public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Sasha Levin <sashal@kernel.org>
To: linux-kernel@vger.kernel.org, stable@vger.kernel.org
Cc: Kaike Wan <kaike.wan@intel.com>,
	Dennis Dalessandro <dennis.dalessandro@intel.com>,
	Jason Gunthorpe <jgg@mellanox.com>,
	Sasha Levin <sashal@kernel.org>,
	linux-rdma@vger.kernel.org
Subject: [PATCH AUTOSEL 5.0 26/79] IB/hfi1: Fix the allocation of RSM table
Date: Fri, 26 Apr 2019 21:37:45 -0400	[thread overview]
Message-ID: <20190427013838.6596-26-sashal@kernel.org> (raw)
In-Reply-To: <20190427013838.6596-1-sashal@kernel.org>

From: Kaike Wan <kaike.wan@intel.com>

[ Upstream commit d0294344470e6b52d097aa7369173f32d11f2f52 ]

The receive side mapping (RSM) on hfi1 hardware is a special
matching mechanism to direct an incoming packet to a given
hardware receive context. It has 4 instances of matching capabilities
(RSM0 - RSM3) that share the same RSM table (RMT). The RMT has a total of
256 entries, each of which points to a receive context.

Currently, three instances of RSM have been used:
1. RSM0 by QOS;
2. RSM1 by PSM FECN;
3. RSM2 by VNIC.

Each RSM instance should reserve enough entries in RMT to function
properly. Since both PSM and VNIC could allocate any receive context
between dd->first_dyn_alloc_ctxt and dd->num_rcv_contexts, PSM FECN must
reserve enough RMT entries to cover the entire receive context index
range (dd->num_rcv_contexts - dd->first_dyn_alloc_ctxt) instead of only
the user receive contexts allocated for PSM
(dd->num_user_contexts). Consequently, the sizing of
dd->num_user_contexts in set_up_context_variables is incorrect.

Fixes: 2280740f01ae ("IB/hfi1: Virtual Network Interface Controller (VNIC) HW support")
Reviewed-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Reviewed-by: Michael J. Ruhl <michael.j.ruhl@intel.com>
Signed-off-by: Kaike Wan <kaike.wan@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/infiniband/hw/hfi1/chip.c | 26 +++++++++++++++++++-------
 1 file changed, 19 insertions(+), 7 deletions(-)

diff --git a/drivers/infiniband/hw/hfi1/chip.c b/drivers/infiniband/hw/hfi1/chip.c
index b443642eac02..0ae05e9249b3 100644
--- a/drivers/infiniband/hw/hfi1/chip.c
+++ b/drivers/infiniband/hw/hfi1/chip.c
@@ -13219,7 +13219,7 @@ static int set_up_context_variables(struct hfi1_devdata *dd)
 	int total_contexts;
 	int ret;
 	unsigned ngroups;
-	int qos_rmt_count;
+	int rmt_count;
 	int user_rmt_reduced;
 	u32 n_usr_ctxts;
 	u32 send_contexts = chip_send_contexts(dd);
@@ -13281,10 +13281,20 @@ static int set_up_context_variables(struct hfi1_devdata *dd)
 		n_usr_ctxts = rcv_contexts - total_contexts;
 	}
 
-	/* each user context requires an entry in the RMT */
-	qos_rmt_count = qos_rmt_entries(dd, NULL, NULL);
-	if (qos_rmt_count + n_usr_ctxts > NUM_MAP_ENTRIES) {
-		user_rmt_reduced = NUM_MAP_ENTRIES - qos_rmt_count;
+	/*
+	 * The RMT entries are currently allocated as shown below:
+	 * 1. QOS (0 to 128 entries);
+	 * 2. FECN for PSM (num_user_contexts + num_vnic_contexts);
+	 * 3. VNIC (num_vnic_contexts).
+	 * It should be noted that PSM FECN oversubscribe num_vnic_contexts
+	 * entries of RMT because both VNIC and PSM could allocate any receive
+	 * context between dd->first_dyn_alloc_text and dd->num_rcv_contexts,
+	 * and PSM FECN must reserve an RMT entry for each possible PSM receive
+	 * context.
+	 */
+	rmt_count = qos_rmt_entries(dd, NULL, NULL) + (num_vnic_contexts * 2);
+	if (rmt_count + n_usr_ctxts > NUM_MAP_ENTRIES) {
+		user_rmt_reduced = NUM_MAP_ENTRIES - rmt_count;
 		dd_dev_err(dd,
 			   "RMT size is reducing the number of user receive contexts from %u to %d\n",
 			   n_usr_ctxts,
@@ -14272,9 +14282,11 @@ static void init_user_fecn_handling(struct hfi1_devdata *dd,
 	u64 reg;
 	int i, idx, regoff, regidx;
 	u8 offset;
+	u32 total_cnt;
 
