From: ebiederm@xmission.com (Eric W. Biederman)
To: "H. Peter Anvin" <hpa@zytor.com>
Cc: riel@redhat.com, hughd@google.com, kexec@lists.infradead.org,
linux-kernel@vger.kernel.org, lisa.mitchell@hp.com,
linux-mm@kvack.org, HATAYAMA Daisuke <d.hatayama@jp.fujitsu.com>,
kumagai-atsushi@mxc.nes.nec.co.jp, jingbai.ma@hp.com,
kosaki.motohiro@jp.fujitsu.com, zhangyanfei@cn.fujitsu.com,
akpm@linux-foundation.org, walken@google.com, cpw@sgi.com,
vgoyal@redhat.com
Subject: Re: [PATCH v6 0/8] kdump, vmcore: support mmap() on /proc/vmcore
Date: Thu, 16 May 2013 19:53:31 -0700 [thread overview]
Message-ID: <87y5bee2qc.fsf@xmission.com> (raw)
In-Reply-To: <51957469.2000008@zytor.com> (H. Peter Anvin's message of "Thu, 16 May 2013 17:06:01 -0700")
"H. Peter Anvin" <hpa@zytor.com> writes:
> On 05/15/2013 02:05 AM, HATAYAMA Daisuke wrote:
>> Currently, read to /proc/vmcore is done by read_oldmem() that uses
>> ioremap/iounmap per a single page. For example, if memory is 1GB,
>> ioremap/iounmap is called (1GB / 4KB)-times, that is, 262144
>> times. This causes big performance degradation.
>
> read_oldmem() is fundamentally broken and unsafe. It needs to be
> unified with the plain /dev/mem code and any missing functionality fixed
> instead of "let's just do a whole new driver".
That is completely and totally orthogonal to this change.
read_oldmem may have problems but in practice on a large systems those
problems are totally dwarfed by real life performance issues that come
from playing too much with the page tables.
I really don't find bringing up whatever foundational issues you have
with read_oldmem() appropriate or relevant here.
Eric
_______________________________________________
kexec mailing list
kexec@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/kexec
WARNING: multiple messages have this Message-ID (diff)
From: ebiederm@xmission.com (Eric W. Biederman)
To: "H. Peter Anvin" <hpa@zytor.com>
Cc: HATAYAMA Daisuke <d.hatayama@jp.fujitsu.com>,
vgoyal@redhat.com, akpm@linux-foundation.org, cpw@sgi.com,
kumagai-atsushi@mxc.nes.nec.co.jp, lisa.mitchell@hp.com,
kexec@lists.infradead.org, linux-kernel@vger.kernel.org,
zhangyanfei@cn.fujitsu.com, jingbai.ma@hp.com,
linux-mm@kvack.org, riel@redhat.com, walken@google.com,
hughd@google.com, kosaki.motohiro@jp.fujitsu.com
Subject: Re: [PATCH v6 0/8] kdump, vmcore: support mmap() on /proc/vmcore
Date: Thu, 16 May 2013 19:53:31 -0700 [thread overview]
Message-ID: <87y5bee2qc.fsf@xmission.com> (raw)
In-Reply-To: <51957469.2000008@zytor.com> (H. Peter Anvin's message of "Thu, 16 May 2013 17:06:01 -0700")
"H. Peter Anvin" <hpa@zytor.com> writes:
> On 05/15/2013 02:05 AM, HATAYAMA Daisuke wrote:
>> Currently, read to /proc/vmcore is done by read_oldmem() that uses
>> ioremap/iounmap per a single page. For example, if memory is 1GB,
>> ioremap/iounmap is called (1GB / 4KB)-times, that is, 262144
>> times. This causes big performance degradation.
>
> read_oldmem() is fundamentally broken and unsafe. It needs to be
> unified with the plain /dev/mem code and any missing functionality fixed
> instead of "let's just do a whole new driver".
That is completely and totally orthogonal to this change.
read_oldmem may have problems but in practice on a large systems those
problems are totally dwarfed by real life performance issues that come
from playing too much with the page tables.
I really don't find bringing up whatever foundational issues you have
with read_oldmem() appropriate or relevant here.
