* RE: [PATCH v3 02/12] x86/hyper-v: Implement hv_is_hibernation_supported()
From: Dexuan Cui @ 2019-08-30 23:37 UTC (permalink / raw)
To: Michael Kelley, linux-hyperv@vger.kernel.org,
gregkh@linuxfoundation.org, Stephen Hemminger, Sasha Levin,
sashal@kernel.org, Haiyang Zhang, KY Srinivasan,
tglx@linutronix.de
Cc: linux-kernel@vger.kernel.org
In-Reply-To: <DM5PR21MB0137A7EC5E51EAF8A0667359D7A40@DM5PR21MB0137.namprd21.prod.outlook.com>
> From: Michael Kelley <mikelley@microsoft.com>
> Sent: Friday, August 23, 2019 12:50 PM
>
> From: Dexuan Cui <decui@microsoft.com> Sent: August 19, 2019 6:52 PM
> >
> > When a Linux VM runs on Hyper-V and hibernates, it must disable the
> > memory hot-add/remove and balloon up/down capabilities in the hv_balloon
> > driver.
>
> I'm unclear on the above statement. I think the requirement is that
> ballooning must not be active when hibernation is initiated. Is hibernation
> blocked in that case? If not, what happens?
Ballooning (and hot-addition of memory) must not be active if the user
wants the Linux VM to support hibernation, not just when hibernation is
initiated. This is because ballooning and hot-addition of memory are
incompatible with hibernation according to Hyper-V team, e.g. upon
hibernation the balloon VSP doesn't save any info about ballooned-out
pages (if any), so after Linux resumes, Linux balloon VSC expects that the
VSP will return the pages if Linux is under memory pressure, but the VSP
will never return the pages, since the VSP thinks it never stole the
pages from the VM.
So, if the user wants Linux VM to support hibernation, Linux must completely
forbid ballooning and hot-addition of memory, and hence the only
functionality of balloon VSC is reporting the memory pressure to the host.
Ideally, when Linux detects that the user wants it to support hibernation,
the balloon VSC should tell the VSP that it does not support ballooning and
hot-addition; however, the current version of the VSP requires
the VSC should support these capabilities, otherwise the capability negotiation
fails and the VSC can not load at all, so in my changes to the VSC driver, I
still report to the VSP that Linux supports the capabilities, but the VSC
ignores the host's requests of balloon up/down and hot add, and returns an
error to the VSP, when applicable.
BTW, in the future the balloon VSP driver will allow the VSC to not support
the capabilities of balloon up/down and hot add.
The remaining problem is: how can Linux know the user wants Linux VM
to support hibernation?
The ACPI S4 state is not a must for hibernation to work, because Linux is
able to hibernate as long as the system can shut down.
My decision is that: we artificially use the presence of the virtual
ACPI S4 state as the indicator of the user's intent of using hibernation.
BTW, I believe the Windows team made the same decision.
Once all the vmbus and VSC patches are accepted, I'll submit a patch to
forbid hibernation if the virtual ACPI S4 state is absent, e.g.
hv_is_hibernation_supported() is false.
> > By default, Hyper-V does not enable the virtual ACPI S4 state for a VM;
> > on recent Hyper-V hosts, the administrator is able to enable the virtual
> > ACPI S4 state for a VM, so we hope to use the presence of the virtual ACPI
>
> "we hope" sounds very indefinite. :-( Does ACPI S4 have to be enabled for
> hibernation to be initiated? Goes back to my first question ....
Technically, we don't have to enable ACPI S4, but in practice, as I mentioned,
I'll submit a patch to forbid hibernation if ACPI S4 is absent.
>
> > S4 state as a hint for hv_balloon to disable the aforementioned
> > capabilities.
> >
> > The new API will be used by hv_balloon.
I'll add my above explanation into the changelog in v4.
