<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
</head>
<body text="#000000" bgcolor="#FFFFFF">
<p>@Gang</p>
<p>Thank you very much!<br>
</p>
<p>I have fixed this issue through upgrading the openmpi.</p>
<p>I have tested this patch using ocfs2-test. It can work fine.</p>
<p>@Andrew Please can you help me to summit this patch?<br>
</p>
<p><br>
</p>
<p>Thanks</p>
<p>Sunny<br>
</p>
<div class="moz-cite-prefix">在 2019/10/24 上午10:02, Gang He 写道:<br>
</div>
<blockquote type="cite"
cite="mid:CH2PR18MB320662998E1EC2BA01BA0B88CF6A0@CH2PR18MB3206.namprd18.prod.outlook.com">
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
<meta name="Generator" content="Microsoft Word 15 (filtered
medium)">
<style><!--
/* Font Definitions */
@font-face
        {font-family:宋体;
        panose-1:2 1 6 0 3 1 1 1 1 1;}
@font-face
        {font-family:"Cambria Math";
        panose-1:2 4 5 3 5 4 6 3 2 4;}
@font-face
        {font-family:Calibri;
        panose-1:2 15 5 2 2 2 4 3 2 4;}
@font-face
        {font-family:"\@宋体";
        panose-1:2 1 6 0 3 1 1 1 1 1;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
        {margin:0cm;
        margin-bottom:.0001pt;
        font-size:12.0pt;
        font-family:宋体;
        color:black;}
a:link, span.MsoHyperlink
        {mso-style-priority:99;
        color:blue;
        text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
        {mso-style-priority:99;
        color:purple;
        text-decoration:underline;}
p
        {mso-style-priority:99;
        mso-margin-top-alt:auto;
        margin-right:0cm;
        mso-margin-bottom-alt:auto;
        margin-left:0cm;
        font-size:12.0pt;
        font-family:宋体;
        color:black;}
span.tlid-translation
        {mso-style-name:tlid-translation;}
span.EmailStyle19
        {mso-style-type:personal-reply;
        font-family:"Calibri",sans-serif;
        color:#1F497D;}
.MsoChpDefault
        {mso-style-type:export-only;
        font-size:10.0pt;}
@page WordSection1
        {size:612.0pt 792.0pt;
        margin:72.0pt 90.0pt 72.0pt 90.0pt;}
div.WordSection1
        {page:WordSection1;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext="edit" spidmax="1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext="edit">
<o:idmap v:ext="edit" data="1" />
</o:shapelayout></xml><![endif]-->
<div class="WordSection1">
<p class="MsoNormal"><span
style="font-size:10.5pt;font-family:"Calibri",sans-serif;color:#1F497D"
lang="EN-US">Hi Sunny,<o:p></o:p></span></p>
<p class="MsoNormal"><span
style="font-size:10.5pt;font-family:"Calibri",sans-serif;color:#1F497D"
lang="EN-US"><o:p> </o:p></span></p>
<p class="MsoNormal"><span
style="font-size:10.5pt;font-family:"Calibri",sans-serif;color:#1F497D"
lang="EN-US">This is a mpi problem in your environment, you
can look at how to run mpi successfully in your Linux
distribution.<o:p></o:p></span></p>
<p class="MsoNormal"><span
