February 2020 - Linux-stable-mirror

stable-rc 5.5.5-rc1: [drm:ade_irq_handler [kirin_drm]] *ERROR* LDI underflow!

by Naresh Kamboju

The arm64 device running LTP hugetlb test suite caused test hang on stable-rc 5.5.5-rc1 due to [drm:ade_irq_handler [kirin_drm]] *ERROR* LDI underflow!. Same problem noticed while running libhugetlbfs test suite. Problematic patch not identified yet. hugemmap05.c:223: INFO: original nr_hugepages is 0 hugemmap05.c:236: INFO: original nr_overcommit_hugepages is 0 hugemmap05.c:104: INFO: check /proc/meminfo before allocation. hugemmap05.c:285: INFO: HugePages_Total is 192. hugemmap05.c:285: INFO: HugePages_Free is 192. hugemmap05.c:285: INFO: HugePages_Surp is 64. hugemmap05.c:285: INFO: HugePages_[ 51.411646] [drm:ade_irq_handler [kirin_drm]] *ERROR* LDI underflow! Rsvd is 192. [ 51.411735] [drm:ade_irq_handler [kirin_drm]] *ERROR* LDI underflow! [ 51.419705] [drm:ade_irq_handler [kirin_drm]] *ERROR* LDI underflow! [ 51.427680] [drm:ade_irq_handler [kirin_drm]] *ERROR* LDI underflow! [ 51.434455] [drm:ade_irq_handler [kirin_drm]] *ERROR* LDI underflow! [ 51.441179] [drm:ade_irq_handler [kirin_drm]] *ERROR* LDI underflow! [ 51.447912] [drm:ade_irq_handler [kirin_drm]] *ERROR* LDI underflow! [ 51.454639] [drm:ade_irq_handler [kirin_drm]] *ERROR* LDI underflow! [ 51.461395] [drm:ade_irq_handler [kirin_drm]] *ERROR* LDI underflow! [ 51.468175] [drm:ade_irq_handler [kirin_drm]] *ERROR* LDI underflow! [ 51.475014] [drm:ade_irq_handler [kirin_drm]] *ERROR* LDI underflow! [ 51.481868] [drm:ade_irq_handler [kirin_drm]] *ERROR* LDI underflow! [ 51.488621] [drm:ade_irq_handler [kirin_drm]] *ERROR* LDI underflow! [ 51.495367] [drm:ade_irq_handler [kirin_drm]] *ERROR* LDI underflow! [ 51.502113] [drm:ade_irq_handler [kirin_drm]] *ERROR* LDI underflow! [ 51.508845] [drm:ade_irq_handler [kirin_drm]] *ERROR* LDI underflow! [ 51.515582] [drm:ade_irq_handler [kirin_drm]] *ERROR* LDI underflow! [ 51.522316] [drm:ade_irq_handler [kirin_drm]] *ERROR* LDI underflow! [ 51.529054] [drm:ade_irq_handler [kirin_drm]] *ERROR* LDI underflow! [ 51.535836] [drm:ade_irq_handler [kirin_drm]] *ERROR* LDI underflow! [ 51.542621] [drm:ade_irq_handler [kirin_drm]] *ERROR* LDI underflow! [ 51.549399] [drm:ade_irq_handler [kirin_drm]] *ERROR* LDI underflow! [ 51.556144] [drm:ade_irq_handler [kirin_drm]] *ERROR* LDI underflow! [ 51.562897] [drm:ade_irq_handler [kirin_drm]] *ERROR* LDI underflow! [ 51.569634] [drm:ade_irq_handler [kirin_drm]] *ERROR* LDI underflow! [ 51.576378] [drm:ade_irq_handler [kirin_drm]] *ERROR* LDI underflow! [ 51.583142] [drm:ade_irq_handler [kirin_drm]] *ERROR* LDI underflow! [ 51.589889] [drm:ade_irq_handler [kirin_drm]] *ERROR* LDI underflow! [ 51.596610] [drm:ade_irq_handler [kirin_drm]] *ERROR* LDI underflow! hugemmap05.c:260:[ 51.603356] [drm:ade_irq_handler [kirin_drm]] *ERROR* LDI underflow! INFO: First hex is 7070707 [ 51.610101] [drm:ade_irq_handler [kirin_drm]] *ERROR* LDI underflow! [ 51.621956] [drm:ade_ldi_set_mode [kirin_drm]] *ERROR* failed to set pixel clk 0Hz (-22) hugemmap05.c:139: INFO: check /proc/meminfo. hugemmap05.c:285: INFO: HugePages_Total is 192. hugemmap05.c:285: INFO: HugePages_Free is 0. hugemmap05.c:285: INFO: HugePages_Surp is 64. hugemmap05.c:285: INFO: HugePages_Rsvd is 0. hugemmap05.c:163: PASS: hugepages overcommit test pass hugemmap05.c:180: INFO: restore nr_hugepages to 0. hugemmap05.c:189: INFO: restore nr_overcommit_hugepages to 0. Summary: passed 1 failed 0 skipped 0 warnings 0 tst_test.c:1217: INFO: Timeout per run is 0h 15m 00s mem.c:817: INFO: set nr_hugepages to 255 [ 54.139687] [drm:ade_irq_handler [kirin_drm]] *ERROR* LDI underflow! [ 54.143384] [drm:ade_ldi_set_mode [kirin_drm]] *ERROR* failed to set pixel clk 0Hz (-22) [ 54.278589] [drm:ade_irq_handler [kirin_drm]] *ERROR* LDI underflow! [ 54.280642] [drm:ade_ldi_set_mode [kirin_drm]] *ERROR* failed to set pixel clk 0Hz (-22) hugemmap06.c:139: PASS: No regression found. [ 54.520165] [drm:ade_irq_handler [kirin_drm]] *ERROR* LDI underflow! [ 54.522769] [drm:ade_ldi_set_mode [kirin_drm]] *ERROR* failed to set pixel clk 0Hz (-22) [ 54.664774] [drm:ade_irq_handler [kirin_drm]] *ERROR* LDI underflow! [ 54.668013] [drm:ade_ldi_set_mode [kirin_drm]] *ERROR* failed to set pixel clk 0Hz (-22) <SYSTEM HANGS HERE> Ref: https://lkft.validation.linaro.org/scheduler/job/1227316#L4288 metadata: git branch: linux-5.5.y git repo: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git git commit: 7d6c8f2632c92635fcef4175921a7742f23947e4 git describe: v5.5.4-46-g7d6c8f2632c9 make_kernelversion: 5.5.5-rc1 kernel-config: http://snapshots.linaro.org/openembedded/lkft/lkft/sumo/hikey/lkft/linux-st… build-location: http://snapshots.linaro.org/openembedded/lkft/lkft/sumo/hikey/lkft/linux-st… -- Linaro LKFT https://lkft.linaro.org

