From: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
To: Christophe Leroy <christophe.leroy@c-s.fr>,
        Benjamin Herrenschmidt <benh@kernel.crashing.org>,
        Paul Mackerras <paulus@samba.org>,
        Michael Ellerman <mpe@ellerman.id.au>, Scott Wood <oss@buserror.net>
Cc: linuxppc-dev@lists.ozlabs.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH] powerpc/mm: some cleanup of do_page_fault()
In-Reply-To: <20170413145342.40ED56E667@localhost.localdomain>
References: <20170413145342.40ED56E667@localhost.localdomain>
Date: Mon, 17 Apr 2017 14:05:05 +0530
MIME-Version: 1.0
Content-Type: text/plain
Message-Id: <87d1cbh1om.fsf@skywalker.in.ibm.com>
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 6199
Lines: 172

Christophe Leroy <christophe.leroy@c-s.fr> writes:

> This patch is a bunch of small cleanups of the do_page_fault()
> function:
> 1/ Function store_updates_sp() checks whether the faulting
> instruction is a store updating r1. Therefore we can limit its calls
> to stores exceptions
> 2/ Only the get_user() in store_updates_sp() has to be done outside
> the mm semaphore. All the comparison can be done within the semaphore,
> so only when really needed.
> 3/ As we got a DSI exception, the address pointed by regs->nip is
> obviously valid, otherwise we would have had a instruction exception.
> So __get_user() can be used instead of get_user()
> 4/ Replaced one duplicate 'trap == 0x400' by 'is_exec'
> 5/ Created a 'is_user = user_mode(regs)' and replaced all
> 'user_mode(regs)'.

If this can be made a seperate patch, I guess it will make this patch
much easier to review.


> Analysis of the assembly code shows that when using user_mode(regs),
> at least the 'andi. r7,r8,16384' is redone several times, and also
> the 'lwz r8,132(r31)' at times. With the new form, the 'is_user'
> is mapped to cr4, then all further use of is_user results in just
> things like 'beq cr4,218 <do_page_fault+0x218>'
> 6/ The 8xx has a dedicated exception for breakpoints, that directly
> calls do_break()
> 7/ stdu and stdux only exist on PPC64, so no need to check for them
> on PPC32
>
> With this patch, on PPC8xx the size of do_page_fault() is reduced by
> 12% (from 1420 bytes to 1248 bytes)
>
> Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>

