From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
To: Kautuk Consul <consul.kautuk@gmail.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>,
linux-kernel@vger.kernel.org,
Christian Dietrich
<christian.dietrich@informatik.uni-erlangen.de>,
Paul Mackerras <paulus@samba.org>,
"Mohd. Faris" <mohdfarisq2010@gmail.com>,
Ingo Molnar <mingo@elte.hu>,
linuxppc-dev@lists.ozlabs.org
Subject: Re: [PATCH 04/20 v3] powerpc/mm/fault.c: Port OOM changes to do_page_fault
Date: Wed, 21 Mar 2012 08:28:56 +1100 [thread overview]
Message-ID: <1332278936.2982.23.camel@pasglop> (raw)
In-Reply-To: <1332259260-1943-1-git-send-email-consul.kautuk@gmail.com>
On Tue, 2012-03-20 at 12:01 -0400, Kautuk Consul wrote:
> Commit d065bd810b6deb67d4897a14bfe21f8eb526ba99
> (mm: retry page fault when blocking on disk transfer) and
> commit 37b23e0525d393d48a7d59f870b3bc061a30ccdb
> (x86,mm: make pagefault killable)
>
> The above commits introduced changes into the x86 pagefault handler
> for making the page fault handler retryable as well as killable.
>
> These changes reduce the mmap_sem hold time, which is crucial
> during OOM killer invocation.
>
> Port these changes to powerpc.
I have done that port already :-) It's in powerpc-next which I will send
to Linus to pull today or tomorrow....
Cheers,
Ben.
> Signed-off-by: Mohd. Faris <mohdfarisq2010@gmail.com>
> Signed-off-by: Kautuk Consul <consul.kautuk@gmail.com>
> ---
> arch/powerpc/mm/fault.c | 51 +++++++++++++++++++++++++++++++++-------------
> 1 files changed, 36 insertions(+), 15 deletions(-)
>
> diff --git a/arch/powerpc/mm/fault.c b/arch/powerpc/mm/fault.c
> index 2f0d1b0..a3b1176 100644
> --- a/arch/powerpc/mm/fault.c
> +++ b/arch/powerpc/mm/fault.c
> @@ -129,6 +129,7 @@ int __kprobes do_page_fault(struct pt_regs *regs, unsigned long address,
> int is_write = 0, ret;
> int trap = TRAP(regs);
> int is_exec = trap == 0x400;
> + unsigned int flags = FAULT_FLAG_ALLOW_RETRY | FAULT_FLAG_KILLABLE;
>
> #if !(defined(CONFIG_4xx) || defined(CONFIG_BOOKE))
> /*
> @@ -212,6 +213,7 @@ int __kprobes do_page_fault(struct pt_regs *regs, unsigned long address,
> if (!user_mode(regs) && !search_exception_tables(regs->nip))
> goto bad_area_nosemaphore;
>
> +retry:
> down_read(&mm->mmap_sem);
> }
>
> @@ -313,6 +315,7 @@ good_area:
> } else if (is_write) {
> if (!(vma->vm_flags & VM_WRITE))
> goto bad_area;
> + flags |= FAULT_FLAG_WRITE;
> /* a read */
> } else {
> /* protection fault */
> @@ -327,7 +330,11 @@ good_area:
> * make sure we exit gracefully rather than endlessly redo
> * the fault.
> */
> - ret = handle_mm_fault(mm, vma, address, is_write ? FAULT_FLAG_WRITE : 0);
> + ret = handle_mm_fault(mm, vma, address, flags);
> +
> + if ((fault & VM_FAULT_RETRY) && fatal_signal_pending(current))
> + return 0;
> +
> if (unlikely(ret & VM_FAULT_ERROR)) {
> if (ret & VM_FAULT_OOM)
> goto out_of_memory;
> @@ -335,22 +342,36 @@ good_area:
> goto do_sigbus;
> BUG();
> }
> - if (ret & VM_FAULT_MAJOR) {
> - current->maj_flt++;
> - perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS_MAJ, 1,
> - regs, address);
> -#ifdef CONFIG_PPC_SMLPAR
> - if (firmware_has_feature(FW_FEATURE_CMO)) {
> - preempt_disable();
> - get_lppaca()->page_ins += (1 << PAGE_FACTOR);
> - preempt_enable();
> + if (flags & FAULT_FLAG_ALLOW_RETRY) {
> + if (ret & VM_FAULT_MAJOR) {
> + current->maj_flt++;
> + perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS_MAJ, 1,
> + regs, address);
> +#ifdef CONFIG_PPC_SMLPAR
> + if (firmware_has_feature(FW_FEATURE_CMO)) {
> + preempt_disable();
> + get_lppaca()->page_ins += (1 << PAGE_FACTOR);
> + preempt_enable();
> + }
> +#endif
> + } else {
> + current->min_flt++;
> + perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS_MIN, 1,
> + regs, address);
> + }
> + if (fault & VM_FAULT_RETRY) {
> + flags &= ~FAULT_FLAG_ALLOW_RETRY;
> +
> + /*
> + * No need to up_read(&mm->mmap_sem) as we would
> + * have already released it in __lock_page_or_retry
> + * in mm/filemap.c.
> + */
> +
> + goto retry;
> }
> -#endif
> - } else {
> - current->min_flt++;
> - perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS_MIN, 1,
> - regs, address);
> }
> +
> up_read(&mm->mmap_sem);
> return 0;
>
next prev parent reply other threads:[~2012-03-20 21:29 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-03-20 16:01 [PATCH 04/20 v3] powerpc/mm/fault.c: Port OOM changes to do_page_fault Kautuk Consul
2012-03-20 21:28 ` Benjamin Herrenschmidt [this message]
2012-03-20 21:29 ` Benjamin Herrenschmidt
2012-03-21 3:27 ` Kautuk Consul
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=1332278936.2982.23.camel@pasglop \
--to=benh@kernel.crashing.org \
--cc=a.p.zijlstra@chello.nl \
--cc=christian.dietrich@informatik.uni-erlangen.de \
--cc=consul.kautuk@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linuxppc-dev@lists.ozlabs.org \
--cc=mingo@elte.hu \
--cc=mohdfarisq2010@gmail.com \
--cc=paulus@samba.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 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).