/usr/src/kernel-patches/lustre/patches/debian-2.6.26.diff is in linux-patch-lustre 1.8.5+dfsg-3ubuntu1.
This file is owned by root:root, with mode 0o644.
The actual contents of the file can be viewed below.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 | diff -u -r debian-2.6.26/Documentation/filesystems/ext2.txt debian-2.6.26_lustre.1.8.2/Documentation/filesystems/ext2.txt
--- debian-2.6.26/Documentation/filesystems/ext2.txt 2008-07-13 23:51:29.000000000 +0200
+++ debian-2.6.26_lustre.1.8.2/Documentation/filesystems/ext2.txt 2010-02-12 15:11:19.000000000 +0100
@@ -58,6 +58,22 @@
xip Use execute in place (no caching) if possible
+iopen Makes an invisible pseudo-directory called
+ __iopen__ available in the root directory
+ of the filesystem. Allows open-by-inode-
+ number. i.e., inode 3145 can be accessed
+ via /mntpt/__iopen__/3145
+
+iopen_nopriv This option makes the iopen directory be
+ world-readable. This may be safer since it
+ allows daemons to run as an unprivileged user,
+ however it significantly changes the security
+ model of a Unix filesystem, since previously
+ all files under a mode 700 directory were not
+ generally avilable even if the
+ permissions on the file itself is
+ world-readable.
+
grpquota,noquota,quota,usrquota Quota options are silently ignored by ext2.
diff -u -r debian-2.6.26/block/blk-core.c debian-2.6.26_lustre.1.8.2/block/blk-core.c
--- debian-2.6.26/block/blk-core.c 2008-07-13 23:51:29.000000000 +0200
+++ debian-2.6.26_lustre.1.8.2/block/blk-core.c 2010-02-12 15:14:32.000000000 +0100
@@ -1270,6 +1270,8 @@
#endif /* CONFIG_FAIL_MAKE_REQUEST */
+int dev_check_rdonly(struct block_device *bdev);
+
/*
* Check whether this bio extends beyond the end of the device.
*/
@@ -1371,6 +1373,12 @@
if (unlikely(test_bit(QUEUE_FLAG_DEAD, &q->queue_flags)))
goto end_io;
+ /* this is cfs's dev_rdonly check */
+ if (bio->bi_rw == WRITE &&
+ dev_check_rdonly(bio->bi_bdev)) {
+ bio_endio(bio, 0);
+ break;
+ }
if (should_fail_request(bio))
goto end_io;
@@ -2028,6 +2036,91 @@
}
EXPORT_SYMBOL(kblockd_flush_work);
+ /*
+ * Debug code for turning block devices "read-only" (will discard writes
+ * silently). This is for filesystem crash/recovery testing.
+ */
+struct deventry {
+ dev_t dev;
+ struct deventry *next;
+};
+
+static struct deventry *devlist = NULL;
+static spinlock_t devlock = SPIN_LOCK_UNLOCKED;
+
+int dev_check_rdonly(struct block_device *bdev)
+{
+ struct deventry *cur;
+ if (!bdev) return 0;
+ spin_lock(&devlock);
+ cur = devlist;
+ while(cur) {
+ if (bdev->bd_dev == cur->dev) {
+ spin_unlock(&devlock);
+ return 1;
+ }
+ cur = cur->next;
+ }
+ spin_unlock(&devlock);
+ return 0;
+}
+
+void dev_set_rdonly(struct block_device *bdev)
+{
+ struct deventry *newdev, *cur;
+
+ if (!bdev)
+ return;
+ newdev = kmalloc(sizeof(struct deventry), GFP_KERNEL);
+ if (!newdev)
+ return;
+
+ spin_lock(&devlock);
+ cur = devlist;
+ while(cur) {
+ if (bdev->bd_dev == cur->dev) {
+ spin_unlock(&devlock);
+ kfree(newdev);
+ return;
+ }
+ cur = cur->next;
+ }
+ newdev->dev = bdev->bd_dev;
+ newdev->next = devlist;
+ devlist = newdev;
+ spin_unlock(&devlock);
+ printk(KERN_WARNING "Turning device %s (%#x) read-only\n",
+ bdev->bd_disk ? bdev->bd_disk->disk_name : "", bdev->bd_dev);
+}
+
+void dev_clear_rdonly(struct block_device *bdev)
+{
+ struct deventry *cur, *last = NULL;
+ if (!bdev) return;
+ spin_lock(&devlock);
+ cur = devlist;
+ while(cur) {
+ if (bdev->bd_dev == cur->dev) {
+ if (last)
+ last->next = cur->next;
+ else
+ devlist = cur->next;
+ spin_unlock(&devlock);
+ kfree(cur);
+ printk(KERN_WARNING "Removing read-only on %s (%#x)\n",
+ bdev->bd_disk ? bdev->bd_disk->disk_name :
+ "unknown block", bdev->bd_dev);
+ return;
+ }
+ last = cur;
+ cur = cur->next;
+ }
+ spin_unlock(&devlock);
+}
+
+EXPORT_SYMBOL(dev_set_rdonly);
+EXPORT_SYMBOL(dev_clear_rdonly);
+
int __init blk_dev_init(void)
{
int i;
diff -u -r debian-2.6.26/drivers/md/raid5.c debian-2.6.26_lustre.1.8.2/drivers/md/raid5.c
--- debian-2.6.26/drivers/md/raid5.c 2008-07-13 23:51:29.000000000 +0200
+++ debian-2.6.26_lustre.1.8.2/drivers/md/raid5.c 2010-02-12 15:19:25.000000000 +0100
@@ -1817,6 +1817,8 @@
bi->bi_next = *bip;
*bip = bi;
bi->bi_phys_segments ++;
+ if (bio_sync(bi) && !forwrite)
+ clear_bit(R5_UPTODATE, &sh->dev[dd_idx].flags); /* force to read from disk. */
spin_unlock_irq(&conf->device_lock);
spin_unlock(&sh->lock);
@@ -3699,6 +3701,8 @@
test_bit(BIO_UPTODATE, &bi->bi_flags)
? 0 : -EIO);
}
+ if (bio_sync(bi))
+ raid5_unplug_device(q);
return 0;
}
diff -u -r debian-2.6.26/drivers/scsi/Kconfig debian-2.6.26_lustre.1.8.2/drivers/scsi/Kconfig
--- debian-2.6.26/drivers/scsi/Kconfig 2009-12-26 09:14:53.000000000 +0100
+++ debian-2.6.26_lustre.1.8.2/drivers/scsi/Kconfig 2010-02-12 15:20:02.000000000 +0100
@@ -81,6 +81,14 @@
In this case, do not compile the driver for your SCSI host adapter
(below) as a module either.
