From: Ulf Hansson ulf.hansson@linaro.org

[ Upstream commit 533a6cfe08f96a7b5c65e06d20916d552c11b256 ]

All callers of __mmc_switch() should now be specifying a valid timeout for the CMD6 command. However, just to be sure, let's print a warning and default to use the generic_cmd6_time in case the provided timeout_ms argument is zero.

In this context, let's also simplify some of the corresponding code and clarify some related comments.

Signed-off-by: Ulf Hansson ulf.hansson@linaro.org Link: https://lore.kernel.org/r/20200122142747.5690-4-ulf.hansson@linaro.org Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/mmc/core/mmc_ops.c | 25 +++++++++++-------------- 1 file changed, 11 insertions(+), 14 deletions(-)

diff --git a/drivers/mmc/core/mmc_ops.c b/drivers/mmc/core/mmc_ops.c index 873b2aa0c1556..76de8d441cce4 100644 --- a/drivers/mmc/core/mmc_ops.c +++ b/drivers/mmc/core/mmc_ops.c @@ -456,10 +456,6 @@ static int mmc_poll_for_busy(struct mmc_card *card, unsigned int timeout_ms, bool expired = false; bool busy = false;

- /* We have an unspecified cmd timeout, use the fallback value. */ - if (!timeout_ms) - timeout_ms = MMC_OPS_TIMEOUT_MS; - /* * In cases when not allowed to poll by using CMD13 or because we aren't * capable of polling by using ->card_busy(), then rely on waiting the @@ -532,14 +528,19 @@ int __mmc_switch(struct mmc_card *card, u8 set, u8 index, u8 value,

mmc_retune_hold(host);

+ if (!timeout_ms) { + pr_warn("%s: unspecified timeout for CMD6 - use generic\n", + mmc_hostname(host)); + timeout_ms = card->ext_csd.generic_cmd6_time; + } + /* - * If the cmd timeout and the max_busy_timeout of the host are both - * specified, let's validate them. A failure means we need to prevent - * the host from doing hw busy detection, which is done by converting - * to a R1 response instead of a R1B. + * If the max_busy_timeout of the host is specified, make sure it's + * enough to fit the used timeout_ms. In case it's not, let's instruct + * the host to avoid HW busy detection, by converting to a R1 response + * instead of a R1B. */ - if (timeout_ms && host->max_busy_timeout && - (timeout_ms > host->max_busy_timeout)) + if (host->max_busy_timeout && (timeout_ms > host->max_busy_timeout)) use_r1b_resp = false;

cmd.opcode = MMC_SWITCH; @@ -550,10 +551,6 @@ int __mmc_switch(struct mmc_card *card, u8 set, u8 index, u8 value, cmd.flags = MMC_CMD_AC; if (use_r1b_resp) { cmd.flags |= MMC_RSP_SPI_R1B | MMC_RSP_R1B; - /* - * A busy_timeout of zero means the host can decide to use - * whatever value it finds suitable. - */ cmd.busy_timeout = timeout_ms; } else { cmd.flags |= MMC_RSP_SPI_R1 | MMC_RSP_R1;

From: Faiz Abbas faiz_abbas@ti.com

[ Upstream commit 961de0a856e3a30c0238d1269c0b17f9b179b6c3 ]

Errata i929 in certain OMAP5/DRA7XX/AM57XX silicon revisions (SPRZ426D - November 2014 - Revised February 2018 [1]) mentions unexpected tuning pattern errors. A small failure band may be present in the tuning range which may be missed by the current algorithm. Furthermore, the failure bands vary with temperature leading to different optimum tuning values for different temperatures.

As suggested in the related Application Report (SPRACA9B - October 2017 - Revised July 2018 [2]), tuning should be done in two stages. In stage 1, assign the optimum ratio in the maximum pass window for the current temperature. In stage 2, if the chosen value is close to the small failure band, move away from it in the appropriate direction.

[1] http://www.ti.com/lit/pdf/sprz426 [2] http://www.ti.com/lit/pdf/SPRACA9

Signed-off-by: Faiz Abbas faiz_abbas@ti.com Acked-by: Adrian Hunter adrian.hunter@intel.com Signed-off-by: Ulf Hansson ulf.hansson@linaro.org Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/mmc/host/Kconfig | 2 + drivers/mmc/host/sdhci-omap.c | 90 ++++++++++++++++++++++++++++++++++- 2 files changed, 91 insertions(+), 1 deletion(-)

diff --git a/drivers/mmc/host/Kconfig b/drivers/mmc/host/Kconfig index 694d0828215d2..79b8ac9cdc744 100644 --- a/drivers/mmc/host/Kconfig +++ b/drivers/mmc/host/Kconfig @@ -935,6 +935,8 @@ config MMC_SDHCI_XENON config MMC_SDHCI_OMAP tristate "TI SDHCI Controller Support" depends on MMC_SDHCI_PLTFM && OF + select THERMAL + select TI_SOC_THERMAL help This selects the Secure Digital Host Controller Interface (SDHCI) support present in TI's DRA7 SOCs. The controller supports diff --git a/drivers/mmc/host/sdhci-omap.c b/drivers/mmc/host/sdhci-omap.c index 8a172575bb64f..5698af25caef6 100644 --- a/drivers/mmc/host/sdhci-omap.c +++ b/drivers/mmc/host/sdhci-omap.c @@ -27,6 +27,7 @@ #include <linux/regulator/consumer.h> #include <linux/pinctrl/consumer.h> #include <linux/sys_soc.h> +#include <linux/thermal.h>

#include "sdhci-pltfm.h"

@@ -290,15 +291,19 @@ static int sdhci_omap_execute_tuning(struct mmc_host *mmc, u32 opcode) struct sdhci_host *host = mmc_priv(mmc); struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host); struct sdhci_omap_host *omap_host = sdhci_pltfm_priv(pltfm_host); + struct thermal_zone_device *thermal_dev; struct device *dev = omap_host->dev; struct mmc_ios *ios = &mmc->ios; u32 start_window = 0, max_window = 0; + bool single_point_failure = false; bool dcrc_was_enabled = false; u8 cur_match, prev_match = 0; u32 length = 0, max_len = 0; u32 phase_delay = 0; + int temperature; int ret = 0; u32 reg; + int i;

pltfm_host = sdhci_priv(host); omap_host = sdhci_pltfm_priv(pltfm_host); @@ -312,6 +317,16 @@ static int sdhci_omap_execute_tuning(struct mmc_host *mmc, u32 opcode) if (ios->timing == MMC_TIMING_UHS_SDR50 && !(reg & CAPA2_TSDR50)) return 0;

+ thermal_dev = thermal_zone_get_zone_by_name("cpu_thermal"); + if (IS_ERR(thermal_dev)) { + dev_err(dev, "Unable to get thermal zone for tuning\n"); + return PTR_ERR(thermal_dev); + } + + ret = thermal_zone_get_temp(thermal_dev, &temperature); + if (ret) + return ret; + reg = sdhci_omap_readl(omap_host, SDHCI_OMAP_DLL); reg |= DLL_SWT; sdhci_omap_writel(omap_host, SDHCI_OMAP_DLL, reg); @@ -329,6 +344,11 @@ static int sdhci_omap_execute_tuning(struct mmc_host *mmc, u32 opcode)

omap_host->is_tuning = true;

+ /* + * Stage 1: Search for a maximum pass window ignoring any + * any single point failures. If the tuning value ends up + * near it, move away from it in stage 2 below + */ while (phase_delay <= MAX_PHASE_DELAY) { sdhci_omap_set_dll(omap_host, phase_delay);

@@ -336,10 +356,15 @@ static int sdhci_omap_execute_tuning(struct mmc_host *mmc, u32 opcode) if (cur_match) { if (prev_match) { length++; + } else if (single_point_failure) { + /* ignore single point failure */ + length++; } else { start_window = phase_delay; length = 1; } + } else { + single_point_failure = prev_match; }

if (length > max_len) { @@ -357,13 +382,76 @@ static int sdhci_omap_execute_tuning(struct mmc_host *mmc, u32 opcode) goto tuning_error; }

+ /* + * Assign tuning value as a ratio of maximum pass window based + * on temperature + */ + if (temperature < -20000) + phase_delay = min(max_window + 4 * max_len - 24, + max_window + + DIV_ROUND_UP(13 * max_len, 16) * 4); + else if (temperature < 20000) + phase_delay = max_window + DIV_ROUND_UP(9 * max_len, 16) * 4; + else if (temperature < 40000) + phase_delay = max_window + DIV_ROUND_UP(8 * max_len, 16) * 4; + else if (temperature < 70000) + phase_delay = max_window + DIV_ROUND_UP(7 * max_len, 16) * 4; + else if (temperature < 90000) + phase_delay = max_window + DIV_ROUND_UP(5 * max_len, 16) * 4; + else if (temperature < 120000) + phase_delay = max_window + DIV_ROUND_UP(4 * max_len, 16) * 4; + else + phase_delay = max_window + DIV_ROUND_UP(3 * max_len, 16) * 4; + + /* + * Stage 2: Search for a single point failure near the chosen tuning + * value in two steps. First in the +3 to +10 range and then in the + * +2 to -10 range. If found, move away from it in the appropriate + * direction by the appropriate amount depending on the temperature. + */ + for (i = 3; i <= 10; i++) { + sdhci_omap_set_dll(omap_host, phase_delay + i); + + if (mmc_send_tuning(mmc, opcode, NULL)) { + if (temperature < 10000) + phase_delay += i + 6; + else if (temperature < 20000) + phase_delay += i - 12; + else if (temperature < 70000) + phase_delay += i - 8; + else + phase_delay += i - 6; + + goto single_failure_found; + } + } + + for (i = 2; i >= -10; i--) { + sdhci_omap_set_dll(omap_host, phase_delay + i); + + if (mmc_send_tuning(mmc, opcode, NULL)) { + if (temperature < 10000) + phase_delay += i + 12; + else if (temperature < 20000) + phase_delay += i + 8; + else if (temperature < 70000) + phase_delay += i + 8; + else if (temperature < 90000) + phase_delay += i + 10; + else + phase_delay += i + 12; + + goto single_failure_found; + } + } + +single_failure_found: reg = sdhci_omap_readl(omap_host, SDHCI_OMAP_AC12); if (!(reg & AC12_SCLK_SEL)) { ret = -EIO; goto tuning_error; }

- phase_delay = max_window + 4 * (max_len >> 1); sdhci_omap_set_dll(omap_host, phase_delay);

omap_host->is_tuning = false;

From: Ulf Hansson ulf.hansson@linaro.org

[ Upstream commit 055e04830d4544c57f2a5192a26c9e25915c29c0 ]

It has turned out that the sdhci-omap controller requires the R1B response, for commands that has this response associated with them. So, converting from an R1B to an R1 response for a CMD6 for example, leads to problems with the HW busy detection support.

Fix this by informing the mmc core about the requirement, via setting the host cap, MMC_CAP_NEED_RSP_BUSY.

Reported-by: Naresh Kamboju naresh.kamboju@linaro.org Reported-by: Anders Roxell anders.roxell@linaro.org Reported-by: Faiz Abbas faiz_abbas@ti.com Cc: stable@vger.kernel.org Tested-by: Anders Roxell anders.roxell@linaro.org Tested-by: Faiz Abbas faiz_abbas@ti.com Signed-off-by: Ulf Hansson ulf.hansson@linaro.org Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/mmc/host/sdhci-omap.c | 3 +++ 1 file changed, 3 insertions(+)

diff --git a/drivers/mmc/host/sdhci-omap.c b/drivers/mmc/host/sdhci-omap.c index 5698af25caef6..79ee5fc5a2013 100644 --- a/drivers/mmc/host/sdhci-omap.c +++ b/drivers/mmc/host/sdhci-omap.c @@ -1109,6 +1109,9 @@ static int sdhci_omap_probe(struct platform_device *pdev) host->mmc_host_ops.execute_tuning = sdhci_omap_execute_tuning; host->mmc_host_ops.enable_sdio_irq = sdhci_omap_enable_sdio_irq;

+ /* R1B responses is required to properly manage HW busy detection. */ + mmc->caps |= MMC_CAP_NEED_RSP_BUSY; + ret = sdhci_setup_host(host); if (ret) goto err_put_sync;

From: Ulf Hansson ulf.hansson@linaro.org

[ Upstream commit 43cc64e5221cc6741252b64bc4531dd1eefb733d ]

The busy timeout that is computed for each erase/trim/discard operation, can become quite long and may thus exceed the host->max_busy_timeout. If that becomes the case, mmc_do_erase() converts from using an R1B response to an R1 response, as to prevent the host from doing HW busy detection.

However, it has turned out that some hosts requires an R1B response no matter what, so let's respect that via checking MMC_CAP_NEED_RSP_BUSY. Note that, if the R1B gets enforced, the host becomes fully responsible of managing the needed busy timeout, in one way or the other.

