search for: 1024

...che; + + cache = kzalloc(sizeof(*cache), GFP_NOFS); + if (!cache) + return NULL; + cache->free_space_ctl = kzalloc(sizeof(*cache->free_space_ctl), + GFP_NOFS); + if (!cache->free_space_ctl) { + kfree(cache); + return NULL; + } + + cache->key.objectid = 0; + cache->key.offset = 1024 * 1024 * 1024; + cache->key.type = BTRFS_BLOCK_GROUP_ITEM_KEY; + cache->sectorsize = 4096; + + spin_lock_init(&cache->lock); + INIT_LIST_HEAD(&cache->list); + INIT_LIST_HEAD(&cache->cluster_list); + INIT_LIST_HEAD(&cache->new_bg_list); + + btrfs_init_free_space_ctl...

...= ? Process 14581 detached Receiver: read(4, "IX\2444\216\243\365\26\222A\311\277\327\336d\375\336v\210"..., 8184) = 8184 time(NULL) = 1142443237 write(5, "\236/os\367&\330\237\6\25\njH\37\267\256~{\307\273i\27"..., 1024) = 1024 write(5, "\35\300,\305N\267\232v\343#\34\226\n\177\26\20\265J\r\t"..., 1024) = 1024 write(5, "K\16\377\0mE\260\274\230\351\0271\327\375\300\271z\177"..., 1024) = 1024 write(5, "\356$\306*\321MI:\325,\270\215\247D\271<\6\35\204\210\254"..., 1024) = 1024 write...

...=3D ? Process 14581 detached Receiver: read(4, "IX\2444\216\243\365\26\222A\311\277\327\336d\375\336v\210"..., = 8184) =3D 8184 time(NULL) =3D 1142443237 write(5, "\236/os\367&\330\237\6\25\njH\37\267\256~{\307\273i\27"..., = 1024) =3D 1024 write(5, = "\35\300,\305N\267\232v\343#\34\226\n\177\26\20\265J\r\t"..., 1024) =3D = 1024 write(5, = "K\16\377\0mE\260\274\230\351\0271\327\375\300\271z\177"..., 1024) =3D = 1024 write(5, = "\356$\306*\321MI:\325,\270\215\247D\271<\6\35\204\210\254"..., 10...

...ke how patches 1-3 look. Could you test performance > with/without to see whether the extra indirection through > use of DMA ops causes a measurable slow-down? I ran this simple DD command 10 times where /dev/vda is a virtio block device of 10GB size. dd if=/dev/zero of=/dev/vda bs=8M count=1024 oflag=direct With and without patches bandwidth which has a bit wide range does not look that different from each other. Without patches =============== ---------- 1 --------- 1024+0 records in 1024+0 records out 8589934592 bytes (8.6 GB, 8.0 GiB) copied, 1.95557 s, 4.4 GB/s ---------- 2 -------...

...ke how patches 1-3 look. Could you test performance > with/without to see whether the extra indirection through > use of DMA ops causes a measurable slow-down? I ran this simple DD command 10 times where /dev/vda is a virtio block device of 10GB size. dd if=/dev/zero of=/dev/vda bs=8M count=1024 oflag=direct With and without patches bandwidth which has a bit wide range does not look that different from each other. Without patches =============== ---------- 1 --------- 1024+0 records in 1024+0 records out 8589934592 bytes (8.6 GB, 8.0 GiB) copied, 1.95557 s, 4.4 GB/s ---------- 2 -------...

...size; - char t; + char *end; + uint64_t scale = 1; - if (sscanf (str, "%" SCNu64 "%c", &size, &t) == 2) { - switch (t) { - case 'b': case 'B': - return (int64_t) size; - case 'k': case 'K': - return (int64_t) size * 1024; - case 'm': case 'M': - return (int64_t) size * 1024 * 1024; - case 'g': case 'G': - return (int64_t) size * 1024 * 1024 * 1024; - case 't': case 'T': - return (int64_t) size * 1024 * 1024 * 1024 * 1024; - case 'p':...

