_bufs);
	for (i = 0; has_lock && i < ctx->nr_user_bufs; i++) {
		struct io_mapped_ubuf *buf = ctx->user_bufs[i];
		unsigned int len = buf->ubuf_end - buf->ubuf;

		seq_printf(m, "%5u: 0x%llx/%u\n", i, buf->ubuf, len);
	}
	if (has_lock && !xa_empty(&ctx->personalities)) {
		unsigned long index;
		const struct cred *cred;

		seq_printf(m, "Personalities:\n");
		xa_for_each(&ctx->personalities, index, cred)
			io_uring_show_cred(m, index, cred);
	}
	if (has_lock)
		mutex_unlock(&ctx->uring_lock);

	seq_puts(m, "PollList:\n");
	for (i = 0; i < (1U << ctx->cancel_table.hash_bits); i++) {
		struct io_hash_bucket *hb = &ctx->cancel_table.hbs[i];
		struct io_kiocb *req;

		spin_lock(&hb->lock);
		hlist_for_each_entry(req, &hb->list, hash_node)
			seq_printf(m, "  op=%d, task_works=%d\n", req->opcode,
					task_work_pending(req->task));
		spin_unlock(&hb->lock);
	}

	seq_puts(m, "CqOverflowList:\n");
	spin_lock(&ctx->completion_lock);
	list_for_each_entry(ocqe, &ctx->cq_overflow_list, list) {
		struct io_uring_cqe *cqe = &ocqe->cqe;

		seq_printf(m, "  user_data=%llu, res=%d, flags=%x\n",
			   cqe->user_data, cqe->res, cqe->flags);