style="font-size:10.5pt;font-family:"Calibri",sans-serif;color:#1F497D"
lang="EN-US">For us, mpi can work normally, e.g. SLES12SP4
SLE15SP1.<o:p></o:p></span></p>
<p class="MsoNormal"><span
style="font-size:10.5pt;font-family:"Calibri",sans-serif;color:#1F497D"
lang="EN-US"><o:p> </o:p></span></p>
<p class="MsoNormal"><span
style="font-size:10.5pt;font-family:"Calibri",sans-serif;color:#1F497D"
lang="EN-US"><o:p> </o:p></span></p>
<p class="MsoNormal"><span
style="font-size:10.5pt;font-family:"Calibri",sans-serif;color:#1F497D"
lang="EN-US">Thanks<o:p></o:p></span></p>
<p class="MsoNormal"><span
style="font-size:10.5pt;font-family:"Calibri",sans-serif;color:#1F497D"
lang="EN-US">Gang<o:p></o:p></span></p>
<p class="MsoNormal"><span
style="font-size:10.5pt;font-family:"Calibri",sans-serif;color:#1F497D"
lang="EN-US"><o:p> </o:p></span></p>
<div style="border:none;border-left:solid blue 1.5pt;padding:0cm
0cm 0cm 4.0pt">
<div>
<div style="border:none;border-top:solid #E1E1E1
1.0pt;padding:3.0pt 0cm 0cm 0cm">
<p class="MsoNormal"><b><span
style="font-size:11.0pt;font-family:"Calibri",sans-serif;color:windowtext"
lang="EN-US">From:</span></b><span
style="font-size:11.0pt;font-family:"Calibri",sans-serif;color:windowtext"
lang="EN-US"> sunnyZhang
[<a class="moz-txt-link-freetext" href="mailto:sunny.s.zhang@oracle.com">mailto:sunny.s.zhang@oracle.com</a>]
<br>
<b>Sent:</b> 2019</span><span
style="font-size:11.0pt;color:windowtext">年</span><span
style="font-size:11.0pt;font-family:"Calibri",sans-serif;color:windowtext"
lang="EN-US">10</span><span
style="font-size:11.0pt;color:windowtext">月</span><span
style="font-size:11.0pt;font-family:"Calibri",sans-serif;color:windowtext"
lang="EN-US">23</span><span
style="font-size:11.0pt;color:windowtext">日</span><span
style="font-size:11.0pt;font-family:"Calibri",sans-serif;color:windowtext"
lang="EN-US"> 17:15<br>
<b>To:</b> Gang He <a class="moz-txt-link-rfc2396E" href="mailto:GHe@suse.com"><GHe@suse.com></a><br>
<b>Subject:</b> Re: [Ocfs2-devel] [PATCH V2 1/2]
ocfs2: protect extent tree in the
ocfs2_prepare_inode_for_write<o:p></o:p></span></p>
</div>
</div>
<p class="MsoNormal"><span lang="EN-US"><o:p> </o:p></span></p>
<p><span lang="EN-US">Hi Gang,<o:p></o:p></span></p>
<p><span lang="EN-US">I try to test these patches using
ocfs2-test. <o:p></o:p></span></p>
<p><span lang="EN-US">But there is a error about openmpi. <o:p></o:p></span></p>
<p><span class="tlid-translation"><span lang="EN">Have you
ever encountered a similar issue before?</span></span><span
lang="EN-US"><o:p></o:p></span></p>
<p><span lang="EN-US">Logs are as following.<o:p></o:p></span></p>
<p><span lang="EN-US">2019/10/23,17:02:38 [*] Remote Mount
among nodes ol6u9ext3,zhangsn:<br>
2019/10/23,17:02:38
/home/ocfs2test/bin/ocfs2/bin/remote_mount.py -l
ocfs2-multi-refcount-tests-x86_64 -m /mnt/ocfs2 -n
ol6u9ext3,zhangsn<br>
+ /usr/lib64/openmpi/bin/mpirun --allow-run-as-root -mca