+config SD_IOSTATS
+ bool "Enable SCSI disk I/O stats"
+ depends on BLK_DEV_SD
+ default y
+ ---help---
+ This enables SCSI disk I/O stats collection. You must also enable
+ /proc file system support if you want this feature.
+
config CHR_DEV_ST
tristate "SCSI tape support"
depends on SCSI
diff -u -r debian-2.6.26/drivers/scsi/scsi_proc.c debian-2.6.26_lustre.1.8.2/drivers/scsi/scsi_proc.c
--- debian-2.6.26/drivers/scsi/scsi_proc.c 2008-07-13 23:51:29.000000000 +0200
+++ debian-2.6.26_lustre.1.8.2/drivers/scsi/scsi_proc.c 2010-02-12 15:22:05.000000000 +0100
@@ -40,7 +40,8 @@
/* 4K page size, but our output routines, use some slack for overruns */
#define PROC_BLOCK_SIZE (3*1024)
-static struct proc_dir_entry *proc_scsi;
+struct proc_dir_entry *proc_scsi;
+EXPORT_SYMBOL(proc_scsi);
/* Protect sht->present and sht->proc_dir */
static DEFINE_MUTEX(global_host_template_mutex);
diff -u -r debian-2.6.26/drivers/scsi/sd.c debian-2.6.26_lustre.1.8.2/drivers/scsi/sd.c
--- debian-2.6.26/drivers/scsi/sd.c 2008-07-13 23:51:29.000000000 +0200
+++ debian-2.6.26_lustre.1.8.2/drivers/scsi/sd.c 2010-02-17 14:13:32.000000000 +0100
@@ -107,6 +107,24 @@
* object after last put) */
static DEFINE_MUTEX(sd_ref_mutex);
+#if (defined(CONFIG_SD_IOSTATS) && defined(CONFIG_PROC_FS))
+# include <linux/proc_fs.h>
+# include <linux/seq_file.h>
+struct proc_dir_entry *sd_iostats_procdir = NULL;
+char sd_iostats_procdir_name[] = "sd_iostats";
+static struct file_operations sd_iostats_proc_fops;
+
+extern void sd_iostats_init(void);
+extern void sd_iostats_fini(void);
+void sd_iostats_start_req(struct scsi_cmnd *SCpnt);
+void sd_iostats_finish_req(struct scsi_cmnd *SCpnt);
+#else
+static inline void sd_iostats_init(void) {}
+static inline void sd_iostats_fini(void) {}
+static inline void sd_iostats_start_req(struct scsi_cmnd *SCpnt) {}
+static inline void sd_iostats_finish_req(struct scsi_cmnd *SCpnt) {}
+#endif
+
static const char *sd_cache_types[] = {
"write through", "none", "write back",
"write back, no read (daft)"
@@ -531,6 +549,8 @@
}
SCpnt->sdb.length = this_count * sdp->sector_size;
+ sd_iostats_start_req(SCpnt);
+
/*
* We shouldn't disconnect in the middle of a sector, so with a dumb
* host adapter, it's safe to assume that we can at least transfer
@@ -667,7 +687,7 @@
int diskinfo[4];
/* default to most commonly used values */
- diskinfo[0] = 0x40; /* 1 << 6 */
+ diskinfo[0] = 0x40; /* 1 << 6 */
diskinfo[1] = 0x20; /* 1 << 5 */
diskinfo[2] = sdkp->capacity >> 11;
@@ -1023,6 +1043,7 @@
break;
}
out:
+ sd_iostats_finish_req(SCpnt);
return good_bytes;
}
@@ -1711,6 +1732,36 @@
gd->flags = GENHD_FL_DRIVERFS;
if (sdp->removable)
gd->flags |= GENHD_FL_REMOVABLE;
+#if (defined(CONFIG_SD_IOSTATS) && defined(CONFIG_PROC_FS))
+ sdkp->stats = kzalloc(sizeof(iostat_stats_t), GFP_KERNEL);
+ if (!sdkp->stats) {
+ printk(KERN_WARNING "cannot allocate iostat structure for"
+ "%s\n", gd->disk_name);
+ } else {
+ do_gettimeofday(&sdkp->stats->iostat_timeval);
+ sdkp->stats->iostat_queue_stamp = jiffies;
+ spin_lock_init(&sdkp->stats->iostat_lock);
+ if (sd_iostats_procdir) {
+ struct proc_dir_entry *pde;
+ pde = create_proc_entry(gd->disk_name, S_IRUGO | S_IWUSR,
+ sd_iostats_procdir);
+ if (!pde) {
+ printk(KERN_WARNING "Can't create /proc/scsi/"
+ "%s/%s\n",
+ sd_iostats_procdir_name,
+ gd->disk_name);
+ kfree(sdkp->stats);
+ sdkp->stats = NULL;
+ } else {
+ pde->proc_fops = &sd_iostats_proc_fops;
+ pde->data = gd;
+ }
+ } else {
+ kfree(sdkp->stats);
+ sdkp->stats = NULL;
+ }
+ }
+#endif
dev_set_drvdata(dev, sdkp);
add_disk(gd);
@@ -1755,6 +1806,366 @@
return 0;
}
+#if (defined(CONFIG_SD_IOSTATS) && defined(CONFIG_PROC_FS))
+static int
+ sd_iostats_seq_show(struct seq_file *seq, void *v)
+ {
+ struct timeval now;
+ struct gendisk *disk = seq->private;
+ iostat_stats_t *stats;
+ unsigned long long read_len;
+ unsigned long long read_len_tot;
+ unsigned long read_num;
+ unsigned long read_num_tot;
+ unsigned long long write_len;
+ unsigned long long write_len_tot;
+ unsigned long write_num;
+ unsigned long write_num_tot;
+ int i;
+ int maxi;
+
