linux-fsdevel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Bharata B Rao <bharata@in.ibm.com>
To: Suparna Bhattacharya <suparna@in.ibm.com>
Cc: linux-aio@kvack.org, akpm@osdl.org, drepper@redhat.com,
	linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org,
	jakub@redhat.com, mingo@elte.hu,
	"Sébastien Dugué" <sebastien.dugue@bull.net>
Subject: [PATCHSET 3][PATCH 1/5][AIO] - Rework compat_sys_io_submit
Date: Thu, 4 Jan 2007 15:00:55 +0530	[thread overview]
Message-ID: <20070104093055.GE9608@in.ibm.com> (raw)
In-Reply-To: <20070104092733.GD9608@in.ibm.com>

[-- Attachment #1: Type: text/plain, Size: 619 bytes --]


                 compat_sys_io_submit() cleanup

  Cleanup compat_sys_io_submit by duplicating some of the native syscall
logic in the compat layer and directly calling io_submit_one() instead
of fooling the syscall into thinking it is called from a native 64-bit
caller.

  This eliminates:

  - the overhead of copying the nr iocb pointers on the userspace stack

  - the PAGE_SIZE/(sizeof(void *)) limit on the number of iocbs that
    can be submitted.

  This is also needed for the completion notification patch to avoid having
to rewrite each iocb on the caller stack for io_submit_one() to find the
sigevents.

[-- Attachment #2: rework-compat-sys-io-submit.patch --]
[-- Type: text/plain, Size: 2607 bytes --]


From: Sébastien Dugué <sebastien.dugue@bull.net>

  Cleanup compat_sys_io_submit by duplicating some of the native syscall
logic in the compat layer and directly calling io_submit_one() instead
of fooling the syscall into thinking it is called from a native 64-bit
caller.

  This eliminates:

  - the overhead of copying the nr iocb pointers on the userspace stack

  - the PAGE_SIZE/(sizeof(void *)) limit on the number of iocbs that
    can be submitted.

  This is also needed for the completion notification patch to avoid having
to rewrite each iocb on the caller stack for io_submit_one() to find the
sigevents.

Signed-off-by: Sébastien Dugué <sebastien.dugue@bull.net>
Signed-off-by: Bharata B Rao <bharata@in.ibm.com>
---

 fs/compat.c |   61 +++++++++++++++++++++++++++++++++---------------------------
 1 files changed, 34 insertions(+), 27 deletions(-)

diff -puN fs/compat.c~rework-compat-sys-io-submit fs/compat.c
--- linux-2.6.20-rc2/fs/compat.c~rework-compat-sys-io-submit	2007-01-03 10:15:03.000000000 +0530
+++ linux-2.6.20-rc2-bharata/fs/compat.c	2007-01-04 13:21:28.000000000 +0530
@@ -644,40 +644,47 @@ out:
 	return ret;
 }
 
-static inline long
-copy_iocb(long nr, u32 __user *ptr32, struct iocb __user * __user *ptr64)
-{
-	compat_uptr_t uptr;
-	int i;
-
-	for (i = 0; i < nr; ++i) {
-		if (get_user(uptr, ptr32 + i))
-			return -EFAULT;
-		if (put_user(compat_ptr(uptr), ptr64 + i))
-			return -EFAULT;
-	}
-	return 0;
-}
-
-#define MAX_AIO_SUBMITS 	(PAGE_SIZE/sizeof(struct iocb *))
-
 asmlinkage long
 compat_sys_io_submit(aio_context_t ctx_id, int nr, u32 __user *iocb)
 {
-	struct iocb __user * __user *iocb64; 
-	long ret;
+	struct kioctx *ctx;
+	long ret = 0;
+	int i;
 
 	if (unlikely(nr < 0))
 		return -EINVAL;
 
-	if (nr > MAX_AIO_SUBMITS)
-		nr = MAX_AIO_SUBMITS;
-	
-	iocb64 = compat_alloc_user_space(nr * sizeof(*iocb64));
-	ret = copy_iocb(nr, iocb, iocb64);
-	if (!ret)
-		ret = sys_io_submit(ctx_id, nr, iocb64);
-	return ret;
+	if (unlikely(!access_ok(VERIFY_READ, iocb, (nr * sizeof(u32)))))
+		return -EFAULT;
+
+	ctx = lookup_ioctx(ctx_id);
+	if (unlikely(!ctx))
+		return -EINVAL;
+
+	for (i=0; i<nr; i++) {
+		compat_uptr_t uptr;
+		struct iocb __user *user_iocb;
+		struct iocb tmp;
+
+		if (unlikely(get_user(uptr, iocb + i))) {
+			ret = -EFAULT;
+			break;
+		}
+
+		user_iocb = compat_ptr(uptr);
+
+		if (unlikely(copy_from_user(&tmp, user_iocb, sizeof(tmp)))) {
+			ret = -EFAULT;
+			break;
+		}
+
+		ret = io_submit_one(ctx, user_iocb, &tmp);
+		if (ret)
+			break;
+	}
+
+	put_ioctx(ctx);
+	return i ? i: ret;
 }
 
 struct compat_ncp_mount_data {
_

  reply	other threads:[~2007-01-04  9:23 UTC|newest]

