linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	stable@vger.kernel.org, Guy Levi <guyle@mellanox.com>,
	Moni Shoua <monis@mellanox.com>,
	Leon Romanovsky <leonro@mellanox.com>,
	Doug Ledford <dledford@redhat.com>,
	Sasha Levin <sashal@kernel.org>
Subject: [PATCH 4.19 47/85] IB/mlx5: Fix MR registration flow to use UMR properly
Date: Thu, 22 Aug 2019 10:19:20 -0700	[thread overview]
Message-ID: <20190822171733.330417503@linuxfoundation.org> (raw)
In-Reply-To: <20190822171731.012687054@linuxfoundation.org>

[ Upstream commit e5366d309a772fef264ec85e858f9ea46f939848 ]

Driver shouldn't allow to use UMR to register a MR when
umr_modify_atomic_disabled is set. Otherwise it will always end up with a
failure in the post send flow which sets the UMR WQE to modify atomic access
right.

Fixes: c8d75a980fab ("IB/mlx5: Respect new UMR capabilities")
Signed-off-by: Guy Levi <guyle@mellanox.com>
Reviewed-by: Moni Shoua <monis@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Link: https://lore.kernel.org/r/20190731081929.32559-1-leon@kernel.org
Signed-off-by: Doug Ledford <dledford@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/infiniband/hw/mlx5/mr.c | 27 +++++++++------------------
 1 file changed, 9 insertions(+), 18 deletions(-)

diff --git a/drivers/infiniband/hw/mlx5/mr.c b/drivers/infiniband/hw/mlx5/mr.c
index 9bab4fb65c688..bd1fdadf7ba01 100644
--- a/drivers/infiniband/hw/mlx5/mr.c
+++ b/drivers/infiniband/hw/mlx5/mr.c
@@ -51,22 +51,12 @@ static void clean_mr(struct mlx5_ib_dev *dev, struct mlx5_ib_mr *mr);
 static void dereg_mr(struct mlx5_ib_dev *dev, struct mlx5_ib_mr *mr);
 static int mr_cache_max_order(struct mlx5_ib_dev *dev);
 static int unreg_umr(struct mlx5_ib_dev *dev, struct mlx5_ib_mr *mr);
-static bool umr_can_modify_entity_size(struct mlx5_ib_dev *dev)
-{
-	return !MLX5_CAP_GEN(dev->mdev, umr_modify_entity_size_disabled);
-}
 
 static bool umr_can_use_indirect_mkey(struct mlx5_ib_dev *dev)
 {
 	return !MLX5_CAP_GEN(dev->mdev, umr_indirect_mkey_disabled);
 }
 
-static bool use_umr(struct mlx5_ib_dev *dev, int order)
-{
-	return order <= mr_cache_max_order(dev) &&
-		umr_can_modify_entity_size(dev);
-}
-
 static int destroy_mkey(struct mlx5_ib_dev *dev, struct mlx5_ib_mr *mr)
 {
 	int err = mlx5_core_destroy_mkey(dev->mdev, &mr->mmkey);
@@ -1305,7 +1295,7 @@ struct ib_mr *mlx5_ib_reg_user_mr(struct ib_pd *pd, u64 start, u64 length,
 {
 	struct mlx5_ib_dev *dev = to_mdev(pd->device);
 	struct mlx5_ib_mr *mr = NULL;
-	bool populate_mtts = false;
+	bool use_umr;
 	struct ib_umem *umem;
 	int page_shift;
 	int npages;
@@ -1338,29 +1328,30 @@ struct ib_mr *mlx5_ib_reg_user_mr(struct ib_pd *pd, u64 start, u64 length,
 	if (err < 0)
 		return ERR_PTR(err);
 
-	if (use_umr(dev, order)) {
+	use_umr = !MLX5_CAP_GEN(dev->mdev, umr_modify_entity_size_disabled) &&
+		  (!MLX5_CAP_GEN(dev->mdev, umr_modify_atomic_disabled) ||
+		   !MLX5_CAP_GEN(dev->mdev, atomic));
+
+	if (order <= mr_cache_max_order(dev) && use_umr) {
 		mr = alloc_mr_from_cache(pd, umem, virt_addr, length, ncont,
 					 page_shift, order, access_flags);
 		if (PTR_ERR(mr) == -EAGAIN) {
 			mlx5_ib_dbg(dev, "cache empty for order %d\n", order);
 			mr = NULL;
 		}
-		populate_mtts = false;
 	} else if (!MLX5_CAP_GEN(dev->mdev, umr_extended_translation_offset)) {
 		if (access_flags & IB_ACCESS_ON_DEMAND) {
 			err = -EINVAL;
 			pr_err("Got MR registration for ODP MR > 512MB, not supported for Connect-IB\n");
 			goto error;
 		}
-		populate_mtts = true;
+		use_umr = false;
 	}
 
 	if (!mr) {
-		if (!umr_can_modify_entity_size(dev))
-			populate_mtts = true;
 		mutex_lock(&dev->slow_path_mutex);
 		mr = reg_create(NULL, pd, virt_addr, length, umem, ncont,
-				page_shift, access_flags, populate_mtts);
+				page_shift, access_flags, !use_umr);
 		mutex_unlock(&dev->slow_path_mutex);
 	}
 