Suggested-by: Sowjanya Komatineni skomatineni@nvidia.com Cc: stable@vger.kernel.org Tested-by: Anders Roxell anders.roxell@linaro.org Tested-by: Sowjanya Komatineni skomatineni@nvidia.com Tested-by: Faiz Abbas faiz_abbas@ti.com Tested-By: Peter Geis pgwipeout@gmail.com Signed-off-by: Ulf Hansson ulf.hansson@linaro.org Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/mmc/core/core.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/mmc/core/core.c b/drivers/mmc/core/core.c index 0a74785e575ba..56f7f3600469a 100644 --- a/drivers/mmc/core/core.c +++ b/drivers/mmc/core/core.c @@ -2043,8 +2043,11 @@ static int mmc_do_erase(struct mmc_card *card, unsigned int from, * the erase operation does not exceed the max_busy_timeout, we should * use R1B response. Or we need to prevent the host from doing hw busy * detection, which is done by converting to a R1 response instead. + * Note, some hosts requires R1B, which also means they are on their own + * when it comes to deal with the busy timeout. */ - if (card->host->max_busy_timeout && + if (!(card->host->caps & MMC_CAP_NEED_RSP_BUSY) && + card->host->max_busy_timeout && busy_timeout > card->host->max_busy_timeout) { cmd.flags = MMC_RSP_SPI_R1 | MMC_RSP_R1 | MMC_CMD_AC; } else {

From: Jean Delvare jdelvare@suse.de

[ Upstream commit 3f9e12e0df012c4a9a7fd7eb0d3ae69b459d6b2c ]

In case the WDAT interface is broken, give the user an option to ignore it to let a native driver bind to the watchdog device instead.

Signed-off-by: Jean Delvare jdelvare@suse.de Acked-by: Mika Westerberg mika.westerberg@linux.intel.com Signed-off-by: Rafael J. Wysocki rafael.j.wysocki@intel.com Signed-off-by: Sasha Levin sashal@kernel.org --- Documentation/admin-guide/kernel-parameters.txt | 4 ++++ drivers/acpi/acpi_watchdog.c | 12 +++++++++++- 2 files changed, 15 insertions(+), 1 deletion(-)

diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt index 8bf0c0532046f..1a5101b7e853c 100644 --- a/Documentation/admin-guide/kernel-parameters.txt +++ b/Documentation/admin-guide/kernel-parameters.txt @@ -136,6 +136,10 @@ dynamic table installation which will install SSDT tables to /sys/firmware/acpi/tables/dynamic.

+ acpi_no_watchdog [HW,ACPI,WDT] + Ignore the ACPI-based watchdog interface (WDAT) and let + a native driver control the watchdog device instead. + acpi_rsdp= [ACPI,EFI,KEXEC] Pass the RSDP address to the kernel, mostly used on machines running EFI runtime service to boot the diff --git a/drivers/acpi/acpi_watchdog.c b/drivers/acpi/acpi_watchdog.c index 23cde3d8e8fbb..0bd1899a287f3 100644 --- a/drivers/acpi/acpi_watchdog.c +++ b/drivers/acpi/acpi_watchdog.c @@ -58,12 +58,14 @@ static bool acpi_watchdog_uses_rtc(const struct acpi_table_wdat *wdat) } #endif

+static bool acpi_no_watchdog; + static const struct acpi_table_wdat *acpi_watchdog_get_wdat(void) { const struct acpi_table_wdat *wdat = NULL; acpi_status status;

- if (acpi_disabled) + if (acpi_disabled || acpi_no_watchdog) return NULL;

status = acpi_get_table(ACPI_SIG_WDAT, 0, @@ -91,6 +93,14 @@ bool acpi_has_watchdog(void) } EXPORT_SYMBOL_GPL(acpi_has_watchdog);

+/* ACPI watchdog can be disabled on boot command line */ +static int __init disable_acpi_watchdog(char *str) +{ + acpi_no_watchdog = true; + return 1; +} +__setup("acpi_no_watchdog", disable_acpi_watchdog); + void __init acpi_watchdog_init(void) { const struct acpi_wdat_entry *entries;

From: Kai-Heng Feng kai.heng.feng@canonical.com

[ Upstream commit be0aba826c4a6ba5929def1962a90d6127871969 ]

The Surfbook E11B uses the SIPODEV SP1064 touchpad, which does not supply descriptors, so it has to be added to the override list.

BugLink: https://bugs.launchpad.net/bugs/1858299 Signed-off-by: Kai-Heng Feng kai.heng.feng@canonical.com Reviewed-by: Hans de Goede hdegoede@redhat.com Signed-off-by: Benjamin Tissoires benjamin.tissoires@redhat.com Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/hid/i2c-hid/i2c-hid-dmi-quirks.c | 8 ++++++++ 1 file changed, 8 insertions(+)

diff --git a/drivers/hid/i2c-hid/i2c-hid-dmi-quirks.c b/drivers/hid/i2c-hid/i2c-hid-dmi-quirks.c index 10af8585c820d..95052373a8282 100644 --- a/drivers/hid/i2c-hid/i2c-hid-dmi-quirks.c +++ b/drivers/hid/i2c-hid/i2c-hid-dmi-quirks.c @@ -341,6 +341,14 @@ static const struct dmi_system_id i2c_hid_dmi_desc_override_table[] = { }, .driver_data = (void *)&sipodev_desc }, + { + .ident = "Trekstor SURFBOOK E11B", + .matches = { + DMI_EXACT_MATCH(DMI_SYS_VENDOR, "TREKSTOR"), + DMI_EXACT_MATCH(DMI_PRODUCT_NAME, "SURFBOOK E11B"), + }, + .driver_data = (void *)&sipodev_desc + }, { .ident = "Direkt-Tek DTLAPY116-2", .matches = {

From: Igor Druzhinin igor.druzhinin@citrix.com

[ Upstream commit ff6993bb79b9f99bdac0b5378169052931b65432 ]

fc_disc_gpn_id_resp() should be the last function using it so free it here to avoid memory leak.

Link: https://lore.kernel.org/r/1579013000-14570-2-git-send-email-igor.druzhinin@c... Reviewed-by: Hannes Reinecke hare@suse.de Signed-off-by: Igor Druzhinin igor.druzhinin@citrix.com Signed-off-by: Martin K. Petersen martin.petersen@oracle.com Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/scsi/libfc/fc_disc.c | 2 ++ 1 file changed, 2 insertions(+)

diff --git a/drivers/scsi/libfc/fc_disc.c b/drivers/scsi/libfc/fc_disc.c index f969a71348ef1..8839f509b19ab 100644 --- a/drivers/scsi/libfc/fc_disc.c +++ b/drivers/scsi/libfc/fc_disc.c @@ -640,6 +640,8 @@ static void fc_disc_gpn_id_resp(struct fc_seq *sp, struct fc_frame *fp, } out: kref_put(&rdata->kref, fc_rport_destroy); + if (!IS_ERR(fp)) + fc_frame_free(fp); }

/**

From: Marek Vasut marex@denx.de

[ Upstream commit 44343418d0f2f623cb9da6f5000df793131cbe3b ]

The KS8851 requires that packet RX and TX are mutually exclusive. Currently, the driver hopes to achieve this by disabling interrupt from the card by writing the card registers and by disabling the interrupt on the interrupt controller. This however is racy on SMP.

Replace this approach by expanding the spinlock used around the ks_start_xmit() TX path to ks_irq() RX path to assure true mutual exclusion and remove the interrupt enabling/disabling, which is now not needed anymore. Furthermore, disable interrupts also in ks_net_stop(), which was missing before.

Note that a massive improvement here would be to re-use the KS8851 driver approach, which is to move the TX path into a worker thread, interrupt handling to threaded interrupt, and synchronize everything with mutexes, but that would be a much bigger rework, for a separate patch.

Signed-off-by: Marek Vasut marex@denx.de Cc: David S. Miller davem@davemloft.net Cc: Lukas Wunner lukas@wunner.de Cc: Petr Stetiar ynezz@true.cz Cc: YueHaibing yuehaibing@huawei.com Signed-off-by: David S. Miller davem@davemloft.net Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/net/ethernet/micrel/ks8851_mll.c | 14 ++++++++------ 1 file changed, 8 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/micrel/ks8851_mll.c b/drivers/net/ethernet/micrel/ks8851_mll.c index 9de59facec218..a5525bf977e2c 100644 --- a/drivers/net/ethernet/micrel/ks8851_mll.c +++ b/drivers/net/ethernet/micrel/ks8851_mll.c @@ -832,14 +832,17 @@ static irqreturn_t ks_irq(int irq, void *pw) { struct net_device *netdev = pw; struct ks_net *ks = netdev_priv(netdev); + unsigned long flags; u16 status;

+ spin_lock_irqsave(&ks->statelock, flags); /*this should be the first in IRQ handler */ ks_save_cmd_reg(ks);

status = ks_rdreg16(ks, KS_ISR); if (unlikely(!status)) { ks_restore_cmd_reg(ks); + spin_unlock_irqrestore(&ks->statelock, flags); return IRQ_NONE; }

@@ -865,6 +868,7 @@ static irqreturn_t ks_irq(int irq, void *pw) ks->netdev->stats.rx_over_errors++; /* this should be the last in IRQ handler*/ ks_restore_cmd_reg(ks); + spin_unlock_irqrestore(&ks->statelock, flags); return IRQ_HANDLED; }

@@ -934,6 +938,7 @@ static int ks_net_stop(struct net_device *netdev)

/* shutdown RX/TX QMU */ ks_disable_qmu(ks); + ks_disable_int(ks);

/* set powermode to soft power down to save power */ ks_set_powermode(ks, PMECR_PM_SOFTDOWN); @@ -990,10 +995,9 @@ static netdev_tx_t ks_start_xmit(struct sk_buff *skb, struct net_device *netdev) { netdev_tx_t retv = NETDEV_TX_OK; struct ks_net *ks = netdev_priv(netdev); + unsigned long flags;

- disable_irq(netdev->irq); - ks_disable_int(ks); - spin_lock(&ks->statelock); + spin_lock_irqsave(&ks->statelock, flags);

/* Extra space are required: * 4 byte for alignment, 4 for status/length, 4 for CRC @@ -1007,9 +1011,7 @@ static netdev_tx_t ks_start_xmit(struct sk_buff *skb, struct net_device *netdev) dev_kfree_skb(skb); } else retv = NETDEV_TX_BUSY; - spin_unlock(&ks->statelock); - ks_enable_int(ks); - enable_irq(netdev->irq); + spin_unlock_irqrestore(&ks->statelock, flags); return retv; }

From: Linus Torvalds torvalds@linux-foundation.org

[ Upstream commit fda31c50292a5062332fa0343c084bd9f46604d9 ]

When queueing a signal, we increment both the users count of pending signals (for RLIMIT_SIGPENDING tracking) and we increment the refcount of the user struct itself (because we keep a reference to the user in the signal structure in order to correctly account for it when freeing).

That turns out to be fairly expensive, because both of them are atomic updates, and particularly under extreme signal handling pressure on big machines, you can get a lot of cache contention on the user struct. That can then cause horrid cacheline ping-pong when you do these multiple accesses.

So change the reference counting to only pin the user for the _first_ pending signal, and to unpin it when the last pending signal is dequeued. That means that when a user sees a lot of concurrent signal queuing - which is the only situation when this matters - the only atomic access needed is generally the 'sigpending' count update.

This was noticed because of a particularly odd timing artifact on a dual-socket 96C/192T Cascade Lake platform: when you get into bad contention, on that machine for some reason seems to be much worse when the contention happens in the upper 32-byte half of the cacheline.

As a result, the kernel test robot will-it-scale 'signal1' benchmark had an odd performance regression simply due to random alignment of the 'struct user_struct' (and pointed to a completely unrelated and apparently nonsensical commit for the regression).

Avoiding the double increments (and decrements on the dequeueing side, of course) makes for much less contention and hugely improved performance on that will-it-scale microbenchmark.

Quoting Feng Tang:

"It makes a big difference, that the performance score is tripled! bump from original 17000 to 54000. Also the gap between 5.0-rc6 and 5.0-rc6+Jiri's patch is reduced to around 2%"

[ The "2% gap" is the odd cacheline placement difference on that platform: under the extreme contention case, the effect of which half of the cacheline was hot was 5%, so with the reduced contention the odd timing artifact is reduced too ]

It does help in the non-contended case too, but is not nearly as noticeable.