4 years, 10 months

1
0
0 0

stable/linux-4.9.y boot: 43 boots: 1 failed, 41 passed with 1 untried/unknown (v4.9.214)

by kernelci.org bot

stable/linux-4.9.y boot: 43 boots: 1 failed, 41 passed with 1 untried/unknown (v4.9.214) Full Boot Summary: https://kernelci.org/boot/all/job/stable/branch/linux-4.9.y/kernel/v4.9.214/ Full Build Summary: https://kernelci.org/build/stable/branch/linux-4.9.y/kernel/v4.9.214/ Tree: stable Branch: linux-4.9.y Git Describe: v4.9.214 Git Commit: 7ce439266f602f60f05dccf964a8685e53684a9a Git URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable.git Tested: 29 unique boards, 15 SoC families, 7 builds out of 136 Boot Regressions Detected: arm: multi_v7_defconfig: gcc-8: omap3-beagle-xm: lab-baylibre: new failure (last pass: v4.9.211) versatile_defconfig: gcc-8: versatile-pb: lab-collabora: new failure (last pass: v4.9.212) Boot Failure Detected: arm: multi_v7_defconfig: gcc-8: omap3-beagle-xm: 1 failed lab --- For more info write to <info(a)kernelci.org>

4 years, 10 months

1
0
0 0

[PATCH 4/8] net: ll_temac: Handle DMA halt condition caused by buffer underrun

by Esben Haabendal

The SDMA engine used by TEMAC halts operation when it has finished processing of the last buffer descriptor in the buffer ring. Unfortunately, no interrupt event is generated when this happens, so we need to setup another mechanism to make sure DMA operation is restarted when enough buffers have been added to the ring. Signed-off-by: Esben Haabendal <esben(a)geanix.com> Cc: stable(a)vger.kernel.org --- drivers/net/ethernet/xilinx/ll_temac.h | 3 ++ drivers/net/ethernet/xilinx/ll_temac_main.c | 47 ++++++++++++++++++++- 2 files changed, 48 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/xilinx/ll_temac.h b/drivers/net/ethernet/xilinx/ll_temac.h index 99fe059e5c7f..53fb8141f1a6 100644 --- a/drivers/net/ethernet/xilinx/ll_temac.h +++ b/drivers/net/ethernet/xilinx/ll_temac.h @@ -380,6 +380,9 @@ struct temac_local { /* DMA channel control setup */ u32 tx_chnl_ctrl; u32 rx_chnl_ctrl; + u8 coalesce_count_rx; + + struct delayed_work restart_work; }; /* Wrappers for temac_ior()/temac_iow() function pointers above */ diff --git a/drivers/net/ethernet/xilinx/ll_temac_main.c b/drivers/net/ethernet/xilinx/ll_temac_main.c index 255207f2fd27..90b486becb5b 100644 --- a/drivers/net/ethernet/xilinx/ll_temac_main.c +++ b/drivers/net/ethernet/xilinx/ll_temac_main.c @@ -51,6 +51,7 @@ #include <linux/ip.h> #include <linux/slab.h> #include <linux/interrupt.h> +#include <linux/workqueue.h> #include <linux/dma-mapping.h> #include <linux/processor.h> #include <linux/platform_data/xilinx-ll-temac.h> @@ -920,6 +921,17 @@ temac_start_xmit(struct sk_buff *skb, struct net_device *ndev) return NETDEV_TX_OK; } +static int ll_temac_recv_buffers_available(struct temac_local *lp) +{ + int available; + + if (!lp->rx_skb[lp->rx_bd_ci]) + return 0; + available = 1 + lp->rx_bd_tail - lp->rx_bd_ci; + if (available <= 0) + available += RX_BD_NUM; + return available; +} static void ll_temac_recv(struct net_device *ndev) { @@ -990,6 +1002,18 @@ static void ll_temac_recv(struct net_device *ndev) lp->rx_bd_ci = 0; } while (rx_bd != lp->rx_bd_tail); + /* DMA operations will halt when the last buffer descriptor is + * processed (ie. the one pointed to by RX_TAILDESC_PTR). + * When that happens, no more interrupt events will be + * generated. No IRQ_COAL or IRQ_DLY, and not even an + * IRQ_ERR. To avoid stalling, we schedule a delayed work + * when there is a potential risk of that happening. The work + * will call this function, and thus re-schedule itself until + * enough buffers are available again. + */ + if (ll_temac_recv_buffers_available(lp) < lp->coalesce_count_rx) + schedule_delayed_work(&lp->restart_work, HZ / 1000); + /* Allocate new buffers for those buffer descriptors that were * passed to network stack. Note that GFP_ATOMIC allocations * can fail (e.g. when a larger burst of GFP_ATOMIC @@ -1045,6 +1069,18 @@ static void ll_temac_recv(struct net_device *ndev) spin_unlock_irqrestore(&lp->rx_lock, flags); } +/* Function scheduled to ensure a restart in case of DMA halt + * condition caused by running out of buffer descriptors. + */ +static void ll_temac_restart_work_func(struct work_struct *work) +{ + struct temac_local *lp = container_of(work, struct temac_local, + restart_work.work); + struct net_device *ndev = lp->ndev; + + ll_temac_recv(ndev); +} + static irqreturn_t ll_temac_tx_irq(int irq, void *_ndev) { struct net_device *ndev = _ndev; @@ -1137,6 +1173,8 @@ static int temac_stop(struct net_device *ndev) dev_dbg(&ndev->dev, "temac_close()\n"); + cancel_delayed_work_sync(&lp->restart_work); + free_irq(lp->tx_irq, ndev); free_irq(lp->rx_irq, ndev); @@ -1258,6 +1296,7 @@ static int temac_probe(struct platform_device *pdev) lp->dev = &pdev->dev; lp->options = XTE_OPTION_DEFAULTS; spin_lock_init(&lp->rx_lock); + INIT_DELAYED_WORK(&lp->restart_work, ll_temac_restart_work_func); /* Setup mutex for synchronization of indirect register access */ if (pdata) { @@ -1364,6 +1403,7 @@ static int temac_probe(struct platform_device *pdev) */ lp->tx_chnl_ctrl = 0x10220000; lp->rx_chnl_ctrl = 0xff070000; + lp->coalesce_count_rx = 0x07; /* Finished with the DMA node; drop the reference */ of_node_put(dma_np); @@ -1395,11 +1435,14 @@ static int temac_probe(struct platform_device *pdev) (pdata->tx_irq_count << 16); else lp->tx_chnl_ctrl = 0x10220000; - if (pdata->rx_irq_timeout || pdata->rx_irq_count) + if (pdata->rx_irq_timeout || pdata->rx_irq_count) { lp->rx_chnl_ctrl = (pdata->rx_irq_timeout << 24) | (pdata->rx_irq_count << 16); - else + lp->coalesce_count_rx = pdata->rx_irq_count; + } else { lp->rx_chnl_ctrl = 0xff070000; + lp->coalesce_count_rx = 0x07; + } } /* Error handle returned DMA RX and TX interrupts */ -- 2.25.0

