X-Git-Url: https://git.kernel.dk/?p=fio.git;a=blobdiff_plain;f=verify.c;h=9a96fbb33a448c679df829c5be808ca00b8db22f;hp=227d220abb983abc82fb00523f827f22254a182c;hb=3c746856fd63f3f7624a34ea70e226a290d9cc01;hpb=61b9861d45dcd6be23c527861b32aa36bcd1b682

diff --git a/verify.c b/verify.c
index 227d220a..9a96fbb3 100644
--- a/verify.c
+++ b/verify.c
@@ -29,10 +29,12 @@
 static void populate_hdr(struct thread_data *td, struct io_u *io_u,
 			 struct verify_header *hdr, unsigned int header_num,
 			 unsigned int header_len);
-static void fill_hdr(struct verify_header *hdr, int verify_type, uint32_t len,
-		     uint64_t rand_seed);
-static void __fill_hdr(struct verify_header *hdr, int verify_type, uint32_t len,
-		       uint64_t rand_seed);
+static void fill_hdr(struct thread_data *td, struct io_u *io_u,
+		     struct verify_header *hdr, unsigned int header_num,
+		     unsigned int header_len, uint64_t rand_seed);
+static void __fill_hdr(struct thread_data *td, struct io_u *io_u,
+		       struct verify_header *hdr, unsigned int header_num,
+		       unsigned int header_len, uint64_t rand_seed);
 
 void fill_buffer_pattern(struct thread_data *td, void *p, unsigned int len)
 {
@@ -141,6 +143,7 @@ static inline unsigned int __hdr_size(int verify_type)
 
 	switch (verify_type) {
 	case VERIFY_NONE:
+	case VERIFY_HDR_ONLY:
 	case VERIFY_NULL:
 	case VERIFY_PATTERN:
 		len = 0;
@@ -171,9 +174,6 @@ static inline unsigned int __hdr_size(int verify_type)
 	case VERIFY_XXHASH:
 		len = sizeof(struct vhdr_xxhash);
 		break;
-	case VERIFY_META:
-		len = sizeof(struct vhdr_meta);
-		break;
 	case VERIFY_SHA1:
 		len = sizeof(struct vhdr_sha1);
 		break;
@@ -323,7 +323,7 @@ static void dump_verify_buffers(struct verify_header *hdr, struct vcont *vc)
 	struct verify_header shdr;
 
 	if (td->o.verify == VERIFY_PATTERN_NO_HDR) {
-		__fill_hdr(&shdr, td->o.verify, vc->io_u->buflen, 0);
+		__fill_hdr(td, vc->io_u, &shdr, 0, vc->io_u->buflen, 0);
 		hdr = &shdr;
 	}
 
@@ -405,43 +405,6 @@ static int verify_io_u_pattern(struct verify_header *hdr, struct vcont *vc)
 	return EILSEQ;
 }
 
-static int verify_io_u_meta(struct verify_header *hdr, struct vcont *vc)
-{
-	struct thread_data *td = vc->td;
-	struct vhdr_meta *vh = hdr_priv(hdr);
-	struct io_u *io_u = vc->io_u;
-	int ret = EILSEQ;
-
-	dprint(FD_VERIFY, "meta verify io_u %p, len %u\n", io_u, hdr->len);
-
-	if (vh->offset == io_u->offset + vc->hdr_num * td->o.verify_interval)
-		ret = 0;
-
-	if (td->o.verify_pattern_bytes)
-		ret |= verify_io_u_pattern(hdr, vc);
-
-	/*
-	 * For read-only workloads, the program cannot be certain of the
-	 * last numberio written to a block. Checking of numberio will be
-	 * done only for workloads that write data.  For verify_only,
-	 * numberio will be checked in the last iteration when the correct
-	 * state of numberio, that would have been written to each block
-	 * in a previous run of fio, has been reached.
-	 */
-	if ((td_write(td) || td_rw(td)) && (td_min_bs(td) == td_max_bs(td)) &&
-	    !td->o.time_based)
-		if (!td->o.verify_only || td->o.loops == 0)
-			if (vh->numberio != io_u->numberio)
-				ret = EILSEQ;
-
-	if (!ret)
-		return 0;
-
-	vc->name = "meta";
-	log_verify_failure(hdr, vc);
-	return ret;
-}
-
 static int verify_io_u_xxhash(struct verify_header *hdr, struct vcont *vc)
 {
 	void *p = io_u_verify_off(hdr, vc);
@@ -698,42 +661,68 @@ int verify_io_u_async(struct thread_data *td, struct io_u **io_u_ptr)
 	return 0;
 }
 
