From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
stable@vger.kernel.org, Ilya Dryomov <idryomov@gmail.com>
Subject: [PATCH 3.14 01/44] crush: ensuring at most num-rep osds are selected
Date: Fri, 19 Jun 2015 13:36:00 -0700 [thread overview]
Message-ID: <20150619203556.235676996@linuxfoundation.org> (raw)
In-Reply-To: <20150619203556.192033912@linuxfoundation.org>
3.14-stable review patch. If anyone has any objections, please let me know.
------------------
From: Ilya Dryomov <idryomov@gmail.com>
commit 45002267e8d2699bf9b022315bee3dd13b044843 upstream.
Crush temporary buffers are allocated as per replica size configured
by the user. When there are more final osds (to be selected as per
rule) than the replicas, buffer overlaps and it causes crash. Now, it
ensures that at most num-rep osds are selected even if more number of
osds are allowed by the rule.
Reflects ceph.git commits 6b4d1aa99718e3b367496326c1e64551330fabc0,
234b066ba04976783d15ff2abc3e81b6cc06fb10.
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
net/ceph/crush/mapper.c | 16 ++++++++++++----
1 file changed, 12 insertions(+), 4 deletions(-)
--- a/net/ceph/crush/mapper.c
+++ b/net/ceph/crush/mapper.c
@@ -290,6 +290,7 @@ static int is_out(const struct crush_map
* @type: the type of item to choose
* @out: pointer to output vector
* @outpos: our position in that vector
+ * @out_size: size of the out vector
* @tries: number of attempts to make
* @recurse_tries: number of attempts to have recursive chooseleaf make
* @local_retries: localized retries
@@ -302,6 +303,7 @@ static int crush_choose_firstn(const str
const __u32 *weight, int weight_max,
int x, int numrep, int type,
int *out, int outpos,
+ int out_size,
unsigned int tries,
unsigned int recurse_tries,
unsigned int local_retries,
@@ -318,11 +320,12 @@ static int crush_choose_firstn(const str
int item = 0;
int itemtype;
int collide, reject;
+ int count = out_size;
dprintk("CHOOSE%s bucket %d x %d outpos %d numrep %d\n", recurse_to_leaf ? "_LEAF" : "",
bucket->id, x, outpos, numrep);
- for (rep = outpos; rep < numrep; rep++) {
+ for (rep = outpos; rep < numrep && count > 0 ; rep++) {
/* keep trying until we get a non-out, non-colliding item */
ftotal = 0;
skip_rep = 0;
@@ -391,7 +394,7 @@ static int crush_choose_firstn(const str
map->buckets[-1-item],
weight, weight_max,
x, outpos+1, 0,
- out2, outpos,
+ out2, outpos, count,
recurse_tries, 0,
local_retries,
local_fallback_retries,
@@ -449,6 +452,7 @@ reject:
dprintk("CHOOSE got %d\n", item);
out[outpos] = item;
outpos++;
+ count--;
}
dprintk("CHOOSE returns %d\n", outpos);
@@ -640,6 +644,7 @@ int crush_do_rule(const struct crush_map
__u32 step;
int i, j;
int numrep;
+ int out_size;
/*
* the original choose_total_tries value was off by one (it
* counted "retries" and not "tries"). add one.
@@ -740,6 +745,7 @@ int crush_do_rule(const struct crush_map
x, numrep,
curstep->arg2,
o+osize, j,
+ result_max-osize,
choose_tries,
recurse_tries,
choose_local_retries,
@@ -747,11 +753,13 @@ int crush_do_rule(const struct crush_map
recurse_to_leaf,
c+osize);
} else {
+ out_size = ((numrep < (result_max-osize)) ?
+ numrep : (result_max-osize));
crush_choose_indep(
map,
map->buckets[-1-w[i]],
weight, weight_max,
- x, numrep, numrep,
+ x, out_size, numrep,
curstep->arg2,
o+osize, j,
choose_tries,
@@ -760,7 +768,7 @@ int crush_do_rule(const struct crush_map
recurse_to_leaf,
c+osize,
0);
- osize += numrep;
+ osize += out_size;
}
}
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
Please read the FAQ at http://www.tux.org/lkml/
next prev parent reply other threads:[~2015-06-19 20:36 UTC|newest]
Thread overview: 46+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-06-19 20:35 [PATCH 3.14 00/44] 3.14.45-stable review Greg Kroah-Hartman
2015-06-19 20:36 ` Greg Kroah-Hartman [this message]
2015-06-19 20:36 ` [PATCH 3.14 02/44] net: core: Correct an over-stringent device loop detection Greg Kroah-Hartman
2015-06-19 20:36 ` [PATCH 3.14 03/44] x86: bpf_jit: fix compilation of large bpf programs Greg Kroah-Hartman
2015-06-19 20:36 ` [PATCH 3.14 04/44] net: phy: Allow EEE for all RGMII variants Greg Kroah-Hartman
2015-06-19 20:36 ` [PATCH 3.14 05/44] tcp/ipv6: fix flow label setting in TIME_WAIT state Greg Kroah-Hartman
2015-06-19 20:36 ` [PATCH 3.14 06/44] ipv4: Avoid crashing in ip_error Greg Kroah-Hartman
