drivers/block/drbd/drbd_actlog.c:1015: warning: Function parameter or member 'peer_device' not described in 'drbd_try_rs_begin_io'

From: kernel test robot
Date: Sat Oct 21 2023 - 00:50:05 EST

Next message: Baolin Wang: "[PATCH v3] mm: migrate: record the mlocked page status to remove unnecessary lru drain"
Previous message: kernel test robot: "vmlinux.o: warning: objtool: cfi_staa_resume+0x169: cfi_build_cmd() is missing a __noreturn annotation"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

Hi Christoph,

FYI, the error/warning still remains.

tree: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
head: 9c5d00cb7b6bbc5a7965d9ab7d223b5402d1f02c
commit: 0d11f3cf279c5ad20a41f29242f170ba3c02f2da drbd: Pass a peer device to the resync and online verify functions
date: 7 months ago
config: x86_64-randconfig-r015-20230828 (https://download.01.org/0day-ci/archive/20231021/202310211241.qQivM1uZ-lkp@xxxxxxxxx/config)
compiler: gcc-12 (Debian 12.2.0-14) 12.2.0
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20231021/202310211241.qQivM1uZ-lkp@xxxxxxxxx/reproduce)

If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <lkp@xxxxxxxxx>
| Closes: https://lore.kernel.org/oe-kbuild-all/202310211241.qQivM1uZ-lkp@xxxxxxxxx/

All warnings (new ones prefixed by >>):

>> drivers/block/drbd/drbd_actlog.c:1015: warning: Function parameter or member 'peer_device' not described in 'drbd_try_rs_begin_io'
>> drivers/block/drbd/drbd_actlog.c:1015: warning: Excess function parameter 'device' description in 'drbd_try_rs_begin_io'