+/*
+ * Thanks Rusty, for spending the time so I don't have to.
+ *
+ * http://rusty.ozlabs.org/?p=560
+ */
+static int mem_is_zero(const void *data, size_t length)
+{
+	const unsigned char *p = data;
+	size_t len;
+
+	/* Check first 16 bytes manually */
+	for (len = 0; len < 16; len++) {
+		if (!length)
+			return 1;
+		if (*p)
+			return 0;
+		p++;
+		length--;
+	}
+
+	/* Now we know that's zero, memcmp with self. */
+	return memcmp(data, p, length) == 0;
+}
+
+static int mem_is_zero_slow(const void *data, size_t length, size_t *offset)
+{
+	const unsigned char *p = data;
+
+	*offset = 0;
+	while (length) {
+		if (*p)
+			break;
+		(*offset)++;
+		length--;
+		p++;
+	}
+
+	return !length;
+}
+
 static int verify_trimmed_io_u(struct thread_data *td, struct io_u *io_u)
 {
-	static char zero_buf[1024];
-	unsigned int this_len, len;
-	int ret = 0;
-	void *p;
+	size_t offset;
 
 	if (!td->o.trim_zero)
 		return 0;
 
-	len = io_u->buflen;
-	p = io_u->buf;
-	do {
-		this_len = sizeof(zero_buf);
-		if (this_len > len)
-			this_len = len;
-		if (memcmp(p, zero_buf, this_len)) {
-			ret = EILSEQ;
-			break;
-		}
-		len -= this_len;
-		p += this_len;
-	} while (len);
-
-	if (!ret)
+	if (mem_is_zero(io_u->buf, io_u->buflen))
 		return 0;
 
+	mem_is_zero_slow(io_u->buf, io_u->buflen, &offset);
+
 	log_err("trim: verify failed at file %s offset %llu, length %lu"
 		", block offset %lu\n",
 			io_u->file->file_name, io_u->offset, io_u->buflen,
-			(unsigned long) (p - io_u->buf));
-	return ret;
+			(unsigned long) offset);
+	return EILSEQ;
 }
 
-static int verify_header(struct io_u *io_u, struct verify_header *hdr,
-			 unsigned int hdr_num, unsigned int hdr_len)
+static int verify_header(struct io_u *io_u, struct thread_data *td,
+			 struct verify_header *hdr, unsigned int hdr_num,
+			 unsigned int hdr_len)
 {
 	void *p = hdr;
 	uint32_t crc;
@@ -754,6 +743,30 @@ static int verify_header(struct io_u *io_u, struct verify_header *hdr,
 			hdr->rand_seed, io_u->rand_seed);
 		goto err;
 	}
+	if (hdr->offset != io_u->offset + hdr_num * td->o.verify_interval) {
+		log_err("verify: bad header offset %"PRIu64
+			", wanted %llu",
+			hdr->offset, io_u->offset);
+		goto err;
+	}
+
+	/*
+	 * For read-only workloads, the program cannot be certain of the
+	 * last numberio written to a block. Checking of numberio will be
+	 * done only for workloads that write data.  For verify_only,
+	 * numberio will be checked in the last iteration when the correct
+	 * state of numberio, that would have been written to each block
+	 * in a previous run of fio, has been reached.
+	 */
+	if ((td_write(td) || td_rw(td)) && (td_min_bs(td) == td_max_bs(td)) &&
+	    !td->o.time_based)
+		if (!td->o.verify_only || td->o.loops == 0)
+			if (hdr->numberio != io_u->numberio) {
+				log_err("verify: bad header numberio %"PRIu16
+					", wanted %"PRIu16,
+					hdr->numberio, io_u->numberio);
+				goto err;
+			}
 
 	crc = fio_crc32c(p, offsetof(struct verify_header, crc32));
 	if (crc != hdr->crc32) {
@@ -767,6 +780,11 @@ err:
 	log_err(" at file %s offset %llu, length %u\n",
 		io_u->file->file_name,
 		io_u->offset + hdr_num * hdr_len, hdr_len);
+
+	if (td->o.verify_dump)
+		dump_buf(p, hdr_len, io_u->offset + hdr_num * hdr_len,
+				"hdr_fail", io_u->file);
+
 	return EILSEQ;
 }
 
