From: Peng Zhang <zhangpeng.00@bytedance.com>
To: "Liam R. Howlett" <Liam.Howlett@Oracle.com>
Cc: Peng Zhang <zhangpeng.00@bytedance.com>,
avagin@gmail.com, npiggin@gmail.com,
mathieu.desnoyers@efficios.com, peterz@infradead.org,
michael.christie@oracle.com, surenb@google.com,
brauner@kernel.org, willy@infradead.org,
akpm@linux-foundation.org, corbet@lwn.net,
linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org,
linux-mm@kvack.org, linux-doc@vger.kernel.org
Subject: Re: [PATCH 06/11] maple_tree: Introduce mas_replace_entry() to directly replace an entry
Date: Mon, 31 Jul 2023 20:39:08 +0800 [thread overview]
Message-ID: <20aab1af-c183-db94-90d7-5e5425e3fd80@bytedance.com> (raw)
In-Reply-To: <20230726160843.hpl4razxiikqbuxy@revolver>
在 2023/7/27 00:08, Liam R. Howlett 写道:
> * Peng Zhang <zhangpeng.00@bytedance.com> [230726 04:10]:
>> If mas has located a specific entry, it may be need to replace this
>> entry, so introduce mas_replace_entry() to do this. mas_replace_entry()
>> will be more efficient than mas_store*() because it doesn't do many
>> unnecessary checks.
>>
>> This function should be inline, but more functions need to be moved to
>> the header file, so I didn't do it for the time being.
>
> I am really nervous having no checks here. I get that this could be
> used for duplicating the tree more efficiently, but having a function
> that just swaps a value in is very dangerous - especially since it is
> decoupled from the tree duplication code.
I've thought about this, and I feel like this is something the user
should be guaranteed. If the user is not sure whether to use it,
mas_store() can be used instead. And we should provide this interface
because it has better performance.
>
>>
>> Signed-off-by: Peng Zhang <zhangpeng.00@bytedance.com>
>> ---
>> include/linux/maple_tree.h | 1 +
>> lib/maple_tree.c | 25 +++++++++++++++++++++++++
>> 2 files changed, 26 insertions(+)
>>
>> diff --git a/include/linux/maple_tree.h b/include/linux/maple_tree.h
>> index 229fe78e4c89..a05e9827d761 100644
>> --- a/include/linux/maple_tree.h
>> +++ b/include/linux/maple_tree.h
>> @@ -462,6 +462,7 @@ struct ma_wr_state {
>>
>> void *mas_walk(struct ma_state *mas);
>> void *mas_store(struct ma_state *mas, void *entry);
>> +void mas_replace_entry(struct ma_state *mas, void *entry);
>> void *mas_erase(struct ma_state *mas);
>> int mas_store_gfp(struct ma_state *mas, void *entry, gfp_t gfp);
>> void mas_store_prealloc(struct ma_state *mas, void *entry);
>> diff --git a/lib/maple_tree.c b/lib/maple_tree.c
>> index efac6761ae37..d58572666a00 100644
>> --- a/lib/maple_tree.c
>> +++ b/lib/maple_tree.c
>> @@ -5600,6 +5600,31 @@ void *mas_store(struct ma_state *mas, void *entry)
>> }
>> EXPORT_SYMBOL_GPL(mas_store);
>>
>> +/**
>> + * mas_replace_entry() - Replace an entry that already exists in the maple tree
>> + * @mas: The maple state
>> + * @entry: The entry to store
>> + *
>> + * Please note that mas must already locate an existing entry, and the new entry
>> + * must not be NULL. If these two points cannot be guaranteed, please use
>> + * mas_store*() instead, otherwise it will cause an internal error in the maple
>> + * tree. This function does not need to allocate memory, so it must succeed.
>> + */
>> +void mas_replace_entry(struct ma_state *mas, void *entry)
>> +{
>> + void __rcu **slots;
>> +
>> +#ifdef CONFIG_DEBUG_MAPLE_TREE
>> + MAS_WARN_ON(mas, !mte_is_leaf(mas->node));
>> + MAS_WARN_ON(mas, !entry);
>> + MAS_WARN_ON(mas, mas->offset >= mt_slots[mte_node_type(mas->node)]);
>> +#endif
>> +
>> + slots = ma_slots(mte_to_node(mas->node), mte_node_type(mas->node));
>> + rcu_assign_pointer(slots[mas->offset], entry);
>> +}
>> +EXPORT_SYMBOL_GPL(mas_replace_entry);
>> +
>> /**
>> * mas_store_gfp() - Store a value into the tree.
