From a318c2432218d3cd189ec8228b8a795666899c2a Mon Sep 17 00:00:00 2001 From: Philipp Zabel Date: Fri, 31 Aug 2018 14:34:02 +0200 Subject: mfd: da9063: Fix DT probing with constraints Commit 1c892e38ce59 ("regulator: da9063: Handle less LDOs on DA9063L") reordered the da9063_regulator_info[] array, but not the DA9063_ID_* regulator ids and not the da9063_matches[] array, because ids are used as indices in the array initializer. This mismatch between regulator id and da9063_regulator_info[] array index causes the driver probe to fail because constraints from DT are not applied to the correct regulator: da9063 0-0058: Device detected (chip-ID: 0x61, var-ID: 0x50) DA9063_BMEM: Bringing 900000uV into 3300000-3300000uV DA9063_LDO9: Bringing 3300000uV into 2500000-2500000uV DA9063_LDO1: Bringing 900000uV into 3300000-3300000uV DA9063_LDO1: failed to apply 3300000-3300000uV constraint(-22) This patch reorders the DA9063_ID_* as apparently intended, and with them the entries in the da90630_matches[] array. Fixes: 1c892e38ce59 ("regulator: da9063: Handle less LDOs on DA9063L") Signed-off-by: Philipp Zabel Reviewed-by: Marek Vasut Reviewed-by: Geert Uytterhoeven Signed-off-by: Lee Jones --- include/linux/mfd/da9063/pdata.h | 16 +++++++++------- 1 file changed, 9 insertions(+), 7 deletions(-) (limited to 'include/linux') diff --git a/include/linux/mfd/da9063/pdata.h b/include/linux/mfd/da9063/pdata.h index 8a125701ef7b..50bed4f89c1a 100644 --- a/include/linux/mfd/da9063/pdata.h +++ b/include/linux/mfd/da9063/pdata.h @@ -21,7 +21,7 @@ /* * Regulator configuration */ -/* DA9063 regulator IDs */ +/* DA9063 and DA9063L regulator IDs */ enum { /* BUCKs */ DA9063_ID_BCORE1, @@ -37,18 +37,20 @@ enum { DA9063_ID_BMEM_BIO_MERGED, /* When two BUCKs are merged, they cannot be reused separately */ - /* LDOs */ + /* LDOs on both DA9063 and DA9063L */ + DA9063_ID_LDO3, + DA9063_ID_LDO7, + DA9063_ID_LDO8, + DA9063_ID_LDO9, + DA9063_ID_LDO11, + + /* DA9063-only LDOs */ DA9063_ID_LDO1, DA9063_ID_LDO2, - DA9063_ID_LDO3, DA9063_ID_LDO4, DA9063_ID_LDO5, DA9063_ID_LDO6, - DA9063_ID_LDO7, - DA9063_ID_LDO8, - DA9063_ID_LDO9, DA9063_ID_LDO10, - DA9063_ID_LDO11, }; /* Regulators platform data */ -- cgit v1.2.3-71-gd317 From 37a3a98ef601f89100e3bb657fb0e190b857028c Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Mon, 10 Sep 2018 16:20:25 +0200 Subject: ALSA: hda - Enable runtime PM only for discrete GPU The recent change of vga_switcheroo allowed the runtime PM for HD-audio on AMD GPUs, but this also resulted in a regression. When the HD-audio controller driver gets runtime-suspended, HD-audio link is turned off, and the hotplug notification is ignored. This leads to the inconsistent audio state (the connection isn't notified and ELD is ignored). The best fix would be to implement the proper ELD notification via the audio component, but it's still not ready. As a quick workaround, this patch adds the check of runtime_idle and allows the runtime suspend only when the vga_switcheroo is bound with discrete GPU. That is, a system with a single GPU and APU would be again without runtime PM to keep the HD-audio link for the hotplug notification and ELD read out. Also, the codec->auto_runtime_pm flag is set only for the discrete GPU at the time GPU gets bound via vga_switcheroo (i.e. only dGPU is forcibly runtime-PM enabled), so that APU can still get the ELD notification. For identifying which GPU is bound, a new vga_switcheroo client callback, gpu_bound, is implemented. The vga_switcheroo simply calls this when GPU is bound, and tells whether it's dGPU or APU. Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=200945 Fixes: 07f4f97d7b4b ("vga_switcheroo: Use device link for HDA controller") Reported-by: Jian-Hong Pan Tested-by: Jian-Hong Pan Acked-by: Lukas Wunner Signed-off-by: Takashi Iwai --- drivers/gpu/vga/vga_switcheroo.c | 2 + include/linux/vga_switcheroo.h | 3 ++ sound/pci/hda/hda_intel.c | 86 +++++++++++++++++++++++++++++----------- sound/pci/hda/hda_intel.h | 1 + 4 files changed, 69 insertions(+), 23 deletions(-) (limited to 'include/linux') diff --git a/drivers/gpu/vga/vga_switcheroo.c b/drivers/gpu/vga/vga_switcheroo.c index a96bf46bc483..cf2a18571d48 100644 --- a/drivers/gpu/vga/vga_switcheroo.c +++ b/drivers/gpu/vga/vga_switcheroo.c @@ -215,6 +215,8 @@ static void vga_switcheroo_enable(void) return; client->id = ret | ID_BIT_AUDIO; + if (client->ops->gpu_bound) + client->ops->gpu_bound(client->pdev, ret); } vga_switcheroo_debugfs_init(&vgasr_priv); diff --git a/include/linux/vga_switcheroo.h b/include/linux/vga_switcheroo.h index a34539b7f750..7e6ac0114d55 100644 --- a/include/linux/vga_switcheroo.h +++ b/include/linux/vga_switcheroo.h @@ -133,15 +133,18 @@ struct vga_switcheroo_handler { * @can_switch: check if the device is in a position to switch now. * Mandatory. The client should return false if a user space process * has one of its device files open + * @gpu_bound: notify the client id to audio client when the GPU is bound. * * Client callbacks. A client can be either a GPU or an audio device on a GPU. * The @set_gpu_state and @can_switch methods are mandatory, @reprobe may be * set to NULL. For audio clients, the @reprobe member is bogus. + * OTOH, @gpu_bound is only for audio clients, and not used for GPU clients. */ struct vga_switcheroo_client_ops { void (*set_gpu_state)(struct pci_dev *dev, enum vga_switcheroo_state); void (*reprobe)(struct pci_dev *dev); bool (*can_switch)(struct pci_dev *dev); + void (*gpu_bound)(struct pci_dev *dev, enum vga_switcheroo_client_id); }; #if defined(CONFIG_VGA_SWITCHEROO) diff --git a/sound/pci/hda/hda_intel.c