btl tcp,self -mca plm_rsh_agent ssh:rsh -np 2 --host
ol6u9ext3,zhangsn /home/ocfs2test/bin/ocfs2/bin/command.py
--mount -l ocfs2-multi-refcount-tests-x86_64 -m /mnt/ocfs2<br>
2019/10/23,17:02:51 [1] Basic Functional Test,
CMD:/usr/lib64/openmpi/bin/mpirun -mca plm_rsh_agent
ssh:rsh -mca btl tcp,self -np 4 --host ol6u9ext3,zhangsn
/home/ocfs2test/bin/ocfs2/bin/multi_reflink_test -i 1 -l
104857600 -n 10 -w /mnt/ocfs2/ocfs2-multi-refcount-tests
-f<br>
[zhangsn][[58639,1],2][btl_tcp_frag.c:215:mca_btl_tcp_frag_recv]
mca_btl_tcp_frag_recv: readv failed: Connection reset by
peer (104)<br>
[*Round 0 Test Running*]<br>
Test 1: Multi-nodes basic refcount test.<br>
*SubTest 1:Prepare original inode
/mnt/ocfs2/ocfs2-multi-refcount-tests/multi_original_basic_refile.<br>
create file
/mnt/ocfs2/ocfs2-multi-refcount-tests/multi_original_basic_refile
failed:13:Permission denied<br>
Rank:0 on Host(ol6u9ext3) abort!<br>
--------------------------------------------------------------------------<br>
MPI_ABORT was invoked on rank 0 in communicator
MPI_COMM_WORLD<br>
with errorcode 1.<br>
<br>
NOTE: invoking MPI_ABORT causes Open MPI to kill all MPI
processes.<br>
You may or may not see output from other processes,
depending on<br>
exactly when Open MPI kills them.<br>
--------------------------------------------------------------------------<br>
[ol6u9ext3][[58639,1],1][btl_tcp_frag.c:215:mca_btl_tcp_frag_recv]
mca_btl_tcp_frag_recv: readv failed: Connection reset by
peer (104)<o:p></o:p></span></p>
<p><span lang="EN-US"><o:p> </o:p></span></p>
<p><span lang="EN-US">Thanks<o:p></o:p></span></p>
<div>
<p class="MsoNormal">在<span lang="EN-US"> 2019/9/27 </span>下午<span
lang="EN-US">3:15, sunnyZhang
</span>写道<span lang="EN-US">:<o:p></o:p></span></p>
</div>
<blockquote style="margin-top:5.0pt;margin-bottom:5.0pt">
<p class="MsoNormal"><span lang="EN-US">Hi Gang, <br>
<br>
Thanks for review. <br>
<br>
I will try to test using ocfs2-test. <br>
<br>
<br>
Thanks <br>
<br>
</span>在<span lang="EN-US"> 2019/9/25 </span>下午<span
lang="EN-US">6:33, Gang He </span>
写道<span lang="EN-US">: <br>
<br>
<o:p></o:p></span></p>
<blockquote style="margin-top:5.0pt;margin-bottom:5.0pt">
<p class="MsoNormal"><span lang="EN-US"><br>
<br>
<o:p></o:p></span></p>
<blockquote style="margin-top:5.0pt;margin-bottom:5.0pt">
<p class="MsoNormal"><span lang="EN-US">-----Original
Message----- <br>
From: <a
href="mailto:ocfs2-devel-bounces@oss.oracle.com"
moz-do-not-send="true">ocfs2-devel-bounces@oss.oracle.com</a>
<br>
[<a href="mailto:ocfs2-devel-bounces@oss.oracle.com"
moz-do-not-send="true">mailto:ocfs2-devel-bounces@oss.oracle.com</a>]
On Behalf Of Shuning Zhang
<br>
Sent: 2019</span>年<span lang="EN-US">9</span>月<span
lang="EN-US">18</span>日<span lang="EN-US"> 10:03
<br>
To: <a href="mailto:ocfs2-devel@oss.oracle.com"
moz-do-not-send="true">ocfs2-devel@oss.oracle.com</a>
<br>
Subject: [Ocfs2-devel] [PATCH V2 1/2] ocfs2: protect
extent tree in the <br>