Eric
--
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: ebiederm@xmission.com (Eric W. Biederman)
To: "H. Peter Anvin" <hpa@zytor.com>
Cc: HATAYAMA Daisuke <d.hatayama@jp.fujitsu.com>,
vgoyal@redhat.com, akpm@linux-foundation.org, cpw@sgi.com,
kumagai-atsushi@mxc.nes.nec.co.jp, lisa.mitchell@hp.com,
kexec@lists.infradead.org, linux-kernel@vger.kernel.org,
zhangyanfei@cn.fujitsu.com, jingbai.ma@hp.com,
linux-mm@kvack.org, riel@redhat.com, walken@google.com,
hughd@google.com, kosaki.motohiro@jp.fujitsu.com
Subject: Re: [PATCH v6 0/8] kdump, vmcore: support mmap() on /proc/vmcore
Date: Thu, 16 May 2013 19:53:31 -0700 [thread overview]
Message-ID: <87y5bee2qc.fsf@xmission.com> (raw)
In-Reply-To: <51957469.2000008@zytor.com> (H. Peter Anvin's message of "Thu, 16 May 2013 17:06:01 -0700")
"H. Peter Anvin" <hpa@zytor.com> writes:
> On 05/15/2013 02:05 AM, HATAYAMA Daisuke wrote:
>> Currently, read to /proc/vmcore is done by read_oldmem() that uses
>> ioremap/iounmap per a single page. For example, if memory is 1GB,
>> ioremap/iounmap is called (1GB / 4KB)-times, that is, 262144
>> times. This causes big performance degradation.
>
> read_oldmem() is fundamentally broken and unsafe. It needs to be
> unified with the plain /dev/mem code and any missing functionality fixed
> instead of "let's just do a whole new driver".
That is completely and totally orthogonal to this change.
read_oldmem may have problems but in practice on a large systems those
problems are totally dwarfed by real life performance issues that come
from playing too much with the page tables.
I really don't find bringing up whatever foundational issues you have
with read_oldmem() appropriate or relevant here.
Eric
next prev parent reply other threads:[~2013-05-17 2:54 UTC|newest]
Thread overview: 84+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-05-15 9:05 [PATCH v6 0/8] kdump, vmcore: support mmap() on /proc/vmcore HATAYAMA Daisuke
2013-05-15 9:05 ` HATAYAMA Daisuke
2013-05-15 9:05 ` HATAYAMA Daisuke
2013-05-15 9:05 ` [PATCH v6 1/8] vmcore: clean up read_vmcore() HATAYAMA Daisuke
2013-05-15 9:05 ` HATAYAMA Daisuke
2013-05-15 9:05 ` HATAYAMA Daisuke
2013-05-15 9:33 ` Zhang Yanfei
2013-05-15 9:33 ` Zhang Yanfei
2013-05-15 9:33 ` Zhang Yanfei
2013-05-15 9:05 ` [PATCH v6 2/8] vmcore: allocate buffer for ELF headers on page-size alignment HATAYAMA Daisuke
2013-05-15 9:05 ` HATAYAMA Daisuke
2013-05-15 9:05 ` HATAYAMA Daisuke
2013-05-16 5:58 ` Zhang Yanfei
2013-05-16 5:58 ` Zhang Yanfei
2013-05-16 5:58 ` Zhang Yanfei
2013-05-16 16:51 ` Vivek Goyal
2013-05-16 16:51 ` Vivek Goyal
2013-05-16 16:51 ` Vivek Goyal
2013-05-17 0:08 ` HATAYAMA Daisuke
2013-05-17 0:08 ` HATAYAMA Daisuke
2013-05-17 0:08 ` HATAYAMA Daisuke
2013-05-15 9:05 ` [PATCH v6 3/8] vmcore: treat memory chunks referenced by PT_LOAD program header entries in page-size boundary in vmcore_list HATAYAMA Daisuke
2013-05-15 9:05 ` HATAYAMA Daisuke
2013-05-15 9:05 ` HATAYAMA Daisuke
2013-05-16 5:59 ` Zhang Yanfei
