From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
stable@vger.kernel.org, Ming Lei <ming.lei@redhat.com>,
Jakub Kicinski <jakub.kicinski@netronome.com>,
"Luis R. Rodriguez" <mcgrof@kernel.org>
Subject: [PATCH 4.12 38/65] firmware: fix batched requests - wake all waiters
Date: Mon, 14 Aug 2017 18:19:29 -0700 [thread overview]
Message-ID: <20170815011943.900270424@linuxfoundation.org> (raw)
In-Reply-To: <20170815011942.395714306@linuxfoundation.org>
4.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Luis R. Rodriguez <mcgrof@kernel.org>
commit e44565f62a72064e686f7a852137595ec94d78f2 upstream.
The firmware cache mechanism serves two purposes, the secondary purpose is
not well documented nor understood. This fixes a regression with the
secondary purpose of the firmware cache mechanism: batched requests on
successful lookups. Without this fix *any* time a batched request is
triggered, secondary requests for which the batched request mechanism
was designed for will seem to last forver and seem to never return.
This issue is present for all kernel builds possible, and a hard reset
is required.
The firmware cache is used for:
1) Addressing races with file lookups during the suspend/resume cycle
by keeping firmware in memory during the suspend/resume cycle
2) Batched requests for the same file rely only on work from the first file
lookup, which keeps the firmware in memory until the last
release_firmware() is called
Batched requests *only* take effect if secondary requests come in prior to
the first user calling release_firmware(). The devres name used for the
internal firmware cache is used as a hint other pending requests are
ongoing, the firmware buffer data is kept in memory until the last user of
the buffer calls release_firmware(), therefore serializing requests and
delaying the release until all requests are done.
Batched requests wait for a wakup or signal so we can rely on the first file
fetch to write to the pending secondary requests. Commit 5b029624948d
("firmware: do not use fw_lock for fw_state protection") ported the firmware
API to use swait, and in doing so failed to convert complete_all() to
swake_up_all() -- it used swake_up(), loosing the ability for *some* batched
requests to take effect.
We *could* fix this by just using swake_up_all() *but* swait is now known
to be very special use case, so its best to just move away from it. So we
just go back to using completions as before commit 5b029624948d ("firmware:
do not use fw_lock for fw_state protection") given this was using
complete_all().
Without this fix it has been reported plugging in two Intel 6260 Wifi cards
on a system will end up enumerating the two devices only 50% of the time
[0]. The ported swake_up() should have actually handled the case with two
devices, however, *if more than two cards are used* the swake_up() would
not have sufficed. This change is only part of the required fixes for
batched requests. Another fix is provided in the next patch.
This particular change should fix the cases where more than three requests
with the same firmware name is used, otherwise batched requests will wait
for MAX_SCHEDULE_TIMEOUT and just timeout eventually.
Below is a summary of tests triggering batched requests on different
kernel builds.
Before this patch:
============================================================================
CONFIG_FW_LOADER_USER_HELPER_FALLBACK=n
CONFIG_FW_LOADER_USER_HELPER=y
Most common Linux distribution setup.
API-type no-firmware-found firmware-found
----------------------------------------------------------------------
request_firmware() FAIL FAIL
request_firmware_direct() FAIL FAIL
request_firmware_nowait(uevent=true) FAIL FAIL
request_firmware_nowait(uevent=false) FAIL FAIL
============================================================================
CONFIG_FW_LOADER_USER_HELPER_FALLBACK=n
CONFIG_FW_LOADER_USER_HELPER=n
Only possible if CONFIG_DELL_RBU=n and CONFIG_LEDS_LP55XX_COMMON=n, rare.
API-type no-firmware-found firmware-found
----------------------------------------------------------------------
request_firmware() FAIL FAIL
request_firmware_direct() FAIL FAIL
request_firmware_nowait(uevent=true) FAIL FAIL
request_firmware_nowait(uevent=false) FAIL FAIL
============================================================================
CONFIG_FW_LOADER_USER_HELPER_FALLBACK=y
CONFIG_FW_LOADER_USER_HELPER=y
Google Android setup.
API-type no-firmware-found firmware-found
----------------------------------------------------------------------
request_firmware() FAIL FAIL
request_firmware_direct() FAIL FAIL
request_firmware_nowait(uevent=true) FAIL FAIL
request_firmware_nowait(uevent=false) FAIL FAIL
============================================================================
After this patch:
============================================================================
CONFIG_FW_LOADER_USER_HELPER_FALLBACK=n
CONFIG_FW_LOADER_USER_HELPER=y
Most common Linux distribution setup.
API-type no-firmware-found firmware-found
----------------------------------------------------------------------
request_firmware() FAIL OK
request_firmware_direct() FAIL OK
request_firmware_nowait(uevent=true) FAIL OK
request_firmware_nowait(uevent=false) FAIL OK
============================================================================
CONFIG_FW_LOADER_USER_HELPER_FALLBACK=n
CONFIG_FW_LOADER_USER_HELPER=n
Only possible if CONFIG_DELL_RBU=n and CONFIG_LEDS_LP55XX_COMMON=n, rare.