+ stats = scsi_disk(disk)->stats;
+ if (stats == NULL) {
+ printk(KERN_ERR "sd_iostats_seq_show: NULL stats entry\n");
+ BUG();
+ }
+
+ do_gettimeofday(&now);
+ now.tv_sec -= stats->iostat_timeval.tv_sec;
+ now.tv_usec -= stats->iostat_timeval.tv_usec;
+ if (now.tv_usec < 0) {
+ now.tv_usec += 1000000;
+ now.tv_sec--;
+ }
+
+ /* this sampling races with updates */
+ seq_printf(seq, "index: %lu snapshot_time: %lu.%06lu\n",
+ (unsigned long) scsi_disk(disk)->index,
+ now.tv_sec, now.tv_usec);
+
+ for (i = IOSTAT_NCOUNTERS - 1; i > 0; i--)
+ if (stats->iostat_read_histogram[i].iostat_count != 0 ||
+ stats->iostat_write_histogram[i].iostat_count != 0)
+ break;
+ maxi = i;
+
+ seq_printf(seq, "%8s %8s %12s %8s %12s\n", "size",
+ "reads", "total", "writes", "total");
+
+ read_len_tot = write_len_tot = 0;
+ read_num_tot = write_num_tot = 0;
+ for (i = 0; i <= maxi; i++) {
+ read_len = stats->iostat_read_histogram[i].iostat_size;
+ read_len_tot += read_len;
+ read_num = stats->iostat_read_histogram[i].iostat_count;
+ read_num_tot += read_num;
+
+ write_len = stats->iostat_write_histogram[i].iostat_size;
+ write_len_tot += write_len;
+ write_num = stats->iostat_write_histogram[i].iostat_count;
+ write_num_tot += write_num;
+
+ seq_printf (seq, "%8d %8lu %12llu %8lu %12llu\n",
+ 512<<i, read_num, read_len, write_num, write_len);
+ }
+
+ seq_printf(seq, "%8s %8lu %12llu %8lu %12llu\n\n", "total",
+ read_num_tot, read_len_tot,
+ write_num_tot, write_len_tot);
+
+ seq_printf(seq, "%8s %8s %8s\n", "qdepth", "ticks", "%");
+ for (i = 0; i < IOSTAT_NCOUNTERS; i++) {
+ unsigned long long ticks, percent;
+ ticks = stats->iostat_queue_ticks[i];
+ if (ticks == 0)
+ continue;
+ percent = stats->iostat_queue_ticks[i] * 100;
+ do_div(percent, stats->iostat_queue_ticks_sum);
+ seq_printf(seq, "%8d %8llu %8llu\n", i, ticks, percent);
+ }
+
+ if (stats->iostat_reqs != 0) {
+ unsigned long long aveseek = 0, percent = 0;
+
+ if (stats->iostat_seeks) {
+ aveseek = stats->iostat_seek_sectors;
+ do_div(aveseek, stats->iostat_seeks);
+ percent = stats->iostat_seeks * 100;
+ do_div(percent, stats->iostat_reqs);
+ }
+
+ seq_printf(seq, "\n%llu sectors in %llu reqs: %llu seek(s) over "
+ "%llu sectors in ave, %llu%% of all reqs\n",
+ stats->iostat_sectors, stats->iostat_reqs,
+ stats->iostat_seeks, aveseek, percent);
+ }
+
+ seq_printf(seq, "\n%16s %8s %8s %8s %8s\n", "process time", "reads",
+ "%%", "writes", "%%");
+ for (i = 0; i < IOSTAT_NCOUNTERS; i++) {
+ unsigned long read_percent = 0, write_percent = 0;
+ if (stats->iostat_wtime[i] == 0 &&
+ stats->iostat_rtime[i] == 0)
+ continue;
+ if (stats->iostat_read_reqs)
+ read_percent = stats->iostat_rtime[i] * 100 /
+ stats->iostat_read_reqs;
+ if (stats->iostat_write_reqs)
+ write_percent = stats->iostat_wtime[i] * 100 /
+ stats->iostat_write_reqs;
+ seq_printf(seq, "%16u %8lu %8lu %8lu %8lu\n",
+ jiffies_to_msecs(((1UL << i) >> 1) << 1),
+ stats->iostat_rtime[i], read_percent,
+ stats->iostat_wtime[i], write_percent);
+ }
+
+ seq_printf(seq, "\n%16s %8s %8s %8s %8s\n", "time in queue", "reads",
+ "%%", "writes", "%%");
+ for (i = 0; i < IOSTAT_NCOUNTERS; i++) {
+ unsigned long read_percent = 0, write_percent = 0;
+ if (stats->iostat_wtime_in_queue[i] == 0 &&
+ stats->iostat_rtime_in_queue[i] == 0)
+ continue;
+ if (stats->iostat_read_reqs)
+ read_percent = stats->iostat_rtime_in_queue[i] * 100 /
+ stats->iostat_read_reqs;
+ if (stats->iostat_write_reqs)
+ write_percent = stats->iostat_wtime_in_queue[i] * 100 /
+ stats->iostat_write_reqs;
+ seq_printf(seq, "%16u %8lu %8lu %8lu %8lu\n",
+ jiffies_to_msecs(((1UL << i) >> 1) << 1),
+ stats->iostat_rtime_in_queue[i],
+ read_percent,
+ stats->iostat_wtime_in_queue[i],
+ write_percent);
+ }
+
+ return 0;
+ }
+
+ static void *
+ sd_iostats_seq_start(struct seq_file *p, loff_t *pos)
+ {
+ return (*pos == 0) ? (void *)1 : NULL;
+ }
+
+ static void *
+ sd_iostats_seq_next(struct seq_file *p, void *v, loff_t *pos)
+ {
+ ++*pos;
+ return NULL;
+ }
+
+ static void
+ sd_iostats_seq_stop(struct seq_file *p, void *v)
+ {
+ }
+
+ static struct seq_operations sd_iostats_seqops = {
