MyRocks: merge the key packing buffer allocation, annotate m_pack_buffer lifetime #1446
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Contributor
Author
|
This resulted from me proving to myself the lifetime of |
Contributor
Author
|
Valgrind testing timed out after a few days, with no errors up to that point (45% of |
|
@luqun has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
luqun
reviewed
Apr 16, 2024
luqun
reviewed
Apr 16, 2024
Contributor
Author
|
Ack, I am waiting for MEM_* macro for ASan test results and will push again
luqun ***@***.***> writes:
… @luqun commented on this pull request.
------------------------------------------------------------------------------------------------------------------------
In storage/rocksdb/ha_rocksdb.cc:
> @@ -9418,48 +9424,57 @@ int ha_rocksdb::alloc_key_buffers(const TABLE &table_arg,
}
}
- m_sk_packed_tuple = reinterpret_cast<uchar *>(
- my_malloc(PSI_NOT_INSTRUMENTED, max_packed_sk_len, MYF(0)));
- m_sk_packed_tuple_old = reinterpret_cast<uchar *>(
- my_malloc(PSI_NOT_INSTRUMENTED, max_packed_sk_len, MYF(0)));
- m_sk_packed_tuple_updated = reinterpret_cast<uchar *>(
- my_malloc(PSI_NOT_INSTRUMENTED, max_packed_sk_len, MYF(0)));
- m_end_key_packed_tuple = reinterpret_cast<uchar *>(
- my_malloc(PSI_NOT_INSTRUMENTED, max_packed_sk_len, MYF(0)));
- m_pack_buffer = reinterpret_cast<uchar *>(
- my_malloc(PSI_NOT_INSTRUMENTED, max_packed_sk_len, MYF(0)));
+ const auto pad1 [[maybe_unused]] = get_next_aligned(buf_size);
my misunderstanding.. the code is doing same as before.
—
Reply to this email directly, view it on GitHub, or unsubscribe.
You are receiving this because you authored the thread.
|
…fer lifetime All the key packing buffers in the MyRocks handler have exactly the same lifetime between alloc_key_buffers and free_key_buffers. Thus allocate a single large buffer and point the key packing buffers to offsets inside it, reducing number of calls to the heap allocator on table opening/close code path. Annotate the padding bytes for Valgrind and AddressSanitizer as inaccessible. Add additional handling for m_pack_buffer: since its use it's fully contained in Rdb_key_def::pack_record, annotate it as such. At the same time extend the memory debugging macros in include/memory_debugging.h to cover AddressSanitizer too in addition to Valgrind.
laurynas-biveinis
force-pushed
the
myr-handler-buffers
branch
from
59f1fb8 to
6df37de
Compare
|
@laurynas-biveinis has updated the pull request. You must reimport the pull request before landing. |
Contributor
Author
|
@luqun , extended the MEM_* macros to cover ASan, no regressions there, & moved the TODO comment. Ready for review |
|
@luqun has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
luqun
reviewed
Apr 19, 2024
|
This pull request has been merged in fdf68f7. |
sunxiayi
pushed a commit
to sunxiayi/mysql-5.6
that referenced
this pull request
Apr 30, 2024
…otate m_pack_buffer lifetime Summary: All the key packing buffers in the MyRocks handler have exactly the same lifetime between alloc_key_buffers and free_key_buffers. Thus allocate a single large buffer and point the key packing buffers to offsets inside it, reducing number of calls to the heap allocator on table opening/close code path. Annotate the padding bytes for Valgrind as inaccessible. Add additional handling for m_pack_buffer: since its use it's fully contained in Rdb_key_def::pack_record, annotate it as such. Pull Request resolved: facebook#1446 GitHub Author: Laurynas Biveinis <laurynas.biveinis@gmail.com> Test Plan: Imported from GitHub, without a `Test Plan:` line. Reviewers: herman, chni, #mysql_eng Reviewed By: chni Differential Revision: https://phabricator.intern.facebook.com/D56051395
inikep
pushed a commit
to inikep/mysql-5.6
that referenced
this pull request
May 17, 2024
…fer lifetime (facebook#1446) Summary: All the key packing buffers in the MyRocks handler have exactly the same lifetime between alloc_key_buffers and free_key_buffers. Thus allocate a single large buffer and point the key packing buffers to offsets inside it, reducing number of calls to the heap allocator on table opening/close code path. Annotate the padding bytes for Valgrind as inaccessible. Add additional handling for m_pack_buffer: since its use it's fully contained in Rdb_key_def::pack_record, annotate it as such. Pull Request resolved: facebook#1446 Differential Revision: D56051395 fbshipit-source-id: f503d56
inikep
pushed a commit
to inikep/mysql-5.6
that referenced
this pull request
May 17, 2024
…fer lifetime (facebook#1446) Summary: All the key packing buffers in the MyRocks handler have exactly the same lifetime between alloc_key_buffers and free_key_buffers. Thus allocate a single large buffer and point the key packing buffers to offsets inside it, reducing number of calls to the heap allocator on table opening/close code path. Annotate the padding bytes for Valgrind as inaccessible. Add additional handling for m_pack_buffer: since its use it's fully contained in Rdb_key_def::pack_record, annotate it as such. Pull Request resolved: facebook#1446 Differential Revision: D56051395 fbshipit-source-id: f503d56