Thanks,
-- Dexuan
^ permalink raw reply
* RE: [PATCH net-next, 2/2] hv_netvsc: Sync offloading features to VF NIC
From: Haiyang Zhang @ 2019-08-30 23:12 UTC (permalink / raw)
To: Jakub Kicinski
Cc: sashal@kernel.org, linux-hyperv@vger.kernel.org,
netdev@vger.kernel.org, KY Srinivasan, Stephen Hemminger,
olaf@aepfle.de, vkuznets, davem@davemloft.net,
linux-kernel@vger.kernel.org, Mark Bloch
In-Reply-To: <20190830160451.43a61cf9@cakuba.netronome.com>
> -----Original Message-----
> From: Jakub Kicinski <jakub.kicinski@netronome.com>
> Sent: Friday, August 30, 2019 4:05 PM
> To: Haiyang Zhang <haiyangz@microsoft.com>
> Cc: sashal@kernel.org; linux-hyperv@vger.kernel.org;
> netdev@vger.kernel.org; KY Srinivasan <kys@microsoft.com>; Stephen
> Hemminger <sthemmin@microsoft.com>; olaf@aepfle.de; vkuznets
> <vkuznets@redhat.com>; davem@davemloft.net; linux-
> kernel@vger.kernel.org; Mark Bloch <markb@mellanox.com>
> Subject: Re: [PATCH net-next, 2/2] hv_netvsc: Sync offloading features to VF
> NIC
>
> On Fri, 30 Aug 2019 03:45:38 +0000, Haiyang Zhang wrote:
> > VF NIC may go down then come up during host servicing events. This
> > causes the VF NIC offloading feature settings to roll back to the
> > defaults. This patch can synchronize features from synthetic NIC to
> > the VF NIC during ndo_set_features (ethtool -K), and
> > netvsc_register_vf when VF comes back after host events.
> >
> > Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
> > Cc: Mark Bloch <markb@mellanox.com>
>
> If we want to make this change in behaviour we should change net_failover
> at the same time.
I will check net_failover. Thanks.
^ permalink raw reply
* Re: [PATCH net-next, 1/2] hv_netvsc: Allow scatter-gather feature to be tunable
From: Jakub Kicinski @ 2019-08-30 23:05 UTC (permalink / raw)
To: Haiyang Zhang
Cc: sashal@kernel.org, linux-hyperv@vger.kernel.org,
netdev@vger.kernel.org, KY Srinivasan, Stephen Hemminger,
olaf@aepfle.de, vkuznets, davem@davemloft.net,
linux-kernel@vger.kernel.org
In-Reply-To: <1567136656-49288-2-git-send-email-haiyangz@microsoft.com>
On Fri, 30 Aug 2019 03:45:24 +0000, Haiyang Zhang wrote:
> In a previous patch, the NETIF_F_SG was missing after the code changes.
> That caused the SG feature to be "fixed". This patch includes it into
> hw_features, so it is tunable again.
>
> Fixes: 23312a3be999 ("netvsc: negotiate checksum and segmentation parameters")
^
Looks like a tab sneaked in there.
> Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
^ permalink raw reply
* Re: [PATCH net-next, 2/2] hv_netvsc: Sync offloading features to VF NIC
From: Jakub Kicinski @ 2019-08-30 23:04 UTC (permalink / raw)
To: Haiyang Zhang
Cc: sashal@kernel.org, linux-hyperv@vger.kernel.org,
netdev@vger.kernel.org, KY Srinivasan, Stephen Hemminger,
olaf@aepfle.de, vkuznets, davem@davemloft.net,
linux-kernel@vger.kernel.org, Mark Bloch
In-Reply-To: <1567136656-49288-3-git-send-email-haiyangz@microsoft.com>
On Fri, 30 Aug 2019 03:45:38 +0000, Haiyang Zhang wrote:
> VF NIC may go down then come up during host servicing events. This
> causes the VF NIC offloading feature settings to roll back to the
> defaults. This patch can synchronize features from synthetic NIC to
> the VF NIC during ndo_set_features (ethtool -K),
> and netvsc_register_vf when VF comes back after host events.
>
> Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
> Cc: Mark Bloch <markb@mellanox.com>
If we want to make this change in behaviour we should change
net_failover at the same time.
^ permalink raw reply
* RE: [PATCH] x86/Hyper-V: Fix overflow issue in the fill_gva_list()
From: Michael Kelley @ 2019-08-30 17:41 UTC (permalink / raw)
To: lantianyu1986@gmail.com, KY Srinivasan, Haiyang Zhang,
Stephen Hemminger, sashal@kernel.org, tglx@linutronix.de,
mingo@redhat.com, bp@alien8.de, hpa@zytor.com, x86@kernel.org,
gregkh@linuxfoundation.org, alex.williamson@redhat.com,
cohuck@redhat.com
Cc: Tianyu Lan, linux-hyperv@vger.kernel.org,
linux-kernel@vger.kernel.org, kvm@vger.kernel.org
In-Reply-To: <20190830061540.211072-1-Tianyu.Lan@microsoft.com>
From: lantianyu1986@gmail.com Sent: Thursday, August 29, 2019 11:16 PM
>
> From: Tianyu Lan <Tianyu.Lan@microsoft.com>
>
> fill_gva_list() populates gva list and adds offset
> HV_TLB_FLUSH_UNIT(0x1000000) to variable "cur"
> in the each loop. When diff between "end" and "cur" is
> less than HV_TLB_FLUSH_UNIT, the gva entry should
> be the last one and the loop should be end.