Reported-and-tested-by: Feng Tang feng.tang@intel.com Cc: Eric W. Biederman ebiederm@xmission.com Cc: Huang, Ying ying.huang@intel.com Cc: Philip Li philip.li@intel.com Cc: Andi Kleen andi.kleen@intel.com Cc: Jiri Olsa jolsa@redhat.com Cc: Peter Zijlstra peterz@infradead.org Signed-off-by: Linus Torvalds torvalds@linux-foundation.org Signed-off-by: Sasha Levin sashal@kernel.org --- kernel/signal.c | 23 ++++++++++++++--------- 1 file changed, 14 insertions(+), 9 deletions(-)

diff --git a/kernel/signal.c b/kernel/signal.c index 08911bb6fe9ab..c42eaf39b5729 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -407,27 +407,32 @@ __sigqueue_alloc(int sig, struct task_struct *t, gfp_t flags, int override_rlimi { struct sigqueue *q = NULL; struct user_struct *user; + int sigpending;

/* * Protect access to @t credentials. This can go away when all * callers hold rcu read lock. + * + * NOTE! A pending signal will hold on to the user refcount, + * and we get/put the refcount only when the sigpending count + * changes from/to zero. */ rcu_read_lock(); - user = get_uid(__task_cred(t)->user); - atomic_inc(&user->sigpending); + user = __task_cred(t)->user; + sigpending = atomic_inc_return(&user->sigpending); + if (sigpending == 1) + get_uid(user); rcu_read_unlock();

- if (override_rlimit || - atomic_read(&user->sigpending) <= - task_rlimit(t, RLIMIT_SIGPENDING)) { + if (override_rlimit || likely(sigpending <= task_rlimit(t, RLIMIT_SIGPENDING))) { q = kmem_cache_alloc(sigqueue_cachep, flags); } else { print_dropped_signal(sig); }

if (unlikely(q == NULL)) { - atomic_dec(&user->sigpending); - free_uid(user); + if (atomic_dec_and_test(&user->sigpending)) + free_uid(user); } else { INIT_LIST_HEAD(&q->list); q->flags = 0; @@ -441,8 +446,8 @@ static void __sigqueue_free(struct sigqueue *q) { if (q->flags & SIGQUEUE_PREALLOC) return; - atomic_dec(&q->user->sigpending); - free_uid(q->user); + if (atomic_dec_and_test(&q->user->sigpending)) + free_uid(q->user); kmem_cache_free(sigqueue_cachep, q); }

From: Luo bin luobin9@huawei.com

[ Upstream commit 0bff777bd0cba73ad4cd0145696ad284d7e6a99f ]

can not use a local variable as an input parameter of irq_set_affinity_hint

Signed-off-by: Luo bin luobin9@huawei.com Signed-off-by: David S. Miller davem@davemloft.net Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/net/ethernet/huawei/hinic/hinic_hw_qp.h | 1 + drivers/net/ethernet/huawei/hinic/hinic_rx.c | 5 ++--- 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_qp.h b/drivers/net/ethernet/huawei/hinic/hinic_hw_qp.h index 6c84f83ec2831..d46cfd4fbbbc5 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_hw_qp.h +++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_qp.h @@ -103,6 +103,7 @@ struct hinic_rq {

struct hinic_wq *wq;

+ struct cpumask affinity_mask; u32 irq; u16 msix_entry;

diff --git a/drivers/net/ethernet/huawei/hinic/hinic_rx.c b/drivers/net/ethernet/huawei/hinic/hinic_rx.c index 06b24a92ed7d4..3467d84d96c39 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_rx.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_rx.c @@ -414,7 +414,6 @@ static int rx_request_irq(struct hinic_rxq *rxq) struct hinic_hwdev *hwdev = nic_dev->hwdev; struct hinic_rq *rq = rxq->rq; struct hinic_qp *qp; - struct cpumask mask; int err;

rx_add_napi(rxq); @@ -431,8 +430,8 @@ static int rx_request_irq(struct hinic_rxq *rxq) }

qp = container_of(rq, struct hinic_qp, rq); - cpumask_set_cpu(qp->q_id % num_online_cpus(), &mask); - return irq_set_affinity_hint(rq->irq, &mask); + cpumask_set_cpu(qp->q_id % num_online_cpus(), &rq->affinity_mask); + return irq_set_affinity_hint(rq->irq, &rq->affinity_mask); }

static void rx_free_irq(struct hinic_rxq *rxq)

From: Taehee Yoo ap420073@gmail.com

[ Upstream commit 93b5cbfa9636d385126f211dca9efa7e3f683202 ]

rmnet registers IFLA_LINK interface as a lower interface. But, IFLA_LINK could be NULL. In the current code, rmnet doesn't check IFLA_LINK. So, panic would occur.

Test commands: modprobe rmnet ip link add rmnet0 type rmnet mux_id 1

Splat looks like: [ 36.826109][ T1115] general protection fault, probably for non-canonical address 0xdffffc0000000000I [ 36.838817][ T1115] KASAN: null-ptr-deref in range [0x0000000000000000-0x0000000000000007] [ 36.839908][ T1115] CPU: 1 PID: 1115 Comm: ip Not tainted 5.6.0-rc1+ #447 [ 36.840569][ T1115] Hardware name: innotek GmbH VirtualBox/VirtualBox, BIOS VirtualBox 12/01/2006 [ 36.841408][ T1115] RIP: 0010:rmnet_newlink+0x54/0x510 [rmnet] [ 36.841986][ T1115] Code: 83 ec 18 48 c1 e9 03 80 3c 01 00 0f 85 d4 03 00 00 48 8b 6a 28 48 b8 00 00 00 00 00 c [ 36.843923][ T1115] RSP: 0018:ffff8880b7e0f1c0 EFLAGS: 00010247 [ 36.844756][ T1115] RAX: dffffc0000000000 RBX: ffff8880d14cca00 RCX: 1ffff11016fc1e99 [ 36.845859][ T1115] RDX: 0000000000000000 RSI: ffff8880c3d04000 RDI: 0000000000000004 [ 36.846961][ T1115] RBP: 0000000000000000 R08: ffff8880b7e0f8b0 R09: ffff8880b6ac2d90 [ 36.848020][ T1115] R10: ffffffffc0589a40 R11: ffffed1016d585b7 R12: ffffffff88ceaf80 [ 36.848788][ T1115] R13: ffff8880c3d04000 R14: ffff8880b7e0f8b0 R15: ffff8880c3d04000 [ 36.849546][ T1115] FS: 00007f50ab3360c0(0000) GS:ffff8880da000000(0000) knlGS:0000000000000000 [ 36.851784][ T1115] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 36.852422][ T1115] CR2: 000055871afe5ab0 CR3: 00000000ae246001 CR4: 00000000000606e0 [ 36.853181][ T1115] Call Trace: [ 36.853514][ T1115] __rtnl_newlink+0xbdb/0x1270 [ 36.853967][ T1115] ? lock_downgrade+0x6e0/0x6e0 [ 36.854420][ T1115] ? rtnl_link_unregister+0x220/0x220 [ 36.854936][ T1115] ? lock_acquire+0x164/0x3b0 [ 36.855376][ T1115] ? is_bpf_image_address+0xff/0x1d0 [ 36.855884][ T1115] ? rtnl_newlink+0x4c/0x90 [ 36.856304][ T1115] ? kernel_text_address+0x111/0x140 [ 36.856857][ T1115] ? __kernel_text_address+0xe/0x30 [ 36.857440][ T1115] ? unwind_get_return_address+0x5f/0xa0 [ 36.858063][ T1115] ? create_prof_cpu_mask+0x20/0x20 [ 36.858644][ T1115] ? arch_stack_walk+0x83/0xb0 [ 36.859171][ T1115] ? stack_trace_save+0x82/0xb0 [ 36.859710][ T1115] ? stack_trace_consume_entry+0x160/0x160 [ 36.860357][ T1115] ? deactivate_slab.isra.78+0x2c5/0x800 [ 36.860928][ T1115] ? kasan_unpoison_shadow+0x30/0x40 [ 36.861520][ T1115] ? kmem_cache_alloc_trace+0x135/0x350 [ 36.862125][ T1115] ? rtnl_newlink+0x4c/0x90 [ 36.864073][ T1115] rtnl_newlink+0x65/0x90 [ ... ]

Fixes: ceed73a2cf4a ("drivers: net: ethernet: qualcomm: rmnet: Initial implementation") Signed-off-by: Taehee Yoo ap420073@gmail.com Signed-off-by: David S. Miller davem@davemloft.net Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/net/ethernet/qualcomm/rmnet/rmnet_config.c | 5 +++++ 1 file changed, 5 insertions(+)

diff --git a/drivers/net/ethernet/qualcomm/rmnet/rmnet_config.c b/drivers/net/ethernet/qualcomm/rmnet/rmnet_config.c index 4c476fac78358..0fb9375b4f69e 100644 --- a/drivers/net/ethernet/qualcomm/rmnet/rmnet_config.c +++ b/drivers/net/ethernet/qualcomm/rmnet/rmnet_config.c @@ -144,6 +144,11 @@ static int rmnet_newlink(struct net *src_net, struct net_device *dev, int err = 0; u16 mux_id;

+ if (!tb[IFLA_LINK]) { + NL_SET_ERR_MSG_MOD(extack, "link not specified"); + return -EINVAL; + } + real_dev = __dev_get_by_index(src_net, nla_get_u32(tb[IFLA_LINK])); if (!real_dev || !dev) return -ENODEV;

From: Taehee Yoo ap420073@gmail.com

[ Upstream commit 102210f7664442d8c0ce332c006ea90626df745b ]

rmnet_get_port() internally calls rcu_dereference_rtnl(), which checks RTNL. But rmnet_get_port() could be called by packet path. The packet path is not protected by RTNL. So, the suspicious RCU usage problem occurs.

Test commands: modprobe rmnet ip netns add nst ip link add veth0 type veth peer name veth1 ip link set veth1 netns nst ip link add rmnet0 link veth0 type rmnet mux_id 1 ip netns exec nst ip link add rmnet1 link veth1 type rmnet mux_id 1 ip netns exec nst ip link set veth1 up ip netns exec nst ip link set rmnet1 up ip netns exec nst ip a a 192.168.100.2/24 dev rmnet1 ip link set veth0 up ip link set rmnet0 up ip a a 192.168.100.1/24 dev rmnet0 ping 192.168.100.2

Splat looks like: [ 146.630958][ T1174] WARNING: suspicious RCU usage [ 146.631735][ T1174] 5.6.0-rc1+ #447 Not tainted [ 146.632387][ T1174] ----------------------------- [ 146.633151][ T1174] drivers/net/ethernet/qualcomm/rmnet/rmnet_config.c:386 suspicious rcu_dereference_check() ! [ 146.634742][ T1174] [ 146.634742][ T1174] other info that might help us debug this: [ 146.634742][ T1174] [ 146.645992][ T1174] [ 146.645992][ T1174] rcu_scheduler_active = 2, debug_locks = 1 [ 146.646937][ T1174] 5 locks held by ping/1174: [ 146.647609][ T1174] #0: ffff8880c31dea70 (sk_lock-AF_INET){+.+.}, at: raw_sendmsg+0xab8/0x2980 [ 146.662463][ T1174] #1: ffffffff93925660 (rcu_read_lock_bh){....}, at: ip_finish_output2+0x243/0x2150 [ 146.671696][ T1174] #2: ffffffff93925660 (rcu_read_lock_bh){....}, at: __dev_queue_xmit+0x213/0x2940 [ 146.673064][ T1174] #3: ffff8880c19ecd58 (&dev->qdisc_running_key#7){+...}, at: ip_finish_output2+0x714/0x2150 [ 146.690358][ T1174] #4: ffff8880c5796898 (&dev->qdisc_xmit_lock_key#3){+.-.}, at: sch_direct_xmit+0x1e2/0x1020 [ 146.699875][ T1174] [ 146.699875][ T1174] stack backtrace: [ 146.701091][ T1174] CPU: 0 PID: 1174 Comm: ping Not tainted 5.6.0-rc1+ #447 [ 146.705215][ T1174] Hardware name: innotek GmbH VirtualBox/VirtualBox, BIOS VirtualBox 12/01/2006 [ 146.706565][ T1174] Call Trace: [ 146.707102][ T1174] dump_stack+0x96/0xdb [ 146.708007][ T1174] rmnet_get_port.part.9+0x76/0x80 [rmnet] [ 146.709233][ T1174] rmnet_egress_handler+0x107/0x420 [rmnet] [ 146.710492][ T1174] ? sch_direct_xmit+0x1e2/0x1020 [ 146.716193][ T1174] rmnet_vnd_start_xmit+0x3d/0xa0 [rmnet] [ 146.717012][ T1174] dev_hard_start_xmit+0x160/0x740 [ 146.717854][ T1174] sch_direct_xmit+0x265/0x1020 [ 146.718577][ T1174] ? register_lock_class+0x14d0/0x14d0 [ 146.719429][ T1174] ? dev_watchdog+0xac0/0xac0 [ 146.723738][ T1174] ? __dev_queue_xmit+0x15fd/0x2940 [ 146.724469][ T1174] ? lock_acquire+0x164/0x3b0 [ 146.725172][ T1174] __dev_queue_xmit+0x20c7/0x2940 [ ... ]

Fixes: ceed73a2cf4a ("drivers: net: ethernet: qualcomm: rmnet: Initial implementation") Signed-off-by: Taehee Yoo ap420073@gmail.com Signed-off-by: David S. Miller davem@davemloft.net Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/net/ethernet/qualcomm/rmnet/rmnet_config.c | 13 ++++++------- drivers/net/ethernet/qualcomm/rmnet/rmnet_config.h | 2 +- .../net/ethernet/qualcomm/rmnet/rmnet_handlers.c | 4 ++-- 3 files changed, 9 insertions(+), 10 deletions(-)

diff --git a/drivers/net/ethernet/qualcomm/rmnet/rmnet_config.c b/drivers/net/ethernet/qualcomm/rmnet/rmnet_config.c index ad46f5aadc0ab..915165cda996f 100644 --- a/drivers/net/ethernet/qualcomm/rmnet/rmnet_config.c +++ b/drivers/net/ethernet/qualcomm/rmnet/rmnet_config.c @@ -391,11 +391,10 @@ struct rtnl_link_ops rmnet_link_ops __read_mostly = { .fill_info = rmnet_fill_info, };

-/* Needs either rcu_read_lock() or rtnl lock */ -struct rmnet_port *rmnet_get_port(struct net_device *real_dev) +struct rmnet_port *rmnet_get_port_rcu(struct net_device *real_dev) { if (rmnet_is_real_dev_registered(real_dev)) - return rcu_dereference_rtnl(real_dev->rx_handler_data); + return rcu_dereference_bh(real_dev->rx_handler_data); else return NULL; } @@ -421,7 +420,7 @@ int rmnet_add_bridge(struct net_device *rmnet_dev, struct rmnet_port *port, *slave_port; int err;

- port = rmnet_get_port(real_dev); + port = rmnet_get_port_rtnl(real_dev);

/* If there is more than one rmnet dev attached, its probably being * used for muxing. Skip the briding in that case @@ -436,7 +435,7 @@ int rmnet_add_bridge(struct net_device *rmnet_dev, if (err) return -EBUSY;

