agora inbox for [email protected]  

help / color / mirror / Atom feed

[PATCH 7/9] fixup recovery tests
17+ messages / 2 participants
[nested] [flat]

* [PATCH 7/9] fixup recovery tests
@ 2021-06-13 13:22  Andrew Dunstan <[email protected]>
  0 siblings, 0 replies; 17+ messages in thread

From: Andrew Dunstan @ 2021-06-13 13:22 UTC (permalink / raw)

---
 src/test/recovery/t/005_replay_delay.pl          | 2 +-
 src/test/recovery/t/025_stuck_on_old_timeline.pl | 6 +++---
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/src/test/recovery/t/005_replay_delay.pl b/src/test/recovery/t/005_replay_delay.pl
index aa418dda8f..0b56380e0a 100644
--- a/src/test/recovery/t/005_replay_delay.pl
+++ b/src/test/recovery/t/005_replay_delay.pl
@@ -58,7 +58,7 @@ ok(time() - $primary_insert_time >= $delay,
 
 
 # Check that recovery can be paused or resumed expectedly.
-my $node_standby2 = get_new_node('standby2');
+my $node_standby2 = PostgresNode->new('standby2');
 $node_standby2->init_from_backup($node_primary, $backup_name,
 	has_streaming => 1);
 $node_standby2->start;
diff --git a/src/test/recovery/t/025_stuck_on_old_timeline.pl b/src/test/recovery/t/025_stuck_on_old_timeline.pl
index fb15f9576b..00ee9fcaed 100644
--- a/src/test/recovery/t/025_stuck_on_old_timeline.pl
+++ b/src/test/recovery/t/025_stuck_on_old_timeline.pl
@@ -16,7 +16,7 @@ use FindBin;
 use Test::More tests => 1;
 
 # Initialize primary node
-my $node_primary = get_new_node('primary');
+my $node_primary = PostgresNode->new('primary');
 
 # Set up an archive command that will copy the history file but not the WAL
 # files. No real archive command should behave this way; the point is to
@@ -47,7 +47,7 @@ my $backup_name = 'my_backup';
 $node_primary->backup($backup_name);
 
 # Create streaming standby linking to primary
-my $node_standby = get_new_node('standby');
+my $node_standby = PostgresNode->new('standby');
 $node_standby->init_from_backup(
 	$node_primary, $backup_name,
 	allows_streaming => 1,
@@ -60,7 +60,7 @@ $node_standby->backup($backup_name, backup_options => ['-Xnone']);
 
 # Create cascading standby but don't start it yet.
 # Must set up both streaming and archiving.
-my $node_cascade = get_new_node('cascade');
+my $node_cascade = PostgresNode->new('cascade');
 $node_cascade->init_from_backup($node_standby, $backup_name,
 	has_streaming => 1);
 $node_cascade->enable_restoring($node_primary);
-- 
2.25.4


--------------2D62665F86298A3CD7493562
Content-Type: text/x-patch; charset=UTF-8;
 name="0008-fixup-README.patch"
Content-Transfer-Encoding: 7bit
Content-Disposition: attachment;
 filename="0008-fixup-README.patch"



^ permalink  raw  reply  [nested|flat] 17+ messages in thread

* [PATCH v2 19/20] Temporary: Increase BAS_BULKREAD size
@ 2024-09-01 04:42  Andres Freund <[email protected]>
  0 siblings, 0 replies; 17+ messages in thread

From: Andres Freund @ 2024-09-01 04:42 UTC (permalink / raw)

Without this we only can execute very little AIO for sequential scans, as
there's just not enough buffers in the ring.  This isn't the right fix, as
just increasing the ring size can have negative performance implications in
workloads where the kernel has all the data cached.