4 years, 10 months

1
0
0 0

[PATCH 3/8] net: ll_temac: Fix RX buffer descriptor handling on GFP_ATOMIC pressure

by Esben Haabendal

Failures caused by GFP_ATOMIC memory pressure have been observed, and due to the missing error handling, results in kernel crash such as [1876998.350133] kernel BUG at mm/slub.c:3952! [1876998.350141] invalid opcode: 0000 [#1] PREEMPT SMP PTI [1876998.350147] CPU: 2 PID: 0 Comm: swapper/2 Not tainted 5.3.0-scnxt #1 [1876998.350150] Hardware name: N/A N/A/COMe-bIP2, BIOS CCR2R920 03/01/2017 [1876998.350160] RIP: 0010:kfree+0x1ca/0x220 [1876998.350164] Code: 85 db 74 49 48 8b 95 68 01 00 00 48 31 c2 48 89 10 e9 d7 fe ff ff 49 8b 04 24 a9 00 00 01 00 75 0b 49 8b 44 24 08 a8 01 75 02 <0f> 0b 49 8b 04 24 31 f6 a9 00 00 01 00 74 06 41 0f b6 74 24 5b [1876998.350172] RSP: 0018:ffffc900000f0df0 EFLAGS: 00010246 [1876998.350177] RAX: ffffea00027f0708 RBX: ffff888008d78000 RCX: 0000000000391372 [1876998.350181] RDX: 0000000000000000 RSI: ffffe8ffffd01400 RDI: ffff888008d78000 [1876998.350185] RBP: ffff8881185a5d00 R08: ffffc90000087dd8 R09: 000000000000280a [1876998.350189] R10: 0000000000000002 R11: 0000000000000000 R12: ffffea0000235e00 [1876998.350193] R13: ffff8881185438a0 R14: 0000000000000000 R15: ffff888118543870 [1876998.350198] FS: 0000000000000000(0000) GS:ffff88811f300000(0000) knlGS:0000000000000000 [1876998.350203] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 s#1 Part1 [1876998.350206] CR2: 00007f8dac7b09f0 CR3: 000000011e20a006 CR4: 00000000001606e0 [1876998.350210] Call Trace: [1876998.350215] <IRQ> [1876998.350224] ? __netif_receive_skb_core+0x70a/0x920 [1876998.350229] kfree_skb+0x32/0xb0 [1876998.350234] __netif_receive_skb_core+0x70a/0x920 [1876998.350240] __netif_receive_skb_one_core+0x36/0x80 [1876998.350245] process_backlog+0x8b/0x150 [1876998.350250] net_rx_action+0xf7/0x340 [1876998.350255] __do_softirq+0x10f/0x353 [1876998.350262] irq_exit+0xb2/0xc0 [1876998.350265] do_IRQ+0x77/0xd0 [1876998.350271] common_interrupt+0xf/0xf [1876998.350274] </IRQ> In order to handle such failures more graceful, this change splits the receive loop into one for consuming the received buffers, and one for allocating new buffers. When GFP_ATOMIC allocations fail, the receive will continue with the buffers that is still there, and with the expectation that the allocations will succeed in a later call to receive. Signed-off-by: Esben Haabendal <esben(a)geanix.com> Cc: stable(a)vger.kernel.org --- drivers/net/ethernet/xilinx/ll_temac.h | 1 + drivers/net/ethernet/xilinx/ll_temac_main.c | 112 ++++++++++++++------ 2 files changed, 82 insertions(+), 31 deletions(-) diff --git a/drivers/net/ethernet/xilinx/ll_temac.h b/drivers/net/ethernet/xilinx/ll_temac.h index 276292bca334..99fe059e5c7f 100644 --- a/drivers/net/ethernet/xilinx/ll_temac.h +++ b/drivers/net/ethernet/xilinx/ll_temac.h @@ -375,6 +375,7 @@ struct temac_local { int tx_bd_next; int tx_bd_tail; int rx_bd_ci; + int rx_bd_tail; /* DMA channel control setup */ u32 tx_chnl_ctrl; diff --git a/drivers/net/ethernet/xilinx/ll_temac_main.c b/drivers/net/ethernet/xilinx/ll_temac_main.c index c368c3914bda..255207f2fd27 100644 --- a/drivers/net/ethernet/xilinx/ll_temac_main.c +++ b/drivers/net/ethernet/xilinx/ll_temac_main.c @@ -389,12 +389,13 @@ static int temac_dma_bd_init(struct net_device *ndev) lp->tx_bd_next = 0; lp->tx_bd_tail = 0; lp->rx_bd_ci = 0; + lp->rx_bd_tail = RX_BD_NUM - 1; /* Enable RX DMA transfers */ wmb(); lp->dma_out(lp, RX_CURDESC_PTR, lp->rx_bd_p); lp->dma_out(lp, RX_TAILDESC_PTR, - lp->rx_bd_p + (sizeof(*lp->rx_bd_v) * (RX_BD_NUM - 1))); + lp->rx_bd_p + (sizeof(*lp->rx_bd_v) * lp->rx_bd_tail)); /* Prepare for TX DMA transfer */ lp->dma_out(lp, TX_CURDESC_PTR, lp->tx_bd_p); @@ -923,27 +924,41 @@ temac_start_xmit(struct sk_buff *skb, struct net_device *ndev) static void ll_temac_recv(struct net_device *ndev) { struct temac_local *lp = netdev_priv(ndev); - struct sk_buff *skb, *new_skb; - unsigned int bdstat; - struct cdmac_bd *cur_p; - dma_addr_t tail_p, skb_dma_addr; - int length; unsigned long flags; + int