- slave_port = rmnet_get_port(slave_dev); + slave_port = rmnet_get_port_rtnl(slave_dev); slave_port->rmnet_mode = RMNET_EPMODE_BRIDGE; slave_port->bridge_ep = real_dev;

@@ -454,11 +453,11 @@ int rmnet_del_bridge(struct net_device *rmnet_dev, struct net_device *real_dev = priv->real_dev; struct rmnet_port *port, *slave_port;

- port = rmnet_get_port(real_dev); + port = rmnet_get_port_rtnl(real_dev); port->rmnet_mode = RMNET_EPMODE_VND; port->bridge_ep = NULL;

- slave_port = rmnet_get_port(slave_dev); + slave_port = rmnet_get_port_rtnl(slave_dev); rmnet_unregister_real_device(slave_dev, slave_port);

netdev_dbg(slave_dev, "removed from rmnet as slave\n"); diff --git a/drivers/net/ethernet/qualcomm/rmnet/rmnet_config.h b/drivers/net/ethernet/qualcomm/rmnet/rmnet_config.h index 34ac45a774e72..6a978f486ce10 100644 --- a/drivers/net/ethernet/qualcomm/rmnet/rmnet_config.h +++ b/drivers/net/ethernet/qualcomm/rmnet/rmnet_config.h @@ -74,7 +74,7 @@ struct rmnet_priv { struct rmnet_priv_stats stats; };

-struct rmnet_port *rmnet_get_port(struct net_device *real_dev); +struct rmnet_port *rmnet_get_port_rcu(struct net_device *real_dev); struct rmnet_endpoint *rmnet_get_endpoint(struct rmnet_port *port, u8 mux_id); int rmnet_add_bridge(struct net_device *rmnet_dev, struct net_device *slave_dev, diff --git a/drivers/net/ethernet/qualcomm/rmnet/rmnet_handlers.c b/drivers/net/ethernet/qualcomm/rmnet/rmnet_handlers.c index 11167abe5934d..5177d0f24947f 100644 --- a/drivers/net/ethernet/qualcomm/rmnet/rmnet_handlers.c +++ b/drivers/net/ethernet/qualcomm/rmnet/rmnet_handlers.c @@ -193,7 +193,7 @@ rx_handler_result_t rmnet_rx_handler(struct sk_buff **pskb) return RX_HANDLER_PASS;

dev = skb->dev; - port = rmnet_get_port(dev); + port = rmnet_get_port_rcu(dev);

switch (port->rmnet_mode) { case RMNET_EPMODE_VND: @@ -226,7 +226,7 @@ void rmnet_egress_handler(struct sk_buff *skb) skb->dev = priv->real_dev; mux_id = priv->mux_id;

- port = rmnet_get_port(skb->dev); + port = rmnet_get_port_rcu(skb->dev); if (!port) goto drop;

From: Taehee Yoo ap420073@gmail.com

[ Upstream commit 1dc49e9d164cd7e11c81279c83db84a147e14740 ]

Basically, duplicate mux id isn't be allowed. So, the creation of rmnet will be failed if there is duplicate mux id is existing. But, changelink routine doesn't check duplicate mux id.

Test commands: modprobe rmnet ip link add dummy0 type dummy ip link add rmnet0 link dummy0 type rmnet mux_id 1 ip link add rmnet1 link dummy0 type rmnet mux_id 2 ip link set rmnet1 type rmnet mux_id 1

Fixes: 23790ef12082 ("net: qualcomm: rmnet: Allow to configure flags for existing devices") Signed-off-by: Taehee Yoo ap420073@gmail.com Signed-off-by: David S. Miller davem@davemloft.net Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/net/ethernet/qualcomm/rmnet/rmnet_config.c | 4 ++++ 1 file changed, 4 insertions(+)

diff --git a/drivers/net/ethernet/qualcomm/rmnet/rmnet_config.c b/drivers/net/ethernet/qualcomm/rmnet/rmnet_config.c index b7acedf76aa0b..272a08f521e11 100644 --- a/drivers/net/ethernet/qualcomm/rmnet/rmnet_config.c +++ b/drivers/net/ethernet/qualcomm/rmnet/rmnet_config.c @@ -315,6 +315,10 @@ static int rmnet_changelink(struct net_device *dev, struct nlattr *tb[],

if (data[IFLA_RMNET_MUX_ID]) { mux_id = nla_get_u16(data[IFLA_RMNET_MUX_ID]); + if (rmnet_get_endpoint(port, mux_id)) { + NL_SET_ERR_MSG_MOD(extack, "MUX ID already exists"); + return -EINVAL; + } ep = rmnet_get_endpoint(port, priv->mux_id); if (!ep) return -ENODEV;

From: Taehee Yoo ap420073@gmail.com

[ Upstream commit d939b6d30bea1a2322bc536b12be0a7c4c2bccd7 ]

In order to attach a bridge interface to the rmnet interface, "master" operation is used. (e.g. ip link set dummy1 master rmnet0) But, in the rmnet_add_bridge(), which is a callback of ->ndo_add_slave() doesn't register lower interface. So, ->ndo_del_slave() doesn't work. There are other problems too. 1. It couldn't detect circular upper/lower interface relationship. 2. It couldn't prevent stack overflow because of too deep depth of upper/lower interface 3. It doesn't check the number of lower interfaces. 4. Panics because of several reasons.

The root problem of these issues is actually the same. So, in this patch, these all problems will be fixed.

Test commands: modprobe rmnet ip link add dummy0 type dummy ip link add rmnet0 link dummy0 type rmnet mux_id 1 ip link add dummy1 master rmnet0 type dummy ip link add dummy2 master rmnet0 type dummy ip link del rmnet0 ip link del dummy2 ip link del dummy1

Splat looks like: [ 41.867595][ T1164] general protection fault, probably for non-canonical address 0xdffffc0000000101I [ 41.869993][ T1164] KASAN: null-ptr-deref in range [0x0000000000000808-0x000000000000080f] [ 41.872950][ T1164] CPU: 0 PID: 1164 Comm: ip Not tainted 5.6.0-rc1+ #447 [ 41.873915][ T1164] Hardware name: innotek GmbH VirtualBox/VirtualBox, BIOS VirtualBox 12/01/2006 [ 41.875161][ T1164] RIP: 0010:rmnet_unregister_bridge.isra.6+0x71/0xf0 [rmnet] [ 41.876178][ T1164] Code: 48 89 ef 48 89 c6 5b 5d e9 fc fe ff ff e8 f7 f3 ff ff 48 8d b8 08 08 00 00 48 ba 00 7 [ 41.878925][ T1164] RSP: 0018:ffff8880c4d0f188 EFLAGS: 00010202 [ 41.879774][ T1164] RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000101 [ 41.887689][ T1164] RDX: dffffc0000000000 RSI: ffffffffb8cf64f0 RDI: 0000000000000808 [ 41.888727][ T1164] RBP: ffff8880c40e4000 R08: ffffed101b3c0e3c R09: 0000000000000001 [ 41.889749][ T1164] R10: 0000000000000001 R11: ffffed101b3c0e3b R12: 1ffff110189a1e3c [ 41.890783][ T1164] R13: ffff8880c4d0f200 R14: ffffffffb8d56160 R15: ffff8880ccc2c000 [ 41.891794][ T1164] FS: 00007f4300edc0c0(0000) GS:ffff8880d9c00000(0000) knlGS:0000000000000000 [ 41.892953][ T1164] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 41.893800][ T1164] CR2: 00007f43003bc8c0 CR3: 00000000ca53e001 CR4: 00000000000606f0 [ 41.894824][ T1164] Call Trace: [ 41.895274][ T1164] ? rcu_is_watching+0x2c/0x80 [ 41.895895][ T1164] rmnet_config_notify_cb+0x1f7/0x590 [rmnet] [ 41.896687][ T1164] ? rmnet_unregister_bridge.isra.6+0xf0/0xf0 [rmnet] [ 41.897611][ T1164] ? rmnet_unregister_bridge.isra.6+0xf0/0xf0 [rmnet] [ 41.898508][ T1164] ? __module_text_address+0x13/0x140 [ 41.899162][ T1164] notifier_call_chain+0x90/0x160 [ 41.899814][ T1164] rollback_registered_many+0x660/0xcf0 [ 41.900544][ T1164] ? netif_set_real_num_tx_queues+0x780/0x780 [ 41.901316][ T1164] ? __lock_acquire+0xdfe/0x3de0 [ 41.901958][ T1164] ? memset+0x1f/0x40 [ 41.902468][ T1164] ? __nla_validate_parse+0x98/0x1ab0 [ 41.903166][ T1164] unregister_netdevice_many.part.133+0x13/0x1b0 [ 41.903988][ T1164] rtnl_delete_link+0xbc/0x100 [ ... ]

Fixes: 60d58f971c10 ("net: qualcomm: rmnet: Implement bridge mode") Signed-off-by: Taehee Yoo ap420073@gmail.com Signed-off-by: David S. Miller davem@davemloft.net Signed-off-by: Sasha Levin sashal@kernel.org --- .../ethernet/qualcomm/rmnet/rmnet_config.c | 131 +++++++++--------- .../ethernet/qualcomm/rmnet/rmnet_config.h | 1 + .../net/ethernet/qualcomm/rmnet/rmnet_vnd.c | 8 -- .../net/ethernet/qualcomm/rmnet/rmnet_vnd.h | 1 - 4 files changed, 64 insertions(+), 77 deletions(-)

diff --git a/drivers/net/ethernet/qualcomm/rmnet/rmnet_config.c b/drivers/net/ethernet/qualcomm/rmnet/rmnet_config.c index 01a1d3ff3c944..37786affa9750 100644 --- a/drivers/net/ethernet/qualcomm/rmnet/rmnet_config.c +++ b/drivers/net/ethernet/qualcomm/rmnet/rmnet_config.c @@ -22,25 +22,6 @@ #include "rmnet_vnd.h" #include "rmnet_private.h"

-/* Locking scheme - - * The shared resource which needs to be protected is realdev->rx_handler_data. - * For the writer path, this is using rtnl_lock(). The writer paths are - * rmnet_newlink(), rmnet_dellink() and rmnet_force_unassociate_device(). These - * paths are already called with rtnl_lock() acquired in. There is also an - * ASSERT_RTNL() to ensure that we are calling with rtnl acquired. For - * dereference here, we will need to use rtnl_dereference(). Dev list writing - * needs to happen with rtnl_lock() acquired for netdev_master_upper_dev_link(). - * For the reader path, the real_dev->rx_handler_data is called in the TX / RX - * path. We only need rcu_read_lock() for these scenarios. In these cases, - * the rcu_read_lock() is held in __dev_queue_xmit() and - * netif_receive_skb_internal(), so readers need to use rcu_dereference_rtnl() - * to get the relevant information. For dev list reading, we again acquire - * rcu_read_lock() in rmnet_dellink() for netdev_master_upper_dev_get_rcu(). - * We also use unregister_netdevice_many() to free all rmnet devices in - * rmnet_force_unassociate_device() so we dont lose the rtnl_lock() and free in - * same context. - */ - /* Local Definitions and Declarations */

static const struct nla_policy rmnet_policy[IFLA_RMNET_MAX + 1] = { @@ -60,9 +41,10 @@ rmnet_get_port_rtnl(const struct net_device *real_dev) return rtnl_dereference(real_dev->rx_handler_data); }

-static int rmnet_unregister_real_device(struct net_device *real_dev, - struct rmnet_port *port) +static int rmnet_unregister_real_device(struct net_device *real_dev) { + struct rmnet_port *port = rmnet_get_port_rtnl(real_dev); + if (port->nr_rmnet_devs) return -EINVAL;

@@ -102,28 +84,33 @@ static int rmnet_register_real_device(struct net_device *real_dev) return 0; }

-static void rmnet_unregister_bridge(struct net_device *dev, - struct rmnet_port *port) +static void rmnet_unregister_bridge(struct rmnet_port *port) { - struct rmnet_port *bridge_port; - struct net_device *bridge_dev; + struct net_device *bridge_dev, *real_dev, *rmnet_dev; + struct rmnet_port *real_port;

if (port->rmnet_mode != RMNET_EPMODE_BRIDGE) return;

- /* bridge slave handling */ + rmnet_dev = port->rmnet_dev; if (!port->nr_rmnet_devs) { - bridge_dev = port->bridge_ep; + /* bridge device */ + real_dev = port->bridge_ep; + bridge_dev = port->dev;

- bridge_port = rmnet_get_port_rtnl(bridge_dev); - bridge_port->bridge_ep = NULL; - bridge_port->rmnet_mode = RMNET_EPMODE_VND; + real_port = rmnet_get_port_rtnl(real_dev); + real_port->bridge_ep = NULL; + real_port->rmnet_mode = RMNET_EPMODE_VND; } else { + /* real device */ bridge_dev = port->bridge_ep;

- bridge_port = rmnet_get_port_rtnl(bridge_dev); - rmnet_unregister_real_device(bridge_dev, bridge_port); + port->bridge_ep = NULL; + port->rmnet_mode = RMNET_EPMODE_VND; } + + netdev_upper_dev_unlink(bridge_dev, rmnet_dev); + rmnet_unregister_real_device(bridge_dev); }

static int rmnet_newlink(struct net *src_net, struct net_device *dev, @@ -170,6 +157,7 @@ static int rmnet_newlink(struct net *src_net, struct net_device *dev, goto err2;

port->rmnet_mode = mode; + port->rmnet_dev = dev;

hlist_add_head_rcu(&ep->hlnode, &port->muxed_ep[mux_id]);