b/sound/pci/hda/hda_intel.c index 1b2ce304152a..aa4c672dbaf7 100644 --- a/sound/pci/hda/hda_intel.c +++ b/sound/pci/hda/hda_intel.c @@ -365,8 +365,10 @@ enum { */ #ifdef SUPPORT_VGA_SWITCHEROO #define use_vga_switcheroo(chip) ((chip)->use_vga_switcheroo) +#define needs_eld_notify_link(chip) ((chip)->need_eld_notify_link) #else #define use_vga_switcheroo(chip) 0 +#define needs_eld_notify_link(chip) false #endif #define CONTROLLER_IN_GPU(pci) (((pci)->device == 0x0a0c) || \ @@ -453,6 +455,7 @@ static inline void mark_runtime_wc(struct azx *chip, struct azx_dev *azx_dev, #endif static int azx_acquire_irq(struct azx *chip, int do_disconnect); +static void set_default_power_save(struct azx *chip); /* * initialize the PCI registers @@ -1201,6 +1204,10 @@ static int azx_runtime_idle(struct device *dev) azx_bus(chip)->codec_powered || !chip->running) return -EBUSY; + /* ELD notification gets broken when HD-audio bus is off */ + if (needs_eld_notify_link(hda)) + return -EBUSY; + return 0; } @@ -1298,6 +1305,36 @@ static bool azx_vs_can_switch(struct pci_dev *pci) return true; } +/* + * The discrete GPU cannot power down unless the HDA controller runtime + * suspends, so activate runtime PM on codecs even if power_save == 0. + */ +static void setup_vga_switcheroo_runtime_pm(struct azx *chip) +{ + struct hda_intel *hda = container_of(chip, struct hda_intel, chip); + struct hda_codec *codec; + + if (hda->use_vga_switcheroo && !hda->need_eld_notify_link) { + list_for_each_codec(codec, &chip->bus) + codec->auto_runtime_pm = 1; + /* reset the power save setup */ + if (chip->running) + set_default_power_save(chip); + } +} + +static void azx_vs_gpu_bound(struct pci_dev *pci, + enum vga_switcheroo_client_id client_id) +{ + struct snd_card *card = pci_get_drvdata(pci); + struct azx *chip = card->private_data; + struct hda_intel *hda = container_of(chip, struct hda_intel, chip); + + if (client_id == VGA_SWITCHEROO_DIS) + hda->need_eld_notify_link = 0; + setup_vga_switcheroo_runtime_pm(chip); +} + static void init_vga_switcheroo(struct azx *chip) { struct hda_intel *hda = container_of(chip, struct hda_intel, chip); @@ -1306,6 +1343,7 @@ static void init_vga_switcheroo(struct azx *chip) dev_info(chip->card->dev, "Handle vga_switcheroo audio client\n"); hda->use_vga_switcheroo = 1; + hda->need_eld_notify_link = 1; /* cleared in gpu_bound op */ chip->driver_caps |= AZX_DCAPS_PM_RUNTIME; pci_dev_put(p); } @@ -1314,6 +1352,7 @@ static void init_vga_switcheroo(struct azx *chip) static const struct vga_switcheroo_client_ops azx_vs_ops = { .set_gpu_state = azx_vs_set_state, .can_switch = azx_vs_can_switch, + .gpu_bound = azx_vs_gpu_bound, }; static int register_vga_switcheroo(struct azx *chip) @@ -1339,6 +1378,7 @@ static int register_vga_switcheroo(struct azx *chip) #define init_vga_switcheroo(chip) /* NOP */ #define register_vga_switcheroo(chip) 0 #define check_hdmi_disabled(pci) false +#define setup_vga_switcheroo_runtime_pm(chip) /* NOP */ #endif /* SUPPORT_VGA_SWITCHER */ /* @@ -1352,6 +1392,7 @@ static int azx_free(struct azx *chip) if (azx_has_pm_runtime(chip) && chip->running) pm_runtime_get_noresume(&pci->dev); + chip->running = 0; azx_del_card_list(chip); @@ -2230,6 +2271,25 @@ static struct snd_pci_quirk power_save_blacklist[] = { }; #endif /* CONFIG_PM */ +static void set_default_power_save(struct azx *chip) +{ + int val = power_save; + +#ifdef CONFIG_PM + if (pm_blacklist) { + const struct snd_pci_quirk *q; + + q = snd_pci_quirk_lookup(chip->pci, power_save_blacklist); + if (q && val) { + dev_info(chip->card->dev, "device %04x:%04x is on the power_save blacklist, forcing power_save to 0\n", + q->subvendor, q->subdevice); + val = 0; + } + } +#endif /* CONFIG_PM */ + snd_hda_set_power_save(&chip->bus, val * 1000); +} + /* number of codec slots for each chipset: 0 = default slots (i.e. 4) */ static unsigned int azx_max_codecs[AZX_NUM_DRIVERS] = { [AZX_DRIVER_NVIDIA] = 8, @@ -2241,9 +2301,7 @@ static int azx_probe_continue(struct azx *chip) struct hda_intel *hda = container_of(chip, struct hda_intel, chip); struct hdac_bus *bus = azx_bus(chip); struct pci_dev *pci = chip->pci; - struct hda_codec *codec; int dev = chip->dev_index; - int val; int err; hda->probe_continued = 1; @@ -2322,31 +2380,13 @@ static int azx_probe_continue(struct azx *chip) if (err < 0) goto out_free; + setup_vga_switcheroo_runtime_pm(chip); + chip->running = 1; azx_add_card_list(chip); - val = power_save; -#ifdef CONFIG_PM - if (pm_blacklist) { - const struct snd_pci_quirk *q; - - q = snd_pci_quirk_lookup(chip->pci, power_save_blacklist); - if (q && val) { - dev_info(chip->card->dev, "device %04x:%04x is on the power_save blacklist, forcing power_save to 0\n", - q->subvendor, q->subdevice); - val = 0; - } - } -#endif /* CONFIG_PM */ - /* - * The discrete GPU cannot power down unless the HDA controller runtime - * suspends, so activate runtime PM on codecs even if power_save == 0. - */ - if (use_vga_switcheroo(hda)) - list_for_each_codec(codec, &chip->bus) - codec->auto_runtime_pm = 1; + set_default_power_save(chip); - snd_hda_set_power_save(&chip->bus, val * 1000); if (azx_has_pm_runtime(chip)) pm_runtime_put_autosuspend(&pci->dev); diff --git a/sound/pci/hda/hda_intel.h b/sound/pci/hda/hda_intel.h index e3a3d318d2e5..f59719e06b91 100644 --- a/sound/pci/hda/hda_intel.h +++ b/sound/pci/hda/hda_intel.h @@ -37,6 +37,7 @@ struct hda_intel { /* vga_switcheroo setup */ unsigned int use_vga_switcheroo:1; + unsigned int need_eld_notify_link:1; unsigned int vga_switcheroo_registered:1; unsigned int init_failed:1; /* delayed init failed */ -- cgit v1.2.3-71-gd317 From 