2015-06-19 20:36 ` [PATCH 3.14 07/44] bridge: fix parsing of MLDv2 reports Greg Kroah-Hartman
2015-06-19 20:36 ` [PATCH 3.14 08/44] net: dp83640: fix broken calibration routine Greg Kroah-Hartman
2015-06-19 20:36 ` [PATCH 3.14 09/44] net: dp83640: reinforce locking rules Greg Kroah-Hartman
2015-06-19 20:36 ` [PATCH 3.14 10/44] unix/caif: sk_socket can disappear when state is unlocked Greg Kroah-Hartman
2015-06-19 20:36 ` [PATCH 3.14 11/44] net_sched: invoke ->attach() after setting dev->qdisc Greg Kroah-Hartman
2015-06-19 20:36 ` [PATCH 3.14 12/44] udp: fix behavior of wrong checksums Greg Kroah-Hartman
2015-06-19 20:36 ` [PATCH 3.14 13/44] xen: netback: read hotplug script once at start of day Greg Kroah-Hartman
2015-06-19 20:36 ` [PATCH 3.14 14/44] ipv4/udp: Verify multicast group is ours in upd_v4_early_demux() Greg Kroah-Hartman
2015-06-19 20:36 ` [PATCH 3.14 15/44] bridge: disable softirqs around br_fdb_update to avoid lockup Greg Kroah-Hartman
2015-06-19 20:36 ` [PATCH 3.14 16/44] iio: adc: twl6030-gpadc: Fix modalias Greg Kroah-Hartman
2015-06-19 20:36 ` [PATCH 3.14 17/44] iio: adis16400: Report pressure channel scale Greg Kroah-Hartman
2015-06-19 20:36 ` [PATCH 3.14 18/44] iio: adis16400: Use != channel indices for the two voltage channels Greg Kroah-Hartman
2015-06-19 20:36 ` [PATCH 3.14 19/44] iio: adis16400: Compute the scan mask from channel indices Greg Kroah-Hartman
2015-06-19 20:36 ` [PATCH 3.14 21/44] ALSA: hda/realtek - Add a fixup for another Acer Aspire 9420 Greg Kroah-Hartman
2015-06-19 20:36 ` [PATCH 3.14 22/44] ALSA: usb-audio: Add mic volume fix quirk for Logitech Quickcam Fusion Greg Kroah-Hartman
2015-06-19 20:36 ` [PATCH 3.14 23/44] ALSA: usb-audio: add MAYA44 USB+ mixer control names Greg Kroah-Hartman
2015-06-19 20:36 ` [PATCH 3.14 24/44] Input: synaptics - add min/max quirk for Lenovo S540 Greg Kroah-Hartman
2015-06-19 20:36 ` [PATCH 3.14 26/44] block: fix ext_dev_lock lockdep report Greg Kroah-Hartman
2015-06-19 20:36 ` [PATCH 3.14 27/44] USB: cp210x: add ID for HubZ dual ZigBee and Z-Wave dongle Greg Kroah-Hartman
2015-06-19 20:36 ` [PATCH 3.14 28/44] USB: serial: ftdi_sio: Add support for a Motion Tracker Development Board Greg Kroah-Hartman
2015-06-19 20:36 ` [PATCH 3.14 29/44] x86/asm/irq: Stop relying on magic JMP behavior for early_idt_handlers Greg Kroah-Hartman
2015-06-19 20:36 ` [PATCH 3.14 30/44] ring-buffer-benchmark: Fix the wrong sched_priority of producer Greg Kroah-Hartman
2015-06-19 20:36 ` [PATCH 3.14 31/44] MIPS: Fix enabling of DEBUG_STACKOVERFLOW Greg Kroah-Hartman
2015-06-19 20:36 ` [PATCH 3.14 32/44] ozwpan: Use proper check to prevent heap overflow Greg Kroah-Hartman
2015-06-19 20:36 ` [PATCH 3.14 33/44] ozwpan: divide-by-zero leading to panic Greg Kroah-Hartman
2015-06-19 20:36 ` [PATCH 3.14 34/44] ozwpan: unchecked signed subtraction leads to DoS Greg Kroah-Hartman
2015-06-19 20:36 ` [PATCH 3.14 35/44] pata_octeon_cf: fix broken build Greg Kroah-Hartman
2015-06-19 20:36 ` [PATCH 3.14 36/44] ARM: dts: am335x-boneblack: disable RTC-only sleep to avoid hardware damage Greg Kroah-Hartman
2015-06-22 8:23 ` Johan Hovold
2015-06-22 15:41 ` Greg Kroah-Hartman
2015-06-19 20:36 ` [PATCH 3.14 37/44] drm/i915/hsw: Fix workaround for server AUX channel clock divisor Greg Kroah-Hartman
2015-06-19 20:36 ` [PATCH 3.14 40/44] serial: imx: Fix DMA handling for IDLE condition aborts Greg Kroah-Hartman
2015-06-19 20:36 ` [PATCH 3.14 41/44] mm/memory_hotplug.c: set zone->wait_table to null after freeing it Greg Kroah-Hartman
2015-06-19 20:36 ` [PATCH 3.14 42/44] cfg80211: wext: clear sinfo struct before calling driver Greg Kroah-Hartman
2015-06-19 20:36 ` [PATCH 3.14 43/44] btrfs: incorrect handling for fiemap_fill_next_extent return Greg Kroah-Hartman
2015-06-19 20:36 ` [PATCH 3.14 44/44] btrfs: cleanup orphans while looking up default subvolume Greg Kroah-Hartman
2015-06-20 1:12 ` [PATCH 3.14 00/44] 3.14.45-stable review Shuah Khan
2015-06-20 1:26 ` Guenter Roeck
2015-06-20 7:49 ` Sudip Mukherjee
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20150619203556.235676996@linuxfoundation.org \
--to=gregkh@linuxfoundation.org \
--cc=idryomov@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=stable@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).