rx_bd; + bool update_tail = false; spin_lock_irqsave(&lp->rx_lock, flags); - tail_p = lp->rx_bd_p + sizeof(*lp->rx_bd_v) * lp->rx_bd_ci; - cur_p = &lp->rx_bd_v[lp->rx_bd_ci]; - - bdstat = be32_to_cpu(cur_p->app0); - while ((bdstat & STS_CTRL_APP0_CMPLT)) { + /* Process all received buffers, passing them on network + * stack. After this, the buffer descriptors will be in an + * un-allocated stage, where no skb is allocated for it, and + * they are therefore not available for TEMAC/DMA. + */ + do { + struct cdmac_bd *bd = &lp->rx_bd_v[lp->rx_bd_ci]; + struct sk_buff *skb = lp->rx_skb[lp->rx_bd_ci]; + unsigned int bdstat = be32_to_cpu(bd->app0); + int length; + + /* While this should not normally happen, we can end + * here when GFP_ATOMIC allocations fail, and we + * therefore have un-allocated buffers. + */ + if (!skb) + break; - skb = lp->rx_skb[lp->rx_bd_ci]; - length = be32_to_cpu(cur_p->app4) & 0x3FFF; + /* Loop over all completed buffer descriptors */ + if (!(bdstat & STS_CTRL_APP0_CMPLT)) + break; - dma_unmap_single(ndev->dev.parent, be32_to_cpu(cur_p->phys), + dma_unmap_single(ndev->dev.parent, be32_to_cpu(bd->phys), XTE_MAX_JUMBO_FRAME_SIZE, DMA_FROM_DEVICE); + /* The buffer is not valid for DMA anymore */ + bd->phys = 0; + bd->len = 0; + length = be32_to_cpu(bd->app4) & 0x3FFF; skb_put(skb, length); skb->protocol = eth_type_trans(skb, ndev); skb_checksum_none_assert(skb); @@ -958,39 +973,74 @@ static void ll_temac_recv(struct net_device *ndev) * (back) for proper IP checksum byte order * (be16). */ - skb->csum = htons(be32_to_cpu(cur_p->app3) & 0xFFFF); + skb->csum = htons(be32_to_cpu(bd->app3) & 0xFFFF); skb->ip_summed = CHECKSUM_COMPLETE; } if (!skb_defer_rx_timestamp(skb)) netif_rx(skb); + /* The skb buffer is now owned by network stack above */ + lp->rx_skb[lp->rx_bd_ci] = NULL; ndev->stats.rx_packets++; ndev->stats.rx_bytes += length; - new_skb = netdev_alloc_skb_ip_align(ndev, - XTE_MAX_JUMBO_FRAME_SIZE); - if (!new_skb) { - spin_unlock_irqrestore(&lp->rx_lock, flags); - return; + rx_bd = lp->rx_bd_ci; + if (++lp->rx_bd_ci >= RX_BD_NUM) + lp->rx_bd_ci = 0; + } while (rx_bd != lp->rx_bd_tail); + + /* Allocate new buffers for those buffer descriptors that were + * passed to network stack. Note that GFP_ATOMIC allocations + * can fail (e.g. when a larger burst of GFP_ATOMIC + * allocations occurs), so while we try to allocate all + * buffers in the same interrupt where they were processed, we + * continue with what we could get in case of allocation + * failure. Allocation of remaining buffers will be retried + * in following calls. + */ + while (1) { + struct sk_buff *skb; + struct cdmac_bd *bd; + dma_addr_t skb_dma_addr; + + rx_bd = lp->rx_bd_tail + 1; + if (rx_bd >= RX_BD_NUM) + rx_bd = 0; + bd = &lp->rx_bd_v[rx_bd]; + + if (bd->phys) + break; /* All skb's allocated */ + + skb = netdev_alloc_skb_ip_align(ndev, XTE_MAX_JUMBO_FRAME_SIZE); + if (!skb) { + dev_warn(&ndev->dev, "skb alloc failed\n"); + break; } - cur_p->app0 = cpu_to_be32(STS_CTRL_APP0_IRQONEND); - skb_dma_addr = dma_map_single(ndev->dev.parent, new_skb->data, + skb_dma_addr = dma_map_single(ndev->dev.parent, skb->data, XTE_MAX_JUMBO_FRAME_SIZE, DMA_FROM_DEVICE); - cur_p->phys = cpu_to_be32(skb_dma_addr); - cur_p->len = cpu_to_be32(XTE_MAX_JUMBO_FRAME_SIZE); - lp->rx_skb[lp->rx_bd_ci] = new_skb; + if (WARN_ON_ONCE(dma_mapping_error(ndev->dev.parent, + skb_dma_addr))) { + dev_kfree_skb_any(skb); + break; + } - lp->rx_bd_ci++; - if (lp->rx_bd_ci >= RX_BD_NUM) - lp->rx_bd_ci = 0; + bd->phys = cpu_to_be32(skb_dma_addr); + bd->len = cpu_to_be32(XTE_MAX_JUMBO_FRAME_SIZE); + bd->app0 = cpu_to_be32(STS_CTRL_APP0_IRQONEND); + lp->rx_skb[rx_bd] = skb; + + lp->rx_bd_tail = rx_bd; + update_tail = true; + } - cur_p = &lp->rx_bd_v[lp->rx_bd_ci]; - bdstat = be32_to_cpu(cur_p->app0); + /* Move tail pointer when buffers have been allocated */ + if (update_tail) { + lp->dma_out(lp, RX_TAILDESC_PTR, + lp->rx_bd_p + sizeof(*lp->rx_bd_v) * lp->rx_bd_tail); } - lp->dma_out(lp, RX_TAILDESC_PTR, tail_p); spin_unlock_irqrestore(&lp->rx_lock, flags); } -- 2.25.0