ocfs2_prepare_inode_for_write <br>
<br>
When the extent tree is modified, it should be
protected by inode cluster lock <br>
and ip_alloc_sem. <br>
<br>
The extent tree is accessed and modified in the
ocfs2_prepare_inode_for_write, <br>
but isn't protected by ip_alloc_sem. <br>
<br>
The following is a case. The function ocfs2_fiemap
is accessing the extent tree, <br>
which is modified at the same time. <br>
<br>
[47145.974472] kernel BUG at
fs/ocfs2/extent_map.c:475! <br>
[47145.974480] invalid opcode: 0000 [#1] SMP
[47145.974489] Modules <br>
linked in: tun ocfs2 ocfs2_nodemanager configfs
ocfs2_stackglue xen_netback <br>
xen_blkback xen_gntalloc xen_gntdev xen_evtchn xenfs
xen_privcmd vfat fat <br>
bnx2fc fcoe libfcoe libfc scsi_transport_fc sunrpc
bridge 8021q mrp garp stp llc <br>
ib_iser rdma_cm ib_cm iw_cm ib_sa ib_mad ib_core
ib_addr dm_round_robin <br>
dm_multipath sg pcspkr raid1 shpchp ipmi_devintf
ipmi_msghandler ext4 <br>
jbd2 mbcache2 sd_mod nvme nvme_core bnxt_en xhci_pci
xhci_hcd <br>
crc32c_intel be2iscsi bnx2i cnic uio cxgb4i cxgb4
cxgb3i libcxgbi ipv6 cxgb3 <br>
mdio qla4xxx wmi dm_mirror dm_region_hash dm_log
dm_mod iscsi_tcp <br>
libiscsi_tcp libiscsi scsi_transport_iscsi
iscsi_ibft iscsi_boot_sysfs <br>
[47145.974636] CPU: 16 PID: 14047 Comm: o2info Not
tainted <br>
4.1.12-124.23.1.el6uek.x86_64 #2 <br>
[47145.974646] Hardware name: Oracle Corporation
ORACLE SERVER <br>
X7-2L/ASM, MB MECH, X7-2L, BIOS 42040600 10/19/2018
[47145.974658] <br>
task: ffff88019487e200 ti: ffff88003daa4000 task.ti:
<br>
ffff88003daa4000 <br>
[47145.974667] RIP: e030:[<ffffffffa05e4840>]
[<ffffffffa05e4840>] <br>
ocfs2_get_clusters_nocache.isra.11+0x390/0x550
[ocfs2] [47145.974708] <br>
RSP: e02b:ffff88003daa7d88 EFLAGS: 00010287
[47145.974713] RAX: <br>
00000000000000de RBX: ffff8801d1104030 RCX: <br>
ffff8801d1104e10 <br>
[47145.974719] RDX: 00000000000000de RSI:
000000000009ec40 RDI: <br>
ffff8801d1104e24 <br>
[47145.974725] RBP: ffff88003daa7df8 R08:
ffff88003daa7e38 R09: <br>
0000000000000000 <br>
[47145.974732] R10: 000000000009ec3f R11:
0000000000000246 R12: <br>
000000000009ec3f <br>
[47145.974739] R13: ffff88004c419000 R14:
0000000000000002 R15: <br>
ffff88003daa7e28 <br>
[47145.974754] FS: 00007fdbccc92720(0000)
GS:ffff880358800000(0000) <br>
knlGS:ffff880358800000 <br>
[47145.974764] CS: e033 DS: 0000 ES: 0000 CR0:
0000000080050033 <br>
[47145.974772] CR2: 00007fd5dfcd8350 CR3:
0000000208677000 CR4: <br>
0000000000042660 <br>
[47145.974785] Stack: <br>
[47145.974790] ffff88003daa7df8 00002000a05e249b
ffff8801d1104000 <br>
ffff88003daa7e2c [47145.974802] ffff88003daa7e38
ffff88000cc484c0 <br>
ffff880145f5b478 <br>
0000000000000000 <br>
[47145.974811] 0000000000002000 ffff88000cc484c0
ffff88003daa7ea0 <br>
0000000000000000 <br>
[47145.974820] Call Trace: <br>
[47145.974837] [<ffffffffa05e53e3>]
ocfs2_fiemap+0x1e3/0x430 [ocfs2] <br>
[47145.974848] [<ffffffff816f644f>] ?