inikep
pushed a commit
to inikep/mysql-5.6
that referenced
this pull request
May 21, 2024
…fer lifetime (facebook#1446) Summary: All the key packing buffers in the MyRocks handler have exactly the same lifetime between alloc_key_buffers and free_key_buffers. Thus allocate a single large buffer and point the key packing buffers to offsets inside it, reducing number of calls to the heap allocator on table opening/close code path. Annotate the padding bytes for Valgrind as inaccessible. Add additional handling for m_pack_buffer: since its use it's fully contained in Rdb_key_def::pack_record, annotate it as such. Pull Request resolved: facebook#1446 Differential Revision: D56051395 fbshipit-source-id: f503d56
inikep
pushed a commit
to inikep/mysql-5.6
that referenced
this pull request
May 21, 2024
…fer lifetime (facebook#1446) Summary: All the key packing buffers in the MyRocks handler have exactly the same lifetime between alloc_key_buffers and free_key_buffers. Thus allocate a single large buffer and point the key packing buffers to offsets inside it, reducing number of calls to the heap allocator on table opening/close code path. Annotate the padding bytes for Valgrind as inaccessible. Add additional handling for m_pack_buffer: since its use it's fully contained in Rdb_key_def::pack_record, annotate it as such. Pull Request resolved: facebook#1446 Differential Revision: D56051395 fbshipit-source-id: f503d56
inikep
pushed a commit
to inikep/mysql-5.6
that referenced
this pull request
May 21, 2024
…fer lifetime (facebook#1446) Summary: All the key packing buffers in the MyRocks handler have exactly the same lifetime between alloc_key_buffers and free_key_buffers. Thus allocate a single large buffer and point the key packing buffers to offsets inside it, reducing number of calls to the heap allocator on table opening/close code path. Annotate the padding bytes for Valgrind as inaccessible. Add additional handling for m_pack_buffer: since its use it's fully contained in Rdb_key_def::pack_record, annotate it as such. Pull Request resolved: facebook#1446 Differential Revision: D56051395 fbshipit-source-id: f503d56
inikep
pushed a commit
to inikep/mysql-5.6
that referenced
this pull request
May 30, 2024
…fer lifetime (facebook#1446) Summary: All the key packing buffers in the MyRocks handler have exactly the same lifetime between alloc_key_buffers and free_key_buffers. Thus allocate a single large buffer and point the key packing buffers to offsets inside it, reducing number of calls to the heap allocator on table opening/close code path. Annotate the padding bytes for Valgrind as inaccessible. Add additional handling for m_pack_buffer: since its use it's fully contained in Rdb_key_def::pack_record, annotate it as such. Pull Request resolved: facebook#1446 Differential Revision: D56051395 fbshipit-source-id: f503d56
inikep
pushed a commit
to inikep/mysql-5.6
that referenced
this pull request
Jul 29, 2024
…fer lifetime (facebook#1446) Summary: All the key packing buffers in the MyRocks handler have exactly the same lifetime between alloc_key_buffers and free_key_buffers. Thus allocate a single large buffer and point the key packing buffers to offsets inside it, reducing number of calls to the heap allocator on table opening/close code path. Annotate the padding bytes for Valgrind as inaccessible. Add additional handling for m_pack_buffer: since its use it's fully contained in Rdb_key_def::pack_record, annotate it as such. Pull Request resolved: facebook#1446 Differential Revision: D56051395 fbshipit-source-id: f503d56
inikep
pushed a commit
to inikep/mysql-5.6
that referenced
this pull request
Jul 30, 2024
…fer lifetime (facebook#1446) Summary: All the key packing buffers in the MyRocks handler have exactly the same lifetime between alloc_key_buffers and free_key_buffers. Thus allocate a single large buffer and point the key packing buffers to offsets inside it, reducing number of calls to the heap allocator on table opening/close code path. Annotate the padding bytes for Valgrind as inaccessible. Add additional handling for m_pack_buffer: since its use it's fully contained in Rdb_key_def::pack_record, annotate it as such. Pull Request resolved: facebook#1446 Differential Revision: D56051395 fbshipit-source-id: f503d56
inikep
pushed a commit
to inikep/mysql-5.6
that referenced
this pull request
Jul 31, 2024