API-type no-firmware-found firmware-found
----------------------------------------------------------------------
request_firmware() FAIL OK
request_firmware_direct() FAIL OK
request_firmware_nowait(uevent=true) FAIL OK
request_firmware_nowait(uevent=false) FAIL OK
============================================================================
CONFIG_FW_LOADER_USER_HELPER_FALLBACK=y
CONFIG_FW_LOADER_USER_HELPER=y
Google Android setup.
API-type no-firmware-found firmware-found
----------------------------------------------------------------------
request_firmware() OK OK
request_firmware_direct() FAIL OK
request_firmware_nowait(uevent=true) OK OK
request_firmware_nowait(uevent=false) OK OK
============================================================================
[0] https://bugzilla.kernel.org/show_bug.cgi?id=195477
Cc: Ming Lei <ming.lei@redhat.com>
Fixes: 5b029624948d ("firmware: do not use fw_lock for fw_state protection")
Reported-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: Luis R. Rodriguez <mcgrof@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
drivers/base/firmware_class.c | 12 +++++-------
1 file changed, 5 insertions(+), 7 deletions(-)
--- a/drivers/base/firmware_class.c
+++ b/drivers/base/firmware_class.c
@@ -30,7 +30,6 @@
#include <linux/syscore_ops.h>
#include <linux/reboot.h>
#include <linux/security.h>
-#include <linux/swait.h>
#include <generated/utsrelease.h>
@@ -112,13 +111,13 @@ static inline long firmware_loading_time
* state of the firmware loading.
*/
struct fw_state {
- struct swait_queue_head wq;
+ struct completion completion;
enum fw_status status;
};
static void fw_state_init(struct fw_state *fw_st)
{
- init_swait_queue_head(&fw_st->wq);
+ init_completion(&fw_st->completion);
fw_st->status = FW_STATUS_UNKNOWN;
}
@@ -131,9 +130,8 @@ static int __fw_state_wait_common(struct
{
long ret;
- ret = swait_event_interruptible_timeout(fw_st->wq,
- __fw_state_is_done(READ_ONCE(fw_st->status)),
- timeout);
+ ret = wait_for_completion_interruptible_timeout(&fw_st->completion,
+ timeout);
if (ret != 0 && fw_st->status == FW_STATUS_ABORTED)
return -ENOENT;
if (!ret)
@@ -148,7 +146,7 @@ static void __fw_state_set(struct fw_sta
WRITE_ONCE(fw_st->status, status);
if (status == FW_STATUS_DONE || status == FW_STATUS_ABORTED)
- swake_up(&fw_st->wq);
+ complete_all(&fw_st->completion);
}
#define fw_state_start(fw_st) \
next prev parent reply other threads:[~2017-08-15 1:21 UTC|newest]
Thread overview: 76+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-08-15 1:18 [PATCH 4.12 00/65] 4.12.8-stable review Greg Kroah-Hartman
2017-08-15 1:18 ` [PATCH 4.12 01/65] mm: ratelimit PFNs busy info message Greg Kroah-Hartman
2017-08-15 1:18 ` [PATCH 4.12 02/65] mm: fix list corruptions on shmem shrinklist Greg Kroah-Hartman
2017-08-15 1:18 ` [PATCH 4.12 03/65] futex: Remove unnecessary warning from get_futex_key Greg Kroah-Hartman
2017-08-15 1:18 ` [PATCH 4.12 04/65] xtensa: fix cache aliasing handling code for WT cache Greg Kroah-Hartman
2017-08-15 1:18 ` [PATCH 4.12 05/65] xtensa: mm/cache: add missing EXPORT_SYMBOLs Greg Kroah-Hartman
2017-08-15 1:18 ` [PATCH 4.12 06/65] xtensa: dont limit csum_partial export by CONFIG_NET Greg Kroah-Hartman
2017-08-15 1:18 ` [PATCH 4.12 07/65] xfs: Fix leak of discard bio Greg Kroah-Hartman
2017-08-15 1:18 ` [PATCH 4.12 08/65] pinctrl: armada-37xx: Fix number of pin in south bridge Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 09/65] mtd: nand: atmel: Fix DT backward compatibility in pmecc.c Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 10/65] mtd: nand: Fix timing setup for NANDs that do not support SET FEATURES Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 11/65] mtd: nand: Declare tBERS, tR and tPROG as u64 to avoid integer overflow Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 12/65] iscsi-target: fix memory leak in iscsit_setup_text_cmd() Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 13/65] iscsi-target: Fix iscsi_np reset hung task during parallel delete Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 14/65] usb-storage: fix deadlock involving host lock and scsi_done Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 15/65] target: Fix node_acl demo-mode + uncached dynamic shutdown regression Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 16/65] fuse: initialize the flock flag in fuse_file on allocation Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 17/65] i2c: designware: Some broken DSTDs use 1MiHz instead of 1MHz Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 18/65] nand: fix wrong default oob layout for small pages using soft ecc Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 19/65] mmc: mmc: correct the logic for setting HS400ES signal voltage Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 20/65] nfs/flexfiles: fix leak of nfs4_ff_ds_version arrays Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 21/65] drm/bridge: tc358767: fix probe without attached output node Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 22/65] drm/etnaviv: Fix off-by-one error in reloc checking Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 23/65] drm/i915: Fix out-of-bounds array access in bdw_load_gamma_lut Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 24/65] USB: serial: option: add D-Link DWM-222 device ID Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 25/65] USB: serial: cp210x: add support for Qivicon USB ZigBee