All of lore.kernel.org
 help / color / mirror / Atom feed
From: Alex Shi <alex.shi@linaro.org>
To: Mel Gorman <mgorman@suse.de>
Cc: H Peter Anvin <hpa@zytor.com>, Linux-X86 <x86@kernel.org>,
	Linux-MM <linux-mm@kvack.org>,
	LKML <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 1/3] x86: mm: Clean up inconsistencies when flushing TLB ranges
Date: Thu, 12 Dec 2013 21:59:33 +0800	[thread overview]
Message-ID: <52A9C145.9050706@linaro.org> (raw)
In-Reply-To: <1386849309-22584-2-git-send-email-mgorman@suse.de>

On 12/12/2013 07:55 PM, Mel Gorman wrote:
> NR_TLB_LOCAL_FLUSH_ALL is not always accounted for correctly and the
> comparison with total_vm is done before taking tlb_flushall_shift into
> account. Clean it up.
> 
> Signed-off-by: Mel Gorman <mgorman@suse.de>

Reviewed-by: Alex Shi
> ---
>  arch/x86/mm/tlb.c | 12 ++++++------
>  1 file changed, 6 insertions(+), 6 deletions(-)
> 
> diff --git a/arch/x86/mm/tlb.c b/arch/x86/mm/tlb.c
> index ae699b3..09b8cb8 100644
> --- a/arch/x86/mm/tlb.c
> +++ b/arch/x86/mm/tlb.c
> @@ -189,6 +189,7 @@ void flush_tlb_mm_range(struct mm_struct *mm, unsigned long start,
>  {
>  	unsigned long addr;
>  	unsigned act_entries, tlb_entries = 0;
> +	unsigned long nr_base_pages;
>  
>  	preempt_disable();
>  	if (current->active_mm != mm)
> @@ -210,18 +211,17 @@ void flush_tlb_mm_range(struct mm_struct *mm, unsigned long start,
>  		tlb_entries = tlb_lli_4k[ENTRIES];
>  	else
>  		tlb_entries = tlb_lld_4k[ENTRIES];
> +
>  	/* Assume all of TLB entries was occupied by this task */

the benchmark break this assumption?
> -	act_entries = mm->total_vm > tlb_entries ? tlb_entries : mm->total_vm;
> +	act_entries = tlb_entries >> tlb_flushall_shift;
> +	act_entries = mm->total_vm > act_entries ? act_entries : mm->total_vm;
> +	nr_base_pages = (end - start) >> PAGE_SHIFT;
>  
>  	/* tlb_flushall_shift is on balance point, details in commit log */
> -	if ((end - start) >> PAGE_SHIFT > act_entries >> tlb_flushall_shift) {
> +	if (nr_base_pages > act_entries || has_large_page(mm, start, end)) {
>  		count_vm_event(NR_TLB_LOCAL_FLUSH_ALL);
>  		local_flush_tlb();
>  	} else {
> -		if (has_large_page(mm, start, end)) {
> -			local_flush_tlb();
> -			goto flush_all;
> -		}
>  		/* flush range by one by one 'invlpg' */
>  		for (addr = start; addr < end;	addr += PAGE_SIZE) {
>  			count_vm_event(NR_TLB_LOCAL_FLUSH_ONE);
> 


-- 
Thanks
    Alex

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

WARNING: multiple messages have this Message-ID (diff)
From: Alex Shi <alex.shi@linaro.org>
To: Mel Gorman <mgorman@suse.de>
Cc: H Peter Anvin <hpa@zytor.com>, Linux-X86 <x86@kernel.org>,
	Linux-MM <linux-mm@kvack.org>,
	LKML <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 1/3] x86: mm: Clean up inconsistencies when flushing TLB ranges
Date: Thu, 12 Dec 2013 21:59:33 +0800	[thread overview]
Message-ID: <52A9C145.9050706@linaro.org> (raw)
In-Reply-To: <1386849309-22584-2-git-send-email-mgorman@suse.de>

On 12/12/2013 07:55 PM, Mel Gorman wrote:
> NR_TLB_LOCAL_FLUSH_ALL is not always accounted for correctly and the
> comparison with total_vm is done before taking tlb_flushall_shift into
> account. Clean it up.
> 
> Signed-off-by: Mel Gorman <mgorman@suse.de>