@@ -820,7 +838,7 @@ int verify_io_u(struct thread_data *td, struct io_u **io_u_ptr)
 			io_u->rand_seed = hdr->rand_seed;
 
 		if (td->o.verify != VERIFY_PATTERN_NO_HDR) {
-			ret = verify_header(io_u, hdr, hdr_num, hdr_inc);
+			ret = verify_header(io_u, td, hdr, hdr_num, hdr_inc);
 			if (ret)
 				return ret;
 		}
@@ -831,6 +849,12 @@ int verify_io_u(struct thread_data *td, struct io_u **io_u_ptr)
 			verify_type = hdr->verify_type;
 
 		switch (verify_type) {
+		case VERIFY_HDR_ONLY:
+			/* Header is always verified, check if pattern is left
+			 * for verification. */
+			if (td->o.verify_pattern_bytes)
+				ret = verify_io_u_pattern(hdr, &vc);
+			break;
 		case VERIFY_MD5:
 			ret = verify_io_u_md5(hdr, &vc);
 			break;
@@ -859,9 +883,6 @@ int verify_io_u(struct thread_data *td, struct io_u **io_u_ptr)
 		case VERIFY_XXHASH:
 			ret = verify_io_u_xxhash(hdr, &vc);
 			break;
-		case VERIFY_META:
-			ret = verify_io_u_meta(hdr, &vc);
-			break;
 		case VERIFY_SHA1:
 			ret = verify_io_u_sha1(hdr, &vc);
 			break;
@@ -886,21 +907,6 @@ done:
 	return ret;
 }
 
-static void fill_meta(struct verify_header *hdr, struct thread_data *td,
-		      struct io_u *io_u, unsigned int header_num)
-{
-	struct vhdr_meta *vh = hdr_priv(hdr);
-
-	vh->thread = td->thread_number;
-
-	vh->time_sec = io_u->start_time.tv_sec;
-	vh->time_usec = io_u->start_time.tv_usec;
-
-	vh->numberio = io_u->numberio;
-
-	vh->offset = io_u->offset + header_num * td->o.verify_interval;
-}
-
 static void fill_xxhash(struct verify_header *hdr, void *p, unsigned int len)
 {
 	struct vhdr_xxhash *vh = hdr_priv(hdr);
@@ -993,24 +999,32 @@ static void fill_md5(struct verify_header *hdr, void *p, unsigned int len)
 	fio_md5_final(&md5_ctx);
 }
 
-static void __fill_hdr(struct verify_header *hdr, int verify_type,
-		       uint32_t len, uint64_t rand_seed)
+static void __fill_hdr(struct thread_data *td, struct io_u *io_u,
+		       struct verify_header *hdr, unsigned int header_num,
+		       unsigned int header_len, uint64_t rand_seed)
 {
 	void *p = hdr;
 
 	hdr->magic = FIO_HDR_MAGIC;
-	hdr->verify_type = verify_type;
-	hdr->len = len;
+	hdr->verify_type = td->o.verify;
+	hdr->len = header_len;
 	hdr->rand_seed = rand_seed;
+	hdr->offset = io_u->offset + header_num * td->o.verify_interval;
+	hdr->time_sec = io_u->start_time.tv_sec;
+	hdr->time_usec = io_u->start_time.tv_usec;
+	hdr->thread = td->thread_number;
+	hdr->numberio = io_u->numberio;
 	hdr->crc32 = fio_crc32c(p, offsetof(struct verify_header, crc32));
 }
 