…fer lifetime (facebook#1446) Summary: All the key packing buffers in the MyRocks handler have exactly the same lifetime between alloc_key_buffers and free_key_buffers. Thus allocate a single large buffer and point the key packing buffers to offsets inside it, reducing number of calls to the heap allocator on table opening/close code path. Annotate the padding bytes for Valgrind as inaccessible. Add additional handling for m_pack_buffer: since its use it's fully contained in Rdb_key_def::pack_record, annotate it as such. Pull Request resolved: facebook#1446 Differential Revision: D56051395 fbshipit-source-id: f503d56
inikep
pushed a commit
to inikep/mysql-5.6
that referenced
this pull request
Aug 2, 2024
…fer lifetime (facebook#1446) Summary: All the key packing buffers in the MyRocks handler have exactly the same lifetime between alloc_key_buffers and free_key_buffers. Thus allocate a single large buffer and point the key packing buffers to offsets inside it, reducing number of calls to the heap allocator on table opening/close code path. Annotate the padding bytes for Valgrind as inaccessible. Add additional handling for m_pack_buffer: since its use it's fully contained in Rdb_key_def::pack_record, annotate it as such. Pull Request resolved: facebook#1446 Differential Revision: D56051395 fbshipit-source-id: f503d56
inikep
pushed a commit
to inikep/mysql-5.6
that referenced
this pull request
Aug 6, 2024
…fer lifetime (facebook#1446) Summary: All the key packing buffers in the MyRocks handler have exactly the same lifetime between alloc_key_buffers and free_key_buffers. Thus allocate a single large buffer and point the key packing buffers to offsets inside it, reducing number of calls to the heap allocator on table opening/close code path. Annotate the padding bytes for Valgrind as inaccessible. Add additional handling for m_pack_buffer: since its use it's fully contained in Rdb_key_def::pack_record, annotate it as such. Pull Request resolved: facebook#1446 Differential Revision: D56051395 fbshipit-source-id: f503d56
inikep
pushed a commit
to inikep/percona-server
that referenced
this pull request
Apr 16, 2025
…fer lifetime (percona#1446) Upstream commit ID: facebook/mysql-5.6@fdf68f720c84 PS-9395: Merge percona-202404 (https://jira.percona.com/browse/PS-9395) Summary: All the key packing buffers in the MyRocks handler have exactly the same lifetime between alloc_key_buffers and free_key_buffers. Thus allocate a single large buffer and point the key packing buffers to offsets inside it, reducing number of calls to the heap allocator on table opening/close code path. Annotate the padding bytes for Valgrind as inaccessible. Add additional handling for m_pack_buffer: since its use it's fully contained in Rdb_key_def::pack_record, annotate it as such. Pull Request resolved: facebook/mysql-5.6#1446 Differential Revision: D56051395 fbshipit-source-id: f503d56efbaee8a2072fe5701eec4783d7c1cce9
inikep
pushed a commit
to inikep/percona-server
that referenced
this pull request
Apr 16, 2025
…fer lifetime (percona#1446) Upstream commit ID: facebook/mysql-5.6@fdf68f720c84 PS-9395: Merge percona-202404 (https://jira.percona.com/browse/PS-9395) Summary: All the key packing buffers in the MyRocks handler have exactly the same lifetime between alloc_key_buffers and free_key_buffers. Thus allocate a single large buffer and point the key packing buffers to offsets inside it, reducing number of calls to the heap allocator on table opening/close code path. Annotate the padding bytes for Valgrind as inaccessible. Add additional handling for m_pack_buffer: since its use it's fully contained in Rdb_key_def::pack_record, annotate it as such. Pull Request resolved: facebook/mysql-5.6#1446 Differential Revision: D56051395 fbshipit-source-id: f503d56efbaee8a2072fe5701eec4783d7c1cce9
percona-ysorokin
pushed a commit
to percona-ysorokin/percona-server
that referenced
this pull request
Apr 18, 2025
…fer lifetime (percona#1446) Upstream commit ID: facebook/mysql-5.6@fdf68f720c84 PS-9395: Merge percona-202404 (https://jira.percona.com/browse/PS-9395) Summary: All the key packing buffers in the MyRocks handler have exactly the same lifetime between alloc_key_buffers and free_key_buffers. Thus allocate a single large buffer and point the key packing buffers to offsets inside it, reducing number of calls to the heap allocator on table opening/close code path. Annotate the padding bytes for Valgrind as inaccessible. Add additional handling for m_pack_buffer: since its use it's fully contained in Rdb_key_def::pack_record, annotate it as such. Pull Request resolved: facebook/mysql-5.6#1446 Differential Revision: D56051395 fbshipit-source-id: f503d56efbaee8a2072fe5701eec4783d7c1cce9
lukin-oleksiy
pushed a commit
to lukin-oleksiy/percona-server
that referenced
this pull request
Apr 28, 2025