 	/* there needs to be enough room in the map table */
-	if (rmt->used + dd->num_user_contexts >= NUM_MAP_ENTRIES) {
+	total_cnt = dd->num_rcv_contexts - dd->first_dyn_alloc_ctxt;
+	if (rmt->used + total_cnt >= NUM_MAP_ENTRIES) {
 		dd_dev_err(dd, "User FECN handling disabled - too many user contexts allocated\n");
 		return;
 	}
@@ -14328,7 +14340,7 @@ static void init_user_fecn_handling(struct hfi1_devdata *dd,
 	/* add rule 1 */
 	add_rsm_rule(dd, RSM_INS_FECN, &rrd);
 
-	rmt->used += dd->num_user_contexts;
+	rmt->used += total_cnt;
 }
 
 /* Initialize RSM for VNIC */
-- 
2.19.1


  parent reply	other threads:[~2019-04-27  1:39 UTC|newest]

Thread overview: 83+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-04-27  1:37 [PATCH AUTOSEL 5.0 01/79] ASoC: tlv320aic3x: fix reset gpio reference counting Sasha Levin
2019-04-27  1:37 ` [PATCH AUTOSEL 5.0 02/79] ASoC: hdmi-codec: fix S/PDIF DAI Sasha Levin
2019-04-27  1:37 ` [PATCH AUTOSEL 5.0 03/79] ASoC: ab8500: Mark expected switch fall-through Sasha Levin
2019-04-27 17:14   ` Mark Brown
2019-04-27 17:31     ` Gustavo A. R. Silva
2019-04-27 18:00       ` Mark Brown
2019-04-28  1:06         ` Sasha Levin
2019-04-27  1:37 ` [PATCH AUTOSEL 5.0 04/79] ASoC: stm32: sai: fix iec958 controls indexation Sasha Levin
2019-04-27  1:37 ` [PATCH AUTOSEL 5.0 05/79] ASoC: stm32: sai: fix exposed capabilities in spdif mode Sasha Levin
2019-04-27  1:37 ` [PATCH AUTOSEL 5.0 06/79] ASoC: stm32: sai: fix race condition in irq handler Sasha Levin
2019-04-27  1:37 ` [PATCH AUTOSEL 5.0 07/79] ASoC:soc-pcm:fix a codec fixup issue in TDM case Sasha Levin
2019-04-27  1:37 ` [PATCH AUTOSEL 5.0 08/79] ASoC:hdac_hda:use correct format to setup hda codec Sasha Levin
2019-04-27  1:37 ` [PATCH AUTOSEL 5.0 09/79] ASoC:intel:skl:fix a simultaneous playback & capture issue on hda platform Sasha Levin
2019-04-27  1:37 ` [PATCH AUTOSEL 5.0 10/79] ASoC: dpcm: prevent snd_soc_dpcm use after free Sasha Levin
2019-04-27  1:37 ` [PATCH AUTOSEL 5.0 11/79] ASoC: nau8824: fix the issue of the widget with prefix name Sasha Levin
2019-04-27  1:37 ` [PATCH AUTOSEL 5.0 12/79] ASoC: nau8810: fix the issue of widget with prefixed name Sasha Levin
2019-04-27  1:37 ` [PATCH AUTOSEL 5.0 13/79] ASoC: samsung: odroid: Fix clock configuration for 44100 sample rate Sasha Levin
2019-04-27  1:37 ` [PATCH AUTOSEL 5.0 14/79] ASoC: rt5682: Check JD status when system resume Sasha Levin
2019-04-27  1:37 ` [PATCH AUTOSEL 5.0 15/79] ASoC: rt5682: fix jack type detection issue Sasha Levin
2019-04-27  1:37 ` [PATCH AUTOSEL 5.0 16/79] ASoC: rt5682: recording has no sound after booting Sasha Levin
2019-04-27  1:37 ` [PATCH AUTOSEL 5.0 17/79] ASoC: wm_adsp: Add locking to wm_adsp2_bus_error Sasha Levin
2019-04-27  1:37 ` [PATCH AUTOSEL 5.0 18/79] clk: meson-gxbb: round the vdec dividers to closest Sasha Levin
2019-04-27  1:37 ` [PATCH AUTOSEL 5.0 19/79] ASoC: stm32: dfsdm: manage multiple prepare Sasha Levin
2019-04-27  1:37 ` [PATCH AUTOSEL 5.0 20/79] ASoC: stm32: dfsdm: fix debugfs warnings on entry creation Sasha Levin
2019-04-27  1:37 ` [PATCH AUTOSEL 5.0 21/79] ASoC: cs4270: Set auto-increment bit for register writes Sasha Levin
2019-04-27  1:37 ` [PATCH AUTOSEL 5.0 22/79] ASoC: dapm: Fix NULL pointer dereference in snd_soc_dapm_free_kcontrol Sasha Levin