Reviewed-by: Alex Shi
> ---
>  arch/x86/mm/tlb.c | 12 ++++++------
>  1 file changed, 6 insertions(+), 6 deletions(-)
> 
> diff --git a/arch/x86/mm/tlb.c b/arch/x86/mm/tlb.c
> index ae699b3..09b8cb8 100644
> --- a/arch/x86/mm/tlb.c
> +++ b/arch/x86/mm/tlb.c
> @@ -189,6 +189,7 @@ void flush_tlb_mm_range(struct mm_struct *mm, unsigned long start,
>  {
>  	unsigned long addr;
>  	unsigned act_entries, tlb_entries = 0;
> +	unsigned long nr_base_pages;
>  
>  	preempt_disable();
>  	if (current->active_mm != mm)
> @@ -210,18 +211,17 @@ void flush_tlb_mm_range(struct mm_struct *mm, unsigned long start,
>  		tlb_entries = tlb_lli_4k[ENTRIES];
>  	else
>  		tlb_entries = tlb_lld_4k[ENTRIES];
> +
>  	/* Assume all of TLB entries was occupied by this task */

the benchmark break this assumption?
> -	act_entries = mm->total_vm > tlb_entries ? tlb_entries : mm->total_vm;
> +	act_entries = tlb_entries >> tlb_flushall_shift;
> +	act_entries = mm->total_vm > act_entries ? act_entries : mm->total_vm;
> +	nr_base_pages = (end - start) >> PAGE_SHIFT;
>  
>  	/* tlb_flushall_shift is on balance point, details in commit log */
> -	if ((end - start) >> PAGE_SHIFT > act_entries >> tlb_flushall_shift) {
> +	if (nr_base_pages > act_entries || has_large_page(mm, start, end)) {
>  		count_vm_event(NR_TLB_LOCAL_FLUSH_ALL);
>  		local_flush_tlb();
>  	} else {
> -		if (has_large_page(mm, start, end)) {
> -			local_flush_tlb();
> -			goto flush_all;
> -		}
>  		/* flush range by one by one 'invlpg' */
>  		for (addr = start; addr < end;	addr += PAGE_SIZE) {
>  			count_vm_event(NR_TLB_LOCAL_FLUSH_ONE);
> 


-- 
Thanks
    Alex

  reply	other threads:[~2013-12-12 13:59 UTC|newest]

Thread overview: 50+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-12-12 11:55 [RFC PATCH 0/3] Fix ebizzy performance regression on IvyBridge due to X86 TLB range flush Mel Gorman
2013-12-12 11:55 ` Mel Gorman
2013-12-12 11:55 ` [PATCH 1/3] x86: mm: Clean up inconsistencies when flushing TLB ranges Mel Gorman
2013-12-12 11:55   ` Mel Gorman
2013-12-12 13:59   ` Alex Shi [this message]
2013-12-12 13:59     ` Alex Shi
2013-12-12 23:53     ` Mel Gorman
2013-12-12 23:53       ` Mel Gorman
2013-12-12 11:55 ` [PATCH 2/3] x86: mm: Change tlb_flushall_shift for IvyBridge Mel Gorman
2013-12-12 11:55   ` Mel Gorman
2013-12-12 13:13   ` Ingo Molnar
2013-12-12 13:13     ` Ingo Molnar
2013-12-12 13:38     ` Alex Shi
2013-12-12 13:38       ` Alex Shi
2013-12-12 14:11       ` Ingo Molnar
2013-12-12 14:11         ` Ingo Molnar
2013-12-13  1:02         ` Alex Shi
2013-12-13  1:02           ` Alex Shi
2013-12-13  2:11           ` Alex Shi
2013-12-13  2:11             ` Alex Shi
2013-12-13 13:43             ` Ingo Molnar
2013-12-13 13:43               ` Ingo Molnar
2013-12-14 11:01               ` Alex Shi
2013-12-14 11:01                 ` Alex Shi
2013-12-14 14:19             ` Peter Zijlstra
2013-12-14 14:19               ` Peter Zijlstra
2013-12-14 14:27               ` Peter Zijlstra
2013-12-14 14:27                 ` Peter Zijlstra
2013-12-16 13:59                 ` Ingo Molnar
2013-12-16 13:59                   ` Ingo Molnar
2013-12-17 11:59                   ` Alex Shi
2013-12-17 11:59                     ` Alex Shi
2013-12-17 13:14                     ` Ingo Molnar
2013-12-17 13:14                       ` Ingo Molnar
2013-12-16  8:26               ` Alex Shi
2013-12-16  8:26                 ` Alex Shi
2013-12-16 10:06                 ` Peter Zijlstra
2013-12-16 10:06                   ` Peter Zijlstra
2013-12-12 13:45   ` Alex Shi
2013-12-12 13:45     ` Alex Shi
2013-12-12 11:55 ` [PATCH 3/3] x86: mm: Account for the of CPUs that must be flushed during a TLB range flush Mel Gorman
2013-12-12 11:55   ` Mel Gorman
2013-12-12 13:41   ` Alex Shi
2013-12-12 13:41     ` Alex Shi
2013-12-12 13:01 ` [RFC PATCH 0/3] Fix ebizzy performance regression on IvyBridge due to X86 " Ingo Molnar
2013-12-12 13:01   ` Ingo Molnar
2013-12-12 14:40   ` Mel Gorman
2013-12-12 14:40     ` Mel Gorman
2013-12-13 13:35     ` Ingo Molnar
2013-12-13 13:35       ` Ingo Molnar

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=52A9C145.9050706@linaro.org \
    --to=alex.shi@linaro.org \
    --cc=hpa@zytor.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mgorman@suse.de \
    --cc=x86@kernel.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.