Take two, this time split into two patches. I liked Rich's suggestion of unit-testing src/ files directly in src/, and automake is a lot happier with this than with my v1 attempt that tried to compile .c files across directories. It's still enough of a change that I'm not pushing it right away. Eric Blake (2): build: Add unit-testing of internal files utils: Revamp

...uot;, 0 }, { " 08", 8 }, + { "1", 1 }, + { "+1", 1 }, + { "1234567890", 1234567890 }, + { "+1234567890", 1234567890 }, { "9223372036854775807", INT64_MAX }, { "1s", 512 }, { "2S", 1024 }, @@ -88,12 +106,14 @@ test_nbdkit_parse_size (void) { "1K", 1024 }, { "1m", 1024 * 1024 }, { "1M", 1024 * 1024 }, + { "+1M", 1024 * 1024 }, { "1g", 1024 * 1024 * 1024 }, { "1G", 1024 * 1024 * 1024 },...

...test performance > > with/without to see whether the extra indirection through > > use of DMA ops causes a measurable slow-down? > > I ran this simple DD command 10 times where /dev/vda is a virtio block > device of 10GB size. > > dd if=/dev/zero of=/dev/vda bs=8M count=1024 oflag=direct > > With and without patches bandwidth which has a bit wide range does not > look that different from each other. > > Without patches > =============== > > ---------- 1 --------- > 1024+0 records in > 1024+0 records out > 8589934592 bytes (8.6 GB, 8....

...uot;, 0 }, { " 08", 8 }, + { "1", 1 }, + { "+1", 1 }, + { "1234567890", 1234567890 }, + { "+1234567890", 1234567890 }, { "9223372036854775807", INT64_MAX }, { "1s", 512 }, { "2S", 1024 }, @@ -88,12 +106,14 @@ test_nbdkit_parse_size (void) { "1K", 1024 }, { "1m", 1024 * 1024 }, { "1M", 1024 * 1024 }, + { "+1M", 1024 * 1024 }, { "1g", 1024 * 1024 * 1024 }, { "1G", 1024 * 1024 * 1024 },...

This is the first part of a pair of patch series which aim to let us use nbdkit_parse_size (or rather, an equivalent common function) in nbdcopy, so we can write: nbdcopy --request-size=32M ... We can't do that now which was annoying me earlier in the week. This commit creates a new function called human_size_parse which is basically nbdkit_parse_size, and turns nbdkit_parse_size into a

...= 300 27542: fcntl(5, F_SETLKW64, 0xEFFFF0A8) = 0 27542: alarm(0) = 5 27542: sigaction(SIGALRM, 0xEFFFEF98, 0x00000000) = 0 27542: fchmod(5, 0600) = 0 27542: read(5, " # S a m b a S M B ".., 1024) = 1024 27542: read(5, " 5 C F A 4 E 6 5 D 6 6 2".., 1024) = 1024 27542: read(5, " 0 E 4 0 1 C : 7 9 7 D E".., 1024) = 1024 27542: read(5, " A B 8 0 1 4 6 F 5 7 4 9".., 1024) = 1024 27542: read(5, " 1 E 3 :\n m a r c e l p".., 1024) = 1024...

...unction. The old comment mentioned the 'human*' interface from gnulib; we can't use that for licensing reasons; so my rewrite is done from scratch. In particular, I know that gnulib's code accepts 'M' as a synonym for '1M'; treats '1k' and '1kiB' as 1024 but '1kB' as 1000, and so on, which I didn't bother to implement here. The unit test can now be adjusted to cover the cases that are now handled correctly. Signed-off-by: Eric Blake <eblake@redhat.com> --- src/test-utils.c | 20 ++++++------ src/utils.c | 95 ++++++++++++++...

...;> with/without to see whether the extra indirection through >>> use of DMA ops causes a measurable slow-down? >> >> I ran this simple DD command 10 times where /dev/vda is a virtio block >> device of 10GB size. >> >> dd if=/dev/zero of=/dev/vda bs=8M count=1024 oflag=direct >> >> With and without patches bandwidth which has a bit wide range does not >> look that different from each other. >> >> Without patches >> =============== >> >> ---------- 1 --------- >> 1024+0 records in >> 1024+0 records...