xen_hypervisor_callback+0x7f/0x120 <br>
[47145.974855] [<ffffffff816f6448>] ? <br>
xen_hypervisor_callback+0x78/0x120 <br>
[47145.974861] [<ffffffff816f64a3>] ? <br>
xen_hypervisor_callback+0xd3/0x120 <br>
[47145.974872] [<ffffffff81220905>]
do_vfs_ioctl+0x155/0x510 <br>
[47145.974878] [<ffffffff81220d41>]
SyS_ioctl+0x81/0xa0 [47145.974885] <br>
[<ffffffff816f1b9f>] ?
system_call_after_swapgs+0xe9/0x190 <br>
[47145.974891] [<ffffffff816f1b98>] ? <br>
system_call_after_swapgs+0xe2/0x190 <br>
[47145.974899] [<ffffffff816f1b91>] ? <br>
system_call_after_swapgs+0xdb/0x190 <br>
[47145.974905] [<ffffffff816f1c5e>]
system_call_fastpath+0x18/0xd8 <br>
[47145.974910] Code: 18 48 c7 c6 60 7f 65 a0 31 c0
bb e2 ff ff ff 48 8b 4a 40 <br>
48 8b 7a 28 48 c7 c2 78 2d 66 a0 e8 38 4f 05 00 e9
28 fe ff ff 0f 1f 00 <0f> <br>
0b 66 0f 1f 44 00 00 bb 86 ff ff ff e9 13 fe ff ff
66 0f 1f [47145.975000] RIP <br>
[<ffffffffa05e4840>] <br>
ocfs2_get_clusters_nocache.isra.11+0x390/0x550
[ocfs2] [47145.975018] <br>
RSP <ffff88003daa7d88> [47145.989999] ---[ end
trace <br>
c8aa0c8180e869dc ]--- [47146.087579] Kernel panic -
not syncing: Fatal <br>
exception [47146.087691] Kernel Offset: disabled <br>
<br>
Signed-off-by: Shuning Zhang <a
href="mailto:sunny.s.zhang@oracle.com"
moz-do-not-send="true"><sunny.s.zhang@oracle.com></a>
<o:p></o:p></span></p>
</blockquote>
<p class="MsoNormal"><span lang="EN-US">The change looks
good. <br>
If possible, please run ocfs2-test test suits to
verify if the change causes any regressive deadlock.
<br>
Reviewed-by: Gang He <a href="mailto:ghe@suse.com"
moz-do-not-send="true"><ghe@suse.com></a> <br>
<br>
<br>
<o:p></o:p></span></p>
<blockquote style="margin-top:5.0pt;margin-bottom:5.0pt">
<p class="MsoNormal"><span lang="EN-US">--- <br>
fs/ocfs2/file.c | 129 <br>
++++++++++++++++++++++++++++++++++++++++++++--------- <br>
--- <br>
1 file changed, 101 insertions(+), 28 deletions(-)
<br>
<br>
diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index
4435df3..dff4ab2 100644 <br>
--- a/fs/ocfs2/file.c <br>
+++ b/fs/ocfs2/file.c <br>
@@ -2119,27 +2119,90 @@ static int <br>
ocfs2_prepare_inode_for_refcount(struct inode
*inode, <br>
return ret; <br>
} <br>
<br>
+static int ocfs2_inode_lock_for_extent_tree(struct
inode *inode, <br>
+ struct buffer_head **di_bh,
<br>
+ int meta_level, <br>
+ int overwrite_io, <br>
+ int write_sem, <br>
+ int wait) <br>
+{ <br>
+ int ret = 0; <br>
+ <br>
+ if (wait) <br>
+ ret = ocfs2_inode_lock(inode, NULL,
meta_level); <br>
+ else <br>