Author:
Reviewed-By:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/backend/storage/buffer/freelist.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/src/backend/storage/buffer/freelist.c b/src/backend/storage/buffer/freelist.c
index dffdd57e9b5..f5795b509c7 100644
--- a/src/backend/storage/buffer/freelist.c
+++ b/src/backend/storage/buffer/freelist.c
@@ -555,7 +555,12 @@ GetAccessStrategy(BufferAccessStrategyType btype)
 			return NULL;
 
 		case BAS_BULKREAD:
-			ring_size_kb = 256;
+
+			/*
+			 * FIXME: Temporary increase to allow large enough streaming reads
+			 * to actually benefit from AIO. This needs a better solution.
+			 */
+			ring_size_kb = 2 * 1024;
 			break;
 		case BAS_BULKWRITE:
 			ring_size_kb = 16 * 1024;
-- 
2.45.2.746.g06e570c0df.dirty


--oy2jwuii6tssbict
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0020-WIP-Use-MAP_POPULATE.patch"



^ permalink  raw  reply  [nested|flat] 17+ messages in thread

* [PATCH v2.3 28/30] Temporary: Increase BAS_BULKREAD size
@ 2024-09-01 04:42  Andres Freund <[email protected]>
  0 siblings, 0 replies; 17+ messages in thread

From: Andres Freund @ 2024-09-01 04:42 UTC (permalink / raw)

Without this we only can execute very little AIO for sequential scans, as
there's just not enough buffers in the ring.  This isn't the right fix, as
just increasing the ring size can have negative performance implications in
workloads where the kernel has all the data cached.

Author:
Reviewed-By:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/backend/storage/buffer/freelist.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/src/backend/storage/buffer/freelist.c b/src/backend/storage/buffer/freelist.c
index 1f757d96f07..ac19fb87433 100644
--- a/src/backend/storage/buffer/freelist.c
+++ b/src/backend/storage/buffer/freelist.c
@@ -555,7 +555,12 @@ GetAccessStrategy(BufferAccessStrategyType btype)
 			return NULL;
 
 		case BAS_BULKREAD:
-			ring_size_kb = 256;
+
+			/*
+			 * FIXME: Temporary increase to allow large enough streaming reads
+			 * to actually benefit from AIO. This needs a better solution.
+			 */
+			ring_size_kb = 2 * 1024;
 			break;
 		case BAS_BULKWRITE:
 			ring_size_kb = 16 * 1024;
-- 
2.48.1.76.g4e746b1a31.dirty


--gbq4ah2rmhae7qhd
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2.3-0029-WIP-Use-MAP_POPULATE.patch"



^ permalink  raw  reply  [nested|flat] 17+ messages in thread

* [PATCH v2.4 27/29] Temporary: Increase BAS_BULKREAD size
@ 2024-09-01 04:42  Andres Freund <[email protected]>
  0 siblings, 0 replies; 17+ messages in thread

From: Andres Freund @ 2024-09-01 04:42 UTC (permalink / raw)

Without this we only can execute very little AIO for sequential scans, as
there's just not enough buffers in the ring.  This isn't the right fix, as
just increasing the ring size can have negative performance implications in
workloads where the kernel has all the data cached.

Author:
Reviewed-By:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/backend/storage/buffer/freelist.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/src/backend/storage/buffer/freelist.c b/src/backend/storage/buffer/freelist.c
index 336715b6c63..b72a5957a20 100644
--- a/src/backend/storage/buffer/freelist.c
+++ b/src/backend/storage/buffer/freelist.c
@@ -555,7 +555,12 @@ GetAccessStrategy(BufferAccessStrategyType btype)
 			return NULL;
 
 		case BAS_BULKREAD:
-			ring_size_kb = 256;
+
+			/*
+			 * FIXME: Temporary increase to allow large enough streaming reads
+			 * to actually benefit from AIO. This needs a better solution.
+			 */
+			ring_size_kb = 2 * 1024;
 			break;
 		case BAS_BULKWRITE:
 			ring_size_kb = 16 * 1024;
-- 
2.48.1.76.g4e746b1a31.dirty


