rhashtable: avoid reschedule loop after rapid growth and shrink
Subsystems (4)
PATCH 1/1 rhashtable: avoid reschedule loop after rapid growth and shrink
When running workloads with large bursts of fragmented packets, we've seen a few machines stuck returning -EEXIST from rht_shrink() and endlessly rescheduling their hash table's deferred work, pegging a CPU core.
Root cause is commit da20420f83ea ("rhashtable: Add nested tables"), which stops ignoring the return code of rhashtable_shrink() and the reallocs used to grow the hashtable. This uncovers a bug in the shrink logic where "needs to shrink" check runs against the last table but the actual shrink operation runs on the first bucket_table in the hashtable (see below):
+-------+ +--------------+ +---------------+
| ht | | "first" tbl | | "last" tbl |
| - tbl ---> | - future_tbl ---------> | - future_tbl ---> NULL
+-------+ +--------------+ +---------------+
^^^ ^^^
used by rhashtable_shrink() used by rht_shrink_below_30()A rehash then stalls out when both the last table needs to shrink, the first table has more elements than the target size, but rht_shrink() hits a non-NULL future_tbl and returns -EEXIST. This skips the item rehashing and kicks off a reschedule loop, as no forward progress can be made while the rhashtable needs to shrink.
Extend rhashtable_shrink() with a "tbl" param to avoid endless exit-and- reschedules after hitting the EEXIST, allowing it to check a future_tbl pointer that can actually be non-NULL and make forward progress when the hashtable needs to shrink.
--- lib/rhashtable.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
lib/rhashtable.c +3 -3
@@ -377,9 +377,9 @@ static int rhashtable_rehash_alloc(struct rhashtable *ht,* It is valid to have concurrent insertions and deletions protected by per * bucket locks or concurrent RCU protected lookups and traversals. */ -static int rhashtable_shrink (struct rhashtable *ht) +static int rhashtable_shrink (struct rhashtable *ht, +struct bucket_table *old_tbl) { -struct bucket_table *old_tbl =rht_dereference (ht->tbl, ht);unsigned int nelems =atomic_read (&ht->nelems);unsigned int size =0 ;
@@ -412,7 +412,7 @@ static void rht_deferred_worker(struct work_struct *work)if (rht_grow_above_75 (ht, tbl)) err =rhashtable_rehash_alloc (ht, tbl, tbl->size *2 );else if (ht->p.automatic_shrinking &&rht_shrink_below_30 (ht, tbl)) - err =rhashtable_shrink (ht); + err =rhashtable_shrink (ht, tbl);else if (tbl->nest) err =rhashtable_rehash_alloc (ht, tbl, tbl->size);
Signature · 1 line
2.19.1
Herbert Xu
maintainerPATCH v2 1/1 rhashtable: Still do rehash when we get EEXIST
Quoted 23 lines · Josh Elsasser
When running workloads with large bursts of fragmented packets, we've seen a few machines stuck returning
-EEXISTfromrht_shrink()and endlessly rescheduling their hash table's deferred work, pegging a CPU core.Root cause is commit
da20420f83ea("rhashtable: Add nested tables"), which stops ignoring the return code ofrhashtable_shrink()and the reallocs used to grow the hashtable. This uncovers a bug in the shrink logic where "needs to shrink" check runs against the last table but the actual shrink operation runs on the firstbucket_tablein the hashtable (see below):+-------+ +--------------+ +---------------+ | ht | | "first" tbl | | "last" tbl | | - tbl ---> | - future_tbl ---------> | - future_tbl ---> NULL +-------+ +--------------+ +---------------+ ^^^ ^^^ used by rhashtable_shrink() used by rht_shrink_below_30()A rehash then stalls out when both the last table needs to shrink, the first table has more elements than the target size, but
rht_shrink()hits a non-NULLfuture_tbland returns-EEXIST. This skips the item rehashing and kicks off a reschedule loop, as no forward progress can be made while the rhashtable needs to shrink.
Quoted 7 lines
Extend
rhashtable_shrink()with a "tbl" param to avoid endless exit-and- reschedules after hitting the EEXIST, allowing it to check afuture_tblpointer that can actually be non-NULL and make forward progress when the hashtable needs to shrink.Fixes da20420f83ea ("rhashtable: Add nested tables")Signed-off-by Josh Elsasser <[email protected]>
Thanks for catching this!
Although I think we should fix this in a different way. The problem here is that the shrink cannot proceed because there was a previous rehash that is still incomplete. We should wait for its completion and then reattempt a shrinnk should it still be necessary.
So something like this:
---8<---
As it stands if a shrink is delayed because of an outstanding rehash, we will go into a rescheduling loop without ever doing the rehash.
This patch fixes this by still carrying out the rehash and then rescheduling so that we can shrink after the completion of the rehash should it still be necessary.
The return value of EEXIST captures this case and other cases (e.g., another thread expanded/rehashed the table at the same time) where we should still proceed with the rehash.
lib/rhashtable.c +6 -2
@@ -416,8 +416,12 @@ static void rht_deferred_worker(struct work_struct *work)else if (tbl->nest) err =rhashtable_rehash_alloc (ht, tbl, tbl->size); -if (!err) - err =rhashtable_rehash_table (ht); +if (!err || err == -EEXIST ) { +int nerr; + + nerr =rhashtable_rehash_table (ht); + err = err ?: nerr; + }mutex_unlock (&ht->mutex);
Signature · 3 lines
Email: Herbert Xu <[email protected]> Home Page: http://gondor.apana.org.au/~herbert/ PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
Re: [v2 PATCH] rhashtable: Still do rehash when we get EEXIST
Quoted 8 lines · Herbert Xu
Thanks for catching this!