vim +1015 drivers/block/drbd/drbd_actlog.c

b411b3637fa71f Philipp Reisner 2009-09-25 1004
b411b3637fa71f Philipp Reisner 2009-09-25 1005 /**
b411b3637fa71f Philipp Reisner 2009-09-25 1006 * drbd_try_rs_begin_io() - Gets an extent in the resync LRU cache, does not sleep
b30ab7913b0a7b Andreas Gruenbacher 2011-07-03 1007 * @device: DRBD device.
b411b3637fa71f Philipp Reisner 2009-09-25 1008 * @sector: The sector number.
b411b3637fa71f Philipp Reisner 2009-09-25 1009 *
b411b3637fa71f Philipp Reisner 2009-09-25 1010 * Gets an extent in the resync LRU cache, sets it to BME_NO_WRITES, then
b411b3637fa71f Philipp Reisner 2009-09-25 1011 * tries to set it to BME_LOCKED. Returns 0 upon success, and -EAGAIN
b411b3637fa71f Philipp Reisner 2009-09-25 1012 * if there is still application IO going on in this area.
b411b3637fa71f Philipp Reisner 2009-09-25 1013 */
0d11f3cf279c5a Christoph Böhmwalder 2023-03-30 1014 int drbd_try_rs_begin_io(struct drbd_peer_device *peer_device, sector_t sector)
b411b3637fa71f Philipp Reisner 2009-09-25 @1015 {
0d11f3cf279c5a Christoph Böhmwalder 2023-03-30 1016 struct drbd_device *device = peer_device->device;
b411b3637fa71f Philipp Reisner 2009-09-25 1017 unsigned int enr = BM_SECT_TO_EXT(sector);
b411b3637fa71f Philipp Reisner 2009-09-25 1018 const unsigned int al_enr = enr*AL_EXT_PER_BM_SECT;
b411b3637fa71f Philipp Reisner 2009-09-25 1019 struct lc_element *e;
b411b3637fa71f Philipp Reisner 2009-09-25 1020 struct bm_extent *bm_ext;
b411b3637fa71f Philipp Reisner 2009-09-25 1021 int i;
0d11f3cf279c5a Christoph Böhmwalder 2023-03-30 1022 bool throttle = drbd_rs_should_slow_down(peer_device, sector, true);
ad3fee790088d3 Lars Ellenberg 2013-12-20 1023
ad3fee790088d3 Lars Ellenberg 2013-12-20 1024 /* If we need to throttle, a half-locked (only marked BME_NO_WRITES,
ad3fee790088d3 Lars Ellenberg 2013-12-20 1025 * not yet BME_LOCKED) extent needs to be kicked out explicitly if we
ad3fee790088d3 Lars Ellenberg 2013-12-20 1026 * need to throttle. There is at most one such half-locked extent,
ad3fee790088d3 Lars Ellenberg 2013-12-20 1027 * which is remembered in resync_wenr. */
ad3fee790088d3 Lars Ellenberg 2013-12-20 1028
ad3fee790088d3 Lars Ellenberg 2013-12-20 1029 if (throttle && device->resync_wenr != enr)
ad3fee790088d3 Lars Ellenberg 2013-12-20 1030 return -EAGAIN;
b411b3637fa71f Philipp Reisner 2009-09-25 1031
b30ab7913b0a7b Andreas Gruenbacher 2011-07-03 1032 spin_lock_irq(&device->al_lock);
b30ab7913b0a7b Andreas Gruenbacher 2011-07-03 1033 if (device->resync_wenr != LC_FREE && device->resync_wenr != enr) {
b411b3637fa71f Philipp Reisner 2009-09-25 1034 /* in case you have very heavy scattered io, it may
b411b3637fa71f Philipp Reisner 2009-09-25 1035 * stall the syncer undefined if we give up the ref count
b411b3637fa71f Philipp Reisner 2009-09-25 1036 * when we try again and requeue.
b411b3637fa71f Philipp Reisner 2009-09-25 1037 *
b411b3637fa71f Philipp Reisner 2009-09-25 1038 * if we don't give up the refcount, but the next time
b411b3637fa71f Philipp Reisner 2009-09-25 1039 * we are scheduled this extent has been "synced" by new
b411b3637fa71f Philipp Reisner 2009-09-25 1040 * application writes, we'd miss the lc_put on the
b411b3637fa71f Philipp Reisner 2009-09-25 1041 * extent we keep the refcount on.
b411b3637fa71f Philipp Reisner 2009-09-25 1042 * so we remembered which extent we had to try again, and
b411b3637fa71f Philipp Reisner 2009-09-25 1043 * if the next requested one is something else, we do
b411b3637fa71f Philipp Reisner 2009-09-25 1044 * the lc_put here...
b411b3637fa71f Philipp Reisner 2009-09-25 1045 * we also have to wake_up
b411b3637fa71f Philipp Reisner 2009-09-25 1046 */