2013-05-16 5:59 ` Zhang Yanfei
2013-05-16 5:59 ` Zhang Yanfei
2013-05-15 9:06 ` [PATCH v6 4/8] vmalloc: make find_vm_area check in range HATAYAMA Daisuke
2013-05-15 9:06 ` HATAYAMA Daisuke
2013-05-15 9:06 ` HATAYAMA Daisuke
2013-05-15 21:37 ` KOSAKI Motohiro
2013-05-15 21:37 ` KOSAKI Motohiro
2013-05-15 21:37 ` KOSAKI Motohiro
2013-05-16 23:45 ` HATAYAMA Daisuke
2013-05-16 23:45 ` HATAYAMA Daisuke
2013-05-16 23:45 ` HATAYAMA Daisuke
2013-05-15 9:06 ` [PATCH v6 5/8] vmalloc: introduce remap_vmalloc_range_partial HATAYAMA Daisuke
2013-05-15 9:06 ` HATAYAMA Daisuke
2013-05-15 9:06 ` HATAYAMA Daisuke
2013-05-15 9:06 ` [PATCH v6 6/8] vmcore: allocate ELF note segment in the 2nd kernel vmalloc memory HATAYAMA Daisuke
2013-05-15 9:06 ` HATAYAMA Daisuke
2013-05-15 9:06 ` HATAYAMA Daisuke
2013-05-16 7:19 ` Zhang Yanfei
2013-05-16 7:19 ` Zhang Yanfei
2013-05-16 7:19 ` Zhang Yanfei
2013-05-16 20:32 ` Vivek Goyal
2013-05-16 20:32 ` Vivek Goyal
2013-05-16 20:32 ` Vivek Goyal
2013-05-16 23:47 ` HATAYAMA Daisuke
2013-05-16 23:47 ` HATAYAMA Daisuke
2013-05-16 23:47 ` HATAYAMA Daisuke
2013-05-15 9:06 ` [PATCH v6 7/8] vmcore: calculate vmcore file size from buffer size and total size of vmcore objects HATAYAMA Daisuke
2013-05-15 9:06 ` HATAYAMA Daisuke
2013-05-15 9:06 ` HATAYAMA Daisuke
2013-05-16 7:19 ` Zhang Yanfei
2013-05-16 7:19 ` Zhang Yanfei
2013-05-16 7:19 ` Zhang Yanfei
2013-05-15 9:06 ` [PATCH v6 8/8] vmcore: support mmap() on /proc/vmcore HATAYAMA Daisuke
2013-05-15 9:06 ` HATAYAMA Daisuke
2013-05-15 9:06 ` HATAYAMA Daisuke
2013-05-16 7:25 ` Zhang Yanfei
2013-05-16 7:25 ` Zhang Yanfei
2013-05-16 7:25 ` Zhang Yanfei
2013-05-16 20:44 ` Vivek Goyal
2013-05-16 20:44 ` Vivek Goyal
2013-05-16 20:44 ` Vivek Goyal
2013-05-17 0:06 ` [PATCH v6 0/8] kdump, " H. Peter Anvin
2013-05-17 0:06 ` H. Peter Anvin
2013-05-17 0:06 ` H. Peter Anvin
2013-05-17 1:45 ` HATAYAMA Daisuke
2013-05-17 1:45 ` HATAYAMA Daisuke
2013-05-17 1:45 ` HATAYAMA Daisuke
2013-05-17 2:53 ` Eric W. Biederman [this message]
2013-05-17 2:53 ` Eric W. Biederman
2013-05-17 2:53 ` Eric W. Biederman
2013-05-17 3:21 ` H. Peter Anvin
2013-05-17 3:21 ` H. Peter Anvin
2013-05-17 3:21 ` H. Peter Anvin
2013-05-17 4:29 ` Eric W. Biederman
2013-05-17 4:29 ` Eric W. Biederman
2013-05-17 4:29 ` Eric W. Biederman
2013-05-17 5:43 ` H. Peter Anvin
2013-05-17 5:43 ` H. Peter Anvin
2013-05-17 5:43 ` H. Peter Anvin
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=87y5bee2qc.fsf@xmission.com \
--to=ebiederm@xmission.com \
--cc=akpm@linux-foundation.org \
--cc=cpw@sgi.com \
--cc=d.hatayama@jp.fujitsu.com \
--cc=hpa@zytor.com \
--cc=hughd@google.com \
--cc=jingbai.ma@hp.com \
--cc=kexec@lists.infradead.org \
--cc=kosaki.motohiro@jp.fujitsu.com \
--cc=kumagai-atsushi@mxc.nes.nec.co.jp \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=lisa.mitchell@hp.com \
--cc=riel@redhat.com \
--cc=vgoyal@redhat.com \
--cc=walken@google.com \
--cc=zhangyanfei@cn.fujitsu.com \
/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.