* [PATCH 1/5] btrfs-progs: check: inode nbytes fix in lowmem
2017-04-11 3:26 [PATCH 0/5] btrfs-progs: check: simple errors repair in lowmem Su Yue
@ 2017-04-11 3:26 ` Su Yue
2017-04-11 3:26 ` [PATCH 2/5] btrfs-progs: check: dir isize " Su Yue
` (4 subsequent siblings)
5 siblings, 0 replies; 8+ messages in thread
From: Su Yue @ 2017-04-11 3:26 UTC (permalink / raw)
To: linux-btrfs
After checking one inode item, we should get the actual nbytes of the
inode item.
Introduce function 'repair_inode_nbytes_lowmem' to set nbytes in struct
btrfs_inode_item to the actual nbytes. After call of the function, the
wrong nbytes should have been corrected.
Signed-off-by: Su Yue <suy.fnst@cn.fujitsu.com>
---
cmds-check.c | 76 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++---
1 file changed, 73 insertions(+), 3 deletions(-)
diff --git a/cmds-check.c b/cmds-check.c
index 17b7efbf..2bdf6a20 100644
--- a/cmds-check.c
+++ b/cmds-check.c
@@ -1907,6 +1907,9 @@ static int process_one_leaf_v2(struct btrfs_root *root, struct btrfs_path *path,
again:
err |= check_inode_item(root, path, ext_ref);
+ /* modified cur since check_inode_item may change path */
+ cur = path->nodes[0];
+
if (err & LAST_ITEM)
goto out;
@@ -2255,6 +2258,7 @@ static int walk_down_tree_v2(struct btrfs_root *root, struct btrfs_path *path,
}
ret = process_one_leaf_v2(root, path, nrefs,
level, ext_ref);
+ cur = path->nodes[*level];
break;
} else {
ret = btrfs_check_node(root, NULL, cur);
@@ -4815,10 +4819,69 @@ static int check_file_extent(struct btrfs_root *root, struct btrfs_key *fkey,
}
/*
+ * Set inode item nbytes to @nbytes
+ *
+ * Returns <0 means on error
+ * Returns 0 means successful repair
+ */
+static int repair_inode_nbytes_lowmem(struct btrfs_root *root,
+ struct btrfs_path *path,
+ u64 ino, u64 nbytes)
+{
+ struct btrfs_trans_handle *trans;
+ struct btrfs_inode_item *ii;
+ struct btrfs_key key;
+ struct btrfs_key research_key;
+ int ret;
+ int ret2;
+
+ key.objectid = ino;
+ key.type = BTRFS_INODE_ITEM_KEY;
+ key.offset = 0;
+ btrfs_item_key_to_cpu(path->nodes[0], &research_key, path->slots[0]);
+ btrfs_release_path(path);
+
+ trans = btrfs_start_transaction(root, 1);
+ if (IS_ERR(trans)) {
+ ret = PTR_ERR(trans);
+ goto out;
+ }
+
+ ret = btrfs_search_slot(trans, root, &key, path, 0, 1);
+ if (ret < 0)
+ goto out;
+ if (ret > 0) {
+ ret = -ENOENT;
+ goto out;
+ }
+
+ ii = btrfs_item_ptr(path->nodes[0], path->slots[0],
+ struct btrfs_inode_item);
+ btrfs_set_inode_nbytes(path->nodes[0], ii, nbytes);
+ btrfs_mark_buffer_dirty(path->nodes[0]);
+
+ printf("reset nbytes for inode %llu root %llu\n", ino,
+ root->root_key.objectid);
+
+ btrfs_commit_transaction(trans, root);
+out:
+ if (ret < 0)
+ error("failed to reset nbytes for inode %llu root %llu due to %s",
+ ino, root->root_key.objectid, strerror(-ret));
+
+ /* research path */
+ btrfs_release_path(path);
+ ret2 = btrfs_search_slot(NULL, root, &research_key, path, 0, 0);
+ return ret2 < 0 ? ret2 : ret;
+}
+
+/*
* Check INODE_ITEM and related ITEMs (the same inode number)
* 1. check link count
* 2. check inode ref/extref
* 3. check dir item/index
+ * Be Careful, if repair is enable, @path may be changed.
+ * Remember to reassign any context about @path in repair mode.
