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, Kirill Smelkov <kirr@nexedi.com>,
	Han-Wen Nienhuys <hanwen@google.com>,
	Jakob Unterwurzacher <jakobunt@gmail.com>,
	Miklos Szeredi <mszeredi@redhat.com>,
	Sasha Levin <sashal@kernel.org>
Subject: [PATCH 4.14 43/81] fuse: retrieve: cap requested size to negotiated max_write
Date: Thu, 13 Jun 2019 10:33:26 +0200	[thread overview]
Message-ID: <20190613075652.485457686@linuxfoundation.org> (raw)
In-Reply-To: <20190613075649.074682929@linuxfoundation.org>

[ Upstream commit 7640682e67b33cab8628729afec8ca92b851394f ]

FUSE filesystem server and kernel client negotiate during initialization
phase, what should be the maximum write size the client will ever issue.
Correspondingly the filesystem server then queues sys_read calls to read
requests with buffer capacity large enough to carry request header + that
max_write bytes. A filesystem server is free to set its max_write in
anywhere in the range between [1*page, fc->max_pages*page]. In particular
go-fuse[2] sets max_write by default as 64K, wheres default fc->max_pages
corresponds to 128K. Libfuse also allows users to configure max_write, but
by default presets it to possible maximum.

If max_write is < fc->max_pages*page, and in NOTIFY_RETRIEVE handler we
allow to retrieve more than max_write bytes, corresponding prepared
NOTIFY_REPLY will be thrown away by fuse_dev_do_read, because the
filesystem server, in full correspondence with server/client contract, will
be only queuing sys_read with ~max_write buffer capacity, and
fuse_dev_do_read throws away requests that cannot fit into server request
buffer. In turn the filesystem server could get stuck waiting indefinitely
for NOTIFY_REPLY since NOTIFY_RETRIEVE handler returned OK which is
understood by clients as that NOTIFY_REPLY was queued and will be sent
back.

Cap requested size to negotiate max_write to avoid the problem.  This
aligns with the way NOTIFY_RETRIEVE handler works, which already
unconditionally caps requested retrieve size to fuse_conn->max_pages.  This
way it should not hurt NOTIFY_RETRIEVE semantic if we return less data than
was originally requested.

Please see [1] for context where the problem of stuck filesystem was hit
for real, how the situation was traced and for more involving patch that
did not make it into the tree.

[1] https://marc.info/?l=linux-fsdevel&m=155057023600853&w=2
[2] https://github.com/hanwen/go-fuse

Signed-off-by: Kirill Smelkov <kirr@nexedi.com>
Cc: Han-Wen Nienhuys <hanwen@google.com>
Cc: Jakob Unterwurzacher <jakobunt@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 fs/fuse/dev.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

--- a/fs/fuse/dev.c
+++ b/fs/fuse/dev.c
@@ -1678,7 +1678,7 @@ static int fuse_retrieve(struct fuse_con
 	offset = outarg->offset & ~PAGE_MASK;
 	file_size = i_size_read(inode);
 
-	num = outarg->size;
+	num = min(outarg->size, fc->max_write);
 	if (outarg->offset > file_size)
 		num = 0;
 	else if (outarg->offset + num > file_size)



  parent reply	other threads:[~2019-06-13 16:23 UTC|newest]