--4ckqsto27zwk2eqr
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2.4-0028-WIP-Use-MAP_POPULATE.patch"



^ permalink  raw  reply  [nested|flat] 17+ messages in thread

* [PATCH v2.5 28/30] Temporary: Increase BAS_BULKREAD size
@ 2024-09-01 04:42  Andres Freund <[email protected]>
  0 siblings, 0 replies; 17+ messages in thread

From: Andres Freund @ 2024-09-01 04:42 UTC (permalink / raw)

Without this we only can execute very little AIO for sequential scans, as
there's just not enough buffers in the ring.  This isn't the right fix, as
just increasing the ring size can have negative performance implications in
workloads where the kernel has all the data cached.

Author:
Reviewed-By:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/backend/storage/buffer/freelist.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/src/backend/storage/buffer/freelist.c b/src/backend/storage/buffer/freelist.c
index 336715b6c63..b72a5957a20 100644
--- a/src/backend/storage/buffer/freelist.c
+++ b/src/backend/storage/buffer/freelist.c
@@ -555,7 +555,12 @@ GetAccessStrategy(BufferAccessStrategyType btype)
 			return NULL;
 
 		case BAS_BULKREAD:
-			ring_size_kb = 256;
+
+			/*
+			 * FIXME: Temporary increase to allow large enough streaming reads
+			 * to actually benefit from AIO. This needs a better solution.
+			 */
+			ring_size_kb = 2 * 1024;
 			break;
 		case BAS_BULKWRITE:
 			ring_size_kb = 16 * 1024;
-- 
2.48.1.76.g4e746b1a31.dirty


--abvteypvk35ocehs
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2.5-0029-WIP-Use-MAP_POPULATE.patch"



^ permalink  raw  reply  [nested|flat] 17+ messages in thread

* [PATCH v2.6 32/34] Temporary: Increase BAS_BULKREAD size
@ 2024-09-01 04:42  Andres Freund <[email protected]>
  0 siblings, 0 replies; 17+ messages in thread

From: Andres Freund @ 2024-09-01 04:42 UTC (permalink / raw)

Without this we only can execute very little AIO for sequential scans, as
there's just not enough buffers in the ring.  This isn't the right fix, as
just increasing the ring size can have negative performance implications in
workloads where the kernel has all the data cached.

Author:
Reviewed-By:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/backend/storage/buffer/freelist.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/src/backend/storage/buffer/freelist.c b/src/backend/storage/buffer/freelist.c
index 336715b6c63..b72a5957a20 100644
--- a/src/backend/storage/buffer/freelist.c
+++ b/src/backend/storage/buffer/freelist.c
@@ -555,7 +555,12 @@ GetAccessStrategy(BufferAccessStrategyType btype)
 			return NULL;
 
 		case BAS_BULKREAD:
-			ring_size_kb = 256;
+
+			/*
+			 * FIXME: Temporary increase to allow large enough streaming reads
+			 * to actually benefit from AIO. This needs a better solution.
+			 */
+			ring_size_kb = 2 * 1024;
 			break;
 		case BAS_BULKWRITE:
 			ring_size_kb = 16 * 1024;
-- 
2.48.1.76.g4e746b1a31.dirty


--6g5xner6ro2tsnwz
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2.6-0033-WIP-Use-MAP_POPULATE.patch"



^ permalink  raw  reply  [nested|flat] 17+ messages in thread

* [PATCH v2.7 33/35] Temporary: Increase BAS_BULKREAD size
@ 2024-09-01 04:42  Andres Freund <[email protected]>
  0 siblings, 0 replies; 17+ messages in thread

From: Andres Freund @ 2024-09-01 04:42 UTC (permalink / raw)

Without this we only can execute very little AIO for sequential scans, as
there's just not enough buffers in the ring.  This isn't the right fix, as
just increasing the ring size can have negative performance implications in
workloads where the kernel has all the data cached.