Reviewed-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>


> ---
>  arch/powerpc/mm/fault.c | 34 ++++++++++++++++++----------------
>  1 file changed, 18 insertions(+), 16 deletions(-)
>
> diff --git a/arch/powerpc/mm/fault.c b/arch/powerpc/mm/fault.c
> index 3a7d580fdc59..5aa43268baca 100644
> --- a/arch/powerpc/mm/fault.c
> +++ b/arch/powerpc/mm/fault.c
> @@ -73,12 +73,8 @@ static inline int notify_page_fault(struct pt_regs *regs)
>   * Check whether the instruction at regs->nip is a store using
>   * an update addressing form which will update r1.
>   */
> -static int store_updates_sp(struct pt_regs *regs)
> +static int store_updates_sp(unsigned int inst)
>  {
> -	unsigned int inst;
> -
> -	if (get_user(inst, (unsigned int __user *)regs->nip))
> -		return 0;
>  	/* check for 1 in the rA field */
>  	if (((inst >> 16) & 0x1f) != 1)
>  		return 0;
> @@ -90,12 +86,16 @@ static int store_updates_sp(struct pt_regs *regs)
>  	case 53:	/* stfsu */
>  	case 55:	/* stfdu */
>  		return 1;
> +#ifdef CONFIG_PPC64
>  	case 62:	/* std or stdu */
>  		return (inst & 3) == 1;
> +#endif
>  	case 31:
>  		/* check minor opcode */
>  		switch ((inst >> 1) & 0x3ff) {
> +#ifdef CONFIG_PPC64
>  		case 181:	/* stdux */
> +#endif
>  		case 183:	/* stwux */
>  		case 247:	/* stbux */
>  		case 439:	/* sthux */
> @@ -207,7 +207,9 @@ int do_page_fault(struct pt_regs *regs, unsigned long address,
>  	int trap = TRAP(regs);
>   	int is_exec = trap == 0x400;
>  	int fault;
> -	int rc = 0, store_update_sp = 0;
> +	int rc = 0;
> +	int is_user = user_mode(regs);
> +	unsigned int inst = 0;
>  
>  #if !(defined(CONFIG_4xx) || defined(CONFIG_BOOKE))
>  	/*
> @@ -216,7 +218,7 @@ int do_page_fault(struct pt_regs *regs, unsigned long address,
>  	 * bits we are interested in.  But there are some bits which
>  	 * indicate errors in DSISR but can validly be set in SRR1.
>  	 */
> -	if (trap == 0x400)
> +	if (is_exec)
>  		error_code &= 0x48200000;
>  	else
>  		is_write = error_code & DSISR_ISSTORE;
> @@ -247,13 +249,13 @@ int do_page_fault(struct pt_regs *regs, unsigned long address,
>  	 * The kernel should never take an execute fault nor should it
>  	 * take a page fault to a kernel address.
>  	 */
> -	if (!user_mode(regs) && (is_exec || (address >= TASK_SIZE))) {
> +	if (!is_user && (is_exec || (address >= TASK_SIZE))) {
>  		rc = SIGSEGV;
>  		goto bail;
>  	}
>  
>  #if !(defined(CONFIG_4xx) || defined(CONFIG_BOOKE) || \
> -			     defined(CONFIG_PPC_BOOK3S_64))
> +      defined(CONFIG_PPC_BOOK3S_64) || defined(CONFIG_PPC_8xx))
>    	if (error_code & DSISR_DABRMATCH) {
>  		/* breakpoint match */
>  		do_break(regs, address, error_code);
> @@ -266,7 +268,7 @@ int do_page_fault(struct pt_regs *regs, unsigned long address,
>  		local_irq_enable();
>  
>  	if (faulthandler_disabled() || mm == NULL) {
> -		if (!user_mode(regs)) {
> +		if (!is_user) {
>  			rc = SIGSEGV;
>  			goto bail;
>  		}
> @@ -287,10 +289,10 @@ int do_page_fault(struct pt_regs *regs, unsigned long address,
>  	 * can result in fault, which will cause a deadlock when called with
>  	 * mmap_sem held
>  	 */
> -	if (!is_exec && user_mode(regs))
> -		store_update_sp = store_updates_sp(regs);
> +	if (is_write && is_user)
> +		__get_user(inst, (unsigned int __user *)regs->nip);
>

Update the comment around to explain that is_write and is_user check .
ie, we can say we will use that inst to check stdu form later and hence
we need todo this only if it is write fault ?


> -	if (user_mode(regs))
> +	if (is_user)
>  		flags |= FAULT_FLAG_USER;
>  
>  	/* When running in the kernel we expect faults to occur only to
> @@ -309,7 +311,7 @@ int do_page_fault(struct pt_regs *regs, unsigned long address,
>  	 * thus avoiding the deadlock.
>  	 */
>  	if (!down_read_trylock(&mm->mmap_sem)) {
> -		if (!user_mode(regs) && !search_exception_tables(regs->nip))
> +		if (!is_user && !search_exception_tables(regs->nip))
>  			goto bad_area_nosemaphore;
>  
>  retry:
> @@ -358,7 +360,7 @@ int do_page_fault(struct pt_regs *regs, unsigned long address,
>  		 * between the last mapped region and the stack will
>  		 * expand the stack rather than segfaulting.
>  		 */
> -		if (address + 2048 < uregs->gpr[1] && !store_update_sp)
> +		if (address + 2048 < uregs->gpr[1] && !store_updates_sp(inst))
>  			goto bad_area;
>  	}
>  	if (expand_stack(vma, address))
> @@ -509,7 +511,7 @@ int do_page_fault(struct pt_regs *regs, unsigned long address,
>  
>  bad_area_nosemaphore:
>  	/* User mode accesses cause a SIGSEGV */
> -	if (user_mode(regs)) {
> +	if (is_user) {
>  		_exception(SIGSEGV, regs, code, address);
>  		goto bail;
>  	}
> -- 
> 2.12.0