8fce3331702316d4bcfeb0771c09ac75d2192bbc Mon Sep 17 00:00:00 2001 From: Jose Abreu Date: Mon, 17 Sep 2018 09:22:56 +0100 Subject: net: stmmac: Rework coalesce timer and fix multi-queue races This follows David Miller advice and tries to fix coalesce timer in multi-queue scenarios. We are now using per-queue coalesce values and per-queue TX timer. Coalesce timer default values was changed to 1ms and the coalesce frames to 25. Tested in B2B setup between XGMAC2 and GMAC5. Signed-off-by: Jose Abreu Fixes: ce736788e8a ("net: stmmac: adding multiple buffers for TX") Cc: Florian Fainelli Cc: Neil Armstrong Cc: Jerome Brunet Cc: Martin Blumenstingl Cc: David S. Miller Cc: Joao Pinto Cc: Giuseppe Cavallaro Cc: Alexandre Torgue Signed-off-by: David S. Miller --- drivers/net/ethernet/stmicro/stmmac/common.h | 4 +- drivers/net/ethernet/stmicro/stmmac/stmmac.h | 14 +- drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 233 ++++++++++++---------- include/linux/stmmac.h | 1 + 4 files changed, 146 insertions(+), 106 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/ethernet/stmicro/stmmac/common.h b/drivers/net/ethernet/stmicro/stmmac/common.h index 1854f270ad66..b1b305f8f414 100644 --- a/drivers/net/ethernet/stmicro/stmmac/common.h +++ b/drivers/net/ethernet/stmicro/stmmac/common.h @@ -258,10 +258,10 @@ struct stmmac_safety_stats { #define MAX_DMA_RIWT 0xff #define MIN_DMA_RIWT 0x20 /* Tx coalesce parameters */ -#define STMMAC_COAL_TX_TIMER 40000 +#define STMMAC_COAL_TX_TIMER 1000 #define STMMAC_MAX_COAL_TX_TICK 100000 #define STMMAC_TX_MAX_FRAMES 256 -#define STMMAC_TX_FRAMES 64 +#define STMMAC_TX_FRAMES 25 /* Packets types */ enum packets_types { diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac.h b/drivers/net/ethernet/stmicro/stmmac/stmmac.h index c0a855b7ab3b..63e1064b27a2 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac.h +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac.h @@ -48,6 +48,8 @@ struct stmmac_tx_info { /* Frequently used values are kept adjacent for cache effect */ struct stmmac_tx_queue { + u32 tx_count_frames; + struct timer_list txtimer; u32 queue_index; struct stmmac_priv *priv_data; struct dma_extended_desc *dma_etx ____cacheline_aligned_in_smp; @@ -73,7 +75,14 @@ struct stmmac_rx_queue { u32 rx_zeroc_thresh; dma_addr_t dma_rx_phy; u32 rx_tail_addr; +}; + +struct stmmac_channel { struct napi_struct napi ____cacheline_aligned_in_smp; + struct stmmac_priv *priv_data; + u32 index; + int has_rx; + int has_tx; }; struct stmmac_tc_entry { @@ -109,14 +118,12 @@ struct stmmac_pps_cfg { struct stmmac_priv { /* Frequently used values are kept adjacent for cache effect */ - u32 tx_count_frames; u32 tx_coal_frames; u32 tx_coal_timer; int tx_coalesce; int hwts_tx_en; bool tx_path_in_lpi_mode; - struct timer_list txtimer; bool tso; unsigned int dma_buf_sz; @@ -137,6 +144,9 @@ struct stmmac_priv { /* TX Queue */ struct stmmac_tx_queue tx_queue[MTL_MAX_TX_QUEUES]; + /* Generic channel for NAPI */ + struct stmmac_channel channel[STMMAC_CH_MAX]; + bool oldlink; int speed; int oldduplex; diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index 9f458bb16f2a..ab9cc0143ff2 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -148,12 +148,14 @@ static void stmmac_verify_args(void) static void stmmac_disable_all_queues(struct stmmac_priv *priv) { u32 rx_queues_cnt = priv->plat->rx_queues_to_use; + u32 tx_queues_cnt = priv->plat->tx_queues_to_use; + u32 maxq = max(rx_queues_cnt, tx_queues_cnt); u32 queue; - for (queue = 0; queue < rx_queues_cnt; queue++) { - struct stmmac_rx_queue *rx_q = &priv->rx_queue[queue]; + for (queue = 0; queue < maxq; queue++) { + struct stmmac_channel *ch = &priv->channel[queue]; - napi_disable(&rx_q->napi); + napi_disable(&ch->napi); } } @@ -164,12 +166,14 @@ static void stmmac_disable_all_queues(struct stmmac_priv *priv) static void stmmac_enable_all_queues(struct stmmac_priv *priv) { u32 rx_queues_cnt = priv->plat->rx_queues_to_use; + u32 tx_queues_cnt = priv->plat->tx_queues_to_use; + u32 maxq = max(rx_queues_cnt, tx_queues_cnt); u32 queue; - for (queue = 0; queue < rx_queues_cnt; queue++) { - struct stmmac_rx_queue *rx_q = &priv->rx_queue[queue]; + for (queue = 0; queue < maxq; queue++) { + struct stmmac_channel *ch = &priv->channel[queue]; - napi_enable(&rx_q->napi); + napi_enable(&ch->napi); } } @@ -1843,18 +1847,18 @@ static void stmmac_dma_operation_mode(struct stmmac_priv *priv) * @queue: TX queue index * Description: it reclaims the transmit resources after transmission completes. */ -static void stmmac_tx_clean(struct stmmac_priv *priv, u32 queue) +static int stmmac_tx_clean(struct stmmac_priv *priv, int budget, u32 queue) { struct stmmac_tx_queue *tx_q = &priv->tx_queue[queue]; unsigned int bytes_compl = 0, pkts_compl = 0; - unsigned int entry; + unsigned int entry, count = 0; - netif_tx_lock(priv->dev); + __netif_tx_lock_bh(netdev_get_tx_queue(priv->dev, queue)); priv->xstats.tx_clean++; entry = tx_q->dirty_tx; - while (entry != tx_q->cur_tx) { + while ((entry != tx_q->cur_tx) && (count < budget)) { struct sk_buff *skb = tx_q->tx_skbuff[entry]; struct dma_desc *p; int status; @@ -1870,6 +1874,8 @@ static void stmmac_tx_clean(struct stmmac_priv *priv, u32 queue) if (unlikely(status & tx_dma_own)) break; + count++; + /* Make sure descriptor fields are read after reading * the own bit. */ @@ -1937,7 +1943,10 @@ static void stmmac_tx_clean(struct stmmac_priv *priv, u32 queue) stmmac_enable_eee_mode(priv); mod_timer(&priv->eee_ctrl_timer, STMMAC_LPI_T(eee_timer)); } - netif_tx_unlock(priv->dev); + + __netif_tx_unlock_bh(netdev_get_tx_queue(priv->dev, queue)); + + return