Author:
Reviewed-By:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/backend/storage/buffer/freelist.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/src/backend/storage/buffer/freelist.c b/src/backend/storage/buffer/freelist.c
index 336715b6c63..b72a5957a20 100644
--- a/src/backend/storage/buffer/freelist.c
+++ b/src/backend/storage/buffer/freelist.c
@@ -555,7 +555,12 @@ GetAccessStrategy(BufferAccessStrategyType btype)
 			return NULL;
 
 		case BAS_BULKREAD:
-			ring_size_kb = 256;
+
+			/*
+			 * FIXME: Temporary increase to allow large enough streaming reads
+			 * to actually benefit from AIO. This needs a better solution.
+			 */
+			ring_size_kb = 2 * 1024;
 			break;
 		case BAS_BULKWRITE:
 			ring_size_kb = 16 * 1024;
-- 
2.48.1.76.g4e746b1a31.dirty


--vsphh7g5lukufvxs--





^ permalink  raw  reply  [nested|flat] 17+ messages in thread

* [PATCH v2.8 35/38] Temporary: Increase BAS_BULKREAD size
@ 2024-09-01 04:42  Andres Freund <[email protected]>
  0 siblings, 0 replies; 17+ messages in thread

From: Andres Freund @ 2024-09-01 04:42 UTC (permalink / raw)

Without this we only can execute very little AIO for sequential scans, as
there's just not enough buffers in the ring.  This isn't the right fix, as
just increasing the ring size can have negative performance implications in
workloads where the kernel has all the data cached.

Author:
Reviewed-By:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/backend/storage/buffer/freelist.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/src/backend/storage/buffer/freelist.c b/src/backend/storage/buffer/freelist.c
index 336715b6c63..b72a5957a20 100644
--- a/src/backend/storage/buffer/freelist.c
+++ b/src/backend/storage/buffer/freelist.c
@@ -555,7 +555,12 @@ GetAccessStrategy(BufferAccessStrategyType btype)
 			return NULL;
 
 		case BAS_BULKREAD:
-			ring_size_kb = 256;
+
+			/*
+			 * FIXME: Temporary increase to allow large enough streaming reads
+			 * to actually benefit from AIO. This needs a better solution.
+			 */
+			ring_size_kb = 2 * 1024;
 			break;
 		case BAS_BULKWRITE:
 			ring_size_kb = 16 * 1024;
-- 
2.48.1.76.g4e746b1a31.dirty


--ow5flh3n247znjrs
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2.8-0036-WIP-Use-MAP_POPULATE.patch"



^ permalink  raw  reply  [nested|flat] 17+ messages in thread

* [PATCH v2.1 20/20] Temporary: Increase BAS_BULKREAD size
@ 2024-09-01 04:42  Andres Freund <[email protected]>
  0 siblings, 0 replies; 17+ messages in thread

From: Andres Freund @ 2024-09-01 04:42 UTC (permalink / raw)

Without this we only can execute very little AIO for sequential scans, as
there's just not enough buffers in the ring.  This isn't the right fix, as
just increasing the ring size can have negative performance implications in
workloads where the kernel has all the data cached.

Author:
Reviewed-By:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/backend/storage/buffer/freelist.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/src/backend/storage/buffer/freelist.c b/src/backend/storage/buffer/freelist.c
index dffdd57e9b5..5be8125ad3a 100644
--- a/src/backend/storage/buffer/freelist.c
+++ b/src/backend/storage/buffer/freelist.c
@@ -555,7 +555,11 @@ GetAccessStrategy(BufferAccessStrategyType btype)
 			return NULL;
 
 		case BAS_BULKREAD:
-			ring_size_kb = 256;
+			/*
+			 * FIXME: Temporary increase to allow large enough streaming reads
+			 * to actually benefit from AIO. This needs a better solution.
+			 */
+			ring_size_kb = 2 * 1024;
 			break;
 		case BAS_BULKWRITE:
 			ring_size_kb = 16 * 1024;
-- 
2.45.2.827.g557ae147e6