b30ab7913b0a7b Andreas Gruenbacher 2011-07-03 1047 e = lc_find(device->resync, device->resync_wenr);
b411b3637fa71f Philipp Reisner 2009-09-25 1048 bm_ext = e ? lc_entry(e, struct bm_extent, lce) : NULL;
b411b3637fa71f Philipp Reisner 2009-09-25 1049 if (bm_ext) {
0b0ba1efc7b887 Andreas Gruenbacher 2011-06-27 1050 D_ASSERT(device, !test_bit(BME_LOCKED, &bm_ext->flags));
0b0ba1efc7b887 Andreas Gruenbacher 2011-06-27 1051 D_ASSERT(device, test_bit(BME_NO_WRITES, &bm_ext->flags));
b411b3637fa71f Philipp Reisner 2009-09-25 1052 clear_bit(BME_NO_WRITES, &bm_ext->flags);
b30ab7913b0a7b Andreas Gruenbacher 2011-07-03 1053 device->resync_wenr = LC_FREE;
ad3fee790088d3 Lars Ellenberg 2013-12-20 1054 if (lc_put(device->resync, &bm_ext->lce) == 0) {
ad3fee790088d3 Lars Ellenberg 2013-12-20 1055 bm_ext->flags = 0;
b30ab7913b0a7b Andreas Gruenbacher 2011-07-03 1056 device->resync_locked--;
ad3fee790088d3 Lars Ellenberg 2013-12-20 1057 }
b30ab7913b0a7b Andreas Gruenbacher 2011-07-03 1058 wake_up(&device->al_wait);
b411b3637fa71f Philipp Reisner 2009-09-25 1059 } else {
d01801710265cf Andreas Gruenbacher 2011-07-03 1060 drbd_alert(device, "LOGIC BUG\n");
b411b3637fa71f Philipp Reisner 2009-09-25 1061 }
b411b3637fa71f Philipp Reisner 2009-09-25 1062 }
b411b3637fa71f Philipp Reisner 2009-09-25 1063 /* TRY. */
b30ab7913b0a7b Andreas Gruenbacher 2011-07-03 1064 e = lc_try_get(device->resync, enr);
b411b3637fa71f Philipp Reisner 2009-09-25 1065 bm_ext = e ? lc_entry(e, struct bm_extent, lce) : NULL;
b411b3637fa71f Philipp Reisner 2009-09-25 1066 if (bm_ext) {
b411b3637fa71f Philipp Reisner 2009-09-25 1067 if (test_bit(BME_LOCKED, &bm_ext->flags))
b411b3637fa71f Philipp Reisner 2009-09-25 1068 goto proceed;
b411b3637fa71f Philipp Reisner 2009-09-25 1069 if (!test_and_set_bit(BME_NO_WRITES, &bm_ext->flags)) {
b30ab7913b0a7b Andreas Gruenbacher 2011-07-03 1070 device->resync_locked++;
b411b3637fa71f Philipp Reisner 2009-09-25 1071 } else {
b411b3637fa71f Philipp Reisner 2009-09-25 1072 /* we did set the BME_NO_WRITES,
b411b3637fa71f Philipp Reisner 2009-09-25 1073 * but then could not set BME_LOCKED,
b411b3637fa71f Philipp Reisner 2009-09-25 1074 * so we tried again.
b411b3637fa71f Philipp Reisner 2009-09-25 1075 * drop the extra reference. */
b411b3637fa71f Philipp Reisner 2009-09-25 1076 bm_ext->lce.refcnt--;
0b0ba1efc7b887 Andreas Gruenbacher 2011-06-27 1077 D_ASSERT(device, bm_ext->lce.refcnt > 0);
b411b3637fa71f Philipp Reisner 2009-09-25 1078 }
b411b3637fa71f Philipp Reisner 2009-09-25 1079 goto check_al;
b411b3637fa71f Philipp Reisner 2009-09-25 1080 } else {
b411b3637fa71f Philipp Reisner 2009-09-25 1081 /* do we rather want to try later? */
b30ab7913b0a7b Andreas Gruenbacher 2011-07-03 1082 if (device->resync_locked > device->resync->nr_elements-3)
b411b3637fa71f Philipp Reisner 2009-09-25 1083 goto try_again;
b411b3637fa71f Philipp Reisner 2009-09-25 1084 /* Do or do not. There is no try. -- Yoda */
b30ab7913b0a7b Andreas Gruenbacher 2011-07-03 1085 e = lc_get(device->resync, enr);
b411b3637fa71f Philipp Reisner 2009-09-25 1086 bm_ext = e ? lc_entry(e, struct bm_extent, lce) : NULL;
b411b3637fa71f Philipp Reisner 2009-09-25 1087 if (!bm_ext) {
b30ab7913b0a7b Andreas Gruenbacher 2011-07-03 1088 const unsigned long rs_flags = device->resync->flags;
b411b3637fa71f Philipp Reisner 2009-09-25 1089 if (rs_flags & LC_STARVING)
d01801710265cf Andreas Gruenbacher 2011-07-03 1090 drbd_warn(device, "Have to wait for element"
b411b3637fa71f Philipp Reisner 2009-09-25 1091 " (resync LRU too small?)\n");
46a15bc3ec425b Lars Ellenberg 2011-02-21 1092 BUG_ON(rs_flags & LC_LOCKED);
b411b3637fa71f Philipp Reisner 2009-09-25 1093 goto try_again;