count; } /** @@ -2020,6 +2029,33 @@ static bool stmmac_safety_feat_interrupt(struct stmmac_priv *priv) return false; } +static int stmmac_napi_check(struct stmmac_priv *priv, u32 chan) +{ + int status = stmmac_dma_interrupt_status(priv, priv->ioaddr, + &priv->xstats, chan); + struct stmmac_channel *ch = &priv->channel[chan]; + bool needs_work = false; + + if ((status & handle_rx) && ch->has_rx) { + needs_work = true; + } else { + status &= ~handle_rx; + } + + if ((status & handle_tx) && ch->has_tx) { + needs_work = true; + } else { + status &= ~handle_tx; + } + + if (needs_work && napi_schedule_prep(&ch->napi)) { + stmmac_disable_dma_irq(priv, priv->ioaddr, chan); + __napi_schedule(&ch->napi); + } + + return status; +} + /** * stmmac_dma_interrupt - DMA ISR * @priv: driver private structure @@ -2034,57 +2070,14 @@ static void stmmac_dma_interrupt(struct stmmac_priv *priv) u32 channels_to_check = tx_channel_count > rx_channel_count ? tx_channel_count : rx_channel_count; u32 chan; - bool poll_scheduled = false; int status[max_t(u32, MTL_MAX_TX_QUEUES, MTL_MAX_RX_QUEUES)]; /* Make sure we never check beyond our status buffer. */ if (WARN_ON_ONCE(channels_to_check > ARRAY_SIZE(status))) channels_to_check = ARRAY_SIZE(status); - /* Each DMA channel can be used for rx and tx simultaneously, yet - * napi_struct is embedded in struct stmmac_rx_queue rather than in a - * stmmac_channel struct. - * Because of this, stmmac_poll currently checks (and possibly wakes) - * all tx queues rather than just a single tx queue. - */ for (chan = 0; chan < channels_to_check; chan++) - status[chan] = stmmac_dma_interrupt_status(priv, priv->ioaddr, - &priv->xstats, chan); - - for (chan = 0; chan < rx_channel_count; chan++) { - if (likely(status[chan] & handle_rx)) { - struct stmmac_rx_queue *rx_q = &priv->rx_queue[chan]; - - if (likely(napi_schedule_prep(&rx_q->napi))) { - stmmac_disable_dma_irq(priv, priv->ioaddr, chan); - __napi_schedule(&rx_q->napi); - poll_scheduled = true; - } - } - } - - /* If we scheduled poll, we already know that tx queues will be checked. - * If we didn't schedule poll, see if any DMA channel (used by tx) has a - * completed transmission, if so, call stmmac_poll (once). - */ - if (!poll_scheduled) { - for (chan = 0; chan < tx_channel_count; chan++) { - if (status[chan] & handle_tx) { - /* It doesn't matter what rx queue we choose - * here. We use 0 since it always exists. - */ - struct stmmac_rx_queue *rx_q = - &priv->rx_queue[0]; - - if (likely(napi_schedule_prep(&rx_q->napi))) { - stmmac_disable_dma_irq(priv, - priv->ioaddr, chan); - __napi_schedule(&rx_q->napi); - } - break; - } - } - } + status[chan] = stmmac_napi_check(priv, chan); for (chan = 0; chan < tx_channel_count; chan++) { if (unlikely(status[chan] & tx_hard_error_bump_tc)) { @@ -2233,6 +2226,13 @@ static int stmmac_init_dma_engine(struct stmmac_priv *priv) return ret; } +static void stmmac_tx_timer_arm(struct stmmac_priv *priv, u32 queue) +{ + struct stmmac_tx_queue *tx_q = &priv->tx_queue[queue]; + + mod_timer(&tx_q->txtimer, STMMAC_COAL_TIMER(priv->tx_coal_timer)); +} + /** * stmmac_tx_timer - mitigation sw timer for tx. * @data: data pointer @@ -2241,13 +2241,14 @@ static int stmmac_init_dma_engine(struct stmmac_priv *priv) */ static void stmmac_tx_timer(struct timer_list *t) { - struct stmmac_priv *priv = from_timer(priv, t, txtimer); - u32 tx_queues_count = priv->plat->tx_queues_to_use; - u32 queue; + struct stmmac_tx_queue *tx_q = from_timer(tx_q, t, txtimer); + struct stmmac_priv *priv = tx_q->priv_data; + struct stmmac_channel *ch; + + ch = &priv->channel[tx_q->queue_index]; - /* let's scan all the tx queues */ - for (queue = 0; queue < tx_queues_count; queue++) - stmmac_tx_clean(priv, queue); + if (likely(napi_schedule_prep(&ch->napi))) + __napi_schedule(&ch->napi); } /** @@ -2260,11 +2261,17 @@ static void stmmac_tx_timer(struct timer_list *t) */ static void stmmac_init_tx_coalesce(struct stmmac_priv *priv) { + u32 tx_channel_count = priv->plat->tx_queues_to_use; + u32 chan; + priv->tx_coal_frames = STMMAC_TX_FRAMES; priv->tx_coal_timer = STMMAC_COAL_TX_TIMER; - timer_setup(&priv->txtimer, stmmac_tx_timer, 0); - priv->txtimer.expires = STMMAC_COAL_TIMER(priv->tx_coal_timer); - add_timer(&priv->txtimer); + + for (chan = 0; chan < tx_channel_count; chan++) { + struct stmmac_tx_queue *tx_q = &priv->tx_queue[chan]; + + timer_setup(&tx_q->txtimer, stmmac_tx_timer, 0); + } } static void stmmac_set_rings_length(struct stmmac_priv *priv) @@ -2592,6 +2599,7 @@ static void stmmac_hw_teardown(struct net_device *dev) static int stmmac_open(struct net_device *dev) { struct stmmac_priv *priv = netdev_priv(dev); + u32 chan; int ret; stmmac_check_ether_addr(priv); @@ -2688,7 +2696,9 @@ irq_error: if (dev->phydev) phy_stop(dev->phydev); - del_timer_sync(&priv->txtimer); + for (chan = 0; chan < priv->plat->tx_queues_to_use; chan++) + del_timer_sync(&priv->tx_queue[chan].txtimer); + stmmac_hw_teardown(dev); init_error: free_dma_desc_resources(priv); @@ -2708,6 +2718,7 @@ dma_desc_error: static int stmmac_release(struct net_device *dev) { struct stmmac_priv *priv = netdev_priv(dev); + u32 chan; if (priv->eee_enabled) del_timer_sync(&priv->eee_ctrl_timer); @@ -2722,7 +2733,8 @@ static int stmmac_release(struct net_device *dev) stmmac_disable_all_queues(priv); - del_timer_sync(&priv->txtimer); + for (chan = 0; chan < priv->plat->tx_queues_to_use; chan++) + del_timer_sync(&priv->tx_queue[chan].txtimer); /* Free the IRQ lines */ free_irq(dev->irq, dev); @@ -2936,14 +2948,13 @@ static netdev_tx_t stmmac_tso_xmit(struct sk_buff *skb, struct net_device *dev) priv->xstats.tx_tso_nfrags += nfrags; /* Manage tx mitigation */ - priv->tx_count_frames += nfrags + 1; - if (likely(priv->tx_coal_frames > priv->tx_count_frames)) { - mod_timer(&priv->txtimer, - STMMAC_COAL_TIMER(priv->tx_coal_timer)); - } else { - priv->tx_count_frames = 0; + tx_q->tx_count_frames += nfrags + 1; + if (priv->tx_coal_frames <= tx_q->tx_count_frames) { stmmac_set_tx_ic(priv, desc); priv->xstats.tx_set_ic_bit++; + tx_q->tx_count_frames = 0; + } else { + stmmac_tx_timer_arm(priv, queue); } skb_tx_timestamp(skb); @@ -3146,14 +3157,13 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev) * This approach takes care about the fragments: desc is the first * element in case of no SG. */ - priv->tx_count_frames += nfrags + 1; - if (likely(priv->tx_coal_frames > priv->tx_count_frames)) { - mod_timer(&priv->txtimer, - STMMAC_COAL_TIMER(priv->tx_coal_timer)); - } else { - priv->tx_count_frames = 0; + tx_q->tx_count_frames += nfrags + 1; + if (priv->tx_coal_frames <= tx_q->tx_count_frames) { stmmac_set_tx_ic(priv, desc); priv->xstats.tx_set_ic_bit++; + tx_q->tx_count_frames = 0; + } else { + stmmac_tx_timer_arm(priv, queue); } skb_tx_timestamp(skb); @@ -3199,6 +3209,7 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev) netdev_tx_sent_queue(netdev_get_tx_queue(dev, queue), skb->len); stmmac_enable_dma_transmission(priv, priv->ioaddr); + stmmac_set_tx_tail_ptr(priv, priv->ioaddr, tx_q->tx_tail_addr, queue); return NETDEV_TX_OK; @@ -3319,6 +3330,7 @@ static inline void stmmac_rx_refill(struct stmmac_priv *priv, u32 queue) static int stmmac_rx(struct stmmac_priv *priv, int limit, u32 queue) { struct stmmac_rx_queue *rx_q = &priv->rx_queue[queue]; + struct stmmac_channel *ch = &priv->channel[queue]; unsigned int entry = rx_q->cur_rx; int coe = priv->hw->rx_csum; unsigned int next_entry; @@ -3491,7 +3503,7 @@ static int stmmac_rx(struct stmmac_priv *priv, int limit, u32 queue) else skb->ip_summed = CHECKSUM_UNNECESSARY; - napi_gro_receive(&rx_q->napi, skb); + napi_gro_receive(&ch->napi, skb); priv->dev->stats.rx_packets++; priv->dev->stats.rx_bytes += frame_len; @@ -3514,27 +3526,33 @@ static int stmmac_rx(struct stmmac_priv *priv, int limit, u32 queue) * Description : * To look at the incoming frames and clear the tx resources. */ -static int stmmac_poll(struct napi_struct *napi, int budget) +static int stmmac_napi_poll(struct napi_struct *napi, int budget) { - struct stmmac_rx_queue *rx_q = - container_of(napi, struct stmmac_rx_queue, napi); - struct stmmac_priv *priv = rx_q->priv_data; - u32 tx_count = priv->plat->tx_queues_to_use; - u32 chan = rx_q->queue_index; - int work_done = 0; - u32 queue; + struct stmmac_channel *ch = + container_of(napi, struct stmmac_channel, napi); + struct stmmac_priv *priv = ch->priv_data; + int work_done = 0, work_rem = budget; + u32 chan = ch->index; priv->xstats.napi_poll++; - /* check all the queues */ - for (queue = 0; queue < tx_count; queue++) - stmmac_tx_clean(priv, queue); + if (ch->has_tx) { + int done = stmmac_tx_clean(priv, work_rem, chan); - work_done = stmmac_rx(priv, budget, rx_q->queue_index); - if (work_done < budget) { - napi_complete_done(napi, work_done); - stmmac_enable_dma_irq(priv, priv->ioaddr, chan); + work_done += done; + work_rem -= done; + } + + if (ch->has_rx) { + int done = stmmac_rx(priv, work_rem, chan); + + work_done += done; + work_rem -= done; } + + if (work_done < budget && napi_complete_done(napi, work_done)) + stmmac_enable_dma_irq(priv, priv->ioaddr, chan); + return work_done; } @@ -4198,8 +4216,8 @@ int stmmac_dvr_probe(struct device *device, { struct net_device *ndev = NULL; struct stmmac_priv *priv; + u32 queue, maxq; int ret = 0; - u32 queue; ndev = alloc_etherdev_mqs(sizeof(struct stmmac_priv), MTL_MAX_TX_QUEUES, @@ -4322,11 +4340,22 @@ int stmmac_dvr_probe(struct device *device, "Enable RX Mitigation via HW Watchdog Timer\n"); } - for (queue = 0; queue < priv->plat->rx_queues_to_use; queue++) { - struct stmmac_rx_queue *rx_q = &priv->rx_queue[queue]; + /* Setup channels NAPI */ + maxq = max(priv->plat->rx_queues_to_use, priv->plat->tx_queues_to_use); - netif_napi_add(ndev, &rx_q->napi, stmmac_poll, - (8 * priv->plat->rx_queues_to_use)); + for (queue = 0; queue < maxq; queue++) { + struct stmmac_channel *ch = &priv->channel[queue]; + + ch->priv_data = priv; + ch->index = queue; + + if (queue < priv->plat->rx_queues_to_use) + ch->has_rx = true; + if (queue < priv->plat->tx_queues_to_use) + ch->has_tx = true; + + netif_napi_add(ndev, &ch->napi, stmmac_napi_poll, + NAPI_POLL_WEIGHT); } mutex_init(&priv->lock); @@ -4372,10 +4401,10 @@ error_netdev_register: priv->hw->pcs != STMMAC_PCS_RTBI) stmmac_mdio_unregister(ndev); error_mdio_register: - for (queue = 0; queue < priv->plat->rx_queues_to_use; queue++) { - struct stmmac_rx_queue *rx_q = &priv->rx_queue[queue]; + for (queue = 0; queue < maxq; queue++) { + struct stmmac_channel *ch = &priv->channel[queue]; - netif_napi_del(&rx_q->napi); + netif_napi_del(&ch->napi); } error_hw_init: destroy_workqueue(priv->wq); diff --git a/include/linux/stmmac.h b/include/linux/stmmac.h index c43e9a01b892..7ddfc65586b0 100644 --- a/include/linux/stmmac.h +++ b/include/linux/stmmac.h @@ -30,6 +30,7 @@ #define MTL_MAX_RX_QUEUES 8 #define MTL_MAX_TX_QUEUES 8 +#define STMMAC_CH_MAX 8 #define STMMAC_RX_COE_NONE 0 #define STMMAC_RX_COE_TYPE1 1 -- cgit v1.2.3-71-gd317 From 822f312d47f0200dc0999c9f006fe94aa43bd0bd Mon Sep 17 00:00:00 2001 From: Sebastian Andrzej Siewior Date: Wed, 12 Sep 2018 15:33:45 +0200 Subject: kvm: x86: make kvm_{load|put}_guest_fpu() static The functions kvm_load_guest_fpu() kvm_put_guest_fpu() are only used