| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Date: Tue, 27 Jun 2023 01:42:05 -0700
From: Souradeep Chakrabarti <schakrabarti@...ux.microsoft.com>
To: Praveen Kumar <kumarpraveen@...ux.microsoft.com>
Cc: kys@...rosoft.com, haiyangz@...rosoft.com, wei.liu@...nel.org,
decui@...rosoft.com, davem@...emloft.net, edumazet@...gle.com,
kuba@...nel.org, pabeni@...hat.com, longli@...rosoft.com,
sharmaajay@...rosoft.com, leon@...nel.org, cai.huoqing@...ux.dev,
ssengar@...ux.microsoft.com, vkuznets@...hat.com,
tglx@...utronix.de, linux-hyperv@...r.kernel.org,
netdev@...r.kernel.org, linux-kernel@...r.kernel.org,
linux-rdma@...r.kernel.org, stable@...r.kernel.org,
schakrabarti@...rosoft.com
Subject: Re: [PATCH 2/2 V3 net] net: mana: Fix MANA VF unload when host is
unresponsive
On Mon, Jun 26, 2023 at 07:43:07PM +0530, Praveen Kumar wrote:
> On 6/26/2023 2:50 PM, souradeep chakrabarti wrote:
> > From: Souradeep Chakrabarti <schakrabarti@...ux.microsoft.com>
> >
> > This is the second part of the fix.
> >
> > Also this patch adds a new attribute in mana_context, which gets set when
> > mana_hwc_send_request() hits a timeout because of host unresponsiveness.
> > This flag then helps to avoid the timeouts in successive calls.
> >
> > Fixes: ca9c54d2d6a5ab2430c4eda364c77125d62e5e0f (net: mana: Add a driver for
> > Microsoft Azure Network Adapter)
> > Signed-off-by: Souradeep Chakrabarti <schakrabarti@...ux.microsoft.com>
> > ---
> > V2 -> V3:
> > * Removed the initialization of vf_unload_timeout
> > * Splitted the patch in two.
> > * Fixed extra space from the commit message.
> > ---
> > drivers/net/ethernet/microsoft/mana/gdma_main.c | 4 +++-
> > drivers/net/ethernet/microsoft/mana/hw_channel.c | 12 +++++++++++-
> > include/net/mana/mana.h | 2 ++
> > 3 files changed, 16 insertions(+), 2 deletions(-)
> >
> > diff --git a/drivers/net/ethernet/microsoft/mana/gdma_main.c b/drivers/net/ethernet/microsoft/mana/gdma_main.c
> > index 8f3f78b68592..6411f01be0d9 100644
> > --- a/drivers/net/ethernet/microsoft/mana/gdma_main.c
> > +++ b/drivers/net/ethernet/microsoft/mana/gdma_main.c
> > @@ -946,10 +946,12 @@ int mana_gd_deregister_device(struct gdma_dev *gd)
> > struct gdma_context *gc = gd->gdma_context;
> > struct gdma_general_resp resp = {};
> > struct gdma_general_req req = {};
> > + struct mana_context *ac;
> > int err;
> >
> > if (gd->pdid == INVALID_PDID)
> > return -EINVAL;
> > + ac = gd->driver_data;
> >
> > mana_gd_init_req_hdr(&req.hdr, GDMA_DEREGISTER_DEVICE, sizeof(req),
> > sizeof(resp));
> > @@ -957,7 +959,7 @@ int mana_gd_deregister_device(struct gdma_dev *gd)
> > req.hdr.dev_id = gd->dev_id;
> >
> > err = mana_gd_send_request(gc, sizeof(req), &req, sizeof(resp), &resp);
> > - if (err || resp.hdr.status) {
> > + if ((err || resp.hdr.status) && !ac->vf_unload_timeout) {
> > dev_err(gc->dev, "Failed to deregister device: %d, 0x%x\n",
> > err, resp.hdr.status);
>
> With !ac->vf_unload_timeout option, this message may not be correctly showing err, status. Probably you want to add explicit information during timeouts so that it give right information ? Or have the err, status field properly updated.
This check !ac->vf_unload_timeout here means if ac->vf_unload_timeout is not yet set,
then only consider the err path, else continue the remaining operation.
>
> > if (!err)
> > diff --git a/drivers/net/ethernet/microsoft/mana/hw_channel.c b/drivers/net/ethernet/microsoft/mana/hw_channel.c
> > index 9d1507eba5b9..492cb2c6e2cb 100644
> > --- a/drivers/net/ethernet/microsoft/mana/hw_channel.c
> > +++ b/drivers/net/ethernet/microsoft/mana/hw_channel.c
> > @@ -1,8 +1,10 @@
> > // SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
> > /* Copyright (c) 2021, Microsoft Corporation. */
> >
> > +#include "asm-generic/errno.h"
> > #include <net/mana/gdma.h>
> > #include <net/mana/hw_channel.h>
> > +#include <net/mana/mana.h>
> >
> > static int mana_hwc_get_msg_index(struct hw_channel_context *hwc, u16 *msg_id)
> > {
> > @@ -786,12 +788,19 @@ int mana_hwc_send_request(struct hw_channel_context *hwc, u32 req_len,
> > struct hwc_wq *txq = hwc->txq;
> > struct gdma_req_hdr *req_msg;
> > struct hwc_caller_ctx *ctx;
> > + struct mana_context *ac;
> > u32 dest_vrcq = 0;
> > u32 dest_vrq = 0;
> > u16 msg_id;
> > int err;
> >
> > mana_hwc_get_msg_index(hwc, &msg_id);
> > + ac = hwc->gdma_dev->driver_data;
>
> Is there a case where gdma_dev be invalid here ? If so, lets check the state and then proceed further ?
I can see Haiyang has already in his comment, responded on the same.
hwc->gdma_dev will be valid here, but as Haiyang pointed we need to use
hwc->gdma_dev->gdma_context->mana.driver_data, or better to relocate the
attribute in gdma_context.
>
> > + if (ac->vf_unload_timeout) {
> > + dev_err(hwc->dev, "HWC: vport is already unloaded.\n");
> > + err = -ETIMEDOUT;
> > + goto out;
> > + }
> >
> > tx_wr = &txq->msg_buf->reqs[msg_id];
> >
> > @@ -825,9 +834,10 @@ int mana_hwc_send_request(struct hw_channel_context *hwc, u32 req_len,
> > goto out;
> > }
> >
> > - if (!wait_for_completion_timeout(&ctx->comp_event, 30 * HZ)) {
> > + if (!wait_for_completion_timeout(&ctx->comp_event, 5 * HZ)) {
>
> IMHO we should have macros instead of magic numbers (5 , 30 or so). But would like others to comment here.
>
> > dev_err(hwc->dev, "HWC: Request timed out!\n");
> > err = -ETIMEDOUT;
> > + ac->vf_unload_timeout = true;
> > goto out;
> > }
> >
> > diff --git a/include/net/mana/mana.h b/include/net/mana/mana.h
> > index 9eef19972845..5f5affdca1eb 100644
> > --- a/include/net/mana/mana.h
> > +++ b/include/net/mana/mana.h
> > @@ -358,6 +358,8 @@ struct mana_context {
> >
> > u16 num_ports;
> >
> > + bool vf_unload_timeout;
> > +
> > struct mana_eq *eqs;
> >
> > struct net_device *ports[MAX_PORTS_IN_MANA_DEV];
Powered by blists - more mailing lists