2019-04-27  1:37 ` [PATCH AUTOSEL 5.0 23/79] drm/omap: hdmi4_cec: Fix CEC clock handling for PM Sasha Levin
2019-04-27  1:37 ` [PATCH AUTOSEL 5.0 24/79] IB/hfi1: Clear the IOWAIT pending bits when QP is put into error state Sasha Levin
2019-04-27  1:37 ` [PATCH AUTOSEL 5.0 25/79] IB/hfi1: Eliminate opcode tests on mr deref Sasha Levin
2019-04-27  1:37 ` Sasha Levin [this message]
2019-04-27  1:37 ` [PATCH AUTOSEL 5.0 27/79] MIPS: KGDB: fix kgdb support for SMP platforms Sasha Levin
2019-04-27  1:37 ` [PATCH AUTOSEL 5.0 28/79] ASoC: tlv320aic32x4: Fix Common Pins Sasha Levin
2019-04-27  1:37 ` [PATCH AUTOSEL 5.0 29/79] drm/mediatek: Fix an error code in mtk_hdmi_dt_parse_pdata() Sasha Levin
2019-04-27  1:37 ` [PATCH AUTOSEL 5.0 30/79] ASoC: dpcm: skip missing substream while applying symmetry Sasha Levin
2019-04-27  1:37 ` [PATCH AUTOSEL 5.0 31/79] perf/x86/intel: Fix handling of wakeup_events for multi-entry PEBS Sasha Levin
2019-04-27  1:37 ` [PATCH AUTOSEL 5.0 32/79] perf/x86/intel: Initialize TFA MSR Sasha Levin
2019-04-27  1:37 ` [PATCH AUTOSEL 5.0 33/79] linux/kernel.h: Use parentheses around argument in u64_to_user_ptr() Sasha Levin
2019-04-27  1:37 ` [PATCH AUTOSEL 5.0 34/79] ALSA: hda/realtek - Move to ACT_INIT state Sasha Levin
2019-04-27  1:37 ` [PATCH AUTOSEL 5.0 35/79] iov_iter: Fix build error without CONFIG_CRYPTO Sasha Levin
2019-04-27  1:37 ` [PATCH AUTOSEL 5.0 36/79] xtensa: fix initialization of pt_regs::syscall in start_thread Sasha Levin
2019-04-27  1:37 ` [PATCH AUTOSEL 5.0 37/79] ASoC: rockchip: pdm: fix regmap_ops hang issue Sasha Levin
2019-04-27  1:37 ` [PATCH AUTOSEL 5.0 38/79] drm/amdkfd: Add picasso pci id Sasha Levin
2019-04-27  1:37 ` [PATCH AUTOSEL 5.0 39/79] drm/amdgpu: Adjust IB test timeout for XGMI configuration Sasha Levin
2019-04-27  1:37 ` [PATCH AUTOSEL 5.0 40/79] drm/amdgpu: amdgpu_device_recover_vram always failed if only one node in shadow_list Sasha Levin
2019-04-27  1:38 ` [PATCH AUTOSEL 5.0 41/79] drm/amd/display: fix cursor black issue Sasha Levin
2019-04-27  1:38 ` [PATCH AUTOSEL 5.0 42/79] ASoC: cs35l35: Disable regulators on driver removal Sasha Levin
2019-04-27  1:38 ` [PATCH AUTOSEL 5.0 43/79] objtool: Add rewind_stack_do_exit() to the noreturn list Sasha Levin
2019-04-27  1:38 ` [PATCH AUTOSEL 5.0 44/79] powerpc/vdso32: fix CLOCK_MONOTONIC on PPC64 Sasha Levin
2019-04-27  1:38 ` [PATCH AUTOSEL 5.0 45/79] slab: fix a crash by reading /proc/slab_allocators Sasha Levin
2019-04-27  1:38 ` [PATCH AUTOSEL 5.0 46/79] ASoC: stm32: fix sai driver name initialisation Sasha Levin
2019-04-27  1:38 ` [PATCH AUTOSEL 5.0 47/79] drm/sun4i: tcon top: Fix NULL/invalid pointer dereference in sun8i_tcon_top_un/bind Sasha Levin
2019-04-27  1:38 ` [PATCH AUTOSEL 5.0 48/79] virtio_pci: fix a NULL pointer reference in vp_del_vqs Sasha Levin
2019-04-27  1:38 ` [PATCH AUTOSEL 5.0 49/79] RDMA/vmw_pvrdma: Fix memory leak on pvrdma_pci_remove Sasha Levin
2019-04-27  1:38 ` [PATCH AUTOSEL 5.0 50/79] RDMA/hns: Fix bug that caused srq creation to fail Sasha Levin
2019-04-27  1:38 ` [PATCH AUTOSEL 5.0 51/79] tpm: fix an invalid condition in tpm_common_poll Sasha Levin
2019-04-27  1:38 ` [PATCH AUTOSEL 5.0 52/79] KEYS: trusted: fix -Wvarags warning Sasha Levin