locally, make them static. This requires also that both functions are moved because they are used before their implementation. Those functions were exported (via EXPORT_SYMBOL) before commit e5bb40251a920 ("KVM: Drop kvm_{load,put}_guest_fpu() exports"). Signed-off-by: Sebastian Andrzej Siewior Signed-off-by: Paolo Bonzini --- arch/x86/kvm/x86.c | 46 +++++++++++++++++++++++----------------------- include/linux/kvm_host.h | 2 -- 2 files changed, 23 insertions(+), 25 deletions(-) (limited to 'include/linux') diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 9d0fda9056de..6f4789398876 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -7835,6 +7835,29 @@ static int complete_emulated_mmio(struct kvm_vcpu *vcpu) return 0; } +/* Swap (qemu) user FPU context for the guest FPU context. */ +static void kvm_load_guest_fpu(struct kvm_vcpu *vcpu) +{ + preempt_disable(); + copy_fpregs_to_fpstate(&vcpu->arch.user_fpu); + /* PKRU is separately restored in kvm_x86_ops->run. */ + __copy_kernel_to_fpregs(&vcpu->arch.guest_fpu.state, + ~XFEATURE_MASK_PKRU); + preempt_enable(); + trace_kvm_fpu(1); +} + +/* When vcpu_run ends, restore user space FPU context. */ +static void kvm_put_guest_fpu(struct kvm_vcpu *vcpu) +{ + preempt_disable(); + copy_fpregs_to_fpstate(&vcpu->arch.guest_fpu); + copy_kernel_to_fpregs(&vcpu->arch.user_fpu.state); + preempt_enable(); + ++vcpu->stat.fpu_reload; + trace_kvm_fpu(0); +} + int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) { int r; @@ -8412,29 +8435,6 @@ static void fx_init(struct kvm_vcpu *vcpu) vcpu->arch.cr0 |= X86_CR0_ET; } -/* Swap (qemu) user FPU context for the guest FPU context. */ -void kvm_load_guest_fpu(struct kvm_vcpu *vcpu) -{ - preempt_disable(); - copy_fpregs_to_fpstate(&vcpu->arch.user_fpu); - /* PKRU is separately restored in kvm_x86_ops->run. */ - __copy_kernel_to_fpregs(&vcpu->arch.guest_fpu.state, - ~XFEATURE_MASK_PKRU); - preempt_enable(); - trace_kvm_fpu(1); -} - -/* When vcpu_run ends, restore user space FPU context. */ -void kvm_put_guest_fpu(struct kvm_vcpu *vcpu) -{ - preempt_disable(); - copy_fpregs_to_fpstate(&vcpu->arch.guest_fpu); - copy_kernel_to_fpregs(&vcpu->arch.user_fpu.state); - preempt_enable(); - ++vcpu->stat.fpu_reload; - trace_kvm_fpu(0); -} - void kvm_arch_vcpu_free(struct kvm_vcpu *vcpu) { void *wbinvd_dirty_mask = vcpu->arch.wbinvd_dirty_mask; diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h index 0205aee44ded..c926698040e0 100644 --- a/include/linux/kvm_host.h +++ b/include/linux/kvm_host.h @@ -733,8 +733,6 @@ bool kvm_vcpu_wake_up(struct kvm_vcpu *vcpu); void kvm_vcpu_kick(struct kvm_vcpu *vcpu); int kvm_vcpu_yield_to(struct kvm_vcpu *target); void kvm_vcpu_on_spin(struct kvm_vcpu *vcpu, bool usermode_vcpu_not_eligible); -void kvm_load_guest_fpu(struct kvm_vcpu *vcpu); -void kvm_put_guest_fpu(struct kvm_vcpu *vcpu); void kvm_flush_remote_tlbs(struct kvm *kvm); void kvm_reload_remote_mmus(struct kvm *kvm); -- cgit v1.2.3-71-gd317 From d124b44f09cab67fc6da4a4513417e3e54b01efc Mon Sep 17 00:00:00 2001 From: Miguel Ojeda Date: Tue, 18 Sep 2018 18:55:41 +0200 Subject: Compiler Attributes: naked was fixed in gcc 4.6 Commit 9c695203a7dd ("compiler-gcc.h: gcc-4.5 needs noclone and noinline on __naked functions") added noinline and noclone as a workaround for a gcc 4.5 bug, which was resolved in 4.6.0. Since now the minimum gcc supported version is 4.6, we can clean it up. See https://gcc.gnu.org/bugzilla/show_bug.cgi?id=44290 and https://godbolt.org/z/h6NMIL Fixes: 815f0ddb346c ("include/linux/compiler*.h: make compiler-*.h mutually exclusive") Cc: Rasmus Villemoes Cc: Eli Friedman Cc: Christopher Li Cc: Kees Cook Cc: Ingo Molnar Cc: Geert Uytterhoeven Cc: Arnd Bergmann Cc: Greg Kroah-Hartman Cc: Masahiro Yamada Cc: Joe Perches Cc: Dominique Martinet Cc: Linus Torvalds Cc: linux-sparse@vger.kernel.org Tested-by: Stefan Agner Reviewed-by: Stefan Agner Reviewed-by: Luc Van Oostenryck Reviewed-by: Nick Desaulniers Signed-off-by: Miguel Ojeda Signed-off-by: Greg Kroah-Hartman --- include/linux/compiler-gcc.h | 8 +------- 1 file changed, 1 insertion(+), 7 deletions(-) (limited to 'include/linux') diff --git a/include/linux/compiler-gcc.h b/include/linux/compiler-gcc.h index 763bbad1e258..25d3dd6b2702 100644 --- a/include/linux/compiler-gcc.h +++ b/include/linux/compiler-gcc.h @@ -84,14 +84,8 @@ * to trace naked functions because then mcount is called without * stack and frame pointer being set up and there is no chance to * restore the lr register to the value before mcount was called. - * - * The asm() bodies of naked functions often depend on standard calling - * conventions, therefore they must be noinline and noclone. - * - * GCC 4.[56] currently fail to enforce this, so we must do so ourselves. - * See GCC PR44290. */ -#define __naked __attribute__((naked)) noinline __noclone notrace +#define __naked __attribute__((naked)) notrace #define __UNIQUE_ID(prefix) __PASTE(__PASTE(__UNIQUE_ID_, prefix), __COUNTER__) -- cgit v1.2.3-71-gd317 From ae596de1a0c8c2c924dc99d23c026259372ab234 Mon Sep 17 00:00:00 2001 From: Miguel Ojeda Date: Tue, 18 Sep 2018 18:55:42 +0200 Subject: Compiler Attributes: naked can be shared The naked attribute is supported by at least gcc >= 4.6 (for ARM, which is the only current user), gcc >= 8 (for x86), clang >= 3.1 and icc >= 13. See https://godbolt.org/z/350Dyc Therefore, move it out of compiler-gcc.h so that the definition is shared by all compilers. This also fixes Clang support for ARM32 --- 815f0ddb346c ("include/linux/compiler*.h: make compiler-*.h mutually exclusive"). Fixes: 815f0ddb346c ("include/linux/compiler*.h: make