b411b3637fa71f Philipp Reisner 2009-09-25 1094 }
b411b3637fa71f Philipp Reisner 2009-09-25 1095 if (bm_ext->lce.lc_number != enr) {
b30ab7913b0a7b Andreas Gruenbacher 2011-07-03 1096 bm_ext->rs_left = drbd_bm_e_weight(device, enr);
b411b3637fa71f Philipp Reisner 2009-09-25 1097 bm_ext->rs_failed = 0;
b30ab7913b0a7b Andreas Gruenbacher 2011-07-03 1098 lc_committed(device->resync);
b30ab7913b0a7b Andreas Gruenbacher 2011-07-03 1099 wake_up(&device->al_wait);
0b0ba1efc7b887 Andreas Gruenbacher 2011-06-27 1100 D_ASSERT(device, test_bit(BME_LOCKED, &bm_ext->flags) == 0);
b411b3637fa71f Philipp Reisner 2009-09-25 1101 }
b411b3637fa71f Philipp Reisner 2009-09-25 1102 set_bit(BME_NO_WRITES, &bm_ext->flags);
0b0ba1efc7b887 Andreas Gruenbacher 2011-06-27 1103 D_ASSERT(device, bm_ext->lce.refcnt == 1);
b30ab7913b0a7b Andreas Gruenbacher 2011-07-03 1104 device->resync_locked++;
b411b3637fa71f Philipp Reisner 2009-09-25 1105 goto check_al;
b411b3637fa71f Philipp Reisner 2009-09-25 1106 }
b411b3637fa71f Philipp Reisner 2009-09-25 1107 check_al:
b411b3637fa71f Philipp Reisner 2009-09-25 1108 for (i = 0; i < AL_EXT_PER_BM_SECT; i++) {
b30ab7913b0a7b Andreas Gruenbacher 2011-07-03 1109 if (lc_is_used(device->act_log, al_enr+i))
b411b3637fa71f Philipp Reisner 2009-09-25 1110 goto try_again;
b411b3637fa71f Philipp Reisner 2009-09-25 1111 }
b411b3637fa71f Philipp Reisner 2009-09-25 1112 set_bit(BME_LOCKED, &bm_ext->flags);
b411b3637fa71f Philipp Reisner 2009-09-25 1113 proceed:
b30ab7913b0a7b Andreas Gruenbacher 2011-07-03 1114 device->resync_wenr = LC_FREE;
b30ab7913b0a7b Andreas Gruenbacher 2011-07-03 1115 spin_unlock_irq(&device->al_lock);
b411b3637fa71f Philipp Reisner 2009-09-25 1116 return 0;
b411b3637fa71f Philipp Reisner 2009-09-25 1117
b411b3637fa71f Philipp Reisner 2009-09-25 1118 try_again:
ad3fee790088d3 Lars Ellenberg 2013-12-20 1119 if (bm_ext) {
ad3fee790088d3 Lars Ellenberg 2013-12-20 1120 if (throttle) {
ad3fee790088d3 Lars Ellenberg 2013-12-20 1121 D_ASSERT(device, !test_bit(BME_LOCKED, &bm_ext->flags));
ad3fee790088d3 Lars Ellenberg 2013-12-20 1122 D_ASSERT(device, test_bit(BME_NO_WRITES, &bm_ext->flags));
ad3fee790088d3 Lars Ellenberg 2013-12-20 1123 clear_bit(BME_NO_WRITES, &bm_ext->flags);
ad3fee790088d3 Lars Ellenberg 2013-12-20 1124 device->resync_wenr = LC_FREE;
ad3fee790088d3 Lars Ellenberg 2013-12-20 1125 if (lc_put(device->resync, &bm_ext->lce) == 0) {
ad3fee790088d3 Lars Ellenberg 2013-12-20 1126 bm_ext->flags = 0;
ad3fee790088d3 Lars Ellenberg 2013-12-20 1127 device->resync_locked--;
ad3fee790088d3 Lars Ellenberg 2013-12-20 1128 }
ad3fee790088d3 Lars Ellenberg 2013-12-20 1129 wake_up(&device->al_wait);
ad3fee790088d3 Lars Ellenberg 2013-12-20 1130 } else
b30ab7913b0a7b Andreas Gruenbacher 2011-07-03 1131 device->resync_wenr = enr;
ad3fee790088d3 Lars Ellenberg 2013-12-20 1132 }
b30ab7913b0a7b Andreas Gruenbacher 2011-07-03 1133 spin_unlock_irq(&device->al_lock);
b411b3637fa71f Philipp Reisner 2009-09-25 1134 return -EAGAIN;
b411b3637fa71f Philipp Reisner 2009-09-25 1135 }
b411b3637fa71f Philipp Reisner 2009-09-25 1136

:::::: The code at line 1015 was first introduced by commit
:::::: b411b3637fa71fce9cf2acf0639009500f5892fe The DRBD driver

:::::: TO: Philipp Reisner <philipp.reisner@xxxxxxxxxx>
:::::: CC: Jens Axboe <jens.axboe@xxxxxxxxxx>

--
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki

Next message: Baolin Wang: "[PATCH v3] mm: migrate: record the mlocked page status to remove unnecessary lru drain"
Previous message: kernel test robot: "vmlinux.o: warning: objtool: cfi_staa_resume+0x169: cfi_build_cmd() is missing a __noreturn annotation"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]