2019-04-27  1:38 ` [PATCH AUTOSEL 5.0 53/79] scsi: csiostor: fix missing data copy in csio_scsi_err_handler() Sasha Levin
2019-04-27  1:38 ` [PATCH AUTOSEL 5.0 54/79] drm/mediatek: fix possible object reference leak Sasha Levin
2019-04-27  1:38 ` [PATCH AUTOSEL 5.0 55/79] drm/mediatek: fix the rate and divder of hdmi phy for MT2701 Sasha Levin
2019-04-27  1:38 ` [PATCH AUTOSEL 5.0 56/79] drm/mediatek: make implementation of recalc_rate() for MT2701 hdmi phy Sasha Levin
2019-04-27  1:38 ` [PATCH AUTOSEL 5.0 57/79] drm/mediatek: remove flag CLK_SET_RATE_PARENT " Sasha Levin
2019-04-27  1:38 ` [PATCH AUTOSEL 5.0 58/79] drm/mediatek: using new factor for tvdpll " Sasha Levin
2019-04-27  1:38 ` [PATCH AUTOSEL 5.0 59/79] drm/mediatek: no change parent rate in round_rate() " Sasha Levin
2019-04-27  1:38 ` [PATCH AUTOSEL 5.0 60/79] Bluetooth: btusb: request wake pin with NOAUTOEN Sasha Levin
2019-04-27  1:38 ` [PATCH AUTOSEL 5.0 61/79] ASoC: Intel: kbl: fix wrong number of channels Sasha Levin
2019-04-27  1:38 ` [PATCH AUTOSEL 5.0 62/79] ASoC: stm32: sai: fix master clock management Sasha Levin
2019-04-27  1:38 ` [PATCH AUTOSEL 5.0 63/79] ALSA: hda: Fix racy display power access Sasha Levin
2019-04-27  1:38 ` [PATCH AUTOSEL 5.0 64/79] block, bfq: fix use after free in bfq_bfqq_expire Sasha Levin
2019-04-27  1:38 ` [PATCH AUTOSEL 5.0 65/79] virtio-blk: limit number of hw queues by nr_cpu_ids Sasha Levin
2019-04-27  1:38 ` [PATCH AUTOSEL 5.0 66/79] blk-mq: introduce blk_mq_complete_request_sync() Sasha Levin
2019-04-27  1:38 ` [PATCH AUTOSEL 5.0 67/79] nvme: cancel request synchronously Sasha Levin
2019-04-27  1:38 ` [PATCH AUTOSEL 5.0 68/79] clk: x86: Add system specific quirk to mark clocks as critical Sasha Levin
2019-04-27  1:38 ` [PATCH AUTOSEL 5.0 69/79] nvme-fc: correct csn initialization and increments on error Sasha Levin
2019-04-27  1:38 ` [PATCH AUTOSEL 5.0 70/79] nvmet: fix discover log page when offsets are used Sasha Levin
2019-04-27  1:38 ` [PATCH AUTOSEL 5.0 71/79] platform/x86: pmc_atom: Drop __initconst on dmi table Sasha Levin
2019-04-27  1:38 ` [PATCH AUTOSEL 5.0 72/79] NFS: Forbid setting AF_INET6 to "struct sockaddr_in"->sin_family Sasha Levin
2019-04-27  1:38 ` [PATCH AUTOSEL 5.0 73/79] NFSv4.1 fix incorrect return value in copy_file_range Sasha Levin
2019-04-27  1:38 ` [PATCH AUTOSEL 5.0 74/79] perf/core: Fix perf_event_disable_inatomic() race Sasha Levin
2019-04-27  1:38 ` [PATCH AUTOSEL 5.0 75/79] iommu/amd: Set exclusion range correctly Sasha Levin
2019-04-27  1:38 ` [PATCH AUTOSEL 5.0 76/79] mm: make page ref count overflow check tighter and more explicit Sasha Levin
2019-04-27  1:38 ` [PATCH AUTOSEL 5.0 77/79] mm: add 'try_get_page()' helper function Sasha Levin
2019-04-27  1:38 ` [PATCH AUTOSEL 5.0 78/79] mm: prevent get_user_pages() from overflowing page refcount Sasha Levin
2019-04-27  1:38 ` [PATCH AUTOSEL 5.0 79/79] fs: prevent page refcount overflow in pipe_buf_get Sasha Levin

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=20190427013838.6596-26-sashal@kernel.org \
    --to=sashal@kernel.org \
    --cc=dennis.dalessandro@intel.com \
    --cc=jgg@mellanox.com \
    --cc=kaike.wan@intel.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-rdma@vger.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