--ggp33qz5xusicvk6--





^ permalink  raw  reply  [nested|flat] 17+ messages in thread

* [PATCH v2.0 17/17] Temporary: Increase BAS_BULKREAD size
@ 2024-09-01 04:42  Andres Freund <[email protected]>
  0 siblings, 0 replies; 17+ messages in thread

From: Andres Freund @ 2024-09-01 04:42 UTC (permalink / raw)

Without this we only can execute very little AIO for sequential scans, as
there's just not enough buffers in the ring.  This isn't the right fix, as
just increasing the ring size can have negative performance implications in
workloads where the kernel has all the data cached.

Author:
Reviewed-By:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/backend/storage/buffer/freelist.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/src/backend/storage/buffer/freelist.c b/src/backend/storage/buffer/freelist.c
index dffdd57e9b5..5be8125ad3a 100644
--- a/src/backend/storage/buffer/freelist.c
+++ b/src/backend/storage/buffer/freelist.c
@@ -555,7 +555,11 @@ GetAccessStrategy(BufferAccessStrategyType btype)
 			return NULL;
 
 		case BAS_BULKREAD:
-			ring_size_kb = 256;
+			/*
+			 * FIXME: Temporary increase to allow large enough streaming reads
+			 * to actually benefit from AIO. This needs a better solution.
+			 */
+			ring_size_kb = 2 * 1024;
 			break;
 		case BAS_BULKWRITE:
 			ring_size_kb = 16 * 1024;
-- 
2.45.2.827.g557ae147e6


--ww2auydviafoh7lh--





^ permalink  raw  reply  [nested|flat] 17+ messages in thread

* [PATCH v2.9 27/30] Temporary: Increase BAS_BULKREAD size
@ 2025-03-15 16:29  Andres Freund <[email protected]>
  0 siblings, 0 replies; 17+ messages in thread

From: Andres Freund @ 2025-03-15 16:29 UTC (permalink / raw)

Without this we only can execute very little AIO for sequential scans, as
there's just not enough buffers in the ring.  This isn't the right fix, as
just increasing the ring size can have negative performance implications in
workloads where the kernel has all the data cached.

Author:
Reviewed-By:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/backend/storage/buffer/freelist.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/src/backend/storage/buffer/freelist.c b/src/backend/storage/buffer/freelist.c
index 336715b6c63..b72a5957a20 100644
--- a/src/backend/storage/buffer/freelist.c
+++ b/src/backend/storage/buffer/freelist.c
@@ -555,7 +555,12 @@ GetAccessStrategy(BufferAccessStrategyType btype)
 			return NULL;
 
 		case BAS_BULKREAD:
-			ring_size_kb = 256;
+
+			/*
+			 * FIXME: Temporary increase to allow large enough streaming reads
+			 * to actually benefit from AIO. This needs a better solution.
+			 */
+			ring_size_kb = 2 * 1024;
 			break;
 		case BAS_BULKWRITE:
 			ring_size_kb = 16 * 1024;
-- 
2.48.1.76.g4e746b1a31.dirty


--23jbdfobqrqxnmx5
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2.9-0028-WIP-Use-MAP_POPULATE.patch"



^ permalink  raw  reply  [nested|flat] 17+ messages in thread

* [PATCH v2.11 25/27] Temporary: Increase BAS_BULKREAD size
@ 2025-03-18 18:40  Andres Freund <[email protected]>
  0 siblings, 0 replies; 17+ messages in thread

From: Andres Freund @ 2025-03-18 18:40 UTC (permalink / raw)

Without this we only can execute very little AIO for sequential scans, as
there's just not enough buffers in the ring.  This isn't the right fix, as
just increasing the ring size can have negative performance implications in
workloads where the kernel has all the data cached.