+ .start = sd_iostats_seq_start,
+ .stop = sd_iostats_seq_stop,
+ .next = sd_iostats_seq_next,
+ .show = sd_iostats_seq_show,
+ };
+
+ static int
+ sd_iostats_seq_open (struct inode *inode, struct file *file)
+ {
+ int rc;
+
+ rc = seq_open(file, &sd_iostats_seqops);
+ if (rc != 0)
+ return rc;
+
+ ((struct seq_file *)file->private_data)->private = PDE(inode)->data;
+ return 0;
+ }
+
+ static ssize_t
+ sd_iostats_seq_write(struct file *file, const char *buffer,
+ size_t len, loff_t *off)
+ {
+ struct seq_file *seq = file->private_data;
+ struct gendisk *disk = seq->private;
+ iostat_stats_t *stats = scsi_disk(disk)->stats;
+ unsigned long flags;
+ unsigned long qdepth;
+
+
+ spin_lock_irqsave (&stats->iostat_lock, flags);
+ qdepth = stats->iostat_queue_depth;
+ memset (stats, 0, offsetof(iostat_stats_t, iostat_lock));
+ do_gettimeofday(&stats->iostat_timeval);
+ stats->iostat_queue_stamp = jiffies;
+ stats->iostat_queue_depth = qdepth;
+ spin_unlock_irqrestore (&stats->iostat_lock, flags);
+
+ return len;
+ }
+
+ static struct file_operations sd_iostats_proc_fops = {
+ .owner = THIS_MODULE,
+ .open = sd_iostats_seq_open,
+ .read = seq_read,
+ .write = sd_iostats_seq_write,
+ .llseek = seq_lseek,
+ .release = seq_release,
+ };
+
+ extern struct proc_dir_entry *proc_scsi;
+
+ void
+ sd_iostats_init(void)
+ {
+ if (proc_scsi == NULL) {
+ printk(KERN_WARNING "No access to sd iostats: "
+ "proc_scsi is NULL\n");
+ return;
+ }
+
+ sd_iostats_procdir = create_proc_entry(sd_iostats_procdir_name,
+ S_IFDIR | S_IRUGO | S_IXUGO,
+ proc_scsi);
+ if (sd_iostats_procdir == NULL) {
+ printk(KERN_WARNING "No access to sd iostats: "
+ "can't create /proc/scsi/%s\n", sd_iostats_procdir_name);
+ return;
+ }
+ }
+
+ void sd_iostats_fini(void)
+ {
+ if (proc_scsi != NULL && sd_iostats_procdir != NULL)
+ remove_proc_entry(sd_iostats_procdir_name, proc_scsi);
+
+ sd_iostats_procdir = NULL;
+ }
+
+ void sd_iostats_finish_req(struct scsi_cmnd *SCpnt)
+ {
+ struct request *rq = SCpnt->request;
+ iostat_stats_t *stats;
+ unsigned long *tcounter;
+ int tbucket;
+ int tmp;
+ unsigned long irqflags;
+ unsigned long i;
+
+ stats = scsi_disk(rq->rq_disk)->stats;
+ if (stats == NULL)
+ return;
+
+ tmp = jiffies - rq->start_time;
+ for (tbucket = 0; tmp > 1; tbucket++)
+ tmp >>= 1;
+ if (tbucket >= IOSTAT_NCOUNTERS)
+ tbucket = IOSTAT_NCOUNTERS - 1;
+ //printk("%u ticks in D to %u\n", jiffies - rq->start_time, tbucket);
+
+ tcounter = rq_data_dir(rq) == WRITE ?
+ &stats->iostat_wtime[tbucket] : &stats->iostat_rtime[tbucket];
+
+ spin_lock_irqsave(&stats->iostat_lock, irqflags);
+
+ /* update delay stats */
+ (*tcounter)++;
+
+ /* update queue depth stats */
+ i = stats->iostat_queue_depth;
+ if (i >= IOSTAT_NCOUNTERS)
+ i = IOSTAT_NCOUNTERS - 1;
+ stats->iostat_queue_ticks[i] += jiffies - stats->iostat_queue_stamp;
+ stats->iostat_queue_ticks_sum += jiffies - stats->iostat_queue_stamp;
+ BUG_ON(stats->iostat_queue_depth == 0);
+ stats->iostat_queue_depth--;
+
+ /* update seek stats. XXX: not sure about nr_sectors */
+ stats->iostat_sectors += rq->nr_sectors;
+ stats->iostat_reqs++;
+ if (rq->sector != stats->iostat_next_sector) {
+ stats->iostat_seek_sectors +=
+ rq->sector > stats->iostat_next_sector ?
+ rq->sector - stats->iostat_next_sector :
+ stats->iostat_next_sector - rq->sector;
+ stats->iostat_seeks++;
+ }
+ stats->iostat_next_sector = rq->sector + rq->nr_sectors;
+
+ stats->iostat_queue_stamp = jiffies;
+
+ spin_unlock_irqrestore(&stats->iostat_lock, irqflags);
+ }
+
+ void sd_iostats_start_req(struct scsi_cmnd *SCpnt)
+ {
+ struct request *rq = SCpnt->request;
+ iostat_stats_t *stats;
+ iostat_counter_t *counter;
+ int bucket;
+ int tbucket;
+ int tmp;
+ unsigned long irqflags;
+ unsigned long i;
+ int nsect;
+
+ stats = scsi_disk(rq->rq_disk)->stats;
+ if (stats == NULL)
+ return;
+
+ nsect = scsi_bufflen(SCpnt) >> 9;
+ for (bucket = 0, tmp = nsect; tmp > 1; bucket++)
+ tmp >>= 1;
+
+ if (bucket >= IOSTAT_NCOUNTERS) {
+ printk (KERN_ERR "sd_iostats_bump: nsect %d too big\n", nsect);
+ BUG();
+ }
+
+ counter = rq_data_dir(rq) == WRITE ?