	}

	spin_unlock(&ctx->completion_lock);
}

__cold void io_uring_show_fdinfo(struct seq_file *m, struct file *f)
{
	struct io_ring_ctx *ctx = f->private_data;

	if (percpu_ref_tryget(&ctx->refs)) {
		__io_uring_show_fdinfo(ctx, m);
		percpu_ref_put(&ctx->refs);
	}
}
#endif
Commit	Line	Data
a4ad4f74 JA	1	// SPDX-License-Identifier: GPL-2.0
	2	#include <linux/kernel.h>
	3	#include <linux/errno.h>
	4	#include <linux/fs.h>
	5	#include <linux/file.h>
	6	#include <linux/proc_fs.h>
	7	#include <linux/seq_file.h>
	8	#include <linux/io_uring.h>
	9
	10	#include <uapi/linux/io_uring.h>
	11
	12	#include "io_uring_types.h"
	13	#include "io_uring.h"
	14	#include "sqpoll.h"
	15	#include "fdinfo.h"
38513c46	16	#include "cancel.h"
a4ad4f74 JA	17
	18	#ifdef CONFIG_PROC_FS
	19	static __cold int io_uring_show_cred(struct seq_file *m, unsigned int id,
	20	const struct cred *cred)
	21	{
	22	struct user_namespace *uns = seq_user_ns(m);
	23	struct group_info *gi;
	24	kernel_cap_t cap;
	25	unsigned __capi;
	26	int g;
	27
	28	seq_printf(m, "%5d\n", id);
	29	seq_put_decimal_ull(m, "\tUid:\t", from_kuid_munged(uns, cred->uid));
	30	seq_put_decimal_ull(m, "\t\t", from_kuid_munged(uns, cred->euid));
	31	seq_put_decimal_ull(m, "\t\t", from_kuid_munged(uns, cred->suid));
	32	seq_put_decimal_ull(m, "\t\t", from_kuid_munged(uns, cred->fsuid));
	33	seq_put_decimal_ull(m, "\n\tGid:\t", from_kgid_munged(uns, cred->gid));
	34	seq_put_decimal_ull(m, "\t\t", from_kgid_munged(uns, cred->egid));
	35	seq_put_decimal_ull(m, "\t\t", from_kgid_munged(uns, cred->sgid));
	36	seq_put_decimal_ull(m, "\t\t", from_kgid_munged(uns, cred->fsgid));
	37	seq_puts(m, "\n\tGroups:\t");
	38	gi = cred->group_info;
	39	for (g = 0; g < gi->ngroups; g++) {
	40	seq_put_decimal_ull(m, g ? " " : "",
	41	from_kgid_munged(uns, gi->gid[g]));
	42	}
	43	seq_puts(m, "\n\tCapEff:\t");
	44	cap = cred->cap_effective;
	45	CAP_FOR_EACH_U32(__capi)
	46	seq_put_hex_ll(m, NULL, cap.cap[CAP_LAST_U32 - __capi], 8);
	47	seq_putc(m, '\n');
	48	return 0;
	49	}
	50
	51	static __cold void __io_uring_show_fdinfo(struct io_ring_ctx *ctx,
	52	struct seq_file *m)
	53	{
	54	struct io_sq_data *sq = NULL;
	55	struct io_overflow_cqe *ocqe;
	56	struct io_rings *r = ctx->rings;
	57	unsigned int sq_mask = ctx->sq_entries - 1, cq_mask = ctx->cq_entries - 1;
	58	unsigned int sq_head = READ_ONCE(r->sq.head);
	59	unsigned int sq_tail = READ_ONCE(r->sq.tail);
	60	unsigned int cq_head = READ_ONCE(r->cq.head);
	61	unsigned int cq_tail = READ_ONCE(r->cq.tail);
	62	unsigned int cq_shift = 0;
	63	unsigned int sq_entries, cq_entries;
	64	bool has_lock;
	65	bool is_cqe32 = (ctx->flags & IORING_SETUP_CQE32);
	66	unsigned int i;
	67
	68	if (is_cqe32)
	69	cq_shift = 1;
	70
	71	/*
	72	* we may get imprecise sqe and cqe info if uring is actively running
	73	* since we get cached_sq_head and cached_cq_tail without uring_lock
	74	* and sq_tail and cq_head are changed by userspace. But it's ok since
	75	* we usually use these info when it is stuck.
	76	*/
	77	seq_printf(m, "SqMask:\t0x%x\n", sq_mask);
	78	seq_printf(m, "SqHead:\t%u\n", sq_head);
	79	seq_printf(m, "SqTail:\t%u\n", sq_tail);
	80	seq_printf(m, "CachedSqHead:\t%u\n", ctx->cached_sq_head);
81	seq_printf(m, "CqMask:\t0x%x\n", cq_mask);
82	seq_printf(m, "CqHead:\t%u\n", cq_head);
83	seq_printf(m, "CqTail:\t%u\n", cq_tail);
84	seq_printf(m, "CachedCqTail:\t%u\n", ctx->cached_cq_tail);
85	seq_printf(m, "SQEs:\t%u\n", sq_tail - ctx->cached_sq_head);
86	sq_entries = min(sq_tail - sq_head, ctx->sq_entries);
87	for (i = 0; i < sq_entries; i++) {
88	unsigned int entry = i + sq_head;
