LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
From: Anshuman Khandual <anshuman.khandual@arm.com>
To: Mark Rutland <mark.rutland@arm.com>
Cc: linux-kernel@vger.kernel.org,
	linux-arm-kernel@lists.infradead.org,
	Catalin Marinas <catalin.marinas@arm.com>,
	Will Deacon <will.deacon@arm.com>,
	James Morse <james.morse@arm.com>,
	Andrey Konovalov <andreyknvl@google.com>
Subject: Re: [PATCH 4/4] arm64/mm: Drop vm_fault_t argument from __do_page_fault()
Date: Fri, 31 May 2019 14:35:23 +0530	[thread overview]
Message-ID: <5e565fd3-0d2b-31a5-8644-c91ccc5fb05e@arm.com> (raw)
In-Reply-To: <20190529151134.GH31777@lakrids.cambridge.arm.com>



On 05/29/2019 08:41 PM, Mark Rutland wrote:
> On Wed, May 29, 2019 at 06:04:45PM +0530, Anshuman Khandual wrote:
>> __do_page_fault() is over complicated with multiple goto statements. This
>> cleans up code flow and while there drops the vm_fault_t argument.
>> Signed-off-by: Anshuman Khandual <anshuman.khandual@arm.com>
>>
>> Cc: Catalin Marinas <catalin.marinas@arm.com>
>> Cc: Will Deacon <will.deacon@arm.com>
>> Cc: Mark Rutland <mark.rutland@arm.com>
>> Cc: James Morse <james.morse@arm.com> 
>> Cc: Andrey Konovalov <andreyknvl@google.com>
>> ---
>>  arch/arm64/mm/fault.c | 38 ++++++++++++++++----------------------
>>  1 file changed, 16 insertions(+), 22 deletions(-)
>>
>> diff --git a/arch/arm64/mm/fault.c b/arch/arm64/mm/fault.c
>> index 170c71f..a53a30e 100644
>> --- a/arch/arm64/mm/fault.c
>> +++ b/arch/arm64/mm/fault.c
>> @@ -397,37 +397,31 @@ static void do_bad_area(unsigned long addr, unsigned int esr, struct pt_regs *re
>>  static vm_fault_t __do_page_fault(struct mm_struct *mm, unsigned long addr,
>>  			   unsigned int mm_flags, unsigned long vm_flags)
>>  {
>> -	struct vm_area_struct *vma;
>> -	vm_fault_t fault;
>> +	struct vm_area_struct *vma = find_vma(mm, addr);
>>  
>> -	vma = find_vma(mm, addr);
>> -	fault = VM_FAULT_BADMAP;
>>  	if (unlikely(!vma))
>> -		goto out;
>> -	if (unlikely(vma->vm_start > addr))
>> -		goto check_stack;
>> +		return VM_FAULT_BADMAP;
>>  
>>  	/*
>> -	 * Ok, we have a good vm_area for this memory access, so we can handle
>> -	 * it.
>> +	 * Check if the VMA has got the required permssion with respect
>> +	 * to the access fault here.
>>  	 */
> 
> We already had a perfectly good comment for this check:
> 
> 	/*
> 	 * Check that the permissions on the VMA allow for the fault which
> 	 * occurred.
> 	 */
> 
> ... so please keep that and minimize the diff.

Sure, will keep all the existing comments here.

> 
>> -good_area:
>> +	if (!(vma->vm_flags & vm_flags))
>> +		return VM_FAULT_BADACCESS;
>> +
>>  	/*
>> -	 * Check that the permissions on the VMA allow for the fault which
>> -	 * occurred.
>> +	 * There is a valid VMA for this access. But before proceeding
>> +	 * make sure that it has required flags if there is an attempt
>> +	 * to expand the stack downwards.
>>  	 */
> 
> I think we can drop this comment, given we didn't have it previously.

Okay.

> 
>> -	if (!(vma->vm_flags & vm_flags)) {
>> -		fault = VM_FAULT_BADACCESS;
>> -		goto out;
>> -	}
>> +	if (unlikely(vma->vm_start > addr)) {
>> +		if (!(vma->vm_flags & VM_GROWSDOWN))
>> +			return VM_FAULT_BADMAP;
>>  
>> +		if (expand_stack(vma, addr))
>> +			return VM_FAULT_BADMAP;
> 
> You can drop the line space between these two if statements.

Will do.

> 
>> +	}
>>  	return handle_mm_fault(vma, addr & PAGE_MASK, mm_flags);
>> -
>> -check_stack:
>> -	if (vma->vm_flags & VM_GROWSDOWN && !expand_stack(vma, addr))
>> -		goto good_area;
>> -out:
>> -	return fault;
> 
> We used to check the stack before the checknig the rest of the vm_flags,
> so this changes the precedence of the VM_FAULT_BADMAP and
> VM_FAULT_BADACCESS return codes.
> 
> Please check the stack before checking the other vm_flags.

Though it makes some sense to move VMA permission check earlier in the function as it
is the quicker one but I understand need to maintain the existing code flow in a clean
up patch like this. Will retain the existing flow.

  reply	other threads:[~2019-05-31  9:05 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-05-29 12:34 [PATCH 0/4] arm64/mm: Fixes and cleanups for do_page_fault() Anshuman Khandual
2019-05-29 12:34 ` [PATCH 1/4] arm64/mm: Drop mmap_sem before calling __do_kernel_fault() Anshuman Khandual
2019-05-29 12:34 ` [PATCH 2/4] arm64/mm: Drop task_struct argument from __do_page_fault() Anshuman Khandual
2019-05-29 12:34 ` [PATCH 3/4] arm64/mm: Consolidate page fault information capture Anshuman Khandual
2019-05-29 14:53   ` Mark Rutland
2019-05-31  9:10     ` Anshuman Khandual
2019-05-29 12:34 ` [PATCH 4/4] arm64/mm: Drop vm_fault_t argument from __do_page_fault() Anshuman Khandual
2019-05-29 15:11   ` Mark Rutland
2019-05-31  9:05     ` Anshuman Khandual [this message]
2019-05-30  6:34   ` Christoph Hellwig
2019-05-31  8:55     ` Anshuman Khandual
2019-05-29 12:41 ` [PATCH 0/4] arm64/mm: Fixes and cleanups for do_page_fault() Will Deacon
2019-05-29 12:59   ` Anshuman Khandual

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=5e565fd3-0d2b-31a5-8644-c91ccc5fb05e@arm.com \
    --to=anshuman.khandual@arm.com \
    --cc=andreyknvl@google.com \
    --cc=catalin.marinas@arm.com \
    --cc=james.morse@arm.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mark.rutland@arm.com \
    --cc=will.deacon@arm.com \
    --subject='Re: [PATCH 4/4] arm64/mm: Drop vm_fault_t argument from __do_page_fault()' \
    /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

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).