 
-static void fill_hdr(struct verify_header *hdr, int verify_type, uint32_t len,
-		     uint64_t rand_seed)
+static void fill_hdr(struct thread_data *td, struct io_u *io_u,
+		     struct verify_header *hdr, unsigned int header_num,
+		     unsigned int header_len, uint64_t rand_seed)
 {
-	if (verify_type != VERIFY_PATTERN_NO_HDR)
-		__fill_hdr(hdr, verify_type, len, rand_seed);
+
+	if (td->o.verify != VERIFY_PATTERN_NO_HDR)
+		__fill_hdr(td, io_u, hdr, header_num, header_len, rand_seed);
 }
 
 static void populate_hdr(struct thread_data *td, struct io_u *io_u,
@@ -1022,7 +1036,7 @@ static void populate_hdr(struct thread_data *td, struct io_u *io_u,
 
 	p = (void *) hdr;
 
-	fill_hdr(hdr, td->o.verify, header_len, io_u->rand_seed);
+	fill_hdr(td, io_u, hdr, header_num, header_len, io_u->rand_seed);
 
 	data_len = header_len - hdr_size(td, hdr);
 
@@ -1074,16 +1088,12 @@ static void populate_hdr(struct thread_data *td, struct io_u *io_u,
 						io_u, hdr->len);
 		fill_xxhash(hdr, data, data_len);
 		break;
-	case VERIFY_META:
-		dprint(FD_VERIFY, "fill meta io_u %p, len %u\n",
-						io_u, hdr->len);
-		fill_meta(hdr, td, io_u, header_num);
-		break;
 	case VERIFY_SHA1:
 		dprint(FD_VERIFY, "fill sha1 io_u %p, len %u\n",
 						io_u, hdr->len);
 		fill_sha1(hdr, data, data_len);
 		break;
+	case VERIFY_HDR_ONLY:
 	case VERIFY_PATTERN:
 	case VERIFY_PATTERN_NO_HDR:
 		/* nothing to do here */
@@ -1280,7 +1290,7 @@ int verify_async_init(struct thread_data *td)
 	pthread_attr_t attr;
 
 	pthread_attr_init(&attr);
-	pthread_attr_setstacksize(&attr, PTHREAD_STACK_MIN);
+	pthread_attr_setstacksize(&attr, 2 * PTHREAD_STACK_MIN);
 
 	td->verify_thread_exit = 0;
 
@@ -1343,6 +1353,47 @@ int paste_blockoff(char *buf, unsigned int len, void *priv)
 	return 0;
 }
 
+static int __fill_file_completions(struct thread_data *td,
+				   struct thread_io_list *s,
+				   struct fio_file *f, unsigned int *index)
+{
+	unsigned int comps;
+	int i, j;
+
+	if (!f->last_write_comp)
+		return 0;
+
+	if (td->io_blocks[DDIR_WRITE] < td->o.iodepth)
+		comps = td->io_blocks[DDIR_WRITE];
+	else
+		comps = td->o.iodepth;
+
+	j = f->last_write_idx - 1;
+	for (i = 0; i < comps; i++) {
+		if (j == -1)
+			j = td->o.iodepth - 1;
+		s->comps[*index].fileno = __cpu_to_le64(f->fileno);
+		s->comps[*index].offset = cpu_to_le64(f->last_write_comp[j]);
+		(*index)++;
+		j--;
+	}
+
+	return comps;
+}
+
+static int fill_file_completions(struct thread_data *td,
+				 struct thread_io_list *s, unsigned int *index)
+{
+	struct fio_file *f;
+	unsigned int i;
+	int comps = 0;
+
+	for_each_file(td, f, i)
+		comps += __fill_file_completions(td, s, f, index);
+
+	return comps;
+}
+
 struct all_io_list *get_all_io_list(int save_mask, size_t *sz)
 {
 	struct all_io_list *rep;
@@ -1364,7 +1415,7 @@ struct all_io_list *get_all_io_list(int save_mask, size_t *sz)
 			continue;
 		td->stop_io = 1;
 		td->flags |= TD_F_VSTATE_SAVED;
-		depth += td->o.iodepth;
+		depth += (td->o.iodepth * td->o.nr_files);
 		nr++;
 	}
 
@@ -1373,39 +1424,25 @@ struct all_io_list *get_all_io_list(int save_mask, size_t *sz)
 