...;> with/without to see whether the extra indirection through >>> use of DMA ops causes a measurable slow-down? >> >> I ran this simple DD command 10 times where /dev/vda is a virtio block >> device of 10GB size. >> >> dd if=/dev/zero of=/dev/vda bs=8M count=1024 oflag=direct >> >> With and without patches bandwidth which has a bit wide range does not >> look that different from each other. >> >> Without patches >> =============== >> >> ---------- 1 --------- >> 1024+0 records in >> 1024+0 records...

...92 100644 --- a/generator/actions.ml +++ b/generator/actions.ml @@ -4780,7 +4780,7 @@ C<blocksize> option of C<guestfs_mkfs>." }; tests = [ InitEmpty, Always, TestResult ( [["blockdev_getsz"; "/dev/sda"]], - "ret == INT64_C(10)*1024*1024*1024/512"), [] + "ret == INT64_C(2)*1024*1024*1024/512"), [] ]; shortdesc = "get total size of device in 512-byte sectors"; longdesc = "\ @@ -4800,7 +4800,7 @@ This uses the L<blockdev(8)> command." }; tests = [ InitE...

...most 53%) while rx is as much as in the past. Most cases the cpu utilization is also improved: Guest TX: size/session/+thu%/+normalize% 64/ 1/ +17%/ +6% 64/ 4/ +9%/ +17% 64/ 8/ +34%/ +21% 512/ 1/ +48%/ +40% 512/ 4/ +31%/ +20% 512/ 8/ +39%/ +22% 1024/ 1/ +158%/ +99% 1024/ 4/ +20%/ +11% 1024/ 8/ +40%/ +18% 2048/ 1/ +108%/ +74% 2048/ 4/ +21%/ +7% 2048/ 8/ +32%/ +14% 4096/ 1/ +94%/ +77% 4096/ 4/ +7%/ -6% 4096/ 8/ +9%/ -4% 16384/ 1/ +33%/ +9% 16384/ 4/ +10%/ -6% 16384/...

...most 53%) while rx is as much as in the past. Most cases the cpu utilization is also improved: Guest TX: size/session/+thu%/+normalize% 64/ 1/ +17%/ +6% 64/ 4/ +9%/ +17% 64/ 8/ +34%/ +21% 512/ 1/ +48%/ +40% 512/ 4/ +31%/ +20% 512/ 8/ +39%/ +22% 1024/ 1/ +158%/ +99% 1024/ 4/ +20%/ +11% 1024/ 8/ +40%/ +18% 2048/ 1/ +108%/ +74% 2048/ 4/ +21%/ +7% 2048/ 8/ +32%/ +14% 4096/ 1/ +94%/ +77% 4096/ 4/ +7%/ -6% 4096/ 8/ +9%/ -4% 16384/ 1/ +33%/ +9% 16384/ 4/ +10%/ -6% 16384/...

Hello, We have a 3-node cluster setup with ocfs2. Since friday one of the nodes went down and would not become clustermember after a reboot because it was unable to write to the ocfs2 filesystem. Message: no space left on device. There is plenty of diskspace though. No problem whatsoever to create a file / directory on the filesystem using one of the other nodes. Today one of the remaining

...stem sometimes did not warn me a tip when the soft-limit of space usage was exceeded.following are the steps to reproduce this issue: 1.mkfs.ocfs2 --fs-features=usrquota,grpquota ... /dev/sdc5 2.mount -t ocfs2 -o usrquota,grpquota /dev/sdc5 /quota 3.useradd -m quotauser 4.setquota -u quotauser 1024 2048 10 20 -a /dev/sdc5 5.quotaon -avug 6.su - quotauser; cd /quota 7. dd if=/dev/zero of=./testfile1 bs=1024 count=1024 1024+0 records in 1024+0 records out 1048576 bytes (1.0 MB) copied, 0.00825239 seconds, 127 MB/s dd if=/dev/zero of=./testfile2bs=1024 count=10 sdc5: warning, user block quo...