Although I think we should fix this in a different way. The problem here is that the shrink cannot proceed because there was a previous rehash that is still incomplete. We should wait for its completion and then reattempt a shrinnk should it still be necessary.
So something like this:
SGTM.
I can't test this right now because our VM server's down after a power
outage this evening, but I tried a similar patch that swallowed the -EEXIST err and even with that oversight the hashtable dodged the
reschedule loop.
- Josh
Josh Elsasser
Tested-byRe: [v2 PATCH] rhashtable: Still do rehash when we get EEXIST
Quoted 10 lines · Josh Elsasser
Quoted 6 lines · Herbert Xu
Thanks for catching this!
Although I think we should fix this in a different way. The problem here is that the shrink cannot proceed because there was a previous rehash that is still incomplete. We should wait for its completion and then reattempt a shrinnk should it still be necessary.
I can't test this right now because our VM server's down
Got one of the poor little reproducer VM's back up and running and loaded up this patch. Works like a charm. For the v2 PATCH, can add my:
Re: [v2 PATCH] rhashtable: Still do rehash when we get EEXIST
Quoted 11 lines · Josh Elsasser
Quoted 9 lines · Josh Elsasser
Quoted 6 lines · Herbert Xu
Thanks for catching this!
Although I think we should fix this in a different way. The problem here is that the shrink cannot proceed because there was a previous rehash that is still incomplete. We should wait for its completion and then reattempt a shrinnk should it still be necessary.
Quoted 6 lines
Quoted 1 line
I can't test this right now because our VM server's down
Got one of the poor little reproducer VM's back up and running and loaded up this patch. Works like a charm. For the v2 PATCH, can add my:
Tested-by Josh Elsasser <[email protected]>
Trying again... gmail sent HTML mail first time.
Herbert
We're seeing this pretty regularly on 4.14 LTS kernels. I didn't see your change in any of the regular trees. Are there plans to submit this? If so, can it get queued up for 4.14 stable too?
Thanks!
Signature · 1 line
Josh
A message that is not in the archive (sent off-list or not yet received): CAKA=qzY4Pzee9BVzRCciW32toeHSz7t0q9LuvQXKLG2fX9fBbg@mail.gmail.com
Herbert Xu
maintainerPATCH v3 1/1 rhashtable: Still do rehash when we get EEXIST
Quoted 6 lines · Josh Hunt
Herbert
We're seeing this pretty regularly on 4.14 LTS kernels. I didn't see your change in any of the regular trees. Are there plans to submit this? If so, can it get queued up for 4.14 stable too?
Hi Josh:
Thanks for reminding me. Looks like this one slipped through the cracks.
Dave, could you please apply this? Thanks!
---8<---
As it stands if a shrink is delayed because of an outstanding rehash, we will go into a rescheduling loop without ever doing the rehash.
This patch fixes this by still carrying out the rehash and then rescheduling so that we can shrink after the completion of the rehash should it still be necessary.
The return value of EEXIST captures this case and other cases (e.g., another thread expanded/rehashed the table at the same time) where we should still proceed with the rehash.
lib/rhashtable.c +6 -2
@@ -416,8 +416,12 @@ static void rht_deferred_worker(struct work_struct *work)else if (tbl->nest) err =rhashtable_rehash_alloc (ht, tbl, tbl->size); -if (!err) - err =rhashtable_rehash_table (ht); +if (!err || err == -EEXIST ) { +int nerr; + + nerr =rhashtable_rehash_table (ht); + err = err ?: nerr; + }mutex_unlock (&ht->mutex);
Signature · 3 lines
Email: Herbert Xu <[email protected]> Home Page: http://gondor.apana.org.au/~herbert/ PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
Re: [v3 PATCH] rhashtable: Still do rehash when we get EEXIST
Quoted 2 lines · Herbert Xu
Dave, could you please apply this? Thanks!
I forgot to include the tested-by so here it is for patchwork:
Signature · 3 lines
Email: Herbert Xu <[email protected]> Home Page: http://gondor.apana.org.au/~herbert/ PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
David Miller
maintainerRe: [v3 PATCH] rhashtable: Still do rehash when we get EEXIST
From: Herbert Xu <[email protected]>
Date: Thu, 21 Mar 2019 09:46:15 +0800
Quoted 7 lines · Herbert Xu
Quoted 2 lines
Dave, could you please apply this? Thanks!
I forgot to include the tested-by so here it is for patchwork:
Tested-by Josh Elsasser <[email protected]>
Thanks for providing this.
David Miller
maintainerRe: [v3 PATCH] rhashtable: Still do rehash when we get EEXIST
In reply to Herbert XuFrom: Herbert Xu <[email protected]>
Date: Thu, 21 Mar 2019 09:39:52 +0800
Quoted 8 lines
As it stands if a shrink is delayed because of an outstanding rehash, we will go into a rescheduling loop without ever doing the rehash.
This patch fixes this by still carrying out the rehash and then rescheduling so that we can shrink after the completion of the rehash should it still be necessary.
Quoted 7 lines
The return value of EEXIST captures this case and other cases (e.g., another thread expanded/rehashed the table at the same time) where we should still proceed with the rehash.
Fixes da20420f83ea ("rhashtable: Add nested tables")Reported-by Josh Elsasser <[email protected]>Signed-off-by Herbert Xu <[email protected]>
Applied and queued up for -stable, thanks Herbert.