linux-kernel - Re: [PATCH 2/2] io_uring: batch getting pcpu references

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <161c01e5-d007-af45-c6a5-ccebf1e00533@gmail.com>
Date:   Wed, 18 Dec 2019 12:23:43 +0300
From:   Pavel Begunkov <asml.silence@...il.com>
To:     Jens Axboe <axboe@...nel.dk>, io-uring@...r.kernel.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH 2/2] io_uring: batch getting pcpu references

On 12/18/2019 2:21 AM, Jens Axboe wrote:
> On 12/17/19 3:28 PM, Pavel Begunkov wrote:
>> percpu_ref_tryget() has its own overhead. Instead getting a reference
>> for each request, grab a bunch once per io_submit_sqes().
>>
>> basic benchmark with submit and wait 128 non-linked nops showed ~5%
>> performance gain. (7044 KIOPS vs 7423)
>>
>> Signed-off-by: Pavel Begunkov <asml.silence@...il.com>
>> ---
>>
>> For notice: it could be done without @extra_refs variable,
>> but looked too tangled because of gotos.
>>
>>
>>  fs/io_uring.c | 11 ++++++++---
>>  1 file changed, 8 insertions(+), 3 deletions(-)
>>
>> diff --git a/fs/io_uring.c b/fs/io_uring.c
>> index cf4138f0e504..6c85dfc62224 100644
>> --- a/fs/io_uring.c
>> +++ b/fs/io_uring.c
>> @@ -845,9 +845,6 @@ static struct io_kiocb *io_get_req(struct io_ring_ctx *ctx,
>>  	gfp_t gfp = GFP_KERNEL | __GFP_NOWARN;
>>  	struct io_kiocb *req;
>>  
>> -	if (!percpu_ref_tryget(&ctx->refs))
>> -		return NULL;
>> -
>>  	if (!state) {
>>  		req = kmem_cache_alloc(req_cachep, gfp);
>>  		if (unlikely(!req))
>> @@ -3929,6 +3926,7 @@ static int io_submit_sqes(struct io_ring_ctx *ctx, unsigned int nr,
>>  	struct io_submit_state state, *statep = NULL;
>>  	struct io_kiocb *link = NULL;
>>  	int i, submitted = 0;
>> +	unsigned int extra_refs;
>>  	bool mm_fault = false;
>>  
>>  	/* if we have a backlog and couldn't flush it all, return BUSY */
>> @@ -3941,6 +3939,10 @@ static int io_submit_sqes(struct io_ring_ctx *ctx, unsigned int nr,
>>  		statep = &state;
>>  	}
>>  
>> +	if (!percpu_ref_tryget_many(&ctx->refs, nr))
>> +		return -EAGAIN;
>> +	extra_refs = nr;
>> +
>>  	for (i = 0; i < nr; i++) {
>>  		struct io_kiocb *req = io_get_req(ctx, statep);
>>  
>> @@ -3949,6 +3951,7 @@ static int io_submit_sqes(struct io_ring_ctx *ctx, unsigned int nr,
>>  				submitted = -EAGAIN;
>>  			break;
>>  		}
>> +		--extra_refs;
>>  		if (!io_get_sqring(ctx, req)) {
>>  			__io_free_req(req);
>>  			break;
>> @@ -3976,6 +3979,8 @@ static int io_submit_sqes(struct io_ring_ctx *ctx, unsigned int nr,
>>  		io_queue_link_head(link);
>>  	if (statep)
>>  		io_submit_state_end(&state);
>> +	if (extra_refs)
>> +		percpu_ref_put_many(&ctx->refs, extra_refs);
> 
> Might be cleaner to introduce a 'ret' variable, and leave submitted to be just
> that, the number submitted. Then you could just do:
> 
> if (submitted != nr)
> 	percpu_ref_put_many(&ctx->refs, nr - submitted);
> 
> and not need that weird extra_refs.
> 
That won't work as is, because  __io_free_req() after io_get_sqring()
puts a ref. And that's the reason why --extra_refs; placed in between
those 2 if's. Also, percpu_ref_put_many() is inline, so I want to have
only 1 call site.

I'll send another version, which won't even need the "if" at the end,
but is more knotty.

-- 
Pavel Begunkov