>> * @mas: The maple state
>> --
>> 2.20.1
>>
next prev parent reply other threads:[~2023-07-31 12:39 UTC|newest]
Thread overview: 42+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-07-26 8:09 [PATCH 00/11] Introduce mt_dup() to improve the performance of fork() Peng Zhang
2023-07-26 8:09 ` [PATCH 01/11] maple_tree: Introduce ma_nonleaf_data_end{_nocheck}() Peng Zhang
2023-07-26 14:58 ` Liam R. Howlett
2023-07-31 9:52 ` Peng Zhang
2023-07-31 16:08 ` Liam R. Howlett
2023-07-26 8:09 ` [PATCH 02/11] maple_tree: Validate MAPLE_ENODE and ma_nonleaf_data_end() Peng Zhang
2023-07-26 8:09 ` [PATCH 03/11] maple_tree: Add some helper functions Peng Zhang
2023-07-26 15:02 ` Liam R. Howlett
2023-07-26 15:08 ` Matthew Wilcox
2023-07-31 11:45 ` Peng Zhang
2023-08-11 17:28 ` Liam R. Howlett
2023-07-31 11:40 ` Peng Zhang
2023-07-26 8:09 ` [PATCH 04/11] maple_tree: Introduce interfaces __mt_dup() and mt_dup() Peng Zhang
2023-07-26 16:03 ` Liam R. Howlett
2023-07-31 12:24 ` Peng Zhang
2023-07-31 16:27 ` Liam R. Howlett
2023-08-16 13:41 ` Peng Zhang
2023-08-16 18:30 ` Liam R. Howlett
2023-08-18 11:53 ` Peng Zhang
2023-08-18 16:13 ` Liam R. Howlett
2023-07-26 8:09 ` [PATCH 05/11] maple_tree: Add test for mt_dup() Peng Zhang
2023-07-26 16:06 ` Liam R. Howlett
2023-07-31 12:32 ` Peng Zhang
2023-07-31 16:41 ` Liam R. Howlett
2023-07-26 8:09 ` [PATCH 06/11] maple_tree: Introduce mas_replace_entry() to directly replace an entry Peng Zhang
2023-07-26 16:08 ` Liam R. Howlett
2023-07-31 12:39 ` Peng Zhang [this message]
2023-07-31 16:48 ` Liam R. Howlett
2023-08-16 13:11 ` Peng Zhang
2023-08-16 17:40 ` Liam R. Howlett
2023-08-18 9:39 ` Peng Zhang
2023-08-18 16:15 ` Liam R. Howlett
2023-07-26 8:09 ` [PATCH 07/11] maple_tree: Update the documentation of maple tree Peng Zhang
2023-07-26 8:09 ` [PATCH 08/11] maple_tree: Skip other tests when BENCH is enabled Peng Zhang
2023-07-26 8:09 ` [PATCH 09/11] maple_tree: Update check_forking() and bench_forking() Peng Zhang
2023-07-26 8:09 ` [PATCH 10/11] MAINTAINERS: Add co-maintainer for maple tree Peng Zhang
2023-07-26 16:39 ` Liam R. Howlett
2023-07-31 12:55 ` Peng Zhang
2023-07-31 20:55 ` Liam R. Howlett
2023-07-26 8:09 ` [PATCH 11/11] fork: Use __mt_dup() to duplicate maple tree in dup_mmap() Peng Zhang
2023-07-26 17:06 ` Liam R. Howlett
2023-07-31 12:59 ` Peng Zhang
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=20aab1af-c183-db94-90d7-5e5425e3fd80@bytedance.com \
--to=zhangpeng.00@bytedance.com \
--cc=Liam.Howlett@Oracle.com \
--cc=akpm@linux-foundation.org \
--cc=avagin@gmail.com \
--cc=brauner@kernel.org \
--cc=corbet@lwn.net \
--cc=linux-doc@vger.kernel.org \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mathieu.desnoyers@efficios.com \
--cc=michael.christie@oracle.com \
--cc=npiggin@gmail.com \
--cc=peterz@infradead.org \
--cc=surenb@google.com \
--cc=willy@infradead.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).