All of lore.kernel.org
 help / color / mirror / Atom feed
From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	stable@vger.kernel.org, Andreas Dilger <adilger@dilger.ca>,
	Roman Gushchin <guro@fb.com>, Theodore Tso <tytso@mit.edu>
Subject: [PATCH 4.19 04/64] ext4: use non-movable memory for superblock readahead
Date: Wed, 22 Apr 2020 11:56:48 +0200	[thread overview]
Message-ID: <20200422095013.021182139@linuxfoundation.org> (raw)
In-Reply-To: <20200422095008.799686511@linuxfoundation.org>

From: Roman Gushchin <guro@fb.com>

commit d87f639258a6a5980183f11876c884931ad93da2 upstream.

Since commit a8ac900b8163 ("ext4: use non-movable memory for the
superblock") buffers for ext4 superblock were allocated using
the sb_bread_unmovable() helper which allocated buffer heads
out of non-movable memory blocks. It was necessarily to not block
page migrations and do not cause cma allocation failures.

However commit 85c8f176a611 ("ext4: preload block group descriptors")
broke this by introducing pre-reading of the ext4 superblock.
The problem is that __breadahead() is using __getblk() underneath,
which allocates buffer heads out of movable memory.

It resulted in page migration failures I've seen on a machine
with an ext4 partition and a preallocated cma area.

Fix this by introducing sb_breadahead_unmovable() and
__breadahead_gfp() helpers which use non-movable memory for buffer
head allocations and use them for the ext4 superblock readahead.

Reviewed-by: Andreas Dilger <adilger@dilger.ca>
Fixes: 85c8f176a611 ("ext4: preload block group descriptors")
Signed-off-by: Roman Gushchin <guro@fb.com>
Link: https://lore.kernel.org/r/20200229001411.128010-1-guro@fb.com
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

---
 fs/buffer.c                 |   11 +++++++++++
 fs/ext4/inode.c             |    2 +-
 fs/ext4/super.c             |    2 +-
 include/linux/buffer_head.h |    8 ++++++++
 4 files changed, 21 insertions(+), 2 deletions(-)

--- a/fs/buffer.c
+++ b/fs/buffer.c
@@ -1336,6 +1336,17 @@ void __breadahead(struct block_device *b
 }
 EXPORT_SYMBOL(__breadahead);
 
+void __breadahead_gfp(struct block_device *bdev, sector_t block, unsigned size,
+		      gfp_t gfp)
+{
+	struct buffer_head *bh = __getblk_gfp(bdev, block, size, gfp);
+	if (likely(bh)) {
+		ll_rw_block(REQ_OP_READ, REQ_RAHEAD, 1, &bh);
+		brelse(bh);
+	}
+}
+EXPORT_SYMBOL(__breadahead_gfp);
+
 /**
  *  __bread_gfp() - reads a specified block and returns the bh
  *  @bdev: the block_device to read from
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -4690,7 +4690,7 @@ make_io:
 			if (end > table)
 				end = table;
 			while (b <= end)
-				sb_breadahead(sb, b++);
+				sb_breadahead_unmovable(sb, b++);
 		}
 
 		/*
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -4207,7 +4207,7 @@ static int ext4_fill_super(struct super_
 	/* Pre-read the descriptors into the buffer cache */
 	for (i = 0; i < db_count; i++) {
 		block = descriptor_loc(sb, logical_sb_block, i);
-		sb_breadahead(sb, block);
+		sb_breadahead_unmovable(sb, block);
 	}
 
 	for (i = 0; i < db_count; i++) {
--- a/include/linux/buffer_head.h
+++ b/include/linux/buffer_head.h
@@ -189,6 +189,8 @@ struct buffer_head *__getblk_gfp(struct
 void __brelse(struct buffer_head *);
 void __bforget(struct buffer_head *);
 void __breadahead(struct block_device *, sector_t block, unsigned int size);
+void __breadahead_gfp(struct block_device *, sector_t block, unsigned int size,
+		  gfp_t gfp);
 struct buffer_head *__bread_gfp(struct block_device *,
 				sector_t block, unsigned size, gfp_t gfp);
 void invalidate_bh_lrus(void);
@@ -319,6 +321,12 @@ sb_breadahead(struct super_block *sb, se
 	__breadahead(sb->s_bdev, block, sb->s_blocksize);
 }
 