+ &stats->iostat_write_histogram[bucket] :
+ &stats->iostat_read_histogram[bucket];
+
+ tmp = jiffies - rq->start_time;
+ for (tbucket = 0; tmp > 1; tbucket++)
+ tmp >>= 1;
+ if (tbucket >= IOSTAT_NCOUNTERS)
+ tbucket = IOSTAT_NCOUNTERS - 1;
+ //printk("%u ticks in Q to %u\n", jiffies - rq->start_time, tbucket);
+
+ /* an ugly hack to know exact processing time. the right
+ * solution is to add one more field to struct request
+ * hopefully it will break nothing ... */
+ rq->start_time = jiffies;
+
+ spin_lock_irqsave(&stats->iostat_lock, irqflags);
+
+ /* update queue depth stats */
+ i = stats->iostat_queue_depth;
+ if (i >= IOSTAT_NCOUNTERS)
+ i = IOSTAT_NCOUNTERS - 1;
+ stats->iostat_queue_ticks[i] += jiffies - stats->iostat_queue_stamp;
+ stats->iostat_queue_ticks_sum += jiffies - stats->iostat_queue_stamp;
+ stats->iostat_queue_depth++;
+
+ /* update delay stats */
+ if (rq_data_dir(rq) == WRITE) {
+ stats->iostat_wtime_in_queue[tbucket]++;
+ stats->iostat_write_reqs++;
+ } else {
+ stats->iostat_rtime_in_queue[tbucket]++;
+ stats->iostat_read_reqs++;
+ }
+
+ /* update size stats */
+ counter->iostat_size += nsect;
+ counter->iostat_count++;
+
+ stats->iostat_queue_stamp = jiffies;
+
+ spin_unlock_irqrestore(&stats->iostat_lock, irqflags);
+}
+#endif
+
/**
* scsi_disk_release - Called to free the scsi_disk structure
* @dev: pointer to embedded class device
@@ -1773,6 +2184,13 @@
idr_remove(&sd_index_idr, sdkp->index);
spin_unlock(&sd_index_lock);
+#if (defined(CONFIG_SD_IOSTATS) && defined(CONFIG_PROC_FS))
+ if (sdkp->stats) {
+ remove_proc_entry(disk->disk_name, sd_iostats_procdir);
+ kfree(sdkp->stats);
+ sdkp->stats = NULL;
+ }
+#endif
disk->private_data = NULL;
put_disk(disk);
put_device(&sdkp->device->sdev_gendev);
@@ -1890,6 +2308,8 @@
if (!majors)
return -ENODEV;
+ sd_iostats_init();
+
err = class_register(&sd_disk_class);
if (err)
goto err_out;
@@ -1905,6 +2325,7 @@
err_out:
for (i = 0; i < SD_MAJORS; i++)
unregister_blkdev(sd_major(i), "sd");
+ sd_iostats_fini();
return err;
}
diff -u -r debian-2.6.26/fs/block_dev.c debian-2.6.26_lustre.1.8.2/fs/block_dev.c
--- debian-2.6.26/fs/block_dev.c 2008-07-13 23:51:29.000000000 +0200
+++ debian-2.6.26_lustre.1.8.2/fs/block_dev.c 2010-02-17 14:19:50.000000000 +0100
@@ -1125,6 +1125,7 @@
if (bdev != bdev->bd_contains)
victim = bdev->bd_contains;
bdev->bd_contains = NULL;
+ dev_clear_rdonly(bdev);
}
unlock_kernel();
mutex_unlock(&bdev->bd_mutex);
diff -u -r debian-2.6.26/fs/dcache.c debian-2.6.26_lustre.1.8.2/fs/dcache.c
--- debian-2.6.26/fs/dcache.c 2009-12-26 09:14:56.000000000 +0100
+++ debian-2.6.26_lustre.1.8.2/fs/dcache.c 2010-02-17 14:24:24.000000000 +0100
@@ -250,6 +250,13 @@
spin_unlock(&dcache_lock);
return 0;
}
+
+ /* network invalidation by Lustre */
+ if (dentry->d_flags & DCACHE_LUSTRE_INVALID) {
+ spin_unlock(&dcache_lock);
+ return 0;
+ }
+
/*
* Check whether to do a partial shrink_dcache
* to get rid of unused child entries.
@@ -1427,14 +1434,23 @@
*
* Adds a dentry to the hash according to its name.
*/
-
+
+void d_rehash_cond(struct dentry * entry, int lock)
+{
+ if (lock)
+ spin_lock(&dcache_lock);
+ spin_lock(&entry->d_lock);
+ _d_rehash(entry);
+ spin_unlock(&entry->d_lock);
+ if (lock)
+ spin_unlock(&dcache_lock);
+}
+
+EXPORT_SYMBOL(d_rehash_cond);
+
void d_rehash(struct dentry * entry)
{
- spin_lock(&dcache_lock);
- spin_lock(&entry->d_lock);
- _d_rehash(entry);
- spin_unlock(&entry->d_lock);
- spin_unlock(&dcache_lock);
+ d_rehash_cond(entry, 1);
}
#define do_switch(x,y) do { \
@@ -1510,7 +1526,7 @@
* Update the dcache to reflect the move of a file name. Negative
* dcache entries should not be moved in this way.
*/
-static void d_move_locked(struct dentry * dentry, struct dentry * target)
+void d_move_locked(struct dentry * dentry, struct dentry * target)
{
struct hlist_head *list;
@@ -1568,6 +1584,7 @@
spin_unlock(&dentry->d_lock);
write_sequnlock(&rename_lock);
}
+EXPORT_SYMBOL(d_move_locked);
/**
* d_move - move a dentry
@@ -2051,6 +2068,7 @@
return result;
}
+EXPORT_SYMBOL(is_subdir);
void d_genocide(struct dentry *root)
{
diff -u -r debian-2.6.26/fs/filesystems.c debian-2.6.26_lustre.1.8.2/fs/filesystems.c
--- debian-2.6.26/fs/filesystems.c 2009-12-26 09:14:56.000000000 +0100
+++ debian-2.6.26_lustre.1.8.2/fs/filesystems.c 2010-02-17 14:26:59.000000000 +0100
@@ -28,7 +28,9 @@
*/
static struct file_system_type *file_systems;
-static DEFINE_RWLOCK(file_systems_lock);
+DEFINE_RWLOCK(file_systems_lock);
+
+EXPORT_SYMBOL(file_systems_lock);
/* WARNING: This can be used only if we _already_ own a reference */
void get_filesystem(struct file_system_type *fs)
diff -u -r debian-2.6.26/fs/inode.c debian-2.6.26_lustre.1.8.2/fs/inode.c
--- debian-2.6.26/fs/inode.c 2008-07-13 23:51:29.000000000 +0200
+++ debian-2.6.26_lustre.1.8.2/fs/inode.c 2010-02-17 14:27:51.000000000 +0100
@@ -409,7 +409,9 @@
int nr_scanned;
unsigned long reap = 0;
- mutex_lock(&iprune_mutex);
+ if (!mutex_trylock(&iprune_mutex))
+ return;
+
spin_lock(&inode_lock);
for (nr_scanned = 0; nr_scanned < nr_to_scan; nr_scanned++) {
struct inode *inode;
diff -u -r debian-2.6.26/fs/jbd2/checkpoint.c debian-2.6.26_lustre.1.8.2/fs/jbd2/checkpoint.c
--- debian-2.6.26/fs/jbd2/checkpoint.c 2008-07-13 23:51:29.000000000 +0200
+++ debian-2.6.26_lustre.1.8.2/fs/jbd2/checkpoint.c 2010-02-17 14:28:49.000000000 +0100
@@ -696,6 +696,7 @@
J_ASSERT(transaction->t_checkpoint_list == NULL);
J_ASSERT(transaction->t_checkpoint_io_list == NULL);
J_ASSERT(transaction->t_updates == 0);
+ J_ASSERT(list_empty(&transaction->t_jcb));
J_ASSERT(journal->j_committing_transaction != transaction);
J_ASSERT(journal->j_running_transaction != transaction);
diff -u -r debian-2.6.26/fs/jbd2/commit.c debian-2.6.26_lustre.1.8.2/fs/jbd2/commit.c
--- debian-2.6.26/fs/jbd2/commit.c 2008-07-13 23:51:29.000000000 +0200
+++ debian-2.6.26_lustre.1.8.2/fs/jbd2/commit.c 2010-02-17 14:29:44.000000000 +0100
@@ -874,6 +874,30 @@
transaction can be removed from any checkpoint list it was on
before. */
+ /*
+ * Call any callbacks that had been registered for handles in this
+ * transaction. It is up to the callback to free any allocated
+ * memory.