@@ -1378,7 +1369,7 @@ struct ib_mr *mlx5_ib_reg_user_mr(struct ib_pd *pd, u64 start, u64 length,
 	update_odp_mr(mr);
 #endif
 
-	if (!populate_mtts) {
+	if (use_umr) {
 		int update_xlt_flags = MLX5_IB_UPD_XLT_ENABLE;
 
 		if (access_flags & IB_ACCESS_ON_DEMAND)
-- 
2.20.1




  parent reply	other threads:[~2019-08-22 17:29 UTC|newest]

Thread overview: 91+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-08-22 17:18 [PATCH 4.19 00/85] 4.19.68-stable review Greg Kroah-Hartman
2019-08-22 17:18 ` [PATCH 4.19 01/85] sh: kernel: hw_breakpoint: Fix missing break in switch statement Greg Kroah-Hartman
2019-08-22 17:18 ` [PATCH 4.19 02/85] seq_file: fix problem when seeking mid-record Greg Kroah-Hartman
2019-08-22 17:18 ` [PATCH 4.19 03/85] mm/hmm: fix bad subpage pointer in try_to_unmap_one Greg Kroah-Hartman
2019-08-22 17:18 ` [PATCH 4.19 04/85] mm: mempolicy: make the behavior consistent when MPOL_MF_MOVE* and MPOL_MF_STRICT were specified Greg Kroah-Hartman
2019-08-22 17:18 ` [PATCH 4.19 05/85] mm: mempolicy: handle vma with unmovable pages mapped correctly in mbind Greg Kroah-Hartman
2019-08-22 17:18 ` [PATCH 4.19 06/85] mm/memcontrol.c: fix use after free in mem_cgroup_iter() Greg Kroah-Hartman
2019-08-22 17:18 ` [PATCH 4.19 07/85] mm/usercopy: use memory range to be accessed for wraparound check Greg Kroah-Hartman
2019-08-22 17:18 ` [PATCH 4.19 08/85] Revert "pwm: Set class for exported channels in sysfs" Greg Kroah-Hartman
2019-08-22 17:18 ` [PATCH 4.19 09/85] cpufreq: schedutil: Dont skip freq update when limits change Greg Kroah-Hartman
2019-08-22 17:18 ` [PATCH 4.19 10/85] xtensa: add missing isync to the cpu_reset TLB code Greg Kroah-Hartman
2019-08-22 17:18 ` [PATCH 4.19 11/85] ALSA: hda/realtek - Add quirk for HP Envy x360 Greg Kroah-Hartman
2019-08-22 17:18 ` [PATCH 4.19 12/85] ALSA: usb-audio: Fix a stack buffer overflow bug in check_input_term Greg Kroah-Hartman
2019-08-22 17:18 ` [PATCH 4.19 13/85] ALSA: usb-audio: Fix an OOB bug in parse_audio_mixer_unit Greg Kroah-Hartman
2019-08-22 17:18 ` [PATCH 4.19 14/85] ALSA: hda - Apply workaround for another AMD chip 1022:1487 Greg Kroah-Hartman
2019-08-22 17:18 ` [PATCH 4.19 15/85] ALSA: hda - Fix a memory leak bug Greg Kroah-Hartman
2019-08-22 17:18 ` [PATCH 4.19 16/85] ALSA: hda - Add a generic reboot_notify Greg Kroah-Hartman
2019-08-22 17:18 ` [PATCH 4.19 17/85] ALSA: hda - Let all conexant codec enter D3 when rebooting Greg Kroah-Hartman
2019-08-22 17:18 ` [PATCH 4.19 18/85] HID: holtek: test for sanity of intfdata Greg Kroah-Hartman
2019-08-22 17:18 ` [PATCH 4.19 19/85] HID: hiddev: avoid opening a disconnected device Greg Kroah-Hartman
2019-08-22 17:18 ` [PATCH 4.19 20/85] HID: hiddev: do cleanup in failure of opening a device Greg Kroah-Hartman
2019-08-22 17:18 ` [PATCH 4.19 21/85] Input: kbtab - sanity check for endpoint type Greg Kroah-Hartman
2019-08-22 17:18 ` [PATCH 4.19 22/85] Input: iforce - add sanity checks Greg Kroah-Hartman
2019-08-22 17:18 ` [PATCH 4.19 23/85] net: usb: pegasus: fix improper read if get_registers() fail Greg Kroah-Hartman
2019-08-22 17:18 ` [PATCH 4.19 24/85] netfilter: ebtables: also count base chain policies Greg Kroah-Hartman
2019-08-22 17:18 ` [PATCH 4.19 25/85] riscv: Make __fstate_clean() work correctly Greg Kroah-Hartman
2019-08-22 17:18 ` [PATCH 4.19 26/85] clk: at91: generated: Truncate divisor to GENERATED_MAX_DIV + 1 Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 27/85] clk: sprd: Select REGMAP_MMIO to avoid compile errors Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 28/85] clk: renesas: cpg-mssr: Fix reset control race condition Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 29/85] xen/pciback: remove set but not used variable old_state Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 30/85] irqchip/gic-v3-its: Free unused vpt_page when alloc vpe table fail Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 31/85] irqchip/irq-imx-gpcv2: Forward irq type to parent Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 32/85] perf header: Fix divide by zero error if f_header.attr_size==0 Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 33/85] perf header: Fix use of unitialized value warning Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 34/85] libata: zpodd: Fix small read overflow in zpodd_get_mech_type() Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 35/85] drm/bridge: lvds-encoder: Fix build error while CONFIG_DRM_KMS_HELPER=m Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 36/85] Btrfs: fix deadlock between fiemap and transaction commits Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 37/85] scsi: hpsa: correct scsi command status issue after reset Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 38/85] scsi: qla2xxx: Fix possible fcport null-pointer dereferences Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 39/85] drm/amdgpu: fix a potential information leaking bug Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 40/85] ata: libahci: do not complain in case of deferred probe Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 41/85] kbuild: modpost: handle KBUILD_EXTRA_SYMBOLS only for external modules Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 42/85] kbuild: Check for unknown options with cc-option usage in Kconfig and clang Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 43/85] arm64/efi: fix variable si set but not used Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 44/85] arm64: unwind: Prohibit probing on return_address() Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 45/85] arm64/mm: fix variable pud set but not used Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 46/85] IB/core: Add mitigation for Spectre V1 Greg Kroah-Hartman