+static inline void
+sb_breadahead_unmovable(struct super_block *sb, sector_t block)
+{
+	__breadahead_gfp(sb->s_bdev, block, sb->s_blocksize, 0);
+}
+
 static inline struct buffer_head *
 sb_getblk(struct super_block *sb, sector_t block)
 {



  parent reply	other threads:[~2020-04-22 10:49 UTC|newest]

Thread overview: 73+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-04-22  9:56 [PATCH 4.19 00/64] 4.19.118-rc1 review Greg Kroah-Hartman
2020-04-22  9:56 ` [PATCH 4.19 01/64] arm, bpf: Fix offset overflow for BPF_MEM BPF_DW Greg Kroah-Hartman
2020-04-22  9:56 ` [PATCH 4.19 02/64] objtool: Fix switch table detection in .text.unlikely Greg Kroah-Hartman
2020-04-22  9:56 ` [PATCH 4.19 03/64] scsi: sg: add sg_remove_request in sg_common_write Greg Kroah-Hartman
2020-04-22  9:56 ` Greg Kroah-Hartman [this message]
2020-04-22  9:56 ` [PATCH 4.19 05/64] watchdog: sp805: fix restart handler Greg Kroah-Hartman
2020-04-22  9:56 ` [PATCH 4.19 06/64] arm, bpf: Fix bugs with ALU64 {RSH, ARSH} BPF_K shift by 0 Greg Kroah-Hartman
2020-04-22  9:56 ` [PATCH 4.19 07/64] ARM: dts: imx6: Use gpc for FEC interrupt controller to fix wake on LAN Greg Kroah-Hartman
2020-04-22  9:56 ` [PATCH 4.19 08/64] netfilter: nf_tables: report EOPNOTSUPP on unsupported flags/object type Greg Kroah-Hartman
2020-04-22  9:56 ` [PATCH 4.19 09/64] irqchip/mbigen: Free msi_desc on device teardown Greg Kroah-Hartman
2020-04-22  9:56 ` [PATCH 4.19 10/64] ALSA: hda: Dont release card at firmware loading error Greg Kroah-Hartman
2020-04-22  9:56 ` [PATCH 4.19 11/64] of: unittest: kmemleak on changeset destroy Greg Kroah-Hartman
2020-04-22  9:56 ` [PATCH 4.19 12/64] of: unittest: kmemleak in of_unittest_platform_populate() Greg Kroah-Hartman
2020-04-22  9:56 ` [PATCH 4.19 13/64] of: unittest: kmemleak in of_unittest_overlay_high_level() Greg Kroah-Hartman
2020-04-22  9:56 ` [PATCH 4.19 14/64] of: overlay: kmemleak in dup_and_fixup_symbol_prop() Greg Kroah-Hartman
2020-04-22  9:56 ` [PATCH 4.19 15/64] x86/Hyper-V: Report crash register data or kmsg before running crash kernel Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 16/64] lib/raid6: use vdupq_n_u8 to avoid endianness warnings Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 17/64] video: fbdev: sis: Remove unnecessary parentheses and commented code Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 18/64] rbd: avoid a deadlock on header_rwsem when flushing notifies Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 19/64] rbd: call rbd_dev_unprobe() after unwatching and " Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 20/64] xsk: Add missing check on user supplied headroom size Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 21/64] x86/Hyper-V: Unload vmbus channel in hv panic callback Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 22/64] x86/Hyper-V: Free hv_panic_page when fail to register kmsg dump Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 23/64] x86/Hyper-V: Trigger crash enlightenment only once during system crash Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 24/64] x86/Hyper-V: Report crash register data when sysctl_record_panic_msg is not set Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 25/64] x86/Hyper-V: Report crash data in die() when panic_on_oops is set Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 26/64] clk: at91: usb: continue if clk_hw_round_rate() return zero Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 27/64] power: supply: bq27xxx_battery: Silence deferred-probe error Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 28/64] clk: tegra: Fix Tegra PMC clock out parents Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 29/64] soc: imx: gpc: fix power up sequencing Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 30/64] rtc: 88pm860x: fix possible race condition Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 31/64] NFSv4/pnfs: Return valid stateids in nfs_layout_find_inode_by_stateid() Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 32/64] NFS: direct.c: Fix memory leak of dreq when nfs_get_lock_context fails Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 33/64] s390/cpuinfo: fix wrong output when CPU0 is