 	*sz = sizeof(*rep);
 	*sz += nr * sizeof(struct thread_io_list);
-	*sz += depth * sizeof(uint64_t);
+	*sz += depth * sizeof(struct file_comp);
 	rep = malloc(*sz);
+	memset(rep, 0, *sz);
 
 	rep->threads = cpu_to_le64((uint64_t) nr);
 
 	next = &rep->state[0];
 	for_each_td(td, i) {
 		struct thread_io_list *s = next;
-		unsigned int comps;
+		unsigned int comps, index = 0;
 
 		if (save_mask != IO_LIST_ALL && (i + 1) != save_mask)
 			continue;
 
-		if (td->last_write_comp) {
-			int j, k;
-
-			if (td->io_blocks[DDIR_WRITE] < td->o.iodepth)
-				comps = td->io_blocks[DDIR_WRITE];
-			else
-				comps = td->o.iodepth;
-
-			k = td->last_write_idx - 1;
-			for (j = 0; j < comps; j++) {
-				if (k == -1)
-					k = td->o.iodepth - 1;
-				s->offsets[j] = cpu_to_le64(td->last_write_comp[k]);
-				k--;
-			}
-		} else
-			comps = 0;
+		comps = fill_file_completions(td, s, &index);
 
 		s->no_comps = cpu_to_le64((uint64_t) comps);
 		s->depth = cpu_to_le64((uint64_t) td->o.iodepth);
+		s->nofiles = cpu_to_le64((uint64_t) td->o.nr_files);
 		s->numberio = cpu_to_le64((uint64_t) td->io_issues[DDIR_WRITE]);
 		s->index = cpu_to_le64((uint64_t) i);
 		if (td->random_state.use64) {
@@ -1434,7 +1471,7 @@ struct all_io_list *get_all_io_list(int save_mask, size_t *sz)
 static int open_state_file(const char *name, const char *prefix, int num,
 			   int for_write)
 {
-	char out[64];
+	char out[PATH_MAX];
 	int flags;
 	int fd;
 
@@ -1448,6 +1485,7 @@ static int open_state_file(const char *name, const char *prefix, int num,
 	fd = open(out, flags, 0644);
 	if (fd == -1) {
 		perror("fio: open state file");
+		log_err("fio: state file: %s (for_write=%d)\n", out, for_write);
 		return -1;
 	}
 
@@ -1525,72 +1563,34 @@ void verify_free_state(struct thread_data *td)
 		free(td->vstate);
 }
 
-static struct thread_io_list *convert_v1_list(struct thread_io_list_v1 *s)
+void verify_assign_state(struct thread_data *td, void *p)
 {
-	struct thread_io_list *til;
+	struct thread_io_list *s = p;
 	int i;
 
-	til = malloc(__thread_io_list_sz(s->no_comps));
-	til->no_comps = s->no_comps;
-	til->depth = s->depth;
-	til->numberio = s->numberio;
-	til->index = s->index;
-	memcpy(til->name, s->name, sizeof(til->name));
-
-	til->rand.use64 = 0;
-	for (i = 0; i < 4; i++)
-		til->rand.state32.s[i] = s->rand.s[i];
-
-	for (i = 0; i < s->no_comps; i++)
-		til->offsets[i] = s->offsets[i];
+	s->no_comps = le64_to_cpu(s->no_comps);
+	s->depth = le32_to_cpu(s->depth);
+	s->nofiles = le32_to_cpu(s->nofiles);
+	s->numberio = le64_to_cpu(s->numberio);
+	s->rand.use64 = le64_to_cpu(s->rand.use64);
 