…fer lifetime (percona#1446) Upstream commit ID: facebook/mysql-5.6@fdf68f720c84 PS-9395: Merge percona-202404 (https://jira.percona.com/browse/PS-9395) Summary: All the key packing buffers in the MyRocks handler have exactly the same lifetime between alloc_key_buffers and free_key_buffers. Thus allocate a single large buffer and point the key packing buffers to offsets inside it, reducing number of calls to the heap allocator on table opening/close code path. Annotate the padding bytes for Valgrind as inaccessible. Add additional handling for m_pack_buffer: since its use it's fully contained in Rdb_key_def::pack_record, annotate it as such. Pull Request resolved: facebook/mysql-5.6#1446 Differential Revision: D56051395 fbshipit-source-id: f503d56efbaee8a2072fe5701eec4783d7c1cce9
lukin-oleksiy
pushed a commit
to lukin-oleksiy/percona-server
that referenced
this pull request
Apr 28, 2025
…fer lifetime (percona#1446) Upstream commit ID: facebook/mysql-5.6@fdf68f720c84 PS-9395: Merge percona-202404 (https://jira.percona.com/browse/PS-9395) Summary: All the key packing buffers in the MyRocks handler have exactly the same lifetime between alloc_key_buffers and free_key_buffers. Thus allocate a single large buffer and point the key packing buffers to offsets inside it, reducing number of calls to the heap allocator on table opening/close code path. Annotate the padding bytes for Valgrind as inaccessible. Add additional handling for m_pack_buffer: since its use it's fully contained in Rdb_key_def::pack_record, annotate it as such. Pull Request resolved: facebook/mysql-5.6#1446 Differential Revision: D56051395 fbshipit-source-id: f503d56efbaee8a2072fe5701eec4783d7c1cce9
lukin-oleksiy
pushed a commit
to lukin-oleksiy/percona-server
that referenced
this pull request
Apr 28, 2025
…fer lifetime (percona#1446) Upstream commit ID: facebook/mysql-5.6@fdf68f720c84 PS-9395: Merge percona-202404 (https://jira.percona.com/browse/PS-9395) Summary: All the key packing buffers in the MyRocks handler have exactly the same lifetime between alloc_key_buffers and free_key_buffers. Thus allocate a single large buffer and point the key packing buffers to offsets inside it, reducing number of calls to the heap allocator on table opening/close code path. Annotate the padding bytes for Valgrind as inaccessible. Add additional handling for m_pack_buffer: since its use it's fully contained in Rdb_key_def::pack_record, annotate it as such. Pull Request resolved: facebook/mysql-5.6#1446 Differential Revision: D56051395 fbshipit-source-id: f503d56efbaee8a2072fe5701eec4783d7c1cce9
lukin-oleksiy
pushed a commit
to lukin-oleksiy/percona-server
that referenced
this pull request
Apr 30, 2025
…fer lifetime (percona#1446) Upstream commit ID: facebook/mysql-5.6@fdf68f720c84 PS-9395: Merge percona-202404 (https://jira.percona.com/browse/PS-9395) Summary: All the key packing buffers in the MyRocks handler have exactly the same lifetime between alloc_key_buffers and free_key_buffers. Thus allocate a single large buffer and point the key packing buffers to offsets inside it, reducing number of calls to the heap allocator on table opening/close code path. Annotate the padding bytes for Valgrind as inaccessible. Add additional handling for m_pack_buffer: since its use it's fully contained in Rdb_key_def::pack_record, annotate it as such. Pull Request resolved: facebook/mysql-5.6#1446 Differential Revision: D56051395 fbshipit-source-id: f503d56efbaee8a2072fe5701eec4783d7c1cce9
lukin-oleksiy
pushed a commit
to lukin-oleksiy/percona-server
that referenced
this pull request
May 4, 2025
…fer lifetime (percona#1446) Upstream commit ID: facebook/mysql-5.6@fdf68f720c84 PS-9395: Merge percona-202404 (https://jira.percona.com/browse/PS-9395) Summary: All the key packing buffers in the MyRocks handler have exactly the same lifetime between alloc_key_buffers and free_key_buffers. Thus allocate a single large buffer and point the key packing buffers to offsets inside it, reducing number of calls to the heap allocator on table opening/close code path. Annotate the padding bytes for Valgrind as inaccessible. Add additional handling for m_pack_buffer: since its use it's fully contained in Rdb_key_def::pack_record, annotate it as such. Pull Request resolved: facebook/mysql-5.6#1446 Differential Revision: D56051395 fbshipit-source-id: f503d56efbaee8a2072fe5701eec4783d7c1cce9
inikep
pushed a commit
to percona/percona-server
that referenced
this pull request
Jun 4, 2025
…fer lifetime (#1446) Upstream commit ID: facebook/mysql-5.6@fdf68f720c84 PS-9395: Merge percona-202404 (https://jira.percona.com/browse/PS-9395) Summary: All the key packing buffers in the MyRocks handler have exactly the same lifetime between alloc_key_buffers and free_key_buffers. Thus allocate a single large buffer and point the key packing buffers to offsets inside it, reducing number of calls to the heap allocator on table opening/close code path. Annotate the padding bytes for Valgrind as inaccessible. Add additional handling for m_pack_buffer: since its use it's fully contained in Rdb_key_def::pack_record, annotate it as such. Pull Request resolved: facebook/mysql-5.6#1446 Differential Revision: D56051395 fbshipit-source-id: f503d56efbaee8a2072fe5701eec4783d7c1cce9