*
* @ext_ref: the EXTENDED_IREF feature
*
@@ -4968,9 +5031,16 @@ out:
}
if (nbytes != extent_size) {
- err |= NBYTES_ERROR;
- error("root %llu INODE[%llu] nbytes(%llu) not equal to extent_size(%llu)",
- root->objectid, inode_id, nbytes, extent_size);
+ if (repair) {
+ ret = repair_inode_nbytes_lowmem(root, path,
+ inode_id, extent_size);
+ }
+ if (!repair || ret) {
+ err |= NBYTES_ERROR;
+ error("root %llu INODE[%llu] nbytes(%llu) not equal to extent_size(%llu)",
+ root->objectid, inode_id, nbytes,
+ extent_size);
+ }
}
}
--
2.12.0
^ permalink raw reply related [flat|nested] 8+ messages in thread* [PATCH 2/5] btrfs-progs: check: dir isize fix in lowmem
2017-04-11 3:26 [PATCH 0/5] btrfs-progs: check: simple errors repair in lowmem Su Yue
2017-04-11 3:26 ` [PATCH 1/5] btrfs-progs: check: inode nbytes fix " Su Yue
@ 2017-04-11 3:26 ` Su Yue
2017-04-11 3:26 ` [PATCH 3/5] btrfs-progs: check: enable lowmem repair Su Yue
` (3 subsequent siblings)
5 siblings, 0 replies; 8+ messages in thread
From: Su Yue @ 2017-04-11 3:26 UTC (permalink / raw)
To: linux-btrfs
After traversal of whole directory, we should get the actual isize.
Introduce function 'repair_dir_isize_lowmem' to set isize in the directory
inode item to actual size. After call of the function, the wrong dir isize
should have been corrected.
Signed-off-by: Su Yue <suy.fnst@cn.fujitsu.com>
---
cmds-check.c | 68 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++---
1 file changed, 65 insertions(+), 3 deletions(-)
diff --git a/cmds-check.c b/cmds-check.c
index 2bdf6a20..6ab18b26 100644
--- a/cmds-check.c
+++ b/cmds-check.c
@@ -4876,6 +4876,62 @@ out:
}
/*
+ * Set dir isize to @isize
+ *
+ * Returns <0 means on error
+ * Returns 0 means successful repair
+ */
+static int repair_dir_isize_lowmem(struct btrfs_root *root,
+ struct btrfs_path *path,
+ u64 ino, u64 isize)
+{
+ struct btrfs_trans_handle *trans;
+ struct btrfs_inode_item *ii;
+ struct btrfs_key key;
+ struct btrfs_key research_key;
+ int ret;
+ int ret2;
+
+ key.objectid = ino;
+ key.type = BTRFS_INODE_ITEM_KEY;
+ key.offset = 0;
+
+ btrfs_item_key_to_cpu(path->nodes[0], &research_key, path->slots[0]);
+ btrfs_release_path(path);
+
+ trans = btrfs_start_transaction(root, 1);
+ if (IS_ERR(trans)) {
+ ret = PTR_ERR(trans);
+ goto out;
+ }
+
+ ret = btrfs_search_slot(trans, root, &key, path, 0, 1);
+ if (ret < 0)
+ goto out;
+ if (ret > 0) {
+ ret = -ENOENT;
+ goto out;
+ }
+
+ ii = btrfs_item_ptr(path->nodes[0], path->slots[0],
+ struct btrfs_inode_item);
+ btrfs_set_inode_size(path->nodes[0], ii, isize);
+ btrfs_mark_buffer_dirty(path->nodes[0]);
+
+ printf("reset isize for inode %llu root %llu\n", ino,
+ root->root_key.objectid);
+
+ btrfs_commit_transaction(trans, root);
+out:
+ if (ret < 0)
+ error("failed to reset isize for inode %llu root %llu due to %s",
+ ino, root->root_key.objectid, strerror(-ret));
+ btrfs_release_path(path);
+ ret2 = btrfs_search_slot(NULL, root, &research_key, path, 0, 0);
+ return ret2 < 0 ? ret2 : ret;
+}
+
+/*
* Check INODE_ITEM and related ITEMs (the same inode number)
* 1. check link count
* 2. check inode ref/extref
@@ -5011,9 +5067,15 @@ out:
}
if (isize != size) {
- err |= ISIZE_ERROR;
- error("root %llu DIR INODE [%llu] size(%llu) not equal to %llu",
- root->objectid, inode_id, isize, size);
+ if (repair)
+ ret = repair_dir_isize_lowmem(root, path,
+ inode_id, size);
+
+ if (!repair || ret) {
+ err |= ISIZE_ERROR;
+ error("root %llu DIR INODE [%llu] size(%llu) not equal to %llu",
+ root->objectid, inode_id, isize, size);
+ }
}
} else {
if (nlink != refs) {
--
2.12.0
^ permalink raw reply related [flat|nested] 8+ messages in thread* [PATCH 3/5] btrfs-progs: check: enable lowmem repair
2017-04-11 3:26 [PATCH 0/5] btrfs-progs: check: simple errors repair in lowmem Su Yue
2017-04-11 3:26 ` [PATCH 1/5] btrfs-progs: check: inode nbytes fix " Su Yue
2017-04-11 3:26 ` [PATCH 2/5] btrfs-progs: check: dir isize " Su Yue
@ 2017-04-11 3:26 ` Su Yue
2017-04-11 3:26 ` [PATCH 4/5] btrfs-progs: fsck-check: Allow fsck check test to repair in lowmem mode for certain test cases Su Yue
` (2 subsequent siblings)
5 siblings, 0 replies; 8+ messages in thread
From: Su Yue @ 2017-04-11 3:26 UTC (permalink / raw)
To: linux-btrfs
Enable btrfsck option '--repair' with option '--mode=lowmem'.