@@ -187,8 +175,9 @@ static int rmnet_newlink(struct net *src_net, struct net_device *dev,

err2: unregister_netdevice(dev); + rmnet_vnd_dellink(mux_id, port, ep); err1: - rmnet_unregister_real_device(real_dev, port); + rmnet_unregister_real_device(real_dev); err0: kfree(ep); return err; @@ -197,30 +186,32 @@ static int rmnet_newlink(struct net *src_net, struct net_device *dev, static void rmnet_dellink(struct net_device *dev, struct list_head *head) { struct rmnet_priv *priv = netdev_priv(dev); - struct net_device *real_dev; + struct net_device *real_dev, *bridge_dev; + struct rmnet_port *real_port, *bridge_port; struct rmnet_endpoint *ep; - struct rmnet_port *port; - u8 mux_id; + u8 mux_id = priv->mux_id;

real_dev = priv->real_dev;

- if (!real_dev || !rmnet_is_real_dev_registered(real_dev)) + if (!rmnet_is_real_dev_registered(real_dev)) return;

- port = rmnet_get_port_rtnl(real_dev); - - mux_id = rmnet_vnd_get_mux(dev); + real_port = rmnet_get_port_rtnl(real_dev); + bridge_dev = real_port->bridge_ep; + if (bridge_dev) { + bridge_port = rmnet_get_port_rtnl(bridge_dev); + rmnet_unregister_bridge(bridge_port); + }

- ep = rmnet_get_endpoint(port, mux_id); + ep = rmnet_get_endpoint(real_port, mux_id); if (ep) { hlist_del_init_rcu(&ep->hlnode); - rmnet_unregister_bridge(dev, port); - rmnet_vnd_dellink(mux_id, port, ep); + rmnet_vnd_dellink(mux_id, real_port, ep); kfree(ep); } - netdev_upper_dev_unlink(real_dev, dev); - rmnet_unregister_real_device(real_dev, port);

+ netdev_upper_dev_unlink(real_dev, dev); + rmnet_unregister_real_device(real_dev); unregister_netdevice_queue(dev, head); }

@@ -232,23 +223,23 @@ static void rmnet_force_unassociate_device(struct net_device *real_dev) unsigned long bkt_ep; LIST_HEAD(list);

- ASSERT_RTNL(); - port = rmnet_get_port_rtnl(real_dev);

- rmnet_unregister_bridge(real_dev, port); - - hash_for_each_safe(port->muxed_ep, bkt_ep, tmp_ep, ep, hlnode) { - netdev_upper_dev_unlink(real_dev, ep->egress_dev); - unregister_netdevice_queue(ep->egress_dev, &list); - rmnet_vnd_dellink(ep->mux_id, port, ep); - hlist_del_init_rcu(&ep->hlnode); - kfree(ep); + if (port->nr_rmnet_devs) { + /* real device */ + rmnet_unregister_bridge(port); + hash_for_each_safe(port->muxed_ep, bkt_ep, tmp_ep, ep, hlnode) { + unregister_netdevice_queue(ep->egress_dev, &list); + netdev_upper_dev_unlink(real_dev, ep->egress_dev); + rmnet_vnd_dellink(ep->mux_id, port, ep); + hlist_del_init_rcu(&ep->hlnode); + kfree(ep); + } + rmnet_unregister_real_device(real_dev); + unregister_netdevice_many(&list); + } else { + rmnet_unregister_bridge(port); } - - unregister_netdevice_many(&list); - - rmnet_unregister_real_device(real_dev, port); }

static int rmnet_config_notify_cb(struct notifier_block *nb, @@ -427,6 +418,9 @@ int rmnet_add_bridge(struct net_device *rmnet_dev, if (port->nr_rmnet_devs > 1) return -EINVAL;

+ if (port->rmnet_mode != RMNET_EPMODE_VND) + return -EINVAL; + if (rmnet_is_real_dev_registered(slave_dev)) return -EBUSY;

@@ -434,9 +428,17 @@ int rmnet_add_bridge(struct net_device *rmnet_dev, if (err) return -EBUSY;

+ err = netdev_master_upper_dev_link(slave_dev, rmnet_dev, NULL, NULL, + extack); + if (err) { + rmnet_unregister_real_device(slave_dev); + return err; + } + slave_port = rmnet_get_port_rtnl(slave_dev); slave_port->rmnet_mode = RMNET_EPMODE_BRIDGE; slave_port->bridge_ep = real_dev; + slave_port->rmnet_dev = rmnet_dev;

port->rmnet_mode = RMNET_EPMODE_BRIDGE; port->bridge_ep = slave_dev; @@ -448,16 +450,9 @@ int rmnet_add_bridge(struct net_device *rmnet_dev, int rmnet_del_bridge(struct net_device *rmnet_dev, struct net_device *slave_dev) { - struct rmnet_priv *priv = netdev_priv(rmnet_dev); - struct net_device *real_dev = priv->real_dev; - struct rmnet_port *port, *slave_port; - - port = rmnet_get_port_rtnl(real_dev); - port->rmnet_mode = RMNET_EPMODE_VND; - port->bridge_ep = NULL; + struct rmnet_port *port = rmnet_get_port_rtnl(slave_dev);

- slave_port = rmnet_get_port_rtnl(slave_dev); - rmnet_unregister_real_device(slave_dev, slave_port); + rmnet_unregister_bridge(port);

netdev_dbg(slave_dev, "removed from rmnet as slave\n"); return 0; diff --git a/drivers/net/ethernet/qualcomm/rmnet/rmnet_config.h b/drivers/net/ethernet/qualcomm/rmnet/rmnet_config.h index 6a978f486ce10..7691d1abe6e2a 100644 --- a/drivers/net/ethernet/qualcomm/rmnet/rmnet_config.h +++ b/drivers/net/ethernet/qualcomm/rmnet/rmnet_config.h @@ -37,6 +37,7 @@ struct rmnet_port { u8 rmnet_mode; struct hlist_head muxed_ep[RMNET_MAX_LOGICAL_EP]; struct net_device *bridge_ep; + struct net_device *rmnet_dev; };

extern struct rtnl_link_ops rmnet_link_ops; diff --git a/drivers/net/ethernet/qualcomm/rmnet/rmnet_vnd.c b/drivers/net/ethernet/qualcomm/rmnet/rmnet_vnd.c index d11c16aeb19ad..ed02926f5fe90 100644 --- a/drivers/net/ethernet/qualcomm/rmnet/rmnet_vnd.c +++ b/drivers/net/ethernet/qualcomm/rmnet/rmnet_vnd.c @@ -276,14 +276,6 @@ int rmnet_vnd_dellink(u8 id, struct rmnet_port *port, return 0; }

-u8 rmnet_vnd_get_mux(struct net_device *rmnet_dev) -{ - struct rmnet_priv *priv; - - priv = netdev_priv(rmnet_dev); - return priv->mux_id; -} - int rmnet_vnd_do_flow_control(struct net_device *rmnet_dev, int enable) { netdev_dbg(rmnet_dev, "Setting VND TX queue state to %d\n", enable); diff --git a/drivers/net/ethernet/qualcomm/rmnet/rmnet_vnd.h b/drivers/net/ethernet/qualcomm/rmnet/rmnet_vnd.h index 71e4c32869512..22cfdfd5625a4 100644 --- a/drivers/net/ethernet/qualcomm/rmnet/rmnet_vnd.h +++ b/drivers/net/ethernet/qualcomm/rmnet/rmnet_vnd.h @@ -25,6 +25,5 @@ int rmnet_vnd_dellink(u8 id, struct rmnet_port *port, struct rmnet_endpoint *ep); void rmnet_vnd_rx_fixup(struct sk_buff *skb, struct net_device *dev); void rmnet_vnd_tx_fixup(struct sk_buff *skb, struct net_device *dev); -u8 rmnet_vnd_get_mux(struct net_device *rmnet_dev); void rmnet_vnd_setup(struct net_device *dev); #endif /* _RMNET_VND_H_ */

From: Alex Maftei (amaftei) amaftei@solarflare.com

[ Upstream commit 23797b98909f34b75fd130369bde86f760db69d0 ]

We can't just use the top bits of the last sync event as they could be off-by-one every 65,536 seconds, giving an error in reconstruction of 65,536 seconds.

This patch uses the difference in the bottom 16 bits (mod 2^16) to calculate an offset that needs to be applied to the last sync event to get to the current time.

Signed-off-by: Alexandru-Mihai Maftei amaftei@solarflare.com Acked-by: Martin Habets mhabets@solarflare.com Signed-off-by: David S. Miller davem@davemloft.net Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/net/ethernet/sfc/ptp.c | 38 +++++++++++++++++++++++++++++++--- 1 file changed, 35 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/sfc/ptp.c b/drivers/net/ethernet/sfc/ptp.c index cc8fbf398c0d7..d47151dbe804d 100644 --- a/drivers/net/ethernet/sfc/ptp.c +++ b/drivers/net/ethernet/sfc/ptp.c @@ -563,13 +563,45 @@ efx_ptp_mac_nic_to_ktime_correction(struct efx_nic *efx, u32 nic_major, u32 nic_minor, s32 correction) { + u32 sync_timestamp; ktime_t kt = { 0 }; + s16 delta;

if (!(nic_major & 0x80000000)) { WARN_ON_ONCE(nic_major >> 16); - /* Use the top bits from the latest sync event. */ - nic_major &= 0xffff; - nic_major |= (last_sync_timestamp_major(efx) & 0xffff0000); + + /* Medford provides 48 bits of timestamp, so we must get the top + * 16 bits from the timesync event state. + * + * We only have the lower 16 bits of the time now, but we do + * have a full resolution timestamp at some point in past. As + * long as the difference between the (real) now and the sync + * is less than 2^15, then we can reconstruct the difference + * between those two numbers using only the lower 16 bits of + * each. + * + * Put another way + * + * a - b = ((a mod k) - b) mod k + * + * when -k/2 < (a-b) < k/2. In our case k is 2^16. We know + * (a mod k) and b, so can calculate the delta, a - b. + * + */ + sync_timestamp = last_sync_timestamp_major(efx); + + /* Because delta is s16 this does an implicit mask down to + * 16 bits which is what we need, assuming + * MEDFORD_TX_SECS_EVENT_BITS is 16. delta is signed so that + * we can deal with the (unlikely) case of sync timestamps + * arriving from the future. + */ + delta = nic_major - sync_timestamp; + + /* Recover the fully specified time now, by applying the offset + * to the (fully specified) sync time. + */ + nic_major = sync_timestamp + delta;

kt = ptp->nic_to_kernel_time(nic_major, nic_minor, correction);

From: Navid Emamdoost navid.emamdoost@gmail.com

[ Upstream commit 2507e6ab7a9a440773be476141a255934468c5ef ]

In i2400m_op_rfkill_sw_toggle cmd buffer should be released along with skb response.

Signed-off-by: Navid Emamdoost navid.emamdoost@gmail.com Signed-off-by: David S. Miller davem@davemloft.net Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/net/wimax/i2400m/op-rfkill.c | 1 + 1 file changed, 1 insertion(+)

diff --git a/drivers/net/wimax/i2400m/op-rfkill.c b/drivers/net/wimax/i2400m/op-rfkill.c index b0dba35a8ad2a..7c92e8ace9c2f 100644 --- a/drivers/net/wimax/i2400m/op-rfkill.c +++ b/drivers/net/wimax/i2400m/op-rfkill.c @@ -142,6 +142,7 @@ int i2400m_op_rfkill_sw_toggle(struct wimax_dev *wimax_dev, "%d\n", result); result = 0; error_cmd: + kfree(cmd); kfree_skb(ack_skb); error_msg_to_dev: error_alloc:

From: Faiz Abbas faiz_abbas@ti.com

[ Upstream commit 5c41ea6d52003b5bc77c2a82fd5ca7d480237d89 ]

commit 5b0d62108b46 ("mmc: sdhci-omap: Add platform specific reset callback") skips data resets during tuning operation. Because of this, a data error or data finish interrupt might still arrive after a command error has been handled and the mrq ended. This ends up with a "mmc0: Got data interrupt 0x00000002 even though no data operation was in progress" error message.

Fix this by adding a platform specific callback for sdhci_irq. Mark the mrq as a failure but wait for a data interrupt instead of calling finish_mrq().