inikep
pushed a commit
to inikep/percona-server
that referenced
this pull request
Jun 4, 2025
…fer lifetime (percona#1446) Upstream commit ID: facebook/mysql-5.6@fdf68f720c84 PS-9395: Merge percona-202404 (https://jira.percona.com/browse/PS-9395) Summary: All the key packing buffers in the MyRocks handler have exactly the same lifetime between alloc_key_buffers and free_key_buffers. Thus allocate a single large buffer and point the key packing buffers to offsets inside it, reducing number of calls to the heap allocator on table opening/close code path. Annotate the padding bytes for Valgrind as inaccessible. Add additional handling for m_pack_buffer: since its use it's fully contained in Rdb_key_def::pack_record, annotate it as such. Pull Request resolved: facebook/mysql-5.6#1446 Differential Revision: D56051395 fbshipit-source-id: f503d56efbaee8a2072fe5701eec4783d7c1cce9
inikep
pushed a commit
to inikep/percona-server
that referenced
this pull request
Jun 5, 2025
…fer lifetime (percona#1446) Upstream commit ID: facebook/mysql-5.6@fdf68f720c84 PS-9395: Merge percona-202404 (https://jira.percona.com/browse/PS-9395) Summary: All the key packing buffers in the MyRocks handler have exactly the same lifetime between alloc_key_buffers and free_key_buffers. Thus allocate a single large buffer and point the key packing buffers to offsets inside it, reducing number of calls to the heap allocator on table opening/close code path. Annotate the padding bytes for Valgrind as inaccessible. Add additional handling for m_pack_buffer: since its use it's fully contained in Rdb_key_def::pack_record, annotate it as such. Pull Request resolved: facebook/mysql-5.6#1446 Differential Revision: D56051395 fbshipit-source-id: f503d56efbaee8a2072fe5701eec4783d7c1cce9
inikep
pushed a commit
to inikep/percona-server
that referenced
this pull request
Jun 5, 2025
…fer lifetime (percona#1446) Upstream commit ID: facebook/mysql-5.6@fdf68f720c84 PS-9395: Merge percona-202404 (https://jira.percona.com/browse/PS-9395) Summary: All the key packing buffers in the MyRocks handler have exactly the same lifetime between alloc_key_buffers and free_key_buffers. Thus allocate a single large buffer and point the key packing buffers to offsets inside it, reducing number of calls to the heap allocator on table opening/close code path. Annotate the padding bytes for Valgrind as inaccessible. Add additional handling for m_pack_buffer: since its use it's fully contained in Rdb_key_def::pack_record, annotate it as such. Pull Request resolved: facebook/mysql-5.6#1446 Differential Revision: D56051395 fbshipit-source-id: f503d56efbaee8a2072fe5701eec4783d7c1cce9
inikep
pushed a commit
to inikep/percona-server
that referenced
this pull request
Jun 5, 2025
…fer lifetime (percona#1446) Upstream commit ID: facebook/mysql-5.6@fdf68f720c84 PS-9395: Merge percona-202404 (https://jira.percona.com/browse/PS-9395) Summary: All the key packing buffers in the MyRocks handler have exactly the same lifetime between alloc_key_buffers and free_key_buffers. Thus allocate a single large buffer and point the key packing buffers to offsets inside it, reducing number of calls to the heap allocator on table opening/close code path. Annotate the padding bytes for Valgrind as inaccessible. Add additional handling for m_pack_buffer: since its use it's fully contained in Rdb_key_def::pack_record, annotate it as such. Pull Request resolved: facebook/mysql-5.6#1446 Differential Revision: D56051395 fbshipit-source-id: f503d56efbaee8a2072fe5701eec4783d7c1cce9
lukin-oleksiy
pushed a commit
to lukin-oleksiy/percona-server
that referenced
this pull request
Jul 31, 2025
…fer lifetime (percona#1446) Upstream commit ID: facebook/mysql-5.6@fdf68f720c84 PS-9395: Merge percona-202404 (https://jira.percona.com/browse/PS-9395) Summary: All the key packing buffers in the MyRocks handler have exactly the same lifetime between alloc_key_buffers and free_key_buffers. Thus allocate a single large buffer and point the key packing buffers to offsets inside it, reducing number of calls to the heap allocator on table opening/close code path. Annotate the padding bytes for Valgrind as inaccessible. Add additional handling for m_pack_buffer: since its use it's fully contained in Rdb_key_def::pack_record, annotate it as such. Pull Request resolved: facebook/mysql-5.6#1446 Differential Revision: D56051395 fbshipit-source-id: f503d56efbaee8a2072fe5701eec4783d7c1cce9
percona-ysorokin
pushed a commit
to percona-ysorokin/percona-server
that referenced
this pull request
Jul 31, 2025