Now lowmem mode only repairs wrong nbytes, dir isize.
Signed-off-by: Su Yue <suy.fnst@cn.fujitsu.com>
---
cmds-check.c | 5 ++---
1 file changed, 2 insertions(+), 3 deletions(-)
diff --git a/cmds-check.c b/cmds-check.c
index 6ab18b26..da5fe75f 100644
--- a/cmds-check.c
+++ b/cmds-check.c
@@ -12925,11 +12925,10 @@ int cmd_check(int argc, char **argv)
}
/*
- * Not supported yet
+ * Support partially
*/
if (repair && check_mode == CHECK_MODE_LOWMEM) {
- error("low memory mode doesn't support repair yet");
- exit(1);
+ warning("low memory mode support repair partially");
}
radix_tree_init();
--
2.12.0
^ permalink raw reply related [flat|nested] 8+ messages in thread* [PATCH 4/5] btrfs-progs: fsck-check: Allow fsck check test to repair in lowmem mode for certain test cases
2017-04-11 3:26 [PATCH 0/5] btrfs-progs: check: simple errors repair in lowmem Su Yue
` (2 preceding siblings ...)
2017-04-11 3:26 ` [PATCH 3/5] btrfs-progs: check: enable lowmem repair Su Yue
@ 2017-04-11 3:26 ` Su Yue
2017-04-11 3:26 ` [PATCH 5/5] btrfs-progs: fsck-check: test cases for nbytes and dir isize Su Yue
2017-05-17 8:25 ` [PATCH 0/5] btrfs-progs: check: simple errors repair in lowmem Su Yue
5 siblings, 0 replies; 8+ messages in thread
From: Su Yue @ 2017-04-11 3:26 UTC (permalink / raw)
To: linux-btrfs; +Cc: Qu Wenruo
From: Qu Wenruo <quwenruo@cn.fujitsu.com>
Since lowmem mode can repair certain corruptions (mostly in fs tree),
insert a beacon into each fsck test cases to allow some of them be
tested for lowmem mode.
With this patch, fsck option override will check the beacon file
".lowmem_repairbale" in the same directory of the test image, and if the
beacon exists, then it will also run lowmem mode repair to repair the
image.
Signed-off-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
---
tests/common.local | 14 +++++++++++++-
1 file changed, 13 insertions(+), 1 deletion(-)
diff --git a/tests/common.local b/tests/common.local
index 4f56bb08..af372f16 100644
--- a/tests/common.local
+++ b/tests/common.local
@@ -15,11 +15,23 @@ TEST_ARGS_CHECK=--mode=lowmem
# gets arguments of a current command and can decide if the argument insertion
# should happen, eg. if some option combination does not make sense or would
# break tests
+#
+# Return 0 if we need to skip option override
+# Return 1 if we don't need to skip option override
_skip_spec()
{
+ beacon=.lowmem_repairable
+
+ # For loemem repair, only support fs tree repair yet
+ # So we place lowmem repair beacon in the same dir of the
+ # test case
if echo "$TEST_ARGS_CHECK" | grep -q 'mode=lowmem' &&
echo "$@" | grep -q -- '--repair'; then
- return 0
+ dir="$(dirname ${@: -1})"
+ if [ -f ${dir}/${beacon} ]; then
+ return 1;
+ fi
+ return 0;
fi
return 1
}
--
2.12.0
^ permalink raw reply related [flat|nested] 8+ messages in thread* [PATCH 5/5] btrfs-progs: fsck-check: test cases for nbytes and dir isize
2017-04-11 3:26 [PATCH 0/5] btrfs-progs: check: simple errors repair in lowmem Su Yue
` (3 preceding siblings ...)