+ ret = ocfs2_try_inode_lock(inode, <br>
+ overwrite_io ? NULL : di_bh,
meta_level); <br>
+ if (ret < 0) { <br>
+ goto out; <br>
+ } <br>
+ <br>
+ if (wait) { <br>
+ if (write_sem) <br>
+
down_write(&OCFS2_I(inode)->ip_alloc_sem); <br>
+ else <br>
+
down_read(&OCFS2_I(inode)->ip_alloc_sem); <br>
+ } else { <br>
+ if (write_sem) <br>
+ ret =
down_write_trylock(&OCFS2_I(inode)->ip_alloc_sem);
<br>
+ else <br>
+ ret =
down_read_trylock(&OCFS2_I(inode)->ip_alloc_sem);
<br>
+ <br>
+ if (!ret) { <br>
+ ret = -EAGAIN; <br>
+ goto out_unlock; <br>
+ } <br>
+ } <br>
+ <br>
+ return ret; <br>
+ <br>
+out_unlock: <br>
+ brelse(*di_bh); <br>
+ ocfs2_inode_unlock(inode, meta_level); <br>
+out: <br>
+ return ret; <br>
+} <br>
+ <br>
+static void
ocfs2_inode_unlock_for_extent_tree(struct inode
*inode, <br>
+ struct buffer_head
**di_bh, <br>
+ int meta_level, <br>
+ int write_sem) <br>
+{ <br>
+ if (write_sem) <br>
+
up_write(&OCFS2_I(inode)->ip_alloc_sem); <br>
+ else <br>
+
up_read(&OCFS2_I(inode)->ip_alloc_sem); <br>
+ <br>
+ brelse(*di_bh); <br>
+ *di_bh = NULL; <br>
+ <br>
+ if (meta_level >= 0) <br>
+ ocfs2_inode_unlock(inode, meta_level); } <br>
+ <br>
static int ocfs2_prepare_inode_for_write(struct
file *file, <br>
loff_t pos, size_t count, int
wait) { <br>
- int ret = 0, meta_level = 0, overwrite_io = 0;
<br>
+ int ret = 0, meta_level = 0, overwrite_io = 0,
write_sem = 0; <br>
struct dentry *dentry =
file->f_path.dentry; <br>
struct inode *inode = d_inode(dentry); <br>
struct buffer_head *di_bh = NULL; <br>
loff_t end; <br>
+ u32 cpos; <br>
+ u32 clusters; <br>
<br>
/* <br>
* We start with a read level meta lock and
only jump to an ex <br>
* if we need to make modifications here. <br>
*/ <br>
for(;;) { <br>
- if (wait) <br>
- ret = ocfs2_inode_lock(inode, NULL,
meta_level); <br>
- else <br>
- ret = ocfs2_try_inode_lock(inode, <br>
- overwrite_io ? NULL : &di_bh,
meta_level); <br>
+ ret =
ocfs2_inode_lock_for_extent_tree(inode, <br>
+ &di_bh, <br>
+ meta_level, <br>
+ overwrite_io, <br>
+ write_sem, <br>
+ wait); <br>
if (ret < 0) { <br>
- meta_level = -1; <br>
if (ret != -EAGAIN) <br>
mlog_errno(ret); <br>
goto out; <br>
@@ -2151,15 +2214,8 @@ static int
ocfs2_prepare_inode_for_write(struct <br>
file *file, <br>
*/ <br>
if (!wait && !overwrite_io) { <br>
overwrite_io = 1; <br>
- if
(!down_read_trylock(&OCFS2_I(inode)->ip_alloc_sem))
{ <br>
- ret = -EAGAIN; <br>
- goto out_unlock; <br>
- } <br>
<br>
ret = ocfs2_overwrite_io(inode, di_bh,
pos, count); <br>
- brelse(di_bh); <br>