offline Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 34/64] powerpc/maple: Fix declaration made after definition Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 35/64] s390/cpum_sf: Fix wrong page count in error message Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 36/64] ext4: do not commit super on read-only bdev Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 37/64] um: ubd: Prevent buffer overrun on command completion Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 38/64] cifs: Allocate encryption header through kmalloc Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 39/64] include/linux/swapops.h: correct guards for non_swap_entry() Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 40/64] percpu_counter: fix a data race at vm_committed_as Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 41/64] compiler.h: fix error in BUILD_BUG_ON() reporting Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 42/64] KVM: s390: vsie: Fix possible race when shadowing region 3 tables Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 43/64] x86: ACPI: fix CPU hotplug deadlock Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 44/64] drm/amdkfd: kfree the wrong pointer Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 45/64] NFS: Fix memory leaks in nfs_pageio_stop_mirroring() Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 46/64] f2fs: fix NULL pointer dereference in f2fs_write_begin() Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 47/64] drm/vc4: Fix HDMI mode validation Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 48/64] iommu/vt-d: Fix mm reference leak Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 49/64] ext2: fix empty body warnings when -Wextra is used Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 50/64] ext2: fix debug reference to ext2_xattr_cache Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 51/64] power: supply: axp288_fuel_gauge: Broaden vendor check for Intel Compute Sticks Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 52/64] libnvdimm: Out of bounds read in __nd_ioctl() Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 53/64] iommu/amd: Fix the configuration of GCR3 table root pointer Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 54/64] f2fs: fix to wait all node page writeback Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 55/64] net: dsa: bcm_sf2: Fix overflow checks Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 56/64] fbdev: potential information leak in do_fb_ioctl() Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 57/64] iio: si1133: read 24-bit signed integer for measurement Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 58/64] tty: evh_bytechan: Fix out of bounds accesses Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 59/64] locktorture: Print ratio of acquisitions, not failures Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 60/64] mtd: spinand: Explicitly use MTD_OPS_RAW to write the bad block marker to OOB Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 61/64] mtd: lpddr: Fix a double free in probe() Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 62/64] mtd: phram: fix a double free issue in error path Greg Kroah-Hartman
2020-04-22  9:57   ` Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 63/64] KEYS: Dont write out to userspace while holding key semaphore Greg Kroah-Hartman
2020-04-22  9:57 ` [PATCH 4.19 64/64] bpf: fix buggy r0 retval refinement for tracing helpers Greg Kroah-Hartman
2020-04-22 11:10 ` [PATCH 4.19 00/64] 4.19.118-rc1 review Chris Paterson
2020-04-22 12:51   ` Greg Kroah-Hartman
2020-04-22 20:35 ` Guenter Roeck
2020-04-23  8:22 ` Naresh Kamboju
     [not found] ` <20200422095008.799686511-hQyY1W1yCW8ekmWlsbkhG0B+6BGkLq7r@public.gmane.org>
2020-04-23 10:22   ` Jon Hunter
2020-04-23 10:22     ` Jon Hunter
2020-04-24 16:39 ` shuah

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=20200422095013.021182139@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=adilger@dilger.ca \
    --cc=guro@fb.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=stable@vger.kernel.org \
    --cc=tytso@mit.edu \
    /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.