2017-04-11 3:26 ` [PATCH 4/5] btrfs-progs: fsck-check: Allow fsck check test to repair in lowmem mode for certain test cases Su Yue
@ 2017-04-11 3:26 ` Su Yue
2017-05-17 8:25 ` [PATCH 0/5] btrfs-progs: check: simple errors repair in lowmem Su Yue
5 siblings, 0 replies; 8+ messages in thread
From: Su Yue @ 2017-04-11 3:26 UTC (permalink / raw)
To: linux-btrfs
Create test case '026-wrong-dir-inode-isize'.
Create becon files '.lowmem_repairable' under tests/fsck-test/016 and 026.
Now 'make test-fsck' will test lowmem repairable test cases if
TEST_ENABLE_OVERRIDE=true.
Signed-off-by: Su Yue <suy.fnst@cn.fujitsu.com>
---
.../fsck-tests/016-wrong-inode-nbytes/.lowmem_repairable | 0
.../026-wrong-dir-inode-isize/.lowmem_repairable | 0
.../026-wrong-dir-inode-isize/default_case.img | Bin 0 -> 4096 bytes
3 files changed, 0 insertions(+), 0 deletions(-)
create mode 100644 tests/fsck-tests/016-wrong-inode-nbytes/.lowmem_repairable
create mode 100644 tests/fsck-tests/026-wrong-dir-inode-isize/.lowmem_repairable
create mode 100644 tests/fsck-tests/026-wrong-dir-inode-isize/default_case.img
diff --git a/tests/fsck-tests/016-wrong-inode-nbytes/.lowmem_repairable b/tests/fsck-tests/016-wrong-inode-nbytes/.lowmem_repairable
new file mode 100644
index 00000000..e69de29b
diff --git a/tests/fsck-tests/026-wrong-dir-inode-isize/.lowmem_repairable b/tests/fsck-tests/026-wrong-dir-inode-isize/.lowmem_repairable
new file mode 100644
index 00000000..e69de29b
diff --git a/tests/fsck-tests/026-wrong-dir-inode-isize/default_case.img b/tests/fsck-tests/026-wrong-dir-inode-isize/default_case.img
new file mode 100644
index 0000000000000000000000000000000000000000..a060cca86e945f2659595d812b3017c66c570d5e
GIT binary patch
literal 4096
zcmeHHX*d*I8)g_@X;2Lby@Ra9kg`N}hBQhtuf2#!WM`~nnW8LZ8=4OjNm<8M%Gk5C
zVH*27BfA*OSf(N7lkff2*Prj-cU^DiI@kT&&vUMG&U4?_xlX(b#&<%}uAb%Z!MPV%
z_van~?=cn~ogc^|zK>1&$ix1l-n`ETe=J76-RI<e-m=dh{@{wOKaj;{zp_t%!*2(E
zJMjPGK+te#-m^?%B&Tti8UFhFsg{Z)QYq5ZR_Q1k+X*WXgGv^kx>O@G8}B*=2wcB~
z^*(@;6D9)F0?a!pI<WZ-=0tn90YvM~4mxWC%(K)Eniy2GHU|bqf95guW^)u&Ft(#5
zFa$C$f5Cqm=xXRCJ15L<B!hxTbB8!{E-Z5CwX(vMatj^wE`2iVxSwKBA+09C?SOs3
zS`0Li34@gb;W7d`>v^`3N!pPJoFh0P`ikNKShI{EaE@j8<#d!~Ws;2m`Fwy*s;k=F
zk6t>HkjRxG@3V==J_OD=!qNhTVl+roTs2j!_L`W{V3r$D4V&V7z%K3Rle__{=j%j7
zmz*U9ad2?V?B0<rU=|i`AM>X{!_<*_6E<DGrd6zgUawT?vy0|!-FiUnbawG5(XBs7
zA1^gut0-S%Es<~#yM=kHC$|JN2TIt)C%KlB7RMx!vms)J1SapMXP~Pq2flInHpAK+