+ *
+ * The spinlocking (t_jcb_lock) here is surely unnecessary...
+ */
+ spin_lock(&commit_transaction->t_jcb_lock);
+ if (!list_empty(&commit_transaction->t_jcb)) {
+ struct list_head *p, *n;
+ int error = is_journal_aborted(journal);
+
+ list_for_each_safe(p, n, &commit_transaction->t_jcb) {
+ struct journal_callback *jcb;
+
+ jcb = list_entry(p, struct journal_callback, jcb_list);
+ list_del(p);
+ spin_unlock(&commit_transaction->t_jcb_lock);
+ jcb->jcb_func(jcb, error);
+ spin_lock(&commit_transaction->t_jcb_lock);
+ }
+ }
+ spin_unlock(&commit_transaction->t_jcb_lock);
+
jbd_debug(3, "JBD: commit phase 7\n");
J_ASSERT(commit_transaction->t_sync_datalist == NULL);
diff -u -r debian-2.6.26/fs/jbd2/journal.c debian-2.6.26_lustre.1.8.2/fs/jbd2/journal.c
--- debian-2.6.26/fs/jbd2/journal.c 2008-07-13 23:51:29.000000000 +0200
+++ debian-2.6.26_lustre.1.8.2/fs/jbd2/journal.c 2010-02-17 14:38:15.000000000 +0100
@@ -82,6 +82,8 @@
EXPORT_SYMBOL(jbd2_journal_invalidatepage);
EXPORT_SYMBOL(jbd2_journal_try_to_free_buffers);
EXPORT_SYMBOL(jbd2_journal_force_commit);
+EXPORT_SYMBOL(jbd2_journal_callback_set);
+EXPORT_SYMBOL(jbd2_journal_bmap);
static int journal_convert_superblock_v1(journal_t *, journal_superblock_t *);
static void __journal_abort_soft (journal_t *journal, int errno);
@@ -460,6 +462,7 @@
spin_unlock(&journal->j_state_lock);
return ret;
}
+EXPORT_SYMBOL(jbd2_log_start_commit);
/*
* Force and wait upon a commit if the calling process is not within
diff -u -r debian-2.6.26/fs/jbd2/transaction.c debian-2.6.26_lustre.1.8.2/fs/jbd2/transaction.c
--- debian-2.6.26/fs/jbd2/transaction.c 2008-07-13 23:51:29.000000000 +0200
+++ debian-2.6.26_lustre.1.8.2/fs/jbd2/transaction.c 2010-02-17 14:42:20.000000000 +0100
@@ -51,10 +51,12 @@
transaction->t_state = T_RUNNING;
transaction->t_tid = journal->j_transaction_sequence++;
transaction->t_expires = jiffies + journal->j_commit_interval;
+ INIT_LIST_HEAD(&transaction->t_jcb);
spin_lock_init(&transaction->t_handle_lock);
+ spin_lock_init(&transaction->t_jcb_lock);
/* Set up the commit timer for the new transaction. */
- journal->j_commit_timer.expires = round_jiffies(transaction->t_expires);
+ journal->j_commit_timer.expires = transaction->t_expires;
add_timer(&journal->j_commit_timer);
J_ASSERT(journal->j_running_transaction == NULL);
@@ -252,6 +254,7 @@
memset(handle, 0, sizeof(*handle));
handle->h_buffer_credits = nblocks;
handle->h_ref = 1;
+ INIT_LIST_HEAD(&handle->h_jcb);
lockdep_init_map(&handle->h_lockdep_map, "jbd2_handle",
&jbd2_handle_key, 0);
@@ -1350,6 +1353,36 @@
}
/**
+ * void jbd2_journal_callback_set() - Register a callback function for this handle.
+ * @handle: handle to attach the callback to.
+ * @func: function to callback.
+ * @jcb: structure with additional information required by func() , and
+ * some space for jbd2 internal information.
+ *
+ * The function will be
+ * called when the transaction that this handle is part of has been
+ * committed to disk with the original callback data struct and the
+ * error status of the journal as parameters. There is no guarantee of
+ * ordering between handles within a single transaction, nor between
+ * callbacks registered on the same handle.
+ *
+ * The caller is responsible for allocating the journal_callback struct.
+ * This is to allow the caller to add as much extra data to the callback
+ * as needed, but reduce the overhead of multiple allocations. The caller
+ * allocated struct must start with a struct journal_callback at offset 0,
+ * and has the caller-specific data afterwards.
+ */
+void jbd2_journal_callback_set(handle_t *handle,
+ void (*func)(struct journal_callback *jcb, int error),
+ struct journal_callback *jcb)
+{
+ spin_lock(&handle->h_transaction->t_jcb_lock);
+ list_add_tail(&jcb->jcb_list, &handle->h_jcb);
+ spin_unlock(&handle->h_transaction->t_jcb_lock);
+ jcb->jcb_func = func;
+}
+
+/**
* int jbd2_journal_stop() - complete a transaction
* @handle: tranaction to complete.