- di_bh = NULL; <br>
-
up_read(&OCFS2_I(inode)->ip_alloc_sem); <br>
if (ret < 0) { <br>
if (ret != -EAGAIN) <br>
mlog_errno(ret); <br>
@@ -2178,7 +2234,10 @@ static int
ocfs2_prepare_inode_for_write(struct <br>
file *file, <br>
* set inode->i_size at the end of a
write. */ <br>
if (should_remove_suid(dentry)) { <br>
if (meta_level == 0) { <br>
- ocfs2_inode_unlock(inode,
meta_level); <br>
+
ocfs2_inode_unlock_for_extent_tree(inode, <br>
+ &di_bh, <br>
+ meta_level, <br>
+ write_sem); <br>
meta_level = 1; <br>
continue; <br>
} <br>
@@ -2194,18 +2253,32 @@ static int
ocfs2_prepare_inode_for_write(struct <br>
file *file, <br>
<br>
ret =
ocfs2_check_range_for_refcount(inode, pos, count); <br>
if (ret == 1) { <br>
- ocfs2_inode_unlock(inode, meta_level);
<br>
- meta_level = -1; <br>
- <br>
- ret =
ocfs2_prepare_inode_for_refcount(inode, <br>
- file, <br>
- pos, <br>
- count, <br>
-
&meta_level); <br>
+
ocfs2_inode_unlock_for_extent_tree(inode, <br>
+ &di_bh, <br>
+ meta_level, <br>
+ write_sem); <br>
+ ret =
ocfs2_inode_lock_for_extent_tree(inode, <br>
+ &di_bh, <br>
+ meta_level, <br>
+ overwrite_io, <br>
+ 1, <br>
+ wait); <br>
+ write_sem = 1; <br>
+ if (ret < 0) { <br>
+ if (ret != -EAGAIN) <br>
+ mlog_errno(ret); <br>
+ goto out; <br>
+ } <br>
+ <br>
+ cpos = pos >>
OCFS2_SB(inode->i_sb)->s_clustersize_bits; <br>
+ clusters = <br>
+
ocfs2_clusters_for_bytes(inode->i_sb, pos +
count) - cpos; <br>
+ ret = ocfs2_refcount_cow(inode, di_bh,
cpos, clusters, UINT_MAX); <br>
} <br>
<br>
if (ret < 0) { <br>
- mlog_errno(ret); <br>
+ if (ret != -EAGAIN) <br>
+ mlog_errno(ret); <br>
goto out_unlock; <br>
} <br>
<br>
@@ -2216,10 +2289,10 @@ static int
ocfs2_prepare_inode_for_write(struct <br>
file *file, <br>
trace_ocfs2_prepare_inode_for_write(OCFS2_I(inode)->ip_blkno,
<br>
pos, count, wait); <br>
<br>
- brelse(di_bh); <br>
- <br>
- if (meta_level >= 0) <br>
- ocfs2_inode_unlock(inode, meta_level); <br>
+ ocfs2_inode_unlock_for_extent_tree(inode, <br>
+ &di_bh, <br>
+ meta_level, <br>
+ write_sem); <br>
<br>
out: <br>
return ret; <br>
-- <br>
2.7.4 <br>
<br>
<br>
_______________________________________________ <br>
Ocfs2-devel mailing list <br>
<a href="mailto:Ocfs2-devel@oss.oracle.com"
moz-do-not-send="true">Ocfs2-devel@oss.oracle.com</a>
<br>
<a
href="https://oss.oracle.com/mailman/listinfo/ocfs2-devel"
moz-do-not-send="true">https://oss.oracle.com/mailman/listinfo/ocfs2-devel</a>
<o:p></o:p></span></p>
</blockquote>
</blockquote>
</blockquote>
</div>
</div>
</blockquote>
</body>
</html>