z4EcC88-9vHYVYjXSWOYk%k$glA)v?&BWk0RQ2S()T_5-0(*>V6mhGl0559Hqvz(Gi
zVH|u1LK^T)a5<qz8o7uZrbhQbl-Gurj_|9T=Q{H1SEOj^vi?5d^)8W!8Qewkf7ei#
zt&*f}mn|CM12(%O7C@G@?#W@Dn4>k|fx7PX;D2cmd#q;)op9}4n%I2G+g{Yn6IA<<
zSDga<uevA3Tl9i~=o-rpTJtH#Q<-Z_7P-VbihB-?UA$$75NulY^OD9sz`~1VY);*9
zD&c*}ZnE`RBHL3=?@%0q;2xhS>T*SS0_Gp5f{zomAG!x(_aGUbG(MD{{XBrA$4!MD
z3Xp6QH(ZGVHivjEGcGOdbd5U0r4Gk3z5E~)Mwez;$I_G>ghE4O+lun&jmn2SPnoh`
z{`0%@cZO|3A?ND0C>N)2`)1f1pY0yoMmx7H69{jhrN_0{`oK(ox5AAwN2iC+$Z69v
z8DG9RN|jHJaW3N$yr2eAWdCuxs<c+TmsT9W-8^b6cc+~dQxS><Emj_j^6V?jeTLWz
zt1@O6fH29)L>Wasdwu&pB>Dm(7ymU3{HF8Q_TOS1(zjD`I#KqJI=nq`>gF~KZ}}WD
zzdg~l1DWd(4T!iCf8Eb0RbFkom8dbf&=rqvb6L`mtkCXBC(xC4<U$L$RcmDtL5Pi6
zq>pdJ?s#WDJyW09K0pd+#Cyz;N5bs`5g544k?A}1z_AB}W3Bl%h)3sRT&Tl0i;8L`
ztagx}PI&k(E>%tX&aZ4}(9#VE-)s8DVqJ@gBh2e}JeiFzk->yYg8w$PQ|T~*tJjlU
zHwE|r&^+nxu`}cRqMqHZy*`B$(eGl=!&j>=cdSn|)yHkn#Ui!ov>Vw1Y?a-M1J%6%
z1h{=$z@NmdW;yp-IOqWSdWM>)P<<duIHk#7#d`?Y<;?a@^4A{afZltq{~3xI8@+0X
z$SL6lTSW2L=Un8MnFB<Sjg49B^TJM=DF*>ZuQ5Ubi%}H22gi$aGN9wTXkBI7N>$g<
zDMpXZBaf|$#-s=Hy8&N6+MPkSSEZw5rr@Ri^Vo1pVrTW?x*g`LnN6lEOiyd`Vei&t
z(rM{vY=2FY6Z(P<!79Cras=pUS}L7!niiCjYxE)P)rZF?3cvUv!W(3*U(@O)Y3qgP
z)y`a*6&-*1uU=t|VhmUYPl@#t_Vr%uYSh#u&BX#R!kNM}S&6lu_QYs`>%-LuA8ANR
zP;mAR$A1eSjv+%QsJ6Gw@T9`L`1~AUhs!Nz&rv%TGVplQna+T#o5>i98XSi`?-aX5
z#Q6hQ3GJFRjrXs5Pd5}VDrXtJgAPx+meo;)HX^b`mha9CM+=l%AWO<uM?OB7xXOtN
z@uw~GOS7TeZHnJzR-fU<R^wC{g+9WvULA`DFXN;}HE3^f12AEnGu8T8aKmnZZ_0)T
zUH)3^t3|b!f3>;mgo`bdD-m7NX7A4)2R97Hwc~5@cS9#YIhq-vxHT)!R7z^T)*UkC
zB4UR@<z~ifH|Ebk6ubm?|MOo)UlC*(_iH$Cwl?Zk?@(V1pRI3lPA)f}S94q8^XlqN
zUvAoA_W1)sLc?S7s^}a8i*MpvV4pDD{LZ;q%@!<0cQ!yoNkmhijBw$^K%(pG`pkY*
zf#y4F&90G(Zcj4gde*>pY;Ibpy6hr!$rLq{lB89iwS@qE5ks7_;(+v71Ag+{Htnu&
z@NA<C6)ALeb<TfHRhzN|wQO^1S*B-{1V?N#%c|&1x&WQl0QqTWzX}jHu4K8di>KKE
z+`ihQbaoIfF2#|TsC&!U%0L{iV+W{}I*e{rHdf`z?ECz9>|Sv>DMWbdkk}0xQyv%8
z6&$<t;s!XKriZ$kH55WDf7NySwcv|Z_1jC#;rjlf9uh%i4#|5?_Si^>dfsNr*MT6^
z;?rv9Zm_KOwd7Q7QUgE6!7)TkSYAXSOPB;wSkAjQIUU>)vlQb9C;94M;s2@Uzta`Y
z!>E-<WqlRZ1#BxT*$n#HbCdM%F63lh3e#Kb%~4zgz4kgQG)X^xPu-`+1G+wf9OIM<
zHHIIn(u7c!?!26wIaF8Rr~^{!W>VktDp_cA&S<JqGP(~GgxlKs8s6xI^Y32KU;At`
zlYb@42;J$XTzq#?sGc@wZ#35C{{-IN;&<ZZq*Iv<^E*bn_D!uOxsnfRNT9}=asSlv
z3bB7aoSrTHqTqPJUN;>TAhaSM(9E_MjE$6+iSmG%sJvGMUs6J-OL&++1}4#P{*_Bl
zmd%k18QXM5SznBP-1eDiXpBWP<7izGq^I^8Bh4Z4+#ipmi0!#0Sr7YFf4cCWm5u{c
R>`I?{{<B2?R{Os>@NZR9TeJWG
literal 0
HcmV?d00001
--
2.12.0