*
@@ -1423,6 +1456,11 @@
wake_up(&journal->j_wait_transaction_locked);
}
+ /* Move callbacks from the handle to the transaction. */
+ spin_lock(&transaction->t_jcb_lock);
+ list_splice(&handle->h_jcb, &transaction->t_jcb);
+ spin_unlock(&transaction->t_jcb_lock);
+
/*
* If the handle is marked SYNC, we need to set another commit
* going! We also want to force a commit if the current
diff -u -r debian-2.6.26/fs/namespace.c debian-2.6.26_lustre.1.8.2/fs/namespace.c
--- debian-2.6.26/fs/namespace.c 2009-12-26 09:14:56.000000000 +0100
+++ debian-2.6.26_lustre.1.8.2/fs/namespace.c 2010-02-17 14:43:07.000000000 +0100
@@ -1660,6 +1660,7 @@
return do_add_mount(mnt, nd, mnt_flags, NULL);
}
+EXPORT_SYMBOL(set_fs_pwd);
/*
* add a mount into a namespace's mount tree
diff -u -r debian-2.6.26/include/linux/blkdev.h debian-2.6.26_lustre.1.8.2/include/linux/blkdev.h
--- debian-2.6.26/include/linux/blkdev.h 2009-12-26 09:14:55.000000000 +0100
+++ debian-2.6.26_lustre.1.8.2/include/linux/blkdev.h 2010-02-18 07:44:31.000000000 +0100
@@ -806,7 +806,7 @@
#define MAX_PHYS_SEGMENTS 128
#define MAX_HW_SEGMENTS 128
#define SAFE_MAX_SECTORS 255
-#define BLK_DEF_MAX_SECTORS 1024
+#define BLK_DEF_MAX_SECTORS 2048
#define MAX_SEGMENT_SIZE 65536
diff -u -r debian-2.6.26/include/linux/dcache.h debian-2.6.26_lustre.1.8.2/include/linux/dcache.h
--- debian-2.6.26/include/linux/dcache.h 2008-07-13 23:51:29.000000000 +0200
+++ debian-2.6.26_lustre.1.8.2/include/linux/dcache.h 2010-02-18 07:46:08.000000000 +0100
@@ -173,6 +173,7 @@
#define DCACHE_REFERENCED 0x0008 /* Recently used, don't discard. */
#define DCACHE_UNHASHED 0x0010
+#define DCACHE_LUSTRE_INVALID 0x0040 /* Lustre invalidated */
#define DCACHE_INOTIFY_PARENT_WATCHED 0x0020 /* Parent inode is watched */
@@ -250,6 +251,7 @@
* This adds the entry to the hash queues.
*/
extern void d_rehash(struct dentry *);
+extern void d_rehash_cond(struct dentry *, int lock);
/**
* d_add - add dentry to hash queues
@@ -285,6 +287,7 @@
/* used for rename() and baskets */
extern void d_move(struct dentry *, struct dentry *);
+extern void d_move_locked(struct dentry *, struct dentry *);
/* appendix may either be NULL or be used for transname suffixes */
extern struct dentry * d_lookup(struct dentry *, struct qstr *);
diff -u -r debian-2.6.26/include/linux/fs.h debian-2.6.26_lustre.1.8.2/include/linux/fs.h
--- debian-2.6.26/include/linux/fs.h 2008-07-13 23:51:29.000000000 +0200
+++ debian-2.6.26_lustre.1.8.2/include/linux/fs.h 2010-02-18 07:57:56.000000000 +0100
@@ -1832,6 +1832,10 @@
extern void submit_bio(int, struct bio *);
extern int bdev_read_only(struct block_device *);
#endif
+#define HAVE_CLEAR_RDONLY_ON_PUT
+extern void dev_set_rdonly(struct block_device *bdev);
+extern int dev_check_rdonly(struct block_device *bdev);
+extern void dev_clear_rdonly(struct block_device *bdev);
extern int set_blocksize(struct block_device *, int);
extern int sb_set_blocksize(struct super_block *, int);
extern int sb_min_blocksize(struct super_block *, int);
@@ -1930,6 +1934,7 @@
extern const struct file_operations generic_ro_fops;
+extern rwlock_t file_systems_lock;
#define special_file(m) (S_ISCHR(m)||S_ISBLK(m)||S_ISFIFO(m)||S_ISSOCK(m))
extern int vfs_readlink(struct dentry *, char __user *, int, const char *);
diff -u -r debian-2.6.26/include/linux/jbd2.h debian-2.6.26_lustre.1.8.2/include/linux/jbd2.h
--- debian-2.6.26/include/linux/jbd2.h 2008-07-13 23:51:29.000000000 +0200
+++ debian-2.6.26_lustre.1.8.2/include/linux/jbd2.h 2010-02-18 08:08:30.000000000 +0100
@@ -379,6 +379,27 @@
bit_spin_unlock(BH_JournalHead, &bh->b_state);
}
+#define HAVE_JOURNAL_CALLBACK_STATUS
+/**
+ * struct journal_callback - Base structure for callback information.
+ * @jcb_list: list information for other callbacks attached to the same handle.
+ * @jcb_func: Function to call with this callback structure.
+ *
+ * This struct is a 'seed' structure for a using with your own callback
+ * structs. If you are using callbacks you must allocate one of these
+ * or another struct of your own definition which has this struct
+ * as it's first element and pass it to journal_callback_set().
+ *
+ * This is used internally by jbd2 to maintain callback information.
+ *
+ * See journal_callback_set for more information.
+ **/
+struct journal_callback {
+ struct list_head jcb_list; /* t_jcb_lock */
+ void (*jcb_func)(struct journal_callback *jcb, int error);
+ /* user data goes here */
+};
+
struct jbd2_revoke_table_s;
/**
@@ -387,6 +408,7 @@
* @h_transaction: Which compound transaction is this update a part of?
* @h_buffer_credits: Number of remaining buffers we are allowed to dirty.
* @h_ref: Reference count on this handle
+ * @h_jcb: List of application registered callbacks for this handle.
* @h_err: Field for caller's use to track errors through large fs operations
* @h_sync: flag for sync-on-close
* @h_jdata: flag to force data journaling
@@ -412,6 +434,13 @@
/* operations */
int h_err;
+ /*
+ * List of application registered callbacks for this handle. The
+ * function(s) will be called after the transaction that this handle is
+ * part of has been committed to disk. [t_jcb_lock]
+ */
+ struct list_head h_jcb;
+
/* Flags [no locking] */
unsigned int h_sync: 1; /* sync-on-close */
unsigned int h_jdata: 1; /* force data journaling */
@@ -467,6 +496,9 @@
* j_state_lock
* ->j_list_lock (journal_unmap_buffer)
*
+ * t_handle_lock
+ * ->t_jcb_lock
+ *
*/
struct transaction_s
@@ -613,6 +645,15 @@
*/
int t_handle_count;
+ /*
+ * Protects the callback list
+ */
+ spinlock_t t_jcb_lock;
+ /*
+ * List of registered callback functions for this transaction.