…fer lifetime (percona#1446) Upstream commit ID: facebook/mysql-5.6@fdf68f720c84 PS-9395: Merge percona-202404 (https://jira.percona.com/browse/PS-9395) Summary: All the key packing buffers in the MyRocks handler have exactly the same lifetime between alloc_key_buffers and free_key_buffers. Thus allocate a single large buffer and point the key packing buffers to offsets inside it, reducing number of calls to the heap allocator on table opening/close code path. Annotate the padding bytes for Valgrind as inaccessible. Add additional handling for m_pack_buffer: since its use it's fully contained in Rdb_key_def::pack_record, annotate it as such. Pull Request resolved: facebook/mysql-5.6#1446 Differential Revision: D56051395 fbshipit-source-id: f503d56efbaee8a2072fe5701eec4783d7c1cce9
lukin-oleksiy
pushed a commit
to lukin-oleksiy/percona-server
that referenced
this pull request
Jul 31, 2025
…fer lifetime (percona#1446) Upstream commit ID: facebook/mysql-5.6@fdf68f720c84 PS-9395: Merge percona-202404 (https://jira.percona.com/browse/PS-9395) Summary: All the key packing buffers in the MyRocks handler have exactly the same lifetime between alloc_key_buffers and free_key_buffers. Thus allocate a single large buffer and point the key packing buffers to offsets inside it, reducing number of calls to the heap allocator on table opening/close code path. Annotate the padding bytes for Valgrind as inaccessible. Add additional handling for m_pack_buffer: since its use it's fully contained in Rdb_key_def::pack_record, annotate it as such. Pull Request resolved: facebook/mysql-5.6#1446 Differential Revision: D56051395 fbshipit-source-id: f503d56efbaee8a2072fe5701eec4783d7c1cce9
inikep
pushed a commit
to inikep/percona-server
that referenced
this pull request
Jul 31, 2025
…fer lifetime (percona#1446) Upstream commit ID: facebook/mysql-5.6@fdf68f720c84 PS-9395: Merge percona-202404 (https://jira.percona.com/browse/PS-9395) Summary: All the key packing buffers in the MyRocks handler have exactly the same lifetime between alloc_key_buffers and free_key_buffers. Thus allocate a single large buffer and point the key packing buffers to offsets inside it, reducing number of calls to the heap allocator on table opening/close code path. Annotate the padding bytes for Valgrind as inaccessible. Add additional handling for m_pack_buffer: since its use it's fully contained in Rdb_key_def::pack_record, annotate it as such. Pull Request resolved: facebook/mysql-5.6#1446 Differential Revision: D56051395 fbshipit-source-id: f503d56efbaee8a2072fe5701eec4783d7c1cce9
inikep
pushed a commit
to inikep/percona-server
that referenced
this pull request
Jul 31, 2025
…fer lifetime (percona#1446) Upstream commit ID: facebook/mysql-5.6@fdf68f720c84 PS-9395: Merge percona-202404 (https://jira.percona.com/browse/PS-9395) Summary: All the key packing buffers in the MyRocks handler have exactly the same lifetime between alloc_key_buffers and free_key_buffers. Thus allocate a single large buffer and point the key packing buffers to offsets inside it, reducing number of calls to the heap allocator on table opening/close code path. Annotate the padding bytes for Valgrind as inaccessible. Add additional handling for m_pack_buffer: since its use it's fully contained in Rdb_key_def::pack_record, annotate it as such. Pull Request resolved: facebook/mysql-5.6#1446 Differential Revision: D56051395 fbshipit-source-id: f503d56efbaee8a2072fe5701eec4783d7c1cce9
percona-ysorokin
pushed a commit
to percona-ysorokin/percona-server
that referenced
this pull request
Aug 2, 2025
…fer lifetime (percona#1446) Upstream commit ID: facebook/mysql-5.6@fdf68f720c84 PS-9395: Merge percona-202404 (https://jira.percona.com/browse/PS-9395) Summary: All the key packing buffers in the MyRocks handler have exactly the same lifetime between alloc_key_buffers and free_key_buffers. Thus allocate a single large buffer and point the key packing buffers to offsets inside it, reducing number of calls to the heap allocator on table opening/close code path. Annotate the padding bytes for Valgrind as inaccessible. Add additional handling for m_pack_buffer: since its use it's fully contained in Rdb_key_def::pack_record, annotate it as such. Pull Request resolved: facebook/mysql-5.6#1446 Differential Revision: D56051395 fbshipit-source-id: f503d56efbaee8a2072fe5701eec4783d7c1cce9
lukin-oleksiy
pushed a commit
to lukin-oleksiy/percona-server
that referenced
this pull request
Aug 5, 2025
…fer lifetime (percona#1446) Upstream commit ID: facebook/mysql-5.6@fdf68f720c84 PS-9395: Merge percona-202404 (https://jira.percona.com/browse/PS-9395) Summary: All the key packing buffers in the MyRocks handler have exactly the same lifetime between alloc_key_buffers and free_key_buffers. Thus allocate a single large buffer and point the key packing buffers to offsets inside it, reducing number of calls to the heap allocator on table opening/close code path. Annotate the padding bytes for Valgrind as inaccessible. Add additional handling for m_pack_buffer: since its use it's fully contained in Rdb_key_def::pack_record, annotate it as such. Pull Request resolved: facebook/mysql-5.6#1446 Differential Revision: D56051395 fbshipit-source-id: f503d56efbaee8a2072fe5701eec4783d7c1cce9