Author:
Reviewed-By:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/backend/storage/buffer/freelist.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/src/backend/storage/buffer/freelist.c b/src/backend/storage/buffer/freelist.c
index 336715b6c63..b72a5957a20 100644
--- a/src/backend/storage/buffer/freelist.c
+++ b/src/backend/storage/buffer/freelist.c
@@ -555,7 +555,12 @@ GetAccessStrategy(BufferAccessStrategyType btype)
 			return NULL;
 
 		case BAS_BULKREAD:
-			ring_size_kb = 256;
+
+			/*
+			 * FIXME: Temporary increase to allow large enough streaming reads
+			 * to actually benefit from AIO. This needs a better solution.
+			 */
+			ring_size_kb = 2 * 1024;
 			break;
 		case BAS_BULKWRITE:
 			ring_size_kb = 16 * 1024;
-- 
2.48.1.76.g4e746b1a31.dirty


--bjnmbpad43bpmfxt
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2.11-0026-WIP-Use-MAP_POPULATE.patch"



^ permalink  raw  reply  [nested|flat] 17+ messages in thread

* [PATCH v2.12 25/28] Temporary: Increase BAS_BULKREAD size
@ 2025-03-18 18:40  Andres Freund <[email protected]>
  0 siblings, 0 replies; 17+ messages in thread

From: Andres Freund @ 2025-03-18 18:40 UTC (permalink / raw)

Without this we only can execute very little AIO for sequential scans, as
there's just not enough buffers in the ring.  This isn't the right fix, as
just increasing the ring size can have negative performance implications in
workloads where the kernel has all the data cached.

Author:
Reviewed-By:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/backend/storage/buffer/freelist.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/src/backend/storage/buffer/freelist.c b/src/backend/storage/buffer/freelist.c
index 336715b6c63..b72a5957a20 100644
--- a/src/backend/storage/buffer/freelist.c
+++ b/src/backend/storage/buffer/freelist.c
@@ -555,7 +555,12 @@ GetAccessStrategy(BufferAccessStrategyType btype)
 			return NULL;
 
 		case BAS_BULKREAD:
-			ring_size_kb = 256;
+
+			/*
+			 * FIXME: Temporary increase to allow large enough streaming reads
+			 * to actually benefit from AIO. This needs a better solution.
+			 */
+			ring_size_kb = 2 * 1024;
 			break;
 		case BAS_BULKWRITE:
 			ring_size_kb = 16 * 1024;
-- 
2.48.1.76.g4e746b1a31.dirty


--5i73spx2p4vwf7fe
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2.12-0026-WIP-Use-MAP_POPULATE.patch"



^ permalink  raw  reply  [nested|flat] 17+ messages in thread

* [PATCH v2.10 24/28] Temporary: Increase BAS_BULKREAD size
@ 2025-03-18 18:40  Andres Freund <[email protected]>
  0 siblings, 0 replies; 17+ messages in thread

From: Andres Freund @ 2025-03-18 18:40 UTC (permalink / raw)

Without this we only can execute very little AIO for sequential scans, as
there's just not enough buffers in the ring.  This isn't the right fix, as
just increasing the ring size can have negative performance implications in
workloads where the kernel has all the data cached.

Author:
Reviewed-By:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/backend/storage/buffer/freelist.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/src/backend/storage/buffer/freelist.c b/src/backend/storage/buffer/freelist.c
index 336715b6c63..b72a5957a20 100644
--- a/src/backend/storage/buffer/freelist.c
+++ b/src/backend/storage/buffer/freelist.c
@@ -555,7 +555,12 @@ GetAccessStrategy(BufferAccessStrategyType btype)
 			return NULL;
 
 		case BAS_BULKREAD:
-			ring_size_kb = 256;
+
+			/*
+			 * FIXME: Temporary increase to allow large enough streaming reads
+			 * to actually benefit from AIO. This needs a better solution.
+			 */
+			ring_size_kb = 2 * 1024;
 			break;
 		case BAS_BULKWRITE:
 			ring_size_kb = 16 * 1024;
-- 
2.48.1.76.g4e746b1a31.dirty


