[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <ba0f252c-345e-4988-8d4f-36116708f19e@rivosinc.com>
Date: Thu, 28 Nov 2024 08:36:04 +0100
From: Clément Léger <cleger@...osinc.com>
To: laokz <zhangkai@...as.ac.cn>, Paul Walmsley <paul.walmsley@...ive.com>,
Palmer Dabbelt <palmer@...belt.com>, Albert Ou <aou@...s.berkeley.edu>,
linux-riscv@...ts.infradead.org, linux-kernel@...r.kernel.org
Cc: Charlie Jenkins <charlie@...osinc.com>,
Andrew Jones <ajones@...tanamicro.com>
Subject: Re: [PATCH v2] riscv: module: use a plain variable for list_head
instead of a pointer
On 28/11/2024 02:01, laokz wrote:
> On Wed, 2024-11-27 at 15:25 +0100, Clément Léger wrote:
>> rel_head's list_head member, rel_entry, doesn't need to be allocated,
>> its storage can just be part of the allocated rel_head. Remove the
>
> Oh my poor English. OK, it's more better than just add the lost kfree.
>
>> pointer which allows to get rid of the allocation as well as an
>> existing
>> memory leak found by Kai Zang using kmemleak.
> ^
> Zhang
>
> BTW. Doesn't it need a fixes tag like what you suggested? (The bug
> might come from 6.7)
Hey Kai,
Apologies for misspelling your name, I'll fix that !
That's a good question, I guess I could have added it indeed but since
it was rather a rework, I left it out. But it's probably better to add
it anyway. I'll send a V3 with the Fixes tag as well as fixing your name
spelling.
Thanks,
Clément
>
>>
>> Reported-by: Kai Zhang <zhangkai@...as.ac.cn>
>> Signed-off-by: Clément Léger <cleger@...osinc.com>
>> Reviewed-by: Andrew Jones <ajones@...tanamicro.com>
>> ---
>>
>> V2:
>> - Add Kai Reported-by
>> - Reword the commit description (Andrew)
>>
>> ---
>> arch/riscv/kernel/module.c | 18 ++++--------------
>> 1 file changed, 4 insertions(+), 14 deletions(-)
>>
>> diff --git a/arch/riscv/kernel/module.c b/arch/riscv/kernel/module.c
>> index 1cd461f3d872..47d0ebeec93c 100644
>> --- a/arch/riscv/kernel/module.c
>> +++ b/arch/riscv/kernel/module.c
>> @@ -23,7 +23,7 @@ struct used_bucket {
>>
>> struct relocation_head {
>> struct hlist_node node;
>> - struct list_head *rel_entry;
>> + struct list_head rel_entry;
>> void *location;
>> };
>>
>> @@ -634,7 +634,7 @@ process_accumulated_relocations(struct module
>> *me,
>> location = rel_head_iter->location;
>> list_for_each_entry_safe(rel_entry_iter,
>> rel_entry_iter_tmp,
>> - rel_head_iter-
>>> rel_entry,
>> + &rel_head_iter-
>>> rel_entry,
>> head) {
>> curr_type = rel_entry_iter->type;
>> reloc_handlers[curr_type].reloc_handl
>> er(
>> @@ -704,16 +704,7 @@ static int add_relocation_to_accumulate(struct
>> module *me, int type,
>> return -ENOMEM;
>> }
>>
>> - rel_head->rel_entry =
>> - kmalloc(sizeof(struct list_head),
>> GFP_KERNEL);
>> -
>> - if (!rel_head->rel_entry) {
>> - kfree(entry);
>> - kfree(rel_head);
>> - return -ENOMEM;
>> - }
>> -
>> - INIT_LIST_HEAD(rel_head->rel_entry);
>> + INIT_LIST_HEAD(&rel_head->rel_entry);
>> rel_head->location = location;
>> INIT_HLIST_NODE(&rel_head->node);
>> if (!current_head->first) {
>> @@ -722,7 +713,6 @@ static int add_relocation_to_accumulate(struct
>> module *me, int type,
>>
>> if (!bucket) {
>> kfree(entry);
>> - kfree(rel_head->rel_entry);
>> kfree(rel_head);
>> return -ENOMEM;
>> }
>> @@ -735,7 +725,7 @@ static int add_relocation_to_accumulate(struct
>> module *me, int type,
>> }
>>
>> /* Add relocation to head of discovered rel_head */
>> - list_add_tail(&entry->head, rel_head->rel_entry);
>> + list_add_tail(&entry->head, &rel_head->rel_entry);
>>
>> return 0;
>> }
>
Powered by blists - more mailing lists