inikep
pushed a commit
to inikep/percona-server
that referenced
this pull request
Oct 22, 2025
…fer lifetime (percona#1446) Upstream commit ID: facebook/mysql-5.6@fdf68f720c84 PS-9395: Merge percona-202404 (https://jira.percona.com/browse/PS-9395) Summary: All the key packing buffers in the MyRocks handler have exactly the same lifetime between alloc_key_buffers and free_key_buffers. Thus allocate a single large buffer and point the key packing buffers to offsets inside it, reducing number of calls to the heap allocator on table opening/close code path. Annotate the padding bytes for Valgrind as inaccessible. Add additional handling for m_pack_buffer: since its use it's fully contained in Rdb_key_def::pack_record, annotate it as such. Pull Request resolved: facebook/mysql-5.6#1446 Differential Revision: D56051395 fbshipit-source-id: f503d56efbaee8a2072fe5701eec4783d7c1cce9
inikep
pushed a commit
to inikep/percona-server
that referenced
this pull request
Oct 27, 2025
…fer lifetime (percona#1446) Upstream commit ID: facebook/mysql-5.6@fdf68f720c84 PS-9395: Merge percona-202404 (https://jira.percona.com/browse/PS-9395) Summary: All the key packing buffers in the MyRocks handler have exactly the same lifetime between alloc_key_buffers and free_key_buffers. Thus allocate a single large buffer and point the key packing buffers to offsets inside it, reducing number of calls to the heap allocator on table opening/close code path. Annotate the padding bytes for Valgrind as inaccessible. Add additional handling for m_pack_buffer: since its use it's fully contained in Rdb_key_def::pack_record, annotate it as such. Pull Request resolved: facebook/mysql-5.6#1446 Differential Revision: D56051395 fbshipit-source-id: f503d56efbaee8a2072fe5701eec4783d7c1cce9
dlenev
pushed a commit
to dlenev/percona-server
that referenced
this pull request
Oct 30, 2025
…fer lifetime (percona#1446) Upstream commit ID: facebook/mysql-5.6@fdf68f720c84 PS-9395: Merge percona-202404 (https://jira.percona.com/browse/PS-9395) Summary: All the key packing buffers in the MyRocks handler have exactly the same lifetime between alloc_key_buffers and free_key_buffers. Thus allocate a single large buffer and point the key packing buffers to offsets inside it, reducing number of calls to the heap allocator on table opening/close code path. Annotate the padding bytes for Valgrind as inaccessible. Add additional handling for m_pack_buffer: since its use it's fully contained in Rdb_key_def::pack_record, annotate it as such. Pull Request resolved: facebook/mysql-5.6#1446 Differential Revision: D56051395 fbshipit-source-id: f503d56efbaee8a2072fe5701eec4783d7c1cce9
lukin-oleksiy
pushed a commit
to lukin-oleksiy/percona-server
that referenced
this pull request
Nov 4, 2025
…fer lifetime (percona#1446) Upstream commit ID: facebook/mysql-5.6@fdf68f720c84 PS-9395: Merge percona-202404 (https://jira.percona.com/browse/PS-9395) Summary: All the key packing buffers in the MyRocks handler have exactly the same lifetime between alloc_key_buffers and free_key_buffers. Thus allocate a single large buffer and point the key packing buffers to offsets inside it, reducing number of calls to the heap allocator on table opening/close code path. Annotate the padding bytes for Valgrind as inaccessible. Add additional handling for m_pack_buffer: since its use it's fully contained in Rdb_key_def::pack_record, annotate it as such. Pull Request resolved: facebook/mysql-5.6#1446 Differential Revision: D56051395 fbshipit-source-id: f503d56efbaee8a2072fe5701eec4783d7c1cce9
lukin-oleksiy
pushed a commit
to lukin-oleksiy/percona-server
that referenced
this pull request
Nov 4, 2025
…fer lifetime (percona#1446) Upstream commit ID: facebook/mysql-5.6@fdf68f720c84 PS-9395: Merge percona-202404 (https://jira.percona.com/browse/PS-9395) Summary: All the key packing buffers in the MyRocks handler have exactly the same lifetime between alloc_key_buffers and free_key_buffers. Thus allocate a single large buffer and point the key packing buffers to offsets inside it, reducing number of calls to the heap allocator on table opening/close code path. Annotate the padding bytes for Valgrind as inaccessible. Add additional handling for m_pack_buffer: since its use it's fully contained in Rdb_key_def::pack_record, annotate it as such. Pull Request resolved: facebook/mysql-5.6#1446 Differential Revision: D56051395 fbshipit-source-id: f503d56efbaee8a2072fe5701eec4783d7c1cce9
lukin-oleksiy
pushed a commit
to lukin-oleksiy/percona-server
that referenced
this pull request
Nov 5, 2025