--w6dfit2y42fwvotd
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2.10-0025-WIP-Use-MAP_POPULATE.patch"



^ permalink  raw  reply  [nested|flat] 17+ messages in thread

* [PATCH v2.14 26/29] Temporary: Increase BAS_BULKREAD size
@ 2025-03-18 18:40  Andres Freund <[email protected]>
  0 siblings, 0 replies; 17+ messages in thread

From: Andres Freund @ 2025-03-18 18:40 UTC (permalink / raw)

Without this we only can execute very little AIO for sequential scans, as
there's just not enough buffers in the ring.  This isn't the right fix, as
just increasing the ring size can have negative performance implications in
workloads where the kernel has all the data cached.

Author:
Reviewed-By:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/backend/storage/buffer/freelist.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/src/backend/storage/buffer/freelist.c b/src/backend/storage/buffer/freelist.c
index 336715b6c63..b72a5957a20 100644
--- a/src/backend/storage/buffer/freelist.c
+++ b/src/backend/storage/buffer/freelist.c
@@ -555,7 +555,12 @@ GetAccessStrategy(BufferAccessStrategyType btype)
 			return NULL;
 
 		case BAS_BULKREAD:
-			ring_size_kb = 256;
+
+			/*
+			 * FIXME: Temporary increase to allow large enough streaming reads
+			 * to actually benefit from AIO. This needs a better solution.
+			 */
+			ring_size_kb = 2 * 1024;
 			break;
 		case BAS_BULKWRITE:
 			ring_size_kb = 16 * 1024;
-- 
2.48.1.76.g4e746b1a31.dirty


--m2na7lgr3zfazgom
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2.14-0027-WIP-Use-MAP_POPULATE.patch"



^ permalink  raw  reply  [nested|flat] 17+ messages in thread

* [PATCH v2.15 15/18] Temporary: Increase BAS_BULKREAD size
@ 2025-03-18 18:40  Andres Freund <[email protected]>
  0 siblings, 0 replies; 17+ messages in thread

From: Andres Freund @ 2025-03-18 18:40 UTC (permalink / raw)

Without this we only can execute very little AIO for sequential scans, as
there's just not enough buffers in the ring.  This isn't the right fix, as
just increasing the ring size can have negative performance implications in
workloads where the kernel has all the data cached.

Author:
Reviewed-By:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/backend/storage/buffer/freelist.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/src/backend/storage/buffer/freelist.c b/src/backend/storage/buffer/freelist.c
index 336715b6c63..b72a5957a20 100644
--- a/src/backend/storage/buffer/freelist.c
+++ b/src/backend/storage/buffer/freelist.c
@@ -555,7 +555,12 @@ GetAccessStrategy(BufferAccessStrategyType btype)
 			return NULL;
 
 		case BAS_BULKREAD:
-			ring_size_kb = 256;
+
+			/*
+			 * FIXME: Temporary increase to allow large enough streaming reads
+			 * to actually benefit from AIO. This needs a better solution.
+			 */
+			ring_size_kb = 2 * 1024;
 			break;
 		case BAS_BULKWRITE:
 			ring_size_kb = 16 * 1024;
-- 
2.48.1.76.g4e746b1a31.dirty


--xevce4sdbnyxplun
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2.15-0016-WIP-Use-MAP_POPULATE.patch"



^ permalink  raw  reply  [nested|flat] 17+ messages in thread

* [PATCH v2.13 25/28] Temporary: Increase BAS_BULKREAD size
@ 2025-03-18 18:40  Andres Freund <[email protected]>
  0 siblings, 0 replies; 17+ messages in thread

From: Andres Freund @ 2025-03-18 18:40 UTC (permalink / raw)

Without this we only can execute very little AIO for sequential scans, as
there's just not enough buffers in the ring.  This isn't the right fix, as
just increasing the ring size can have negative performance implications in
workloads where the kernel has all the data cached.