Fixes: 5b0d62108b46 ("mmc: sdhci-omap: Add platform specific reset callback") Signed-off-by: Faiz Abbas faiz_abbas@ti.com Acked-by: Adrian Hunter adrian.hunter@intel.com Cc: stable@vger.kernel.org Signed-off-by: Ulf Hansson ulf.hansson@linaro.org Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/mmc/host/sdhci-omap.c | 38 +++++++++++++++++++++++++++++++++++ 1 file changed, 38 insertions(+)

diff --git a/drivers/mmc/host/sdhci-omap.c b/drivers/mmc/host/sdhci-omap.c index 79ee5fc5a2013..833e13cabd2a8 100644 --- a/drivers/mmc/host/sdhci-omap.c +++ b/drivers/mmc/host/sdhci-omap.c @@ -801,6 +801,43 @@ void sdhci_omap_reset(struct sdhci_host *host, u8 mask) sdhci_reset(host, mask); }

+#define CMD_ERR_MASK (SDHCI_INT_CRC | SDHCI_INT_END_BIT | SDHCI_INT_INDEX |\ + SDHCI_INT_TIMEOUT) +#define CMD_MASK (CMD_ERR_MASK | SDHCI_INT_RESPONSE) + +static u32 sdhci_omap_irq(struct sdhci_host *host, u32 intmask) +{ + struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host); + struct sdhci_omap_host *omap_host = sdhci_pltfm_priv(pltfm_host); + + if (omap_host->is_tuning && host->cmd && !host->data_early && + (intmask & CMD_ERR_MASK)) { + + /* + * Since we are not resetting data lines during tuning + * operation, data error or data complete interrupts + * might still arrive. Mark this request as a failure + * but still wait for the data interrupt + */ + if (intmask & SDHCI_INT_TIMEOUT) + host->cmd->error = -ETIMEDOUT; + else + host->cmd->error = -EILSEQ; + + host->cmd = NULL; + + /* + * Sometimes command error interrupts and command complete + * interrupt will arrive together. Clear all command related + * interrupts here. + */ + sdhci_writel(host, intmask & CMD_MASK, SDHCI_INT_STATUS); + intmask &= ~CMD_MASK; + } + + return intmask; +} + static struct sdhci_ops sdhci_omap_ops = { .set_clock = sdhci_omap_set_clock, .set_power = sdhci_omap_set_power, @@ -811,6 +848,7 @@ static struct sdhci_ops sdhci_omap_ops = { .platform_send_init_74_clocks = sdhci_omap_init_74_clocks, .reset = sdhci_omap_reset, .set_uhs_signaling = sdhci_omap_set_uhs_signaling, + .irq = sdhci_omap_irq, };

static int sdhci_omap_set_capabilities(struct sdhci_omap_host *omap_host)

From: Yong Wu yong.wu@mediatek.com

commit 0fe6f7874d467456da6f6a221dd92499a3ab1780 upstream.

DL_FLAG_AUTOREMOVE_CONSUMER/SUPPLIER means "Remove the link automatically on consumer/supplier driver unbind", that means we should remove whole the device_link when there is no this driver no matter what the ref_count of the link is.

CC: Greg Kroah-Hartman gregkh@linuxfoundation.org Signed-off-by: Yong Wu yong.wu@mediatek.com Signed-off-by: Saravana Kannan saravanak@google.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org

--- drivers/base/core.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)

--- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -546,7 +546,7 @@ static void __device_links_no_driver(str continue;

if (link->flags & DL_FLAG_AUTOREMOVE_CONSUMER) - kref_put(&link->kref, __device_link_del); + __device_link_del(&link->kref); else if (link->status != DL_STATE_SUPPLIER_UNBIND) WRITE_ONCE(link->status, DL_STATE_AVAILABLE); } @@ -591,7 +591,7 @@ void device_links_driver_cleanup(struct */ if (link->status == DL_STATE_SUPPLIER_UNBIND && link->flags & DL_FLAG_AUTOREMOVE_SUPPLIER) - kref_put(&link->kref, __device_link_del); + __device_link_del(&link->kref);

WRITE_ONCE(link->status, DL_STATE_DORMANT); }

From: Rafael J. Wysocki rafael.j.wysocki@intel.com

commit 72175d4ea4c442d95cf690c3e968eeee90fd43ca upstream.

Even though stateful device links are managed by the driver core in principle, their creators are allowed and sometimes even expected to drop references to them via device_link_del() or device_link_remove(), but that doesn't really play well with the "persistent" link concept.

If "persistent" managed device links are created from driver probe callbacks, device_link_add() called to do that will take a new reference on the link each time the callback runs and those references will never be dropped, which kind of isn't nice.

This issues arises because of the link reference counting carried out by device_link_add() for existing links, but that is only done to avoid deleting device links that may still be necessary, which shouldn't be a concern for managed (stateful) links. These device links are managed by the driver core and whoever creates one of them will need it at least as long as until the consumer driver is detached from its device and deleting it may be left to the driver core just fine.

For this reason, rework device_link_add() to apply the reference counting to stateless links only and make device_link_del() and device_link_remove() drop references to stateless links only too. After this change, if called to add a stateful device link for a consumer-supplier pair for which a stateful device link is present already, device_link_add() will return the existing link without incrementing its reference counter. Accordingly, device_link_del() and device_link_remove() will WARN() and do nothing when called to drop a reference to a stateful link. Thus, effectively, all stateful device links will be owned by the driver core.

In addition, clean up the handling of the link management flags, DL_FLAG_AUTOREMOVE_CONSUMER and DL_FLAG_AUTOREMOVE_SUPPLIER, so that (a) they are never set at the same time and (b) if device_link_add() is called for a consumer-supplier pair with an existing stateful link between them, the flags of that link will be combined with the flags passed to device_link_add() to ensure that the life time of the link is sufficient for all of the callers of device_link_add() for the same consumer-supplier pair.

Update the device_link_add() kerneldoc comment to reflect the above changes.

Signed-off-by: Rafael J. Wysocki rafael.j.wysocki@intel.com Signed-off-by: Saravana Kannan saravanak@google.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org

--- Documentation/driver-api/device_link.rst | 42 +++++++++++------- drivers/base/core.c | 69 ++++++++++++++++++++++++------- 2 files changed, 79 insertions(+), 32 deletions(-)

--- a/Documentation/driver-api/device_link.rst +++ b/Documentation/driver-api/device_link.rst @@ -25,8 +25,8 @@ suspend/resume and shutdown ordering.

Device links allow representation of such dependencies in the driver core.

-In its standard form, a device link combines *both* dependency types: -It guarantees correct suspend/resume and shutdown ordering between a +In its standard or *managed* form, a device link combines *both* dependency +types: It guarantees correct suspend/resume and shutdown ordering between a "supplier" device and its "consumer" devices, and it guarantees driver presence on the supplier. The consumer devices are not probed before the supplier is bound to a driver, and they're unbound before the supplier @@ -69,12 +69,14 @@ know that the supplier is functional alr the case, for instance, if the consumer has just acquired some resources that would not have been available had the supplier not been functional then).]

-If a device link is added in the ``->probe`` callback of the supplier or -consumer driver, it is typically deleted in its ``->remove`` callback for -symmetry. That way, if the driver is compiled as a module, the device -link is added on module load and orderly deleted on unload. The same -restrictions that apply to device link addition (e.g. exclusion of a -parallel suspend/resume transition) apply equally to deletion. +If a device link with ``DL_FLAG_STATELESS`` set (i.e. a stateless device link) +is added in the ``->probe`` callback of the supplier or consumer driver, it is +typically deleted in its ``->remove`` callback for symmetry. That way, if the +driver is compiled as a module, the device link is added on module load and +orderly deleted on unload. The same restrictions that apply to device link +addition (e.g. exclusion of a parallel suspend/resume transition) apply equally +to deletion. Device links with ``DL_FLAG_STATELESS`` unset (i.e. managed +device links) are deleted automatically by the driver core.

Several flags may be specified on device link addition, two of which have already been mentioned above: ``DL_FLAG_STATELESS`` to express that no @@ -87,8 +89,6 @@ link is added from the consumer's ``->pr can be specified to runtime resume the supplier upon addition of the device link. ``DL_FLAG_AUTOREMOVE_CONSUMER`` causes the device link to be automatically purged when the consumer fails to probe or later unbinds. -This obviates the need to explicitly delete the link in the ``->remove`` -callback or in the error path of the ``->probe`` callback.

Similarly, when the device link is added from supplier's ``->probe`` callback, ``DL_FLAG_AUTOREMOVE_SUPPLIER`` causes the device link to be automatically @@ -97,12 +97,20 @@ purged when the supplier fails to probe Limitations ===========

-Driver authors should be aware that a driver presence dependency (i.e. when -``DL_FLAG_STATELESS`` is not specified on link addition) may cause probing of -the consumer to be deferred indefinitely. This can become a problem if the -consumer is required to probe before a certain initcall level is reached. -Worse, if the supplier driver is blacklisted or missing, the consumer will -never be probed. +Driver authors should be aware that a driver presence dependency for managed +device links (i.e. when ``DL_FLAG_STATELESS`` is not specified on link addition) +may cause probing of the consumer to be deferred indefinitely. This can become +a problem if the consumer is required to probe before a certain initcall level +is reached. Worse, if the supplier driver is blacklisted or missing, the +consumer will never be probed. + +Moreover, managed device links cannot be deleted directly. They are deleted +by the driver core when they are not necessary any more in accordance with the +``DL_FLAG_AUTOREMOVE_CONSUMER`` and ``DL_FLAG_AUTOREMOVE_SUPPLIER`` flags. +However, stateless device links (i.e. device links with ``DL_FLAG_STATELESS`` +set) are expected to be removed by whoever called :c:func:`device_link_add()` +to add them with the help of either :c:func:`device_link_del()` or +:c:func:`device_link_remove()`.

Sometimes drivers depend on optional resources. They are able to operate in a degraded mode (reduced feature set or performance) when those resources @@ -286,4 +294,4 @@ API ===

.. kernel-doc:: drivers/base/core.c - :functions: device_link_add device_link_del + :functions: device_link_add device_link_del device_link_remove --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -179,10 +179,21 @@ void device_pm_move_to_tail(struct devic * of the link. If DL_FLAG_PM_RUNTIME is not set, DL_FLAG_RPM_ACTIVE will be * ignored. * - * If the DL_FLAG_AUTOREMOVE_CONSUMER flag is set, the link will be removed - * automatically when the consumer device driver unbinds from it. Analogously, - * if DL_FLAG_AUTOREMOVE_SUPPLIER is set in @flags, the link will be removed - * automatically when the supplier device driver unbinds from it. + * If DL_FLAG_STATELESS is set in @flags, the link is not going to be managed by + * the driver core and, in particular, the caller of this function is expected + * to drop the reference to the link acquired by it directly. + * + * If that flag is not set, however, the caller of this function is handing the + * management of the link over to the driver core entirely and its return value + * can only be used to check whether or not the link is present. In that case, + * the DL_FLAG_AUTOREMOVE_CONSUMER and DL_FLAG_AUTOREMOVE_SUPPLIER device link + * flags can be used to indicate to the driver core when the link can be safely + * deleted. Namely, setting one of them in @flags indicates to the driver core + * that the link is not going to be used (by the given caller of this function) + * after unbinding the consumer or supplier driver, respectively, from its + * device, so the link can be deleted at that point. If none of them is set, + * the link will be maintained until one of the devices pointed to by it (either + * the consumer or the supplier) is unregistered. * * The combination of DL_FLAG_STATELESS and either DL_FLAG_AUTOREMOVE_CONSUMER * or DL_FLAG_AUTOREMOVE_SUPPLIER set in @flags at the same time is invalid and @@ -228,6 +239,14 @@ struct device_link *device_link_add(stru goto out; }

+ /* + * DL_FLAG_AUTOREMOVE_SUPPLIER indicates that the link will be needed + * longer than for DL_FLAG_AUTOREMOVE_CONSUMER and setting them both + * together doesn't make sense, so prefer DL_FLAG_AUTOREMOVE_SUPPLIER. + */ + if (flags & DL_FLAG_AUTOREMOVE_SUPPLIER) + flags &= ~DL_FLAG_AUTOREMOVE_CONSUMER; + list_for_each_entry(link, &supplier->links.consumers, s_node) { if (link->consumer != consumer) continue; @@ -241,12 +260,6 @@ struct device_link *device_link_add(stru goto out; }

- if (flags & DL_FLAG_AUTOREMOVE_CONSUMER) - link->flags |= DL_FLAG_AUTOREMOVE_CONSUMER; - - if (flags & DL_FLAG_AUTOREMOVE_SUPPLIER) - link->flags |= DL_FLAG_AUTOREMOVE_SUPPLIER; - if (flags & DL_FLAG_PM_RUNTIME) { if (!(link->flags & DL_FLAG_PM_RUNTIME)) { pm_runtime_new_link(consumer); @@ -256,7 +269,25 @@ struct device_link *device_link_add(stru refcount_inc(&link->rpm_active); }

- kref_get(&link->kref); + if (flags & DL_FLAG_STATELESS) { + kref_get(&link->kref); + goto out; + } + + /* + * If the life time of the link following from the new flags is + * longer than indicated by the flags of the existing link, + * update the existing link to stay around longer. + */ + if (flags & DL_FLAG_AUTOREMOVE_SUPPLIER) { + if (link->flags & DL_FLAG_AUTOREMOVE_CONSUMER) { + link->flags &= ~DL_FLAG_AUTOREMOVE_CONSUMER; + link->flags |= DL_FLAG_AUTOREMOVE_SUPPLIER; + } + } else if (!(flags & DL_FLAG_AUTOREMOVE_CONSUMER)) { + link->flags &= ~(DL_FLAG_AUTOREMOVE_CONSUMER | + DL_FLAG_AUTOREMOVE_SUPPLIER); + } goto out; }

@@ -406,8 +437,16 @@ static void __device_link_del(struct kre } #endif /* !CONFIG_SRCU */

+static void device_link_put_kref(struct device_link *link) +{ + if (link->flags & DL_FLAG_STATELESS) + kref_put(&link->kref, __device_link_del); + else + WARN(1, "Unable to drop a managed device link reference\n"); +} + /** - * device_link_del - Delete a link between two devices. + * device_link_del - Delete a stateless link between two devices. * @link: Device link to delete. * * The caller must ensure proper synchronization of this function with runtime @@ -419,14 +458,14 @@ void device_link_del(struct device_link { device_links_write_lock(); device_pm_lock(); - kref_put(&link->kref, __device_link_del); + device_link_put_kref(link); device_pm_unlock(); device_links_write_unlock(); } EXPORT_SYMBOL_GPL(device_link_del);

/** - * device_link_remove - remove a link between two devices. + * device_link_remove - Delete a stateless link between two devices. * @consumer: Consumer end of the link. * @supplier: Supplier end of the link. * @@ -445,7 +484,7 @@ void device_link_remove(void *consumer,

list_for_each_entry(link, &supplier->links.consumers, s_node) { if (link->consumer == consumer) { - kref_put(&link->kref, __device_link_del); + device_link_put_kref(link); break; } }

From: Rafael J. Wysocki rafael.j.wysocki@intel.com

commit 515db266a9dace92b0cbaed9a6044dd5304b8ca9 upstream.