dongle Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 26/65] USB: serial: pl2303: add new ATEN device id Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 27/65] usb: musb: fix tx fifo flush handling again Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 28/65] USB: hcd: Mark secondary HCD as dead if the primary one died Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 29/65] staging:iio:resolver:ad2s1210 fix negative IIO_ANGL_VEL read Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 30/65] iio: aspeed-adc: wait for initial sequence Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 31/65] iio: accel: st_accel: add SPI-3wire support Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 32/65] iio: accel: bmc150: Always restore device to normal mode after suspend-resume Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 33/65] iio: pressure: st_pressure_core: disable multiread by default for LPS22HB Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 34/65] iio: light: tsl2563: use correct event code Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 35/65] iio: adc: Revert "axp288: Drop bogus AXP288_ADC_TS_PIN_CTRL register modifications" Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 36/65] staging: comedi: comedi_fops: do not call blocking ops when !TASK_RUNNING Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 37/65] uas: Add US_FL_IGNORE_RESIDUE for Initio Corporation INIC-3069 Greg Kroah-Hartman
2017-08-15 1:19 ` Greg Kroah-Hartman [this message]
2017-08-15 1:19 ` [PATCH 4.12 39/65] firmware: fix batched requests - send wake up on failure on direct lookups Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 40/65] firmware: avoid invalid fallback aborts by using killable wait Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 41/65] block: Make blk_mq_delay_kick_requeue_list() rerun the queue at a quiet time Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 42/65] usb: gadget: udc: renesas_usb3: Fix usb_gadget_giveback_request() calling Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 43/65] usb: renesas_usbhs: Fix UGCTRL2 value for R-Car Gen3 Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 44/65] USB: Check for dropped connection before switching to full speed Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 45/65] usb: core: unlink urbs from the tail of the endpoints urb_list Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 46/65] usb: quirks: Add no-lpm quirk for Moshi USB to Ethernet Adapter Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 47/65] usb:xhci:Add quirk for Certain failing HP keyboard on reset after resume Greg Kroah-Hartman
2017-08-18 20:02 ` Ben Hutchings
2017-08-18 22:50 ` Greg Kroah-Hartman
2017-08-21 9:55 ` Sandeep Singh
2017-08-15 1:19 ` [PATCH 4.12 48/65] PCI: Protect pci_error_handlers->reset_notify() usage with device_lock() Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 49/65] PCI: Remove __pci_dev_reset() and pci_dev_reset() Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 50/65] PCI: Add pci_reset_function_locked() Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 51/65] xhci: Reset Renesas uPD72020x USB controller for 32-bit DMA issue Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 52/65] iio: adc: vf610_adc: Fix VALT selection value for REFSEL bits Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 53/65] pnfs/blocklayout: require 64-bit sector_t Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 54/65] pinctrl: cherryview: Add Setzer models to the Chromebook DMI quirk Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 55/65] pinctrl: sunxi: add a missing function of A10/A20 pinctrl driver Greg Kroah-Hartman
2017-08-18 20:07 ` Ben Hutchings
2017-08-22 9:11 ` Chen-Yu Tsai
2017-08-22 15:51 ` Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 56/65] pinctrl: intel: merrifield: Correct UART pin lists Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 57/65] pinctrl: uniphier: fix WARN_ON() of pingroups dump on LD11 Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 58/65] pinctrl: uniphier: fix WARN_ON() of pingroups dump on LD20 Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 59/65] pinctrl: samsung: Remove bogus irq_[un]mask from resource management Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 60/65] pinctrl: meson-gxbb: Add missing GPIODV_18 pin entry Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 61/65] pinctrl: meson-gxl: " Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 62/65] MIPS: DEC: Fix an int-handler.S CPU_DADDI_WORKAROUNDS regression Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 63/65] Revert "MIPS: Dont unnecessarily include kmalloc.h into <asm/cache.h>." Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 64/65] MIPS: Octeon: Fix broken EDAC driver Greg Kroah-Hartman
2017-08-15 1:19 ` [PATCH 4.12 65/65] powerpc: Fix /proc/cpuinfo revision for POWER9 DD2 Greg Kroah-Hartman
2017-08-15 10:45 ` [PATCH 4.12 00/65] 4.12.8-stable review Guenter Roeck
2017-08-15 14:34 ` Greg Kroah-Hartman
2017-08-15 18:09 ` Shuah Khan
2017-08-15 18:10 ` Greg Kroah-Hartman
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=20170815011943.900270424@linuxfoundation.org \
--to=gregkh@linuxfoundation.org \
--cc=jakub.kicinski@netronome.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mcgrof@kernel.org \
--cc=ming.lei@redhat.com \
--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 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).