^ permalink raw reply [flat|nested] 8+ messages in thread* Re: [PATCH 0/5] btrfs-progs: check: simple errors repair in lowmem
2017-04-11 3:26 [PATCH 0/5] btrfs-progs: check: simple errors repair in lowmem Su Yue
` (4 preceding siblings ...)
2017-04-11 3:26 ` [PATCH 5/5] btrfs-progs: fsck-check: test cases for nbytes and dir isize Su Yue
@ 2017-05-17 8:25 ` Su Yue
2017-06-01 14:47 ` David Sterba
5 siblings, 1 reply; 8+ messages in thread
From: Su Yue @ 2017-05-17 8:25 UTC (permalink / raw)
To: linux-btrfs; +Cc: David Sterba
Hi
This patchset was sent a month ago.
Since it's a start of lowmem repair patches, I hope to get some
opinions about it.
Thanks
Su Yue
On 04/11/2017 11:26 AM, Su Yue wrote:
> The series include following contents:
>
> 1) Repair wrong nbytes of file inode item.
> After traversal of extents in one file, we should get the actual nbytes
> of the file. If nbytes in the file inode differs from the actual value,
> set the value to actual one.
>
> The wrong nbytes of file inode case corresponds to fsck-test/016.
>
> 2) Repair wrong iszie of directory inode item.
> After traversal of dir_index and dir_item in one dir, we should get
> the actual isize of the dirctory. If size in the inode item differs
> from the actual value, set the value to actual one.
>
> New test case 'fsck-test/026' is added for the case wrong isize of
> directory inode.
>
> 3) Allow fsck check test to repair in lowmem mode for certain test cases
> if TEST_ENABLE_OVERRIDE=true.
>
> Qu Wenruo (1):
> btrfs-progs: fsck-check: Allow fsck check test to repair in lowmem
> mode for certain test cases
>
> Su Yue (4):
> btrfs-progs: check: inode nbytes fix in lowmem
> btrfs-progs: check: dir isize fix in lowmem
> btrfs-progs: check: enable lowmem repair
> btrfs-progs: fsck-check: test cases for nbytes and dir isize
>
> cmds-check.c | 149 +++++++++++++++++++--
> tests/common.local | 14 +-
> .../016-wrong-inode-nbytes/.lowmem_repairable | 0
> .../026-wrong-dir-inode-isize/.lowmem_repairable | 0
> .../026-wrong-dir-inode-isize/default_case.img | Bin 0 -> 4096 bytes
> 5 files changed, 153 insertions(+), 10 deletions(-)
> create mode 100644 tests/fsck-tests/016-wrong-inode-nbytes/.lowmem_repairable
> create mode 100644 tests/fsck-tests/026-wrong-dir-inode-isize/.lowmem_repairable
> create mode 100644 tests/fsck-tests/026-wrong-dir-inode-isize/default_case.img
>
^ permalink raw reply [flat|nested] 8+ messages in thread