+ * Called when the transaction is committed. [t_jcb_lock]
+ */
+ struct list_head t_jcb;
};
struct transaction_run_stats_s {
@@ -1016,6 +1057,9 @@
extern int jbd2_journal_flush (journal_t *);
extern void jbd2_journal_lock_updates (journal_t *);
extern void jbd2_journal_unlock_updates (journal_t *);
+extern void jbd2_journal_callback_set(handle_t *handle,
+ void (*fn)(struct journal_callback *,int),
+ struct journal_callback *jcb);
extern journal_t * jbd2_journal_init_dev(struct block_device *bdev,
struct block_device *fs_dev,
diff -u -r debian-2.6.26/include/linux/mm.h debian-2.6.26_lustre.1.8.2/include/linux/mm.h
--- debian-2.6.26/include/linux/mm.h 2009-12-26 09:14:57.000000000 +0100
+++ debian-2.6.26_lustre.1.8.2/include/linux/mm.h 2010-02-18 08:10:11.000000000 +0100
@@ -564,6 +564,8 @@
{
return __va(page_to_pfn(page) << PAGE_SHIFT);
}
+/* truncate.c */
+extern void truncate_complete_page(struct address_space *mapping,struct page *);
#if defined(CONFIG_HIGHMEM) && !defined(WANT_PAGE_VIRTUAL)
#define HASHED_PAGE_VIRTUAL
diff -u -r debian-2.6.26/include/scsi/sd.h debian-2.6.26_lustre.1.8.2/include/scsi/sd.h
--- debian-2.6.26/include/scsi/sd.h 2008-07-13 23:51:29.000000000 +0200
+++ debian-2.6.26_lustre.1.8.2/include/scsi/sd.h 2010-02-17 14:18:52.000000000 +0100
@@ -31,6 +31,47 @@
*/
#define SD_BUF_SIZE 512
+#if (defined(CONFIG_SD_IOSTATS) && defined(CONFIG_PROC_FS))
+typedef struct {
+ unsigned long long iostat_size;
+ unsigned long long iostat_count;
+} iostat_counter_t;
+
+#define IOSTAT_NCOUNTERS 16
+typedef struct {
+ iostat_counter_t iostat_read_histogram[IOSTAT_NCOUNTERS];
+ iostat_counter_t iostat_write_histogram[IOSTAT_NCOUNTERS];
+ struct timeval iostat_timeval;
+
+ /* queue depth: how well the pipe is filled up */
+ unsigned long long iostat_queue_ticks[IOSTAT_NCOUNTERS];
+ unsigned long long iostat_queue_ticks_sum;
+ unsigned long iostat_queue_depth;
+ unsigned long iostat_queue_stamp;
+
+ /* seeks: how linear the traffic is */
+ unsigned long long iostat_next_sector;
+ unsigned long long iostat_seek_sectors;
+ unsigned long long iostat_seeks;
+ unsigned long long iostat_sectors;
+ unsigned long long iostat_reqs;
+ unsigned long iostat_read_reqs;
+ unsigned long iostat_write_reqs;
+
+ /* process time: how long it takes to process requests */
+ unsigned long iostat_rtime[IOSTAT_NCOUNTERS];
+ unsigned long iostat_wtime[IOSTAT_NCOUNTERS];
+
+ /* queue time: how long process spent in elevator's queue */
+ unsigned long iostat_rtime_in_queue[IOSTAT_NCOUNTERS];
+ unsigned long iostat_wtime_in_queue[IOSTAT_NCOUNTERS];
+
+ /* must be the last field, as it's used to know size to be memset'ed */
+ spinlock_t iostat_lock;
+} ____cacheline_aligned_in_smp iostat_stats_t;
+#endif
+
+
struct scsi_disk {
struct scsi_driver *driver; /* always &sd_template */
struct scsi_device *device;
@@ -45,6 +86,9 @@
unsigned WCE : 1; /* state of disk WCE bit */
unsigned RCD : 1; /* state of disk RCD bit, unused */
unsigned DPOFUA : 1; /* state of disk DPOFUA bit */
+#if (defined(CONFIG_SD_IOSTATS) && defined(CONFIG_PROC_FS))
+ iostat_stats_t *stats; /* scsi disk statistics */
+#endif
};
#define to_scsi_disk(obj) container_of(obj,struct scsi_disk,dev)
diff -u -r debian-2.6.26/kernel/sched.c debian-2.6.26_lustre.1.8.2/kernel/sched.c
--- debian-2.6.26/kernel/sched.c 2009-12-26 09:14:56.000000000 +0100
+++ debian-2.6.26_lustre.1.8.2/kernel/sched.c 2010-02-18 08:11:02.000000000 +0100
@@ -5477,6 +5477,7 @@
show_stack(p, NULL);
}
+EXPORT_SYMBOL(sched_show_task);
void show_state_filter(unsigned long state_filter)
{
diff -u -r debian-2.6.26/mm/truncate.c debian-2.6.26_lustre.1.8.2/mm/truncate.c
--- debian-2.6.26/mm/truncate.c 2008-07-13 23:51:29.000000000 +0200
+++ debian-2.6.26_lustre.1.8.2/mm/truncate.c 2010-02-18 08:12:58.000000000 +0100
@@ -92,7 +92,7 @@
* its lock, b) when a concurrent invalidate_mapping_pages got there first and
* c) when tmpfs swizzles a page between a tmpfs inode and swapper_space.
*/
-static void
+void
truncate_complete_page(struct address_space *mapping, struct page *page)
{
if (page->mapping != mapping)
@@ -108,6 +108,7 @@
ClearPageMappedToDisk(page);
page_cache_release(page); /* pagecache ref */
}
+EXPORT_SYMBOL_GPL(truncate_complete_page);
/*
* This is for invalidate_mapping_pages(). That function can be called at
diff -u -r debian-2.6.26/security/security.c debian-2.6.26_lustre.1.8.2/security/security.c
--- debian-2.6.26/security/security.c 2009-12-26 09:14:57.000000000 +0100
+++ debian-2.6.26_lustre.1.8.2/security/security.c 2010-02-18 08:13:38.000000000 +0100
@@ -406,6 +406,7 @@
return 0;
return security_ops->inode_unlink(dir, dentry);
}
+EXPORT_SYMBOL(security_inode_unlink);
int security_inode_symlink(struct inode *dir, struct dentry *dentry,
const char *old_name)
|