summaryrefslogtreecommitdiffstats
path: root/net/ceph/ceph_common.c
diff options
context:
space:
mode:
authorIlya Dryomov <idryomov@gmail.com>2021-12-30 15:13:32 +0100
committerIlya Dryomov <idryomov@gmail.com>2022-02-02 18:50:36 +0100
commit038b8d1d1ab1cce11a158d30bf080ff41a2cfd15 (patch)
tree83f509c79a7d909bd440882f0f39492b08dbc31c /net/ceph/ceph_common.c
parentlibceph: make recv path in secure mode work the same as send path (diff)
downloadlinux-038b8d1d1ab1cce11a158d30bf080ff41a2cfd15.tar.xz
linux-038b8d1d1ab1cce11a158d30bf080ff41a2cfd15.zip
libceph: optionally use bounce buffer on recv path in crc mode
Both msgr1 and msgr2 in crc mode are zero copy in the sense that message data is read from the socket directly into the destination buffer. We assume that the destination buffer is stable (i.e. remains unchanged while it is being read to) though. Otherwise, CRC errors ensue: libceph: read_partial_message 0000000048edf8ad data crc 1063286393 != exp. 228122706 libceph: osd1 (1)192.168.122.1:6843 bad crc/signature libceph: bad data crc, calculated 57958023, expected 1805382778 libceph: osd2 (2)192.168.122.1:6876 integrity error, bad crc Introduce rxbounce option to enable use of a bounce buffer when receiving message data. In particular this is needed if a mapped image is a Windows VM disk, passed to QEMU. Windows has a system-wide "dummy" page that may be mapped into the destination buffer (potentially more than once into the same buffer) by the Windows Memory Manager in an effort to generate a single large I/O [1][2]. QEMU makes a point of preserving overlap relationships when cloning I/O vectors, so krbd gets exposed to this behaviour. [1] "What Is Really in That MDL?" https://docs.microsoft.com/en-us/previous-versions/windows/hardware/design/dn614012(v=vs.85) [2] https://blogs.msmvps.com/kernelmustard/2005/05/04/dummy-pages/ URL: https://bugzilla.redhat.com/show_bug.cgi?id=1973317 Signed-off-by: Ilya Dryomov <idryomov@gmail.com> Reviewed-by: Jeff Layton <jlayton@kernel.org>
Diffstat (limited to 'net/ceph/ceph_common.c')
-rw-r--r--net/ceph/ceph_common.c7
1 files changed, 7 insertions, 0 deletions
diff --git a/net/ceph/ceph_common.c b/net/ceph/ceph_common.c
index ecc400a0b7bb..4c6441536d55 100644
--- a/net/ceph/ceph_common.c
+++ b/net/ceph/ceph_common.c
@@ -246,6 +246,7 @@ enum {
Opt_cephx_sign_messages,
Opt_tcp_nodelay,
Opt_abort_on_full,
+ Opt_rxbounce,
};
enum {
@@ -295,6 +296,7 @@ static const struct fs_parameter_spec ceph_parameters[] = {
fsparam_u32 ("osdkeepalive", Opt_osdkeepalivetimeout),
fsparam_enum ("read_from_replica", Opt_read_from_replica,
ceph_param_read_from_replica),
+ fsparam_flag ("rxbounce", Opt_rxbounce),
fsparam_enum ("ms_mode", Opt_ms_mode,
ceph_param_ms_mode),
fsparam_string ("secret", Opt_secret),
@@ -584,6 +586,9 @@ int ceph_parse_param(struct fs_parameter *param, struct ceph_options *opt,
case Opt_abort_on_full:
opt->flags |= CEPH_OPT_ABORT_ON_FULL;
break;
+ case Opt_rxbounce:
+ opt->flags |= CEPH_OPT_RXBOUNCE;
+ break;
default:
BUG();
@@ -660,6 +665,8 @@ int ceph_print_client_options(struct seq_file *m, struct ceph_client *client,
seq_puts(m, "notcp_nodelay,");
if (show_all && (opt->flags & CEPH_OPT_ABORT_ON_FULL))
seq_puts(m, "abort_on_full,");
+ if (opt->flags & CEPH_OPT_RXBOUNCE)
+ seq_puts(m, "rxbounce,");
if (opt->mount_timeout != CEPH_MOUNT_TIMEOUT_DEFAULT)
seq_printf(m, "mount_timeout=%d,",