4 years, 10 months

1
0
0 0

[PATCH 2/8] net: ll_temac: Add more error handling of dma_map_single() calls

by Esben Haabendal

This adds error handling to the remaining dma_map_single() calls, so that behavior is well defined if/when we run out of DMA memory. Signed-off-by: Esben Haabendal <esben(a)geanix.com> Cc: stable(a)vger.kernel.org --- drivers/net/ethernet/xilinx/ll_temac_main.c | 26 +++++++++++++++++++-- 1 file changed, 24 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/xilinx/ll_temac_main.c b/drivers/net/ethernet/xilinx/ll_temac_main.c index 996004ef8bd4..c368c3914bda 100644 --- a/drivers/net/ethernet/xilinx/ll_temac_main.c +++ b/drivers/net/ethernet/xilinx/ll_temac_main.c @@ -367,6 +367,8 @@ static int temac_dma_bd_init(struct net_device *ndev) skb_dma_addr = dma_map_single(ndev->dev.parent, skb->data, XTE_MAX_JUMBO_FRAME_SIZE, DMA_FROM_DEVICE); + if (dma_mapping_error(ndev->dev.parent, skb_dma_addr)) + goto out; lp->rx_bd_v[i].phys = cpu_to_be32(skb_dma_addr); lp->rx_bd_v[i].len = cpu_to_be32(XTE_MAX_JUMBO_FRAME_SIZE); lp->rx_bd_v[i].app0 = cpu_to_be32(STS_CTRL_APP0_IRQONEND); @@ -863,12 +865,13 @@ temac_start_xmit(struct sk_buff *skb, struct net_device *ndev) skb_dma_addr = dma_map_single(ndev->dev.parent, skb->data, skb_headlen(skb), DMA_TO_DEVICE); cur_p->len = cpu_to_be32(skb_headlen(skb)); + if (WARN_ON_ONCE(dma_mapping_error(ndev->dev.parent, skb_dma_addr))) + return NETDEV_TX_BUSY; cur_p->phys = cpu_to_be32(skb_dma_addr); ptr_to_txbd((void *)skb, cur_p); for (ii = 0; ii < num_frag; ii++) { - lp->tx_bd_tail++; - if (lp->tx_bd_tail >= TX_BD_NUM) + if (++lp->tx_bd_tail >= TX_BD_NUM) lp->tx_bd_tail = 0; cur_p = &lp->tx_bd_v[lp->tx_bd_tail]; @@ -876,6 +879,25 @@ temac_start_xmit(struct sk_buff *skb, struct net_device *ndev) skb_frag_address(frag), skb_frag_size(frag), DMA_TO_DEVICE); + if (dma_mapping_error(ndev->dev.parent, skb_dma_addr)) { + if (--lp->tx_bd_tail < 0) + lp->tx_bd_tail = TX_BD_NUM - 1; + cur_p = &lp->tx_bd_v[lp->tx_bd_tail]; + while (--ii >= 0) { + --frag; + dma_unmap_single(ndev->dev.parent, + be32_to_cpu(cur_p->phys), + skb_frag_size(frag), + DMA_TO_DEVICE); + if (--lp->tx_bd_tail < 0) + lp->tx_bd_tail = TX_BD_NUM - 1; + cur_p = &lp->tx_bd_v[lp->tx_bd_tail]; + } + dma_unmap_single(ndev->dev.parent, + be32_to_cpu(cur_p->phys), + skb_headlen(skb), DMA_TO_DEVICE); + return NETDEV_TX_BUSY; + } cur_p->phys = cpu_to_be32(skb_dma_addr); cur_p->len = cpu_to_be32(skb_frag_size(frag)); cur_p->app0 = 0; -- 2.25.0