If device_link_add() is called for a consumer/supplier pair with an existing device link between them and the existing link's type is not in agreement with the flags passed to that function by its caller, NULL will be returned. That is seriously inconvenient, because it forces the callers of device_link_add() to worry about what others may or may not do even if that is not relevant to them for any other reasons.

It turns out, however, that this limitation can be made go away relatively easily.

The underlying observation is that if DL_FLAG_STATELESS has been passed to device_link_add() in flags for the given consumer/supplier pair at least once, calling either device_link_del() or device_link_remove() to release the link returned by it should work, but there are no other requirements associated with that flag. In turn, if at least one of the callers of device_link_add() for the given consumer/supplier pair has not passed DL_FLAG_STATELESS to it in flags, the driver core should track the status of the link and act on it as appropriate (ie. the link should be treated as "managed"). This means that DL_FLAG_STATELESS needs to be set for managed device links and it should be valid to call device_link_del() or device_link_remove() to drop references to them in certain sutiations.

To allow that to happen, introduce a new (internal) device link flag called DL_FLAG_MANAGED and make device_link_add() set it automatically whenever DL_FLAG_STATELESS is not passed to it. Also make it take additional references to existing device links that were previously stateless (that is, with DL_FLAG_STATELESS set and DL_FLAG_MANAGED unset) and will need to be managed going forward and initialize their status (which has been DL_STATE_NONE so far).

Accordingly, when a managed device link is dropped automatically by the driver core, make it clear DL_FLAG_MANAGED, reset the link's status back to DL_STATE_NONE and drop the reference to it associated with DL_FLAG_MANAGED instead of just deleting it right away (to allow it to stay around in case it still needs to be released explicitly by someone).

With that, since setting DL_FLAG_STATELESS doesn't mean that the device link in question is not managed any more, replace all of the status-tracking checks against DL_FLAG_STATELESS with analogous checks against DL_FLAG_MANAGED and update the documentation to reflect these changes.

While at it, make device_link_add() reject flags that it does not recognize, including DL_FLAG_MANAGED.

Signed-off-by: Rafael J. Wysocki rafael.j.wysocki@intel.com Reviewed-by: Saravana Kannan saravanak@google.com Tested-by: Marek Szyprowski m.szyprowski@samsung.com Review-by: Saravana Kannan saravanak@google.com Link: https://lore.kernel.org/r/2305283.AStDPdUUnE@kreacher Signed-off-by: Saravana Kannan saravanak@google.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org

--- Documentation/driver-api/device_link.rst | 4 drivers/base/core.c | 176 +++++++++++++++++-------------- drivers/base/power/runtime.c | 4 include/linux/device.h | 4 4 files changed, 106 insertions(+), 82 deletions(-)

--- a/Documentation/driver-api/device_link.rst +++ b/Documentation/driver-api/device_link.rst @@ -75,8 +75,8 @@ typically deleted in its ``->remove`` ca driver is compiled as a module, the device link is added on module load and orderly deleted on unload. The same restrictions that apply to device link addition (e.g. exclusion of a parallel suspend/resume transition) apply equally -to deletion. Device links with ``DL_FLAG_STATELESS`` unset (i.e. managed -device links) are deleted automatically by the driver core. +to deletion. Device links managed by the driver core are deleted automatically +by it.

Several flags may be specified on device link addition, two of which have already been mentioned above: ``DL_FLAG_STATELESS`` to express that no --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -124,6 +124,50 @@ static int device_is_dependent(struct de return ret; }

+static void device_link_init_status(struct device_link *link, + struct device *consumer, + struct device *supplier) +{ + switch (supplier->links.status) { + case DL_DEV_PROBING: + switch (consumer->links.status) { + case DL_DEV_PROBING: + /* + * A consumer driver can create a link to a supplier + * that has not completed its probing yet as long as it + * knows that the supplier is already functional (for + * example, it has just acquired some resources from the + * supplier). + */ + link->status = DL_STATE_CONSUMER_PROBE; + break; + default: + link->status = DL_STATE_DORMANT; + break; + } + break; + case DL_DEV_DRIVER_BOUND: + switch (consumer->links.status) { + case DL_DEV_PROBING: + link->status = DL_STATE_CONSUMER_PROBE; + break; + case DL_DEV_DRIVER_BOUND: + link->status = DL_STATE_ACTIVE; + break; + default: + link->status = DL_STATE_AVAILABLE; + break; + } + break; + case DL_DEV_UNBINDING: + link->status = DL_STATE_SUPPLIER_UNBIND; + break; + default: + link->status = DL_STATE_DORMANT; + break; + } +} + static int device_reorder_to_tail(struct device *dev, void *not_used) { struct device_link *link; @@ -165,6 +209,10 @@ void device_pm_move_to_tail(struct devic device_links_read_unlock(idx); }

+#define DL_MANAGED_LINK_FLAGS (DL_FLAG_AUTOREMOVE_CONSUMER | \ + DL_FLAG_AUTOREMOVE_SUPPLIER | \ + DL_FLAG_AUTOPROBE_CONSUMER) + /** * device_link_add - Create a link between two devices. * @consumer: Consumer end of the link. @@ -179,9 +227,9 @@ void device_pm_move_to_tail(struct devic * of the link. If DL_FLAG_PM_RUNTIME is not set, DL_FLAG_RPM_ACTIVE will be * ignored. * - * If DL_FLAG_STATELESS is set in @flags, the link is not going to be managed by - * the driver core and, in particular, the caller of this function is expected - * to drop the reference to the link acquired by it directly. + * If DL_FLAG_STATELESS is set in @flags, the caller of this function is + * expected to release the link returned by it directly with the help of either + * device_link_del() or device_link_remove(). * * If that flag is not set, however, the caller of this function is handing the * management of the link over to the driver core entirely and its return value @@ -201,9 +249,16 @@ void device_pm_move_to_tail(struct devic * be used to request the driver core to automaticall probe for a consmer * driver after successfully binding a driver to the supplier device. * - * The combination of DL_FLAG_STATELESS and either DL_FLAG_AUTOREMOVE_CONSUMER - * or DL_FLAG_AUTOREMOVE_SUPPLIER set in @flags at the same time is invalid and - * will cause NULL to be returned upfront. + * The combination of DL_FLAG_STATELESS and one of DL_FLAG_AUTOREMOVE_CONSUMER, + * DL_FLAG_AUTOREMOVE_SUPPLIER, or DL_FLAG_AUTOPROBE_CONSUMER set in @flags at + * the same time is invalid and will cause NULL to be returned upfront. + * However, if a device link between the given @consumer and @supplier pair + * exists already when this function is called for them, the existing link will + * be returned regardless of its current type and status (the link's flags may + * be modified then). The caller of this function is then expected to treat + * the link as though it has just been created, so (in particular) if + * DL_FLAG_STATELESS was passed in @flags, the link needs to be released + * explicitly when not needed any more (as stated above). * * A side effect of the link creation is re-ordering of dpm_list and the * devices_kset list by moving the consumer device and all devices depending @@ -220,10 +275,8 @@ struct device_link *device_link_add(stru struct device_link *link;

if (!consumer || !supplier || - (flags & DL_FLAG_STATELESS && - flags & (DL_FLAG_AUTOREMOVE_CONSUMER | - DL_FLAG_AUTOREMOVE_SUPPLIER | - DL_FLAG_AUTOPROBE_CONSUMER)) || + (flags & ~(DL_FLAG_STATELESS | DL_MANAGED_LINK_FLAGS)) || + (flags & DL_FLAG_STATELESS && flags & DL_MANAGED_LINK_FLAGS) || (flags & DL_FLAG_AUTOPROBE_CONSUMER && flags & (DL_FLAG_AUTOREMOVE_CONSUMER | DL_FLAG_AUTOREMOVE_SUPPLIER))) @@ -236,6 +289,9 @@ struct device_link *device_link_add(stru } }

+ if (!(flags & DL_FLAG_STATELESS)) + flags |= DL_FLAG_MANAGED; + device_links_write_lock(); device_pm_lock();

@@ -262,15 +318,6 @@ struct device_link *device_link_add(stru if (link->consumer != consumer) continue;

- /* - * Don't return a stateless link if the caller wants a stateful - * one and vice versa. - */ - if (WARN_ON((flags & DL_FLAG_STATELESS) != (link->flags & DL_FLAG_STATELESS))) { - link = NULL; - goto out; - } - if (flags & DL_FLAG_PM_RUNTIME) { if (!(link->flags & DL_FLAG_PM_RUNTIME)) { pm_runtime_new_link(consumer); @@ -281,6 +328,7 @@ struct device_link *device_link_add(stru }

if (flags & DL_FLAG_STATELESS) { + link->flags |= DL_FLAG_STATELESS; kref_get(&link->kref); goto out; } @@ -299,6 +347,11 @@ struct device_link *device_link_add(stru link->flags &= ~(DL_FLAG_AUTOREMOVE_CONSUMER | DL_FLAG_AUTOREMOVE_SUPPLIER); } + if (!(link->flags & DL_FLAG_MANAGED)) { + kref_get(&link->kref); + link->flags |= DL_FLAG_MANAGED; + device_link_init_status(link, consumer, supplier); + } goto out; }

@@ -325,48 +378,10 @@ struct device_link *device_link_add(stru kref_init(&link->kref);

/* Determine the initial link state. */ - if (flags & DL_FLAG_STATELESS) { + if (flags & DL_FLAG_STATELESS) link->status = DL_STATE_NONE; - } else { - switch (supplier->links.status) { - case DL_DEV_PROBING: - switch (consumer->links.status) { - case DL_DEV_PROBING: - /* - * A consumer driver can create a link to a - * supplier that has not completed its probing - * yet as long as it knows that the supplier is - * already functional (for example, it has just - * acquired some resources from the supplier). - */ - link->status = DL_STATE_CONSUMER_PROBE; - break; - default: - link->status = DL_STATE_DORMANT; - break; - } - break; - case DL_DEV_DRIVER_BOUND: - switch (consumer->links.status) { - case DL_DEV_PROBING: - link->status = DL_STATE_CONSUMER_PROBE; - break; - case DL_DEV_DRIVER_BOUND: - link->status = DL_STATE_ACTIVE; - break; - default: - link->status = DL_STATE_AVAILABLE; - break; - } - break; - case DL_DEV_UNBINDING: - link->status = DL_STATE_SUPPLIER_UNBIND; - break; - default: - link->status = DL_STATE_DORMANT; - break; - } - } + else + device_link_init_status(link, consumer, supplier);

/* * Some callers expect the link creation during consumer driver probe to @@ -528,7 +543,7 @@ static void device_links_missing_supplie * mark the link as "consumer probe in progress" to make the supplier removal * wait for us to complete (or bad things may happen). * - * Links with the DL_FLAG_STATELESS flag set are ignored. + * Links without the DL_FLAG_MANAGED flag set are ignored. */ int device_links_check_suppliers(struct device *dev) { @@ -538,7 +553,7 @@ int device_links_check_suppliers(struct device_links_write_lock();

list_for_each_entry(link, &dev->links.suppliers, c_node) { - if (link->flags & DL_FLAG_STATELESS) + if (!(link->flags & DL_FLAG_MANAGED)) continue;

if (link->status != DL_STATE_AVAILABLE) { @@ -563,7 +578,7 @@ int device_links_check_suppliers(struct * * Also change the status of @dev's links to suppliers to "active". * - * Links with the DL_FLAG_STATELESS flag set are ignored. + * Links without the DL_FLAG_MANAGED flag set are ignored. */ void device_links_driver_bound(struct device *dev) { @@ -572,7 +587,7 @@ void device_links_driver_bound(struct de device_links_write_lock();

list_for_each_entry(link, &dev->links.consumers, s_node) { - if (link->flags & DL_FLAG_STATELESS) + if (!(link->flags & DL_FLAG_MANAGED)) continue;

/* @@ -593,7 +608,7 @@ void device_links_driver_bound(struct de }

list_for_each_entry(link, &dev->links.suppliers, c_node) { - if (link->flags & DL_FLAG_STATELESS) + if (!(link->flags & DL_FLAG_MANAGED)) continue;

WARN_ON(link->status != DL_STATE_CONSUMER_PROBE); @@ -605,6 +620,13 @@ void device_links_driver_bound(struct de device_links_write_unlock(); }