Thread overview: 107+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-06-13  8:32 [PATCH 4.14 00/81] 4.14.126-stable review Greg Kroah-Hartman
2019-06-13  8:32 ` [PATCH 4.14 01/81] rapidio: fix a NULL pointer dereference when create_workqueue() fails Greg Kroah-Hartman
2019-06-13  8:32 ` [PATCH 4.14 02/81] fs/fat/file.c: issue flush after the writeback of FAT Greg Kroah-Hartman
2019-06-13  8:32 ` [PATCH 4.14 03/81] sysctl: return -EINVAL if val violates minmax Greg Kroah-Hartman
2019-06-13  8:32 ` [PATCH 4.14 04/81] ipc: prevent lockup on alloc_msg and free_msg Greg Kroah-Hartman
2019-06-13  8:32 ` [PATCH 4.14 05/81] ARM: prevent tracing IPI_CPU_BACKTRACE Greg Kroah-Hartman
2019-06-13  8:32 ` [PATCH 4.14 06/81] mm/hmm: select mmu notifier when selecting HMM Greg Kroah-Hartman
2019-06-13  8:32 ` [PATCH 4.14 07/81] hugetlbfs: on restore reserve error path retain subpool reservation Greg Kroah-Hartman
2019-06-13  8:32 ` [PATCH 4.14 08/81] mem-hotplug: fix node spanned pages when we have a node with only ZONE_MOVABLE Greg Kroah-Hartman
2019-06-13  8:32 ` [PATCH 4.14 09/81] mm/cma.c: fix crash on CMA allocation if bitmap allocation fails Greg Kroah-Hartman
2019-06-13  8:32 ` [PATCH 4.14 10/81] mm/cma.c: fix the bitmap status to show failed allocation reason Greg Kroah-Hartman
2019-06-13  8:32 ` [PATCH 4.14 11/81] mm/cma_debug.c: fix the break condition in cma_maxchunk_get() Greg Kroah-Hartman
2019-06-13  8:32 ` [PATCH 4.14 12/81] mm/slab.c: fix an infinite loop in leaks_show() Greg Kroah-Hartman
2019-06-13  8:32 ` [PATCH 4.14 13/81] kernel/sys.c: prctl: fix false positive in validate_prctl_map() Greg Kroah-Hartman
2019-06-13  8:32 ` [PATCH 4.14 14/81] thermal: rcar_gen3_thermal: disable interrupt in .remove Greg Kroah-Hartman
2019-06-13  8:32 ` [PATCH 4.14 15/81] drivers: thermal: tsens: Dont print error message on -EPROBE_DEFER Greg Kroah-Hartman
2019-06-13  8:32 ` [PATCH 4.14 16/81] mfd: tps65912-spi: Add missing of table registration Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 17/81] mfd: intel-lpss: Set the device in reset state when init Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 18/81] drm/nouveau/disp/dp: respect sink limits when selecting failsafe link configuration Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 19/81] mfd: twl6040: Fix device init errors for ACCCTL register Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 20/81] perf/x86/intel: Allow PEBS multi-entry in watermark mode Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 21/81] drm/bridge: adv7511: Fix low refresh rate selection Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 22/81] objtool: Dont use ignore flag for fake jumps Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 23/81] EDAC/mpc85xx: Prevent building as a module Greg Kroah-Hartman
2019-06-13  8:33   ` Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 24/81] pwm: meson: Use the spin-lock only to protect register modifications Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 25/81] ntp: Allow TAI-UTC offset to be set to zero Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 26/81] f2fs: fix to avoid panic in do_recover_data() Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 27/81] f2fs: fix to clear dirty inode in error path of f2fs_iget() Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 28/81] f2fs: fix to avoid panic in dec_valid_block_count() Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 29/81] f2fs: fix to do sanity check on valid block count of segment Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 30/81] percpu: remove spurious lock dependency between percpu and sched Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 31/81] configfs: fix possible use-after-free in configfs_register_group Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 32/81] uml: fix a boot splat wrt use of cpu_all_mask Greg Kroah-Hartman
2019-06-13  8:33   ` Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 33/81] mmc: mmci: Prevent polling for busy detection in IRQ context Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 34/81] mips: Make sure dt memory regions are valid Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 35/81] watchdog: imx2_wdt: Fix set_timeout for big timeout values Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 36/81] watchdog: fix compile time error of pretimeout governors Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 37/81] blk-mq: move cancel of requeue_work into blk_mq_release Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 38/81] iommu/vt-d: Set intel_iommu_gfx_mapped correctly Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 39/81] misc: pci_endpoint_test: Fix test_reg_bar to be updated in pci_endpoint_test Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 40/81] nvme-pci: unquiesce admin queue on shutdown Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 41/81] ALSA: hda - Register irq handler after the chip initialization Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 42/81] nvmem: core: fix read buffer in place Greg Kroah-Hartman
2019-06-13  8:33 ` Greg Kroah-Hartman [this message]