4 years, 10 months

1
0
0 0

[PATCH 1/8] net: ll_temac: Fix race condition causing TX hang

by Esben Haabendal

It is possible that the interrupt handler fires and frees up space in the TX ring in between checking for sufficient TX ring space and stopping the TX queue in temac_start_xmit. If this happens, the queue wake from the interrupt handler will occur before the queue is stopped, causing a lost wakeup and the adapter's transmit hanging. To avoid this, after stopping the queue, check again whether there is sufficient space in the TX ring. If so, wake up the queue again. This is a port of the similar fix in axienet driver, commit 7de44285c1f6 ("net: axienet: Fix race condition causing TX hang"). Signed-off-by: Esben Haabendal <esben(a)geanix.com> Cc: stable(a)vger.kernel.org --- drivers/net/ethernet/xilinx/ll_temac_main.c | 19 ++++++++++++++++--- 1 file changed, 16 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/xilinx/ll_temac_main.c b/drivers/net/ethernet/xilinx/ll_temac_main.c index 6f11f52c9a9e..996004ef8bd4 100644 --- a/drivers/net/ethernet/xilinx/ll_temac_main.c +++ b/drivers/net/ethernet/xilinx/ll_temac_main.c @@ -788,6 +788,9 @@ static void temac_start_xmit_done(struct net_device *ndev) stat = be32_to_cpu(cur_p->app0); } + /* Matches barrier in temac_start_xmit */ + smp_mb(); + netif_wake_queue(ndev); } @@ -830,9 +833,19 @@ temac_start_xmit(struct sk_buff *skb, struct net_device *ndev) cur_p = &lp->tx_bd_v[lp->tx_bd_tail]; if (temac_check_tx_bd_space(lp, num_frag + 1)) { - if (!netif_queue_stopped(ndev)) - netif_stop_queue(ndev); - return NETDEV_TX_BUSY; + if (netif_queue_stopped(ndev)) + return NETDEV_TX_BUSY; + + netif_stop_queue(ndev); + + /* Matches barrier in temac_start_xmit_done */ + smp_mb(); + + /* Space might have just been freed - check again */ + if (temac_check_tx_bd_space(lp, num_frag)) + return NETDEV_TX_BUSY; + + netif_wake_queue(ndev); } cur_p->app0 = 0; -- 2.25.0

4 years, 10 months

1
0
0 0

FAILED: patch "[PATCH] NFSv4: Ensure the delegation is pinned in" failed to apply to 5.5-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 5.5-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. thanks, greg k-h ------------------ original commit in Linus's tree ------------------ >From 8c75593c6eee0f661ddf25dfde0e6ad2a84be7a9 Mon Sep 17 00:00:00 2001 From: Trond Myklebust <trondmy(a)gmail.com> Date: Thu, 13 Feb 2020 14:51:06 -0500 Subject: [PATCH] NFSv4: Ensure the delegation is pinned in nfs_do_return_delegation() The call to nfs_do_return_delegation() needs to be taken without any RCU locks. Add a refcount to make sure the delegation remains pinned in memory until we're done. Fixes: ee05f456772d ("NFSv4: Fix races between open and delegreturn") Signed-off-by: Trond Myklebust <trond.myklebust(a)hammerspace.com> Signed-off-by: Anna Schumaker <Anna.Schumaker(a)Netapp.com> diff --git a/fs/nfs/delegation.c b/fs/nfs/delegation.c index d856326836a2..c17ff826e7e9 100644 --- a/fs/nfs/delegation.c +++ b/fs/nfs/delegation.c @@ -47,10 +47,22 @@ static void nfs_mark_delegation_revoked(struct nfs_delegation *delegation) } } +static struct nfs_delegation *nfs_get_delegation(struct nfs_delegation *delegation) +{ + refcount_inc(&delegation->refcount); + return delegation; +} + +static void nfs_put_delegation(struct nfs_delegation *delegation) +{ + if (refcount_dec_and_test(&delegation->refcount)) + __nfs_free_delegation(delegation); +} + static void nfs_free_delegation(struct nfs_delegation *delegation) { nfs_mark_delegation_revoked(delegation); - __nfs_free_delegation(delegation); + nfs_put_delegation(delegation); } /** @@ -275,8 +287,10 @@ nfs_start_delegation_return_locked(struct nfs_inode *nfsi) if (delegation == NULL) goto out; spin_lock(&delegation->lock); - if (!test_and_set_bit(NFS_DELEGATION_RETURNING, &delegation->flags)) - ret = delegation; + if (!test_and_set_bit(NFS_DELEGATION_RETURNING, &delegation->flags)) { + /* Refcount matched in nfs_end_delegation_return() */ + ret = nfs_get_delegation(delegation); + } spin_unlock(&delegation->lock); if (ret) nfs_clear_verifier_delegated(&nfsi->vfs_inode); @@ -397,6 +411,7 @@ int nfs_inode_set_delegation(struct inode *inode, const struct cred *cred, if (delegation == NULL) return -ENOMEM; nfs4_stateid_copy(&delegation->stateid, stateid); + refcount_set(&delegation->refcount, 1); delegation->type = type; delegation->pagemod_limit = pagemod_limit; delegation->change_attr = inode_peek_iversion_raw(inode); @@ -496,6 +511,8 @@ static int nfs_end_delegation_return(struct inode *inode, struct nfs_delegation err = nfs_do_return_delegation(inode, delegation, issync); out: + /* Refcount matched in nfs_start_delegation_return_locked() */ + nfs_put_delegation(delegation); return err; } @@ -690,7 +707,8 @@ void nfs4_inode_return_delegation_on_close(struct inode *inode) list_empty(&NFS_I(inode)->open_files) && !test_and_set_bit(NFS_DELEGATION_RETURNING, &delegation->flags)) { clear_bit(NFS_DELEGATION_RETURN_IF_CLOSED, &delegation->flags); - ret = delegation; + /* Refcount matched in nfs_end_delegation_return() */ + ret = nfs_get_delegation(delegation); } spin_unlock(&delegation->lock); if (ret) @@ -1094,10 +1112,11 @@ void nfs_delegation_reap_unclaimed(struct nfs_client *clp) delegation = nfs_start_delegation_return_locked(NFS_I(inode)); rcu_read_unlock(); if (delegation != NULL) { - delegation = nfs_detach_delegation(NFS_I(inode), - delegation, server); - if (delegation != NULL) + if (nfs_detach_delegation(NFS_I(inode), delegation, + server) != NULL) nfs_free_delegation(delegation); + /* Match nfs_start_delegation_return_locked */ + nfs_put_delegation(delegation); } iput(inode); nfs_sb_deactive(server->super); diff --git a/fs/nfs/delegation.h b/fs/nfs/delegation.h index 31b84604d383..9b00a0b7f832 100644 --- a/fs/nfs/delegation.h +++ b/fs/nfs/delegation.h @@ -22,6 +22,7 @@ struct nfs_delegation { unsigned long pagemod_limit; __u64 change_attr; unsigned long flags; + refcount_t refcount; spinlock_t lock; struct rcu_head rcu; };