-	return til;
-}
-
-void verify_convert_assign_state(struct thread_data *td, void *p, int version)
-{
-	struct thread_io_list *til;
-	int i;
-
-	if (version == 1) {
-		struct thread_io_list_v1 *s = p;
-
-		s->no_comps = le64_to_cpu(s->no_comps);
-		s->depth = le64_to_cpu(s->depth);
-		s->numberio = le64_to_cpu(s->numberio);
-		for (i = 0; i < 4; i++)
-			s->rand.s[i] = le32_to_cpu(s->rand.s[i]);
-		for (i = 0; i < s->no_comps; i++)
-			s->offsets[i] = le64_to_cpu(s->offsets[i]);
-
-		til = convert_v1_list(s);
-		free(s);
+	if (s->rand.use64) {
+		for (i = 0; i < 6; i++)
+			s->rand.state64.s[i] = le64_to_cpu(s->rand.state64.s[i]);
 	} else {
-		struct thread_io_list *s = p;
-
-		s->no_comps = le64_to_cpu(s->no_comps);
-		s->depth = le64_to_cpu(s->depth);
-		s->numberio = le64_to_cpu(s->numberio);
-		s->rand.use64 = le64_to_cpu(s->rand.use64);
-
-		if (s->rand.use64) {
-			for (i = 0; i < 6; i++)
-				s->rand.state64.s[i] = le64_to_cpu(s->rand.state64.s[i]);
-		} else {
-			for (i = 0; i < 4; i++)
-				s->rand.state32.s[i] = le32_to_cpu(s->rand.state32.s[i]);
-		}
-		for (i = 0; i < s->no_comps; i++)
-			s->offsets[i] = le64_to_cpu(s->offsets[i]);
+		for (i = 0; i < 4; i++)
+			s->rand.state32.s[i] = le32_to_cpu(s->rand.state32.s[i]);
+	}
 
-		til = p;
+	for (i = 0; i < s->no_comps; i++) {
+		s->comps[i].fileno = le64_to_cpu(s->comps[i].fileno);
+		s->comps[i].offset = le64_to_cpu(s->comps[i].offset);
 	}
 
-	td->vstate = til;
+	td->vstate = p;
 }
 
-int verify_state_hdr(struct verify_state_hdr *hdr, struct thread_io_list *s,
-		     int *version)
+int verify_state_hdr(struct verify_state_hdr *hdr, struct thread_io_list *s)
 {
 	uint64_t crc;
 
@@ -1598,15 +1598,13 @@ int verify_state_hdr(struct verify_state_hdr *hdr, struct thread_io_list *s,
 	hdr->size = le64_to_cpu(hdr->size);
 	hdr->crc = le64_to_cpu(hdr->crc);
 
-	if (hdr->version != VSTATE_HDR_VERSION ||
-	    hdr->version != VSTATE_HDR_VERSION_V1)
+	if (hdr->version != VSTATE_HDR_VERSION)
 		return 1;
 
 	crc = fio_crc32c((void *)s, hdr->size);
 	if (crc != hdr->crc)
 		return 1;
 
-	*version = hdr->version;
 	return 0;
 }
 
@@ -1637,9 +1635,9 @@ int verify_load_state(struct thread_data *td, const char *prefix)
 	hdr.size = le64_to_cpu(hdr.size);
 	hdr.crc = le64_to_cpu(hdr.crc);
 
-	if (hdr.version != VSTATE_HDR_VERSION &&
-	    hdr.version != VSTATE_HDR_VERSION_V1) {
-		log_err("fio: bad version in verify state header\n");
+	if (hdr.version != VSTATE_HDR_VERSION) {
+		log_err("fio: unsupported (%d) version in verify state header\n",
+				(unsigned int) hdr.version);
 		goto err;
 	}
 
@@ -1660,7 +1658,7 @@ int verify_load_state(struct thread_data *td, const char *prefix)
 
 	close(fd);
 
-	verify_convert_assign_state(td, s, hdr.version);
+	verify_assign_state(td, s);
 	return 0;
 err:
 	if (s)
@@ -1675,9 +1673,10 @@ err:
 int verify_state_should_stop(struct thread_data *td, struct io_u *io_u)
 {
 	struct thread_io_list *s = td->vstate;
+	struct fio_file *f = io_u->file;
 	int i;
 
-	if (!s)
+	if (!s || !f)
 		return 0;
 
 	/*
@@ -1694,9 +1693,12 @@ int verify_state_should_stop(struct thread_data *td, struct io_u *io_u)
 	 * completed or not. If the IO was seen as completed, then
 	 * lets verify it.
 	 */
-	for (i = 0; i < s->no_comps; i++)
-		if (io_u->offset == s->offsets[i])
+	for (i = 0; i < s->no_comps; i++) {
+		if (s->comps[i].fileno != f->fileno)
+			continue;
+		if (io_u->offset == s->comps[i].offset)
 			return 0;
+	}
 
 	/*
 	 * Not found, we have to stop