2019-08-22 17:19 ` Greg Kroah-Hartman [this message]
2019-08-22 17:19 ` [PATCH 4.19 48/85] IB/mad: Fix use-after-free in ib mad completion handling Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 49/85] drm: msm: Fix add_gpu_components Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 50/85] drm/exynos: fix missing decrement of retry counter Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 51/85] Revert "kmemleak: allow to coexist with fault injection" Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 52/85] ocfs2: remove set but not used variable last_hash Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 53/85] asm-generic: fix -Wtype-limits compiler warnings Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 54/85] arm64: KVM: regmap: Fix unexpected switch fall-through Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 55/85] KVM: arm/arm64: Sync ICH_VMCR_EL2 back when about to block Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 56/85] staging: comedi: dt3000: Fix signed integer overflow divider * base Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 57/85] staging: comedi: dt3000: Fix rounding up of timer divisor Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 58/85] iio: adc: max9611: Fix temperature reading in probe Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 59/85] USB: core: Fix races in character device registration and deregistraion Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 60/85] usb: gadget: udc: renesas_usb3: Fix sysfs interface of "role" Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 61/85] usb: cdc-acm: make sure a refcount is taken early enough Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 62/85] USB: CDC: fix sanity checks in CDC union parser Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 63/85] USB: serial: option: add D-Link DWM-222 device ID Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 64/85] USB: serial: option: Add support for ZTE MF871A Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 65/85] USB: serial: option: add the BroadMobi BM818 card Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 66/85] USB: serial: option: Add Motorola modem UARTs Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 67/85] drm/i915/cfl: Add a new CFL PCI ID Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 68/85] dm: disable DISCARD if the underlying storage no longer supports it Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 69/85] arm64: ftrace: Ensure module ftrace trampoline is coherent with I-side Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 70/85] netfilter: conntrack: Use consistent ct id hash calculation Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 71/85] Input: psmouse - fix build error of multiple definition Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 72/85] iommu/amd: Move iommu_init_pci() to .init section Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 73/85] bnx2x: Fix VFs VLAN reconfiguration in reload Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 74/85] bonding: Add vlan tx offload to hw_enc_features Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 75/85] net: dsa: Check existence of .port_mdb_add callback before calling it Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 76/85] net/mlx4_en: fix a memory leak bug Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 77/85] net/packet: fix race in tpacket_snd() Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 78/85] sctp: fix memleak in sctp_send_reset_streams Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 79/85] sctp: fix the transport error_count check Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 80/85] team: Add vlan tx offload to hw_enc_features Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 81/85] tipc: initialise addr_trail_end when setting node addresses Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 82/85] xen/netback: Reset nr_frags before freeing skb Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 83/85] net/mlx5e: Only support tx/rx pause setting for port owner Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 84/85] net/mlx5e: Use flow keys dissector to parse packets for ARFS Greg Kroah-Hartman
2019-08-22 17:19 ` [PATCH 4.19 85/85] mmc: sdhci-of-arasan: Do now show error message in case of deffered probe Greg Kroah-Hartman
2019-08-22 21:17 ` [PATCH 4.19 00/85] 4.19.68-stable review kernelci.org bot
2019-08-23  2:08 ` Jon Hunter
2019-08-23  8:06 ` Naresh Kamboju
2019-08-23 14:28 ` Guenter Roeck
2019-08-24 17:51 ` shuah

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20190822171733.330417503@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=dledford@redhat.com \
    --cc=guyle@mellanox.com \
    --cc=leonro@mellanox.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=monis@mellanox.com \
    --cc=sashal@kernel.org \
    --cc=stable@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).