4 years, 10 months

1
0
0 0

FAILED: patch "[PATCH] KVM: nVMX: Handle pending #DB when injecting INIT VM-exit" failed to apply to 5.4-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 5.4-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. thanks, greg k-h ------------------ original commit in Linus's tree ------------------ >From 684c0422da71da0cd81319c90b8099b563b13da4 Mon Sep 17 00:00:00 2001 From: Oliver Upton <oupton(a)google.com> Date: Fri, 7 Feb 2020 02:36:05 -0800 Subject: [PATCH] KVM: nVMX: Handle pending #DB when injecting INIT VM-exit SDM 27.3.4 states that the 'pending debug exceptions' VMCS field will be populated if a VM-exit caused by an INIT signal takes priority over a debug-trap. Emulate this behavior when synthesizing an INIT signal VM-exit into L1. Fixes: 4b9852f4f389 ("KVM: x86: Fix INIT signal handling in various CPU states") Signed-off-by: Oliver Upton <oupton(a)google.com> Signed-off-by: Paolo Bonzini <pbonzini(a)redhat.com> diff --git a/arch/x86/kvm/vmx/nested.c b/arch/x86/kvm/vmx/nested.c index 657c2eda357c..1586aaae3a6f 100644 --- a/arch/x86/kvm/vmx/nested.c +++ b/arch/x86/kvm/vmx/nested.c @@ -3575,6 +3575,33 @@ static void nested_vmx_inject_exception_vmexit(struct kvm_vcpu *vcpu, nested_vmx_vmexit(vcpu, EXIT_REASON_EXCEPTION_NMI, intr_info, exit_qual); } +/* + * Returns true if a debug trap is pending delivery. + * + * In KVM, debug traps bear an exception payload. As such, the class of a #DB + * exception may be inferred from the presence of an exception payload. + */ +static inline bool vmx_pending_dbg_trap(struct kvm_vcpu *vcpu) +{ + return vcpu->arch.exception.pending && + vcpu->arch.exception.nr == DB_VECTOR && + vcpu->arch.exception.payload; +} + +/* + * Certain VM-exits set the 'pending debug exceptions' field to indicate a + * recognized #DB (data or single-step) that has yet to be delivered. Since KVM + * represents these debug traps with a payload that is said to be compatible + * with the 'pending debug exceptions' field, write the payload to the VMCS + * field if a VM-exit is delivered before the debug trap. + */ +static void nested_vmx_update_pending_dbg(struct kvm_vcpu *vcpu) +{ + if (vmx_pending_dbg_trap(vcpu)) + vmcs_writel(GUEST_PENDING_DBG_EXCEPTIONS, + vcpu->arch.exception.payload); +} + static int vmx_check_nested_events(struct kvm_vcpu *vcpu, bool external_intr) { struct vcpu_vmx *vmx = to_vmx(vcpu); @@ -3587,6 +3614,7 @@ static int vmx_check_nested_events(struct kvm_vcpu *vcpu, bool external_intr) test_bit(KVM_APIC_INIT, &apic->pending_events)) { if (block_nested_events) return -EBUSY; + nested_vmx_update_pending_dbg(vcpu); clear_bit(KVM_APIC_INIT, &apic->pending_events); nested_vmx_vmexit(vcpu, EXIT_REASON_INIT_SIGNAL, 0, 0); return 0;

