Skip to contentSkip to conversation

Future of Linux 2.6.22.y series

DISCUSSION Greg Kroah-Hartman 4 participants 4 messages Nov 5, 2007 – Nov 6, 2007 Block layer

lore
Share & tools
lore
#0 lore raw

For the last release, I stated that I thought the 2.6.22.12 release would be the last one in the 2.6.22.y series. Since then, I've received a number of other patches that would be nice to have in the .22.y tree.

So, for a while, I'll keep the 2.6.22.y tree open, doing new releases every once in a while as they accumulate. I do this, for no other than the selfish reason that I use it every day on my openSuSE 10.3 boxes as that is the kernel base that release is on :)

If anyone has any objections or questions about this, please let me know.

thanks,

greg k-h

#1 lore raw
Quoted 8 lines

For the last release, I stated that I thought the 2.6.22.12 release would be the last one in the 2.6.22.y series. Since then, I've received a number of other patches that would be nice to have in the .22.y tree.

So, for a while, I'll keep the 2.6.22.y tree open, doing new releases every once in a while as they accumulate. I do this, for no other than the selfish reason that I use it every day on my openSuSE 10.3 boxes as that is the kernel base that release is on :)

In that case you might be interested in picking up the attached fix :)
I don't know if you already didn't get it...

Honza

Signature · 2 lines
Jan Kara <[email protected]>
SuSE CR Labs

Subject: [PATCH] rd: fix data corruption on memory pressure
From: Christian Borntraeger <[email protected]>

We have seen ramdisk based install systems, where some pages of mapped libraries and programs were suddendly zeroed under memory pressure. This should not happen, as the ramdisk avoids freeing its pages by keeping them dirty all the time.

It turns out that there is a case, where the VM makes a ramdisk page clean, without telling the ramdisk driver.
On memory pressure shrink_zone runs and it starts to run shrink_active_list. There is a check for buffer_heads_over_limit, and if true, pagevec_strip is called. pagevec_strip calls try_to_release_page. If the mapping has no releasepage callback, try_to_free_buffers is called. try_to_free_buffers has now a special logic for some file systems to make a dirty page clean, if all buffers are clean. Thats what happened in our test case.

The simplest solution is to provide a noop-releasepage callback for the ramdisk driver. This avoids try_to_free_buffers for ramdisk pages.

---
 drivers/block/rd.c |   13 +++++++++++++
 1 file changed, 13 insertions(+)
drivers/block/rd.c +13 -0
Index: linux-2.6/drivers/block/rd.c
===================================================================
@@ -189,6 +189,18 @@ static int ramdisk_set_page_dirty(struct
 	return 0;
 }
 
+/*
+ * releasepage is called by pagevec_strip/try_to_release_page if
+ * buffers_heads_over_limit is true. Without a releasepage function
+ * try_to_free_buffers is called instead. That can unset the dirty
+ * bit of our ram disk pages, which will be eventually freed, even
+ * if the page is still in use.
+ */
+static int ramdisk_releasepage(struct page *page, gfp_t dummy)
+{
+	return 0;
+}
+
 static const struct address_space_operations ramdisk_aops = {
 	.readpage	= ramdisk_readpage,
 	.prepare_write	= ramdisk_prepare_write,
@@ -196,6 +208,7 @@ static const struct address_space_operat
 	.writepage	= ramdisk_writepage,
 	.set_page_dirty	= ramdisk_set_page_dirty,
 	.writepages	= ramdisk_writepages,
+	.releasepage	= ramdisk_releasepage,
 };
 
 static int rd_blkdev_pagecache_IO(int rw, struct bio_vec *vec, sector_t sector,
#2 lore raw

Am Dienstag, 6. November 2007 schrieb Jan Kara:

Quoted 2 lines

In that case you might be interested in picking up the attached fix :)
I don't know if you already didn't get it...

---
[PATCH] rd: fix data corruption on memory pressure
---

The patch is currently in mm but not yet upstream. As soon as it is merged, this patch is a good candidate for stable.

Christian

#3 lore raw
In reply to Greg Kroah-Hartman
Quoted 8 lines · Greg Kroah-Hartman

For the last release, I stated that I thought the 2.6.22.12 release would be the last one in the 2.6.22.y series. Since then, I've received a number of other patches that would be nice to have in the .22.y tree.

So, for a while, I'll keep the 2.6.22.y tree open, doing new releases every once in a while as they accumulate. I do this, for no other than the selfish reason that I use it every day on my openSuSE 10.3 boxes as that is the kernel base that release is on :)

It certainly is one of the most valuable reasons. If all people were "selfish" enough to contribute what they do only for themselves, we would possibly have less out-of-tree code.

Quoted 2 lines

If anyone has any objections or questions about this, please let me know.

Oh no, please keep it going on for as long as you want, I'll rebase my version on it instead of 2.6.20.y ;-)

Thanks,
Willy