Author:
Reviewed-By:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/backend/storage/buffer/freelist.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/src/backend/storage/buffer/freelist.c b/src/backend/storage/buffer/freelist.c
index 336715b6c63..b72a5957a20 100644
--- a/src/backend/storage/buffer/freelist.c
+++ b/src/backend/storage/buffer/freelist.c
@@ -555,7 +555,12 @@ GetAccessStrategy(BufferAccessStrategyType btype)
 			return NULL;
 
 		case BAS_BULKREAD:
-			ring_size_kb = 256;
+
+			/*
+			 * FIXME: Temporary increase to allow large enough streaming reads
+			 * to actually benefit from AIO. This needs a better solution.
+			 */
+			ring_size_kb = 2 * 1024;
 			break;
 		case BAS_BULKWRITE:
 			ring_size_kb = 16 * 1024;
-- 
2.48.1.76.g4e746b1a31.dirty


--pro7bqageygxfsvg
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2.13-0026-WIP-Use-MAP_POPULATE.patch"



^ permalink  raw  reply  [nested|flat] 17+ messages in thread

end of thread, other threads:[~2025-03-18 18:40 UTC | newest]

Thread overview: 17+ messages (download: mbox mbox.gz follow: Atom feed)
-- links below jump to the message on this page --
2021-06-13 13:22 [PATCH 7/9] fixup recovery tests Andrew Dunstan <[email protected]>
2024-09-01 04:42 [PATCH v2.7 33/35] Temporary: Increase BAS_BULKREAD size Andres Freund <[email protected]>
2024-09-01 04:42 [PATCH v2.0 17/17] Temporary: Increase BAS_BULKREAD size Andres Freund <[email protected]>
2024-09-01 04:42 [PATCH v2.1 20/20] Temporary: Increase BAS_BULKREAD size Andres Freund <[email protected]>
2024-09-01 04:42 [PATCH v2 19/20] Temporary: Increase BAS_BULKREAD size Andres Freund <[email protected]>
2024-09-01 04:42 [PATCH v2.4 27/29] Temporary: Increase BAS_BULKREAD size Andres Freund <[email protected]>
2024-09-01 04:42 [PATCH v2.5 28/30] Temporary: Increase BAS_BULKREAD size Andres Freund <[email protected]>
2024-09-01 04:42 [PATCH v2.8 35/38] Temporary: Increase BAS_BULKREAD size Andres Freund <[email protected]>
2024-09-01 04:42 [PATCH v2.6 32/34] Temporary: Increase BAS_BULKREAD size Andres Freund <[email protected]>
2024-09-01 04:42 [PATCH v2.3 28/30] Temporary: Increase BAS_BULKREAD size Andres Freund <[email protected]>
2025-03-15 16:29 [PATCH v2.9 27/30] Temporary: Increase BAS_BULKREAD size Andres Freund <[email protected]>
2025-03-18 18:40 [PATCH v2.12 25/28] Temporary: Increase BAS_BULKREAD size Andres Freund <[email protected]>
2025-03-18 18:40 [PATCH v2.15 15/18] Temporary: Increase BAS_BULKREAD size Andres Freund <[email protected]>
2025-03-18 18:40 [PATCH v2.13 25/28] Temporary: Increase BAS_BULKREAD size Andres Freund <[email protected]>
2025-03-18 18:40 [PATCH v2.11 25/27] Temporary: Increase BAS_BULKREAD size Andres Freund <[email protected]>
2025-03-18 18:40 [PATCH v2.14 26/29] Temporary: Increase BAS_BULKREAD size Andres Freund <[email protected]>
2025-03-18 18:40 [PATCH v2.10 24/28] Temporary: Increase BAS_BULKREAD size Andres Freund <[email protected]>

This inbox is served by agora; see mirroring instructions
for how to clone and mirror all data and code used for this inbox