Thread overview: 60+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-12-27 15:38 [RFC] Heads up on a series of AIO patchsets Suparna Bhattacharya
2006-12-27 16:25 ` Christoph Hellwig
2006-12-27 16:55   ` Ingo Molnar
2006-12-27 17:18     ` Ingo Molnar
2006-12-28 11:41   ` Evgeniy Polyakov
2007-01-02 21:38     ` Dan Williams
2007-01-03 13:35       ` Evgeniy Polyakov
2006-12-28  8:23 ` [PATCHSET 1][PATCH 0/6] Filesystem AIO read/write Suparna Bhattacharya
2006-12-28  8:34   ` [FSAIO][PATCH 1/6] Add a wait queue parameter to the wait_bit action routine Suparna Bhattacharya
2006-12-28  8:46     ` Suparna Bhattacharya
2006-12-28  8:36   ` [FSAIO][PATCH 2/8] Rename __lock_page to lock_page_slow Suparna Bhattacharya
2006-12-28  8:39   ` [FSAIO][PATCH 3/8] Routines to initialize and test a wait bit key Suparna Bhattacharya
2006-12-28 22:42     ` Andrew Morton
2006-12-28  8:39   ` [FSAIO][PATCH 4/8] Add a default io wait bit field in task struct Suparna Bhattacharya
2006-12-28  8:40   ` [FSAIO][PATCH 5/8] Enable wait bit based filtered wakeups to work for AIO Suparna Bhattacharya
2006-12-28  8:41   ` [FSAIO][PATCH 6/8] Enable asynchronous wait page and lock page Suparna Bhattacharya
2006-12-28 11:55     ` Christoph Hellwig
2006-12-28 14:47       ` Suparna Bhattacharya
2007-01-02 14:26         ` Christoph Hellwig
2007-01-04  6:50           ` Nick Piggin
2006-12-28  8:42   ` [FSAIO][PATCH 7/8] Filesystem AIO read Suparna Bhattacharya
2006-12-28 11:57     ` Christoph Hellwig
2006-12-28 14:15       ` Christoph Hellwig
2006-12-28 15:18       ` Suparna Bhattacharya
2007-01-02 14:29         ` Christoph Hellwig
2006-12-28 16:22       ` Jan Engelhardt
2006-12-28 16:56         ` Randy Dunlap
2006-12-28  8:44   ` [FSAIO][PATCH 8/8] AIO O_SYNC filesystem write Suparna Bhattacharya
2006-12-28  9:52   ` [PATCHSET 1][PATCH 0/6] Filesystem AIO read/write Ingo Molnar
2006-12-28 22:53   ` Andrew Morton
2007-01-03 22:15   ` Andrew Morton
2007-01-04  4:56     ` Suparna Bhattacharya
2007-01-04  5:51       ` Nick Piggin
2007-01-04  6:26         ` Suparna Bhattacharya
2007-01-04  6:50           ` Nick Piggin
2007-01-04 11:24             ` Suparna Bhattacharya
2007-01-05  4:56               ` Nick Piggin
2007-01-04 17:02       ` Andrew Morton
2007-01-04 17:49         ` Jens Axboe
2007-01-05  6:28         ` Suparna Bhattacharya
2007-01-05  7:02           ` Jens Axboe
2007-01-05  8:08             ` Suparna Bhattacharya
2007-01-05  8:32               ` Jens Axboe
2007-01-10  5:44         ` Suparna Bhattacharya
2007-01-11  1:08           ` Andrew Morton
2007-01-11  3:13             ` Suparna Bhattacharya
2007-01-11  4:52               ` Andrew Morton
2007-01-02 23:56 ` [RFC] Heads up on a series of AIO patchsets Zach Brown
     [not found]   ` <6f703f960701021640y444bc537w549fd6d74f3e9529@mail.gmail.com>
     [not found]     ` <A85B8249-FC4E-4612-8B28-02BC680DC812@oracle.com>
2007-01-03  1:18       ` Kent Overstreet
2007-01-04 20:33         ` Pavel Machek
2007-01-03  5:03   ` Suparna Bhattacharya
2007-01-05  0:36     ` Zach Brown
2007-01-03  7:23 ` [PATCHSET 2][PATCH 1/1] Combining epoll and disk file AIO Suparna Bhattacharya
2007-01-04  9:27 ` [PATCHSET 3][PATCH 0/5][AIO] - AIO completion signal notification v4 Bharata B Rao
2007-01-04  9:30   ` Bharata B Rao [this message]
2007-01-04  9:32   ` [PATCHSET 3][PATCH 2/5][AIO] - fix aio.h includes Bharata B Rao
2007-01-04  9:34   ` [PATCHSET 3][PATCH 3/5][AIO] - Make good_sigevent non-static Bharata B Rao
2007-01-04  9:38   ` [PATCHSET 3][PATCH 4/5][AIO] - AIO completion signal notification Bharata B Rao
2007-01-04  9:40   ` [PATCHSET 3][PATCH 5/5][AIO] - Add listio support Bharata B Rao
2007-01-05  5:32 ` [PATCHSET 4][PATCH 1/1] AIO fallback for pipes, sockets and pollable fds Suparna Bhattacharya

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=20070104093055.GE9608@in.ibm.com \
    --to=bharata@in.ibm.com \
    --cc=akpm@osdl.org \
    --cc=drepper@redhat.com \
    --cc=jakub@redhat.com \
    --cc=linux-aio@kvack.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=sebastien.dugue@bull.net \
    --cc=suparna@in.ibm.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 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).