4 years, 10 months

1
0
0 0

FAILED: patch "[PATCH] RDMA/rxe: Fix soft lockup problem due to using tasklets in" failed to apply to 4.9-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 4.9-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. thanks, greg k-h ------------------ original commit in Linus's tree ------------------ >From 8ac0e6641c7ca14833a2a8c6f13d8e0a435e535c Mon Sep 17 00:00:00 2001 From: Zhu Yanjun <yanjunz(a)mellanox.com> Date: Wed, 12 Feb 2020 09:26:33 +0200 Subject: [PATCH] RDMA/rxe: Fix soft lockup problem due to using tasklets in softirq When run stress tests with RXE, the following Call Traces often occur watchdog: BUG: soft lockup - CPU#2 stuck for 22s! [swapper/2:0] ... Call Trace: <IRQ> create_object+0x3f/0x3b0 kmem_cache_alloc_node_trace+0x129/0x2d0 __kmalloc_reserve.isra.52+0x2e/0x80 __alloc_skb+0x83/0x270 rxe_init_packet+0x99/0x150 [rdma_rxe] rxe_requester+0x34e/0x11a0 [rdma_rxe] rxe_do_task+0x85/0xf0 [rdma_rxe] tasklet_action_common.isra.21+0xeb/0x100 __do_softirq+0xd0/0x298 irq_exit+0xc5/0xd0 smp_apic_timer_interrupt+0x68/0x120 apic_timer_interrupt+0xf/0x20 </IRQ> ... The root cause is that tasklet is actually a softirq. In a tasklet handler, another softirq handler is triggered. Usually these softirq handlers run on the same cpu core. So this will cause "soft lockup Bug". Fixes: 8700e3e7c485 ("Soft RoCE driver") Link: https://lore.kernel.org/r/20200212072635.682689-8-leon@kernel.org Signed-off-by: Zhu Yanjun <yanjunz(a)mellanox.com> Signed-off-by: Leon Romanovsky <leonro(a)mellanox.com> Signed-off-by: Jason Gunthorpe <jgg(a)mellanox.com> diff --git a/drivers/infiniband/sw/rxe/rxe_comp.c b/drivers/infiniband/sw/rxe/rxe_comp.c index 116cafc9afcf..4bc88708b355 100644 --- a/drivers/infiniband/sw/rxe/rxe_comp.c +++ b/drivers/infiniband/sw/rxe/rxe_comp.c @@ -329,7 +329,7 @@ static inline enum comp_state check_ack(struct rxe_qp *qp, qp->comp.psn = pkt->psn; if (qp->req.wait_psn) { qp->req.wait_psn = 0; - rxe_run_task(&qp->req.task, 1); + rxe_run_task(&qp->req.task, 0); } } return COMPST_ERROR_RETRY; @@ -463,7 +463,7 @@ static void do_complete(struct rxe_qp *qp, struct rxe_send_wqe *wqe) */ if (qp->req.wait_fence) { qp->req.wait_fence = 0; - rxe_run_task(&qp->req.task, 1); + rxe_run_task(&qp->req.task, 0); } } @@ -479,7 +479,7 @@ static inline enum comp_state complete_ack(struct rxe_qp *qp, if (qp->req.need_rd_atomic) { qp->comp.timeout_retry = 0; qp->req.need_rd_atomic = 0; - rxe_run_task(&qp->req.task, 1); + rxe_run_task(&qp->req.task, 0); } } @@ -725,7 +725,7 @@ int rxe_completer(void *arg) RXE_CNT_COMP_RETRY); qp->req.need_retry = 1; qp->comp.started_retry = 1; - rxe_run_task(&qp->req.task, 1); + rxe_run_task(&qp->req.task, 0); } if (pkt) {

4 years, 10 months

1
0
0 0

FAILED: patch "[PATCH] RDMA/rxe: Fix soft lockup problem due to using tasklets in" failed to apply to 4.14-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 4.14-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. thanks, greg k-h ------------------ original commit in Linus's tree ------------------ >From 8ac0e6641c7ca14833a2a8c6f13d8e0a435e535c Mon Sep 17 00:00:00 2001 From: Zhu Yanjun <yanjunz(a)mellanox.com> Date: Wed, 12 Feb 2020 09:26:33 +0200 Subject: [PATCH] RDMA/rxe: Fix soft lockup problem due to using tasklets in softirq When run stress tests with RXE, the following Call Traces often occur watchdog: BUG: soft lockup - CPU#2 stuck for 22s! [swapper/2:0] ... Call Trace: <IRQ> create_object+0x3f/0x3b0 kmem_cache_alloc_node_trace+0x129/0x2d0 __kmalloc_reserve.isra.52+0x2e/0x80 __alloc_skb+0x83/0x270 rxe_init_packet+0x99/0x150 [rdma_rxe] rxe_requester+0x34e/0x11a0 [rdma_rxe] rxe_do_task+0x85/0xf0 [rdma_rxe] tasklet_action_common.isra.21+0xeb/0x100 __do_softirq+0xd0/0x298 irq_exit+0xc5/0xd0 smp_apic_timer_interrupt+0x68/0x120 apic_timer_interrupt+0xf/0x20 </IRQ> ... The root cause is that tasklet is actually a softirq. In a tasklet handler, another softirq handler is triggered. Usually these softirq handlers run on the same cpu core. So this will cause "soft lockup Bug". Fixes: 8700e3e7c485 ("Soft RoCE driver") Link: https://lore.kernel.org/r/20200212072635.682689-8-leon@kernel.org Signed-off-by: Zhu Yanjun <yanjunz(a)mellanox.com> Signed-off-by: Leon Romanovsky <leonro(a)mellanox.com> Signed-off-by: Jason Gunthorpe <jgg(a)mellanox.com> diff --git a/drivers/infiniband/sw/rxe/rxe_comp.c b/drivers/infiniband/sw/rxe/rxe_comp.c index 116cafc9afcf..4bc88708b355 100644 --- a/drivers/infiniband/sw/rxe/rxe_comp.c +++ b/drivers/infiniband/sw/rxe/rxe_comp.c @@ -329,7 +329,7 @@ static inline enum comp_state check_ack(struct rxe_qp *qp, qp->comp.psn = pkt->psn; if (qp->req.wait_psn) { qp->req.wait_psn = 0; - rxe_run_task(&qp->req.task, 1); + rxe_run_task(&qp->req.task, 0); } } return COMPST_ERROR_RETRY; @@ -463,7 +463,7 @@ static void do_complete(struct rxe_qp *qp, struct rxe_send_wqe *wqe) */ if (qp->req.wait_fence) { qp->req.wait_fence = 0; - rxe_run_task(&qp->req.task, 1); + rxe_run_task(&qp->req.task, 0); } } @@ -479,7 +479,7 @@ static inline enum comp_state complete_ack(struct rxe_qp *qp, if (qp->req.need_rd_atomic) { qp->comp.timeout_retry = 0; qp->req.need_rd_atomic = 0; - rxe_run_task(&qp->req.task, 1); + rxe_run_task(&qp->req.task, 0); } } @@ -725,7 +725,7 @@ int rxe_completer(void *arg) RXE_CNT_COMP_RETRY); qp->req.need_retry = 1; qp->comp.started_retry = 1; - rxe_run_task(&qp->req.task, 1); + rxe_run_task(&qp->req.task, 0); } if (pkt) {

4 years, 10 months

1
0
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror February 2020