compiler-*.h mutually exclusive") Cc: Rasmus Villemoes Cc: Eli Friedman Cc: Christopher Li Cc: Kees Cook Cc: Ingo Molnar Cc: Geert Uytterhoeven Cc: Greg Kroah-Hartman Cc: Masahiro Yamada Cc: Joe Perches Cc: Dominique Martinet Cc: Linus Torvalds Cc: linux-sparse@vger.kernel.org Suggested-by: Arnd Bergmann Tested-by: Stefan Agner Reviewed-by: Stefan Agner Reviewed-by: Luc Van Oostenryck Reviewed-by: Nick Desaulniers Signed-off-by: Miguel Ojeda Signed-off-by: Greg Kroah-Hartman --- include/linux/compiler-gcc.h | 8 -------- include/linux/compiler_types.h | 8 ++++++++ 2 files changed, 8 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/include/linux/compiler-gcc.h b/include/linux/compiler-gcc.h index 25d3dd6b2702..4d36b27214fd 100644 --- a/include/linux/compiler-gcc.h +++ b/include/linux/compiler-gcc.h @@ -79,14 +79,6 @@ #define __noretpoline __attribute__((indirect_branch("keep"))) #endif -/* - * it doesn't make sense on ARM (currently the only user of __naked) - * to trace naked functions because then mcount is called without - * stack and frame pointer being set up and there is no chance to - * restore the lr register to the value before mcount was called. - */ -#define __naked __attribute__((naked)) notrace - #define __UNIQUE_ID(prefix) __PASTE(__PASTE(__UNIQUE_ID_, prefix), __COUNTER__) #define __optimize(level) __attribute__((__optimize__(level))) diff --git a/include/linux/compiler_types.h b/include/linux/compiler_types.h index 3525c179698c..db192becfec4 100644 --- a/include/linux/compiler_types.h +++ b/include/linux/compiler_types.h @@ -226,6 +226,14 @@ struct ftrace_likely_data { #define notrace __attribute__((no_instrument_function)) #endif +/* + * it doesn't make sense on ARM (currently the only user of __naked) + * to trace naked functions because then mcount is called without + * stack and frame pointer being set up and there is no chance to + * restore the lr register to the value before mcount was called. + */ +#define __naked __attribute__((naked)) notrace + #define __compiler_offsetof(a, b) __builtin_offsetof(a, b) /* -- cgit v1.2.3-71-gd317 From b57e99b4b8b0ebdf9707424e7ddc0c392bdc5fe6 Mon Sep 17 00:00:00 2001 From: Omar Sandoval Date: Fri, 21 Sep 2018 16:44:34 -0700 Subject: block: use nanosecond resolution for iostat Klaus Kusche reported that the I/O busy time in /proc/diskstats was not updating properly on 4.18. This is because we started using ktime to track elapsed time, and we convert nanoseconds to jiffies when we update the partition counter. However, this gets rounded down, so any I/Os that take less than a jiffy are not accounted for. Previously in this case, the value of jiffies would sometimes increment while we were doing I/O, so at least some I/Os were accounted for. Let's convert the stats to use nanoseconds internally. We still report milliseconds as before, now more accurately than ever. The value is still truncated to 32 bits for backwards compatibility. Fixes: 522a777566f5 ("block: consolidate struct request timestamp fields") Cc: stable@vger.kernel.org Reported-by: Klaus Kusche Signed-off-by: Omar Sandoval Signed-off-by: Jens Axboe --- block/bio.c | 2 +- block/blk-core.c | 4 +--- block/genhd.c | 6 +++--- block/partition-generic.c | 6 +++--- include/linux/genhd.h | 5 ++++- 5 files changed, 12 insertions(+), 11 deletions(-) (limited to 'include/linux') diff --git a/block/bio.c b/block/bio.c index 8c680a776171..0093bed81c0e 100644 --- a/block/bio.c +++ b/block/bio.c @@ -1684,7 +1684,7 @@ void generic_end_io_acct(struct request_queue *q, int req_op, const int sgrp = op_stat_group(req_op); int cpu = part_stat_lock(); - part_stat_add(cpu, part, ticks[sgrp], duration); + part_stat_add(cpu, part, nsecs[sgrp], jiffies_to_nsecs(duration)); part_round_stats(q, cpu, part); part_dec_in_flight(q, part, op_is_write(req_op)); diff --git a/block/blk-core.c b/block/blk-core.c index 4dbc93f43b38..cff0a60ee200 100644 --- a/block/blk-core.c +++ b/block/blk-core.c @@ -2733,17 +2733,15 @@ void blk_account_io_done(struct request *req, u64 now) * containing request is enough. */ if (blk_do_io_stat(req) && !(req->rq_flags & RQF_FLUSH_SEQ)) { - unsigned long duration; const int sgrp = op_stat_group(req_op(req)); struct hd_struct *part; int cpu; - duration = nsecs_to_jiffies(now - req->start_time_ns); cpu = part_stat_lock(); part = req->part; part_stat_inc(cpu, part, ios[sgrp]); - part_stat_add(cpu, part, ticks[sgrp], duration); + part_stat_add(cpu, part, nsecs[sgrp], now - req->start_time_ns); part_round_stats(req->q, cpu, part); part_dec_in_flight(req->q, part, rq_data_dir(req)); diff --git a/block/genhd.c b/block/genhd.c index 8cc719a37b32..be5bab20b2ab 100644 --- a/block/genhd.c +++ b/block/genhd.c @@ -1343,18 +1343,18 @@ static int diskstats_show(struct seq_file *seqf, void *v) part_stat_read(hd, ios[STAT_READ]), part_stat_read(hd, merges[STAT_READ]), part_stat_read(hd, sectors[STAT_READ]), - jiffies_to_msecs(part_stat_read(hd, ticks[STAT_READ])), + (unsigned int)part_stat_read_msecs(hd, STAT_READ), part_stat_read(hd, ios[STAT_WRITE]), part_stat_read(hd, merges[STAT_WRITE]), part_stat_read(hd, sectors[STAT_WRITE]), - jiffies_to_msecs(part_stat_read(hd, ticks[STAT_WRITE])), + (unsigned int)part_stat_read_msecs(hd, STAT_WRITE), inflight[0], jiffies_to_msecs(part_stat_read(hd, io_ticks)), jiffies_to_msecs(part_stat_read(hd, time_in_queue)), part_stat_read(hd, ios[STAT_DISCARD]), part_stat_read(hd, merges[STAT_DISCARD]), part_stat_read(hd, sectors[STAT_DISCARD]), - jiffies_to_msecs(part_stat_read(hd, ticks[STAT_DISCARD])) + (unsigned int)part_stat_read_msecs(hd, STAT_DISCARD) ); } disk_part_iter_exit(&piter); diff --git a/block/partition-generic.c b/block/partition-generic.c index 5a8975a1201c..d3d14e81fb12 100644 --- a/block/partition-generic.c +++ b/block/partition-generic.c @@ -136,18 +136,18 @@ ssize_t part_stat_show(struct device *dev, part_stat_read(p, ios[STAT_READ]), part_stat_read(p, merges[STAT_READ]), (unsigned long long)part_stat_read(p, sectors[STAT_READ]), - jiffies_to_msecs(part_stat_read(p, ticks[STAT_READ])), + (unsigned int)part_stat_read_msecs(p, STAT_READ), part_stat_read(p, ios[STAT_WRITE]), part_stat_read(p, merges[STAT_WRITE]), (unsigned long long)part_stat_read(p, sectors[STAT_WRITE]), - jiffies_to_msecs(part_stat_read(p, ticks[STAT_WRITE])), + (unsigned int)part_stat_read_msecs(p, STAT_WRITE), inflight[0], jiffies_to_msecs(part_stat_read(p, io_ticks)), jiffies_to_msecs(part_stat_read(p, time_in_queue)), part_stat_read(p, ios[STAT_DISCARD]), part_stat_read(p, merges[STAT_DISCARD]), (unsigned long long)part_stat_read(p, sectors[STAT_DISCARD]), - jiffies_to_msecs(part_stat_read(p, ticks[STAT_DISCARD]))); + (unsigned int)part_stat_read_msecs(p, STAT_DISCARD)); } ssize_t part_inflight_show(struct device *dev, struct device_attribute *attr, diff --git a/include/linux/genhd.h b/include/linux/genhd.h index 57864422a2c8..25c08c6c7f99 100644 --- a/include/linux/genhd.h +++ b/include/linux/genhd.h @@ -83,10 +83,10 @@ struct partition { } __attribute__((packed)); struct disk_stats { + u64 nsecs[NR_STAT_GROUPS]; unsigned long sectors[NR_STAT_GROUPS]; unsigned long ios[NR_STAT_GROUPS]; unsigned long merges[NR_STAT_GROUPS]; - unsigned long ticks[NR_STAT_GROUPS]; unsigned long io_ticks; unsigned long time_in_queue; }; @@ -354,6 +354,9 @@ static inline void free_part_stats(struct hd_struct *part) #endif /* CONFIG_SMP */ +#define part_stat_read_msecs(part, which) \ + div_u64(part_stat_read(part, nsecs[which]), NSEC_PER_MSEC) + #define part_stat_read_accum(part, field) \ (part_stat_read(part, field[STAT_READ]) + \ part_stat_read(part, field[STAT_WRITE]) + \ -- cgit v1.2.3-71-gd317 From ac3d9dd034e565df2c034ab2ca71f0a9f69153c1 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 21 Sep 2018 15:27:38 -0700 Subject: netpoll: make ndo_poll_controller() optional As diagnosed by Song Liu, ndo_poll_controller() can be very dangerous on loaded hosts, since the cpu calling ndo_poll_controller() might steal all NAPI contexts (for all RX/TX queues of the NIC). This capture can last for unlimited amount of time, since one cpu is generally not able to drain all the queues under load. It seems that all networking drivers that do use NAPI for their TX completions, should not provide a ndo_poll_controller(). NAPI drivers have netpoll support already handled in core networking stack, since netpoll_poll_dev() uses poll_napi(dev) to iterate through registered NAPI contexts for a device. This patch allows netpoll_poll_dev() to process NAPI contexts even for drivers not providing ndo_poll_controller(), allowing for following patches in NAPI drivers. Also we export netpoll_poll_dev() so that it can be called by bonding/team drivers in following patches. Reported-by: Song Liu Signed-off-by: Eric Dumazet Tested-by: Song Liu Signed-off-by: David S. Miller --- include/linux/netpoll.h | 5 +++-- net/core/netpoll.c | 19 +++++++------------ 2 files changed, 10 insertions(+), 14 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netpoll.h b/include/linux/netpoll.h index 67662d01130a..3ef82d3a78db 100644 --- a/include/linux/netpoll.h +++ b/include/linux/netpoll.h @@ -49,8 +49,9 @@ struct netpoll_info { }; #ifdef CONFIG_NETPOLL -extern void netpoll_poll_disable(struct net_device *dev); -extern void netpoll_poll_enable(struct net_device *dev); +void netpoll_poll_dev(struct net_device *dev); +void netpoll_poll_disable(struct net_device *dev); +void netpoll_poll_enable(struct net_device *dev); #else static inline void netpoll_poll_disable(struct net_device *dev) { return; } static inline void netpoll_poll_enable(struct net_device *dev) { return; } diff --git a/net/core/netpoll.c b/net/core/netpoll.c index 57557a6a950c..3219a2932463 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -187,16 +187,16 @@ static void poll_napi(struct net_device *dev) } } -static void netpoll_poll_dev(struct net_device *dev) +void netpoll_poll_dev(struct net_device *dev) { - const struct net_device_ops *ops; struct netpoll_info *ni = rcu_dereference_bh(dev->npinfo); + const struct net_device_ops *ops; /* Don't do any rx activity if the dev_lock mutex is held * the dev_open/close paths use this to block netpoll activity * while changing device state */ - if (down_trylock(&ni->dev_lock)) + if (!ni || down_trylock(&ni->dev_lock)) return; if (!netif_running(dev)) { @@ -205,13 +205,8 @@ static void netpoll_poll_dev(struct net_device *dev) } ops = dev->netdev_ops; - if (!ops->ndo_poll_controller) { - up(&ni->dev_lock); - return; - } - - /* Process pending work on NIC */ - ops->ndo_poll_controller(dev); + if (ops->ndo_poll_controller) + ops->ndo_poll_controller(dev); poll_napi(dev); @@ -219,6 +214,7 @@ static void netpoll_poll_dev(struct net_device *dev) zap_completion_queue(); } +EXPORT_SYMBOL(netpoll_poll_dev); void netpoll_poll_disable(struct net_device *dev) { @@ -613,8 +609,7 @@ int __netpoll_setup(struct netpoll *np, struct net_device *ndev) strlcpy(np->dev_name, ndev->name, IFNAMSIZ); INIT_WORK(&np->cleanup_work, netpoll_async_cleanup); - if ((ndev->priv_flags & IFF_DISABLE_NETPOLL) || - !ndev->netdev_ops->ndo_poll_controller) { + if (ndev->priv_flags & IFF_DISABLE_NETPOLL) { np_err(np, "%s doesn't support polling, aborting\n", np->dev_name); err = -ENOTSUPP; -- cgit v1.2.3-71-gd317