…fer lifetime (percona#1446) Upstream commit ID: facebook/mysql-5.6@fdf68f720c84 PS-9395: Merge percona-202404 (https://jira.percona.com/browse/PS-9395) Summary: All the key packing buffers in the MyRocks handler have exactly the same lifetime between alloc_key_buffers and free_key_buffers. Thus allocate a single large buffer and point the key packing buffers to offsets inside it, reducing number of calls to the heap allocator on table opening/close code path. Annotate the padding bytes for Valgrind as inaccessible. Add additional handling for m_pack_buffer: since its use it's fully contained in Rdb_key_def::pack_record, annotate it as such. Pull Request resolved: facebook/mysql-5.6#1446 Differential Revision: D56051395 fbshipit-source-id: f503d56efbaee8a2072fe5701eec4783d7c1cce9
lukin-oleksiy
pushed a commit
to lukin-oleksiy/percona-server
that referenced
this pull request
Nov 5, 2025
…fer lifetime (percona#1446) Upstream commit ID: facebook/mysql-5.6@fdf68f720c84 PS-9395: Merge percona-202404 (https://jira.percona.com/browse/PS-9395) Summary: All the key packing buffers in the MyRocks handler have exactly the same lifetime between alloc_key_buffers and free_key_buffers. Thus allocate a single large buffer and point the key packing buffers to offsets inside it, reducing number of calls to the heap allocator on table opening/close code path. Annotate the padding bytes for Valgrind as inaccessible. Add additional handling for m_pack_buffer: since its use it's fully contained in Rdb_key_def::pack_record, annotate it as such. Pull Request resolved: facebook/mysql-5.6#1446 Differential Revision: D56051395 fbshipit-source-id: f503d56efbaee8a2072fe5701eec4783d7c1cce9
lukin-oleksiy
pushed a commit
to lukin-oleksiy/percona-server
that referenced
this pull request
Nov 6, 2025
…fer lifetime (percona#1446) Upstream commit ID: facebook/mysql-5.6@fdf68f720c84 PS-9395: Merge percona-202404 (https://jira.percona.com/browse/PS-9395) Summary: All the key packing buffers in the MyRocks handler have exactly the same lifetime between alloc_key_buffers and free_key_buffers. Thus allocate a single large buffer and point the key packing buffers to offsets inside it, reducing number of calls to the heap allocator on table opening/close code path. Annotate the padding bytes for Valgrind as inaccessible. Add additional handling for m_pack_buffer: since its use it's fully contained in Rdb_key_def::pack_record, annotate it as such. Pull Request resolved: facebook/mysql-5.6#1446 Differential Revision: D56051395 fbshipit-source-id: f503d56efbaee8a2072fe5701eec4783d7c1cce9
dlenev
pushed a commit
to dlenev/percona-server
that referenced
this pull request
Nov 7, 2025
…fer lifetime (percona#1446) Upstream commit ID: facebook/mysql-5.6@fdf68f720c84 PS-9395: Merge percona-202404 (https://jira.percona.com/browse/PS-9395) Summary: All the key packing buffers in the MyRocks handler have exactly the same lifetime between alloc_key_buffers and free_key_buffers. Thus allocate a single large buffer and point the key packing buffers to offsets inside it, reducing number of calls to the heap allocator on table opening/close code path. Annotate the padding bytes for Valgrind as inaccessible. Add additional handling for m_pack_buffer: since its use it's fully contained in Rdb_key_def::pack_record, annotate it as such. Pull Request resolved: facebook/mysql-5.6#1446 Differential Revision: D56051395 fbshipit-source-id: f503d56efbaee8a2072fe5701eec4783d7c1cce9
lukin-oleksiy
pushed a commit
to lukin-oleksiy/percona-server
that referenced
this pull request
Nov 11, 2025
…fer lifetime (percona#1446) Upstream commit ID: facebook/mysql-5.6@fdf68f720c84 PS-9395: Merge percona-202404 (https://jira.percona.com/browse/PS-9395) Summary: All the key packing buffers in the MyRocks handler have exactly the same lifetime between alloc_key_buffers and free_key_buffers. Thus allocate a single large buffer and point the key packing buffers to offsets inside it, reducing number of calls to the heap allocator on table opening/close code path. Annotate the padding bytes for Valgrind as inaccessible. Add additional handling for m_pack_buffer: since its use it's fully contained in Rdb_key_def::pack_record, annotate it as such. Pull Request resolved: facebook/mysql-5.6#1446 Differential Revision: D56051395 fbshipit-source-id: f503d56efbaee8a2072fe5701eec4783d7c1cce9
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
All the key packing buffers in the MyRocks handler have exactly the same
lifetime between alloc_key_buffers and free_key_buffers. Thus allocate a single
large buffer and point the key packing buffers to offsets inside it, reducing
number of calls to the heap allocator on table opening/close code path. Annotate
the padding bytes for Valgrind as inaccessible. Add additional handling for
m_pack_buffer: since its use it's fully contained in Rdb_key_def::pack_record,
annotate it as such.