2019-06-13  8:33 ` [PATCH 4.14 44/81] nfsd: allow fh_want_write to be called twice Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 45/81] vfio: Fix WARNING "do not call blocking ops when !TASK_RUNNING" Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 46/81] x86/PCI: Fix PCI IRQ routing table memory leak Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 47/81] platform/chrome: cros_ec_proto: check for NULL transfer function Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 48/81] PCI: keystone: Prevent ARM32 specific code to be compiled for ARM64 Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 49/81] soc: mediatek: pwrap: Zero initialize rdata in pwrap_init_cipher Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 50/81] clk: rockchip: Turn on "aclk_dmac1" for suspend on rk3288 Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 51/81] soc: rockchip: Set the proper PWM for rk3288 Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 52/81] ARM: dts: imx51: Specify IMX5_CLK_IPG as "ahb" clock to SDMA Greg Kroah-Hartman
2019-06-13  8:33   ` Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 53/81] ARM: dts: imx50: " Greg Kroah-Hartman
2019-06-13  8:33   ` Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 54/81] ARM: dts: imx53: " Greg Kroah-Hartman
2019-06-13  8:33   ` Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 55/81] ARM: dts: imx6sx: Specify IMX6SX_CLK_IPG " Greg Kroah-Hartman
2019-06-13  8:33   ` Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 56/81] ARM: dts: imx7d: Specify IMX7D_CLK_IPG as "ipg" " Greg Kroah-Hartman
2019-06-13  8:33   ` Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 57/81] ARM: dts: imx6ul: Specify IMX6UL_CLK_IPG " Greg Kroah-Hartman
2019-06-13  8:33   ` Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 58/81] ARM: dts: imx6sx: Specify IMX6SX_CLK_IPG " Greg Kroah-Hartman
2019-06-13  8:33   ` Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 59/81] ARM: dts: imx6qdl: Specify IMX6QDL_CLK_IPG " Greg Kroah-Hartman
2019-06-13  8:33   ` Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 60/81] PCI: rpadlpar: Fix leaked device_node references in add/remove paths Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 61/81] ALSA: seq: Protect in-kernel ioctl calls with mutex Greg Kroah-Hartman
2019-06-13  9:02   ` Takashi Iwai
2019-06-13  9:11     ` Greg Kroah-Hartman
2019-06-13  9:13       ` Takashi Iwai
2019-06-13 15:39         ` Sasha Levin
2019-06-13 15:44           ` Takashi Iwai
2019-06-13 16:28             ` Sasha Levin
2019-06-13  8:33 ` [PATCH 4.14 62/81] platform/x86: intel_pmc_ipc: adding error handling Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 63/81] power: supply: max14656: fix potential use-before-alloc Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 64/81] PCI: rcar: Fix a potential NULL pointer dereference Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 65/81] PCI: rcar: Fix 64bit MSI message address handling Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 66/81] video: hgafb: fix potential NULL pointer dereference Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 67/81] video: imsttfb: fix potential NULL pointer dereferences Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 68/81] block, bfq: increase idling for weight-raised queues Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 69/81] PCI: xilinx: Check for __get_free_pages() failure Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 70/81] gpio: gpio-omap: add check for off wake capable gpios Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 71/81] dmaengine: idma64: Use actual device for DMA transfers Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 72/81] pwm: tiehrpwm: Update shadow register for disabling PWMs Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 73/81] ARM: dts: exynos: Always enable necessary APIO_1V8 and ABB_1V8 regulators on Arndale Octa Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 74/81] pwm: Fix deadlock warning when removing PWM device Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 75/81] ARM: exynos: Fix undefined instruction during Exynos5422 resume Greg Kroah-Hartman
2019-06-13  8:33 ` [PATCH 4.14 76/81] usb: typec: fusb302: Check vconn is off when we start toggling Greg Kroah-Hartman
2019-06-13  8:34 ` [PATCH 4.14 77/81] gpio: vf610: Do not share irq_chip Greg Kroah-Hartman
2019-06-13  8:34 ` [PATCH 4.14 78/81] percpu: do not search past bitmap when allocating an area Greg Kroah-Hartman
2019-06-13  8:34 ` [PATCH 4.14 79/81] Revert "Bluetooth: Align minimum encryption key size for LE and BR/EDR connections" Greg Kroah-Hartman
2019-06-13  8:34 ` [PATCH 4.14 80/81] Revert "drm/nouveau: add kconfig option to turn off nouveau legacy contexts. (v3)" Greg Kroah-Hartman
2019-06-13  8:34 ` [PATCH 4.14 81/81] drm: dont block fb changes for async plane updates Greg Kroah-Hartman
2019-06-13 15:11 ` [PATCH 4.14 00/81] 4.14.126-stable review Guenter Roeck
2019-06-13 15:37   ` Greg Kroah-Hartman
2019-06-13 16:38     ` Sasha Levin
2019-06-13 19:33     ` Naresh Kamboju
2019-06-13 16:30 ` kernelci.org bot
2019-06-13 22:38 ` Guenter Roeck
2019-06-14  2:38 ` shuah
2019-06-14 10:28 ` Jon Hunter
2019-06-14 10:28   ` Jon Hunter

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=20190613075652.485457686@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=hanwen@google.com \
    --cc=jakobunt@gmail.com \
    --cc=kirr@nexedi.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mszeredi@redhat.com \
    --cc=sashal@kernel.org \
    --cc=stable@vger.kernel.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 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.