89	unsigned int sq_idx = READ_ONCE(ctx->sq_array[entry & sq_mask]);
90	struct io_uring_sqe *sqe;
91
92	if (sq_idx > sq_mask)
93	continue;
94	sqe = &ctx->sq_sqes[sq_idx];
95	seq_printf(m, "%5u: opcode:%d, fd:%d, flags:%x, user_data:%llu\n",
96	sq_idx, sqe->opcode, sqe->fd, sqe->flags,
97	sqe->user_data);
98	}
99	seq_printf(m, "CQEs:\t%u\n", cq_tail - cq_head);
100	cq_entries = min(cq_tail - cq_head, ctx->cq_entries);
101	for (i = 0; i < cq_entries; i++) {
102	unsigned int entry = i + cq_head;
103	struct io_uring_cqe *cqe = &r->cqes[(entry & cq_mask) << cq_shift];
104
105	if (!is_cqe32) {
106	seq_printf(m, "%5u: user_data:%llu, res:%d, flag:%x\n",
107	entry & cq_mask, cqe->user_data, cqe->res,
108	cqe->flags);
109	} else {
110	seq_printf(m, "%5u: user_data:%llu, res:%d, flag:%x, "
111	"extra1:%llu, extra2:%llu\n",
112	entry & cq_mask, cqe->user_data, cqe->res,
113	cqe->flags, cqe->big_cqe[0], cqe->big_cqe[1]);
114	}
115	}
116
117	/*
118	* Avoid ABBA deadlock between the seq lock and the io_uring mutex,
119	* since fdinfo case grabs it in the opposite direction of normal use
120	* cases. If we fail to get the lock, we just don't iterate any
121	* structures that could be going away outside the io_uring mutex.
122	*/
123	has_lock = mutex_trylock(&ctx->uring_lock);
124
125	if (has_lock && (ctx->flags & IORING_SETUP_SQPOLL)) {
126	sq = ctx->sq_data;
127	if (!sq->thread)
128	sq = NULL;
129	}
130
131	seq_printf(m, "SqThread:\t%d\n", sq ? task_pid_nr(sq->thread) : -1);
132	seq_printf(m, "SqThreadCpu:\t%d\n", sq ? task_cpu(sq->thread) : -1);
133	seq_printf(m, "UserFiles:\t%u\n", ctx->nr_user_files);
134	for (i = 0; has_lock && i < ctx->nr_user_files; i++) {
135	struct file *f = io_file_from_index(&ctx->file_table, i);
136
137	if (f)
138	seq_printf(m, "%5u: %s\n", i, file_dentry(f)->d_iname);
139	else
140	seq_printf(m, "%5u: <none>\n", i);
141	}
142	seq_printf(m, "UserBufs:\t%u\n", ctx->nr_user_bufs);
143	for (i = 0; has_lock && i < ctx->nr_user_bufs; i++) {
144	struct io_mapped_ubuf *buf = ctx->user_bufs[i];
145	unsigned int len = buf->ubuf_end - buf->ubuf;
146
147	seq_printf(m, "%5u: 0x%llx/%u\n", i, buf->ubuf, len);
148	}
149	if (has_lock && !xa_empty(&ctx->personalities)) {
150	unsigned long index;
151	const struct cred *cred;
152
153	seq_printf(m, "Personalities:\n");
154	xa_for_each(&ctx->personalities, index, cred)
155	io_uring_show_cred(m, index, cred);
156	}
157	if (has_lock)
158	mutex_unlock(&ctx->uring_lock);
159
160	seq_puts(m, "PollList:\n");
e6f89be6 PB	161	for (i = 0; i < (1U << ctx->cancel_table.hash_bits); i++) {
e6f89be6 PB	162	struct io_hash_bucket *hb = &ctx->cancel_table.hbs[i];
a4ad4f74 JA	163	struct io_kiocb *req;
a4ad4f74 JA	164
38513c46 HX	165	spin_lock(&hb->lock);
38513c46 HX	166	hlist_for_each_entry(req, &hb->list, hash_node)
a4ad4f74 JA	167	seq_printf(m, " op=%d, task_works=%d\n", req->opcode,
a4ad4f74 JA	168	task_work_pending(req->task));
38513c46	169	spin_unlock(&hb->lock);
a4ad4f74 JA	170	}
	171
	172	seq_puts(m, "CqOverflowList:\n");
38513c46	173	spin_lock(&ctx->completion_lock);
a4ad4f74 JA	174	list_for_each_entry(ocqe, &ctx->cq_overflow_list, list) {
	175	struct io_uring_cqe *cqe = &ocqe->cqe;
	176
	177	seq_printf(m, " user_data=%llu, res=%d, flags=%x\n",
	178	cqe->user_data, cqe->res, cqe->flags);
	179
	180	}
	181
	182	spin_unlock(&ctx->completion_lock);
	183	}
	184
	185	__cold void io_uring_show_fdinfo(struct seq_file m, struct file f)
	186	{
	187	struct io_ring_ctx *ctx = f->private_data;
	188
	189	if (percpu_ref_tryget(&ctx->refs)) {
	190	__io_uring_show_fdinfo(ctx, m);
	191	percpu_ref_put(&ctx->refs);
	192	}
	193	}
	194	#endif