+static void device_link_drop_managed(struct device_link *link) +{ + link->flags &= ~DL_FLAG_MANAGED; + WRITE_ONCE(link->status, DL_STATE_NONE); + kref_put(&link->kref, __device_link_del); +} + /** * __device_links_no_driver - Update links of a device without a driver. * @dev: Device without a drvier. @@ -615,18 +637,18 @@ void device_links_driver_bound(struct de * unless they already are in the "supplier unbind in progress" state in which * case they need not be updated. * - * Links with the DL_FLAG_STATELESS flag set are ignored. + * Links without the DL_FLAG_MANAGED flag set are ignored. */ static void __device_links_no_driver(struct device *dev) { struct device_link *link, *ln;

list_for_each_entry_safe_reverse(link, ln, &dev->links.suppliers, c_node) { - if (link->flags & DL_FLAG_STATELESS) + if (!(link->flags & DL_FLAG_MANAGED)) continue;

if (link->flags & DL_FLAG_AUTOREMOVE_CONSUMER) - __device_link_del(&link->kref); + device_link_drop_managed(link); else if (link->status == DL_STATE_CONSUMER_PROBE || link->status == DL_STATE_ACTIVE) WRITE_ONCE(link->status, DL_STATE_AVAILABLE); @@ -643,7 +665,7 @@ static void __device_links_no_driver(str * %__device_links_no_driver() to update links to suppliers for it as * appropriate. * - * Links with the DL_FLAG_STATELESS flag set are ignored. + * Links without the DL_FLAG_MANAGED flag set are ignored. */ void device_links_no_driver(struct device *dev) { @@ -652,7 +674,7 @@ void device_links_no_driver(struct devic device_links_write_lock();

list_for_each_entry(link, &dev->links.consumers, s_node) { - if (link->flags & DL_FLAG_STATELESS) + if (!(link->flags & DL_FLAG_MANAGED)) continue;

/* @@ -680,7 +702,7 @@ void device_links_no_driver(struct devic * invoke %__device_links_no_driver() to update links to suppliers for it as * appropriate. * - * Links with the DL_FLAG_STATELESS flag set are ignored. + * Links without the DL_FLAG_MANAGED flag set are ignored. */ void device_links_driver_cleanup(struct device *dev) { @@ -689,7 +711,7 @@ void device_links_driver_cleanup(struct device_links_write_lock();

list_for_each_entry_safe(link, ln, &dev->links.consumers, s_node) { - if (link->flags & DL_FLAG_STATELESS) + if (!(link->flags & DL_FLAG_MANAGED)) continue;

WARN_ON(link->flags & DL_FLAG_AUTOREMOVE_CONSUMER); @@ -702,7 +724,7 @@ void device_links_driver_cleanup(struct */ if (link->status == DL_STATE_SUPPLIER_UNBIND && link->flags & DL_FLAG_AUTOREMOVE_SUPPLIER) - __device_link_del(&link->kref); + device_link_drop_managed(link);

WRITE_ONCE(link->status, DL_STATE_DORMANT); } @@ -724,7 +746,7 @@ void device_links_driver_cleanup(struct * * Return 'false' if there are no probing or active consumers. * - * Links with the DL_FLAG_STATELESS flag set are ignored. + * Links without the DL_FLAG_MANAGED flag set are ignored. */ bool device_links_busy(struct device *dev) { @@ -734,7 +756,7 @@ bool device_links_busy(struct device *de device_links_write_lock();

list_for_each_entry(link, &dev->links.consumers, s_node) { - if (link->flags & DL_FLAG_STATELESS) + if (!(link->flags & DL_FLAG_MANAGED)) continue;

if (link->status == DL_STATE_CONSUMER_PROBE @@ -764,7 +786,7 @@ bool device_links_busy(struct device *de * driver to unbind and start over (the consumer will not re-probe as we have * changed the state of the link already). * - * Links with the DL_FLAG_STATELESS flag set are ignored. + * Links without the DL_FLAG_MANAGED flag set are ignored. */ void device_links_unbind_consumers(struct device *dev) { @@ -776,7 +798,7 @@ void device_links_unbind_consumers(struc list_for_each_entry(link, &dev->links.consumers, s_node) { enum device_link_state status;

- if (link->flags & DL_FLAG_STATELESS) + if (!(link->flags & DL_FLAG_MANAGED)) continue;

status = link->status; --- a/drivers/base/power/runtime.c +++ b/drivers/base/power/runtime.c @@ -1531,7 +1531,7 @@ void pm_runtime_remove(struct device *de * runtime PM references to the device, drop the usage counter of the device * (as many times as needed). * - * Links with the DL_FLAG_STATELESS flag set are ignored. + * Links with the DL_FLAG_MANAGED flag unset are ignored. * * Since the device is guaranteed to be runtime-active at the point this is * called, nothing else needs to be done here. @@ -1548,7 +1548,7 @@ void pm_runtime_clean_up_links(struct de idx = device_links_read_lock();

list_for_each_entry_rcu(link, &dev->links.consumers, s_node) { - if (link->flags & DL_FLAG_STATELESS) + if (!(link->flags & DL_FLAG_MANAGED)) continue;

while (refcount_dec_not_one(&link->rpm_active)) --- a/include/linux/device.h +++ b/include/linux/device.h @@ -820,12 +820,13 @@ enum device_link_state { /* * Device link flags. * - * STATELESS: The core won't track the presence of supplier/consumer drivers. + * STATELESS: The core will not remove this link automatically. * AUTOREMOVE_CONSUMER: Remove the link automatically on consumer driver unbind. * PM_RUNTIME: If set, the runtime PM framework will use this link. * RPM_ACTIVE: Run pm_runtime_get_sync() on the supplier during link creation. * AUTOREMOVE_SUPPLIER: Remove the link automatically on supplier driver unbind. * AUTOPROBE_CONSUMER: Probe consumer driver automatically after supplier binds. + * MANAGED: The core tracks presence of supplier/consumer drivers (internal). */ #define DL_FLAG_STATELESS BIT(0) #define DL_FLAG_AUTOREMOVE_CONSUMER BIT(1) @@ -833,6 +834,7 @@ enum device_link_state { #define DL_FLAG_RPM_ACTIVE BIT(3) #define DL_FLAG_AUTOREMOVE_SUPPLIER BIT(4) #define DL_FLAG_AUTOPROBE_CONSUMER BIT(5) +#define DL_FLAG_MANAGED BIT(6)

/** * struct device_link - Device link representation.

From: Carl Huang cjhuang@codeaurora.org

commit ce57785bf91b1ceaef4f4bffed8a47dc0919c8da upstream.

The len used for skb_put_padto is wrong, it need to add len of hdr.

In qrtr_node_enqueue, local variable size_t len is assign with skb->len, then skb_push(skb, sizeof(*hdr)) will add skb->len with sizeof(*hdr), so local variable size_t len is not same with skb->len after skb_push(skb, sizeof(*hdr)).

Then the purpose of skb_put_padto(skb, ALIGN(len, 4)) is to add add pad to the end of the skb's data if skb->len is not aligned to 4, but unfortunately it use len instead of skb->len, at this line, skb->len is 32 bytes(sizeof(*hdr)) more than len, for example, len is 3 bytes, then skb->len is 35 bytes(3 + 32), and ALIGN(len, 4) is 4 bytes, so __skb_put_padto will do nothing after check size(35) < len(4), the correct value should be 36(sizeof(*hdr) + ALIGN(len, 4) = 32 + 4), then __skb_put_padto will pass check size(35) < len(36) and add 1 byte to the end of skb's data, then logic is correct.

function of skb_push: void *skb_push(struct sk_buff *skb, unsigned int len) { skb->data -= len; skb->len += len; if (unlikely(skb->data < skb->head)) skb_under_panic(skb, len, __builtin_return_address(0)); return skb->data; }

function of skb_put_padto static inline int skb_put_padto(struct sk_buff *skb, unsigned int len) { return __skb_put_padto(skb, len, true); }

function of __skb_put_padto static inline int __skb_put_padto(struct sk_buff *skb, unsigned int len, bool free_on_error) { unsigned int size = skb->len;

if (unlikely(size < len)) { len -= size; if (__skb_pad(skb, len, free_on_error)) return -ENOMEM; __skb_put(skb, len); } return 0; }

Signed-off-by: Carl Huang cjhuang@codeaurora.org Signed-off-by: Wen Gong wgong@codeaurora.org Cc: Doug Anderson dianders@chromium.org Signed-off-by: David S. Miller davem@davemloft.net Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org

--- net/qrtr/qrtr.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)

--- a/net/qrtr/qrtr.c +++ b/net/qrtr/qrtr.c @@ -203,7 +203,7 @@ static int qrtr_node_enqueue(struct qrtr hdr->size = cpu_to_le32(len); hdr->confirm_rx = 0;

- skb_put_padto(skb, ALIGN(len, 4)); + skb_put_padto(skb, ALIGN(len, 4) + sizeof(*hdr));

mutex_lock(&node->ep_lock); if (node->ep)

From: Kees Cook keescook@chromium.org

commit f87b1c49bc675da30d8e1e8f4b60b800312c7b90 upstream.

When the uaccess .fixup section was renamed to .text.fixup, one case was missed. Under ld.bfd, the orphaned section was moved close to .text (since they share the "ax" bits), so things would work normally on uaccess faults. Under ld.lld, the orphaned section was placed outside the .text section, making it unreachable.

Link: https://github.com/ClangBuiltLinux/linux/issues/282 Link: https://bugs.chromium.org/p/chromium/issues/detail?id=1020633#c44 Link: https://lore.kernel.org/r/nycvar.YSQ.7.76.1912032147340.17114@knanqh.ubzr Link: https://lore.kernel.org/lkml/202002071754.F5F073F1D@keescook/

Fixes: c4a84ae39b4a5 ("ARM: 8322/1: keep .text and .fixup regions closer together") Cc: stable@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org Reviewed-by: Ard Biesheuvel ardb@kernel.org Reviewed-by: Nick Desaulniers ndesaulniers@google.com Signed-off-by: Russell King rmk+kernel@armlinux.org.uk Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org

--- arch/arm/lib/copy_from_user.S | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)

--- a/arch/arm/lib/copy_from_user.S +++ b/arch/arm/lib/copy_from_user.S @@ -100,7 +100,7 @@ ENTRY(arm_copy_from_user)

ENDPROC(arm_copy_from_user)

- .pushsection .fixup,"ax" + .pushsection .text.fixup,"ax" .align 0 copy_abort_preamble ldmfd sp!, {r1, r2, r3}

From: Chen-Tsung Hsieh chentsung@chromium.org

commit 58322a1590fc189a8e1e349d309637d4a4942840 upstream.

Add 1 additional hammer-like device.

Signed-off-by: Chen-Tsung Hsieh chentsung@chromium.org Reviewed-by: Nicolas Boichat drinkcat@chromium.org Signed-off-by: Jiri Kosina jkosina@suse.cz Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org

--- drivers/hid/hid-google-hammer.c | 2 ++ drivers/hid/hid-ids.h | 1 + 2 files changed, 3 insertions(+)

--- a/drivers/hid/hid-google-hammer.c +++ b/drivers/hid/hid-google-hammer.c @@ -125,6 +125,8 @@ static const struct hid_device_id hammer { HID_DEVICE(BUS_USB, HID_GROUP_GENERIC, USB_VENDOR_ID_GOOGLE, USB_DEVICE_ID_GOOGLE_MASTERBALL) }, { HID_DEVICE(BUS_USB, HID_GROUP_GENERIC, + USB_VENDOR_ID_GOOGLE, USB_DEVICE_ID_GOOGLE_MOONBALL) }, + { HID_DEVICE(BUS_USB, HID_GROUP_GENERIC, USB_VENDOR_ID_GOOGLE, USB_DEVICE_ID_GOOGLE_STAFF) }, { HID_DEVICE(BUS_USB, HID_GROUP_GENERIC, USB_VENDOR_ID_GOOGLE, USB_DEVICE_ID_GOOGLE_WAND) }, --- a/drivers/hid/hid-ids.h +++ b/drivers/hid/hid-ids.h @@ -468,6 +468,7 @@ #define USB_DEVICE_ID_GOOGLE_WHISKERS 0x5030 #define USB_DEVICE_ID_GOOGLE_MASTERBALL 0x503c #define USB_DEVICE_ID_GOOGLE_MAGNEMITE 0x503d +#define USB_DEVICE_ID_GOOGLE_MOONBALL 0x5044

#define USB_VENDOR_ID_GOTOP 0x08f2 #define USB_DEVICE_ID_SUPER_Q2 0x007f

From: Matteo Croce mcroce@redhat.com

commit 3e72dfdf8227b052393f71d820ec7599909dddc2 upstream.

Similarly to commit c543cb4a5f07 ("ipv4: ensure rcu_read_lock() in ipv4_link_failure()"), __ip_options_compile() must be called under rcu protection.

Fixes: 3da1ed7ac398 ("net: avoid use IPCB in cipso_v4_error") Suggested-by: Guillaume Nault gnault@redhat.com Signed-off-by: Matteo Croce mcroce@redhat.com Acked-by: Paul Moore paul@paul-moore.com Signed-off-by: David S. Miller davem@davemloft.net Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org

--- net/ipv4/cipso_ipv4.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-)

--- a/net/ipv4/cipso_ipv4.c +++ b/net/ipv4/cipso_ipv4.c @@ -1738,6 +1738,7 @@ void cipso_v4_error(struct sk_buff *skb, { unsigned char optbuf[sizeof(struct ip_options) + 40]; struct ip_options *opt = (struct ip_options *)optbuf; + int res;

if (ip_hdr(skb)->protocol == IPPROTO_ICMP || error != -EACCES) return; @@ -1749,7 +1750,11 @@ void cipso_v4_error(struct sk_buff *skb,

memset(opt, 0, sizeof(struct ip_options)); opt->optlen = ip_hdr(skb)->ihl*4 - sizeof(struct iphdr); - if (__ip_options_compile(dev_net(skb->dev), opt, skb, NULL)) + rcu_read_lock(); + res = __ip_options_compile(dev_net(skb->dev), opt, skb, NULL); + rcu_read_unlock(); + + if (res) return;

if (gateway)