>
> If cur is equal or greater than 0xFF000000 on 32-bit
> mode, "cur" will be overflow after adding HV_TLB_FLUSH_UNIT.
> Its value will be wrapped and less than "end". fill_gva_list()
> falls into an infinite loop and fill gva list out of
> border finally.
>
> Set "cur" to be "end" to make loop end when diff is
> less than HV_TLB_FLUSH_UNIT and add HV_TLB_FLUSH_UNIT to
> "cur" when diff is equal or greater than HV_TLB_FLUSH_UNIT.
> Fix the overflow issue.
Let me suggest simplifying the commit message a bit. It
doesn't need to describe every line of the code change. I think
it should also make clear that the same problem could occur on
64-bit systems with the right "start" address. My suggestion:
When the 'start' parameter is >= 0xFF000000 on 32-bit
systems, or >= 0xFFFFFFFF'FF000000 on 64-bit systems,
fill_gva_list gets into an infinite loop. With such inputs,
'cur' overflows after adding HV_TLB_FLUSH_UNIT and always
compares as less than end. Memory is filled with guest virtual
addresses until the system crashes
Fix this by never incrementing 'cur' to be larger than 'end'.
>
> Reported-by: Jong Hyun Park <park.jonghyun@yonsei.ac.kr>
> Signed-off-by: Tianyu Lan <Tianyu.Lan@microsoft.com>
> Fixes: 2ffd9e33ce4a ("x86/hyper-v: Use hypercall for remote
> TLB flush")
The "Fixes:" line needs to not wrap. It's exempt from the
"wrap at 75 columns" rule in order to simplify parsing scripts.
The code itself looks good.
Michael
^ permalink raw reply
* [PATCH] x86/Hyper-V: Fix overflow issue in the fill_gva_list()
From: lantianyu1986 @ 2019-08-30 6:15 UTC (permalink / raw)
To: kys, haiyangz, sthemmin, sashal, tglx, mingo, bp, hpa, x86,
gregkh, alex.williamson, cohuck, michael.h.kelley
Cc: Tianyu Lan, linux-hyperv, linux-kernel, kvm
From: Tianyu Lan <Tianyu.Lan@microsoft.com>
fill_gva_list() populates gva list and adds offset
HV_TLB_FLUSH_UNIT(0x1000000) to variable "cur"
in the each loop. When diff between "end" and "cur" is
less than HV_TLB_FLUSH_UNIT, the gva entry should
be the last one and the loop should be end.
If cur is equal or greater than 0xFF000000 on 32-bit
mode, "cur" will be overflow after adding HV_TLB_FLUSH_UNIT.
Its value will be wrapped and less than "end". fill_gva_list()
falls into an infinite loop and fill gva list out of
border finally.
Set "cur" to be "end" to make loop end when diff is
less than HV_TLB_FLUSH_UNIT and add HV_TLB_FLUSH_UNIT to
"cur" when diff is equal or greater than HV_TLB_FLUSH_UNIT.
Fix the overflow issue.
Reported-by: Jong Hyun Park <park.jonghyun@yonsei.ac.kr>
Signed-off-by: Tianyu Lan <Tianyu.Lan@microsoft.com>
Fixes: 2ffd9e33ce4a ("x86/hyper-v: Use hypercall for remote
TLB flush")
---
arch/x86/hyperv/mmu.c | 8 +++++---
1 file changed, 5 insertions(+), 3 deletions(-)
diff --git a/arch/x86/hyperv/mmu.c b/arch/x86/hyperv/mmu.c
index e65d7fe6489f..5208ba49c89a 100644
--- a/arch/x86/hyperv/mmu.c
+++ b/arch/x86/hyperv/mmu.c
@@ -37,12 +37,14 @@ static inline int fill_gva_list(u64 gva_list[], int offset,
* Lower 12 bits encode the number of additional
* pages to flush (in addition to the 'cur' page).
*/
- if (diff >= HV_TLB_FLUSH_UNIT)
+ if (diff >= HV_TLB_FLUSH_UNIT) {
gva_list[gva_n] |= ~PAGE_MASK;
- else if (diff)
+ cur += HV_TLB_FLUSH_UNIT;
+ } else if (diff) {
gva_list[gva_n] |= (diff - 1) >> PAGE_SHIFT;
+ cur = end;
+ }
- cur += HV_TLB_FLUSH_UNIT;
gva_n++;
} while (cur < end);
--
2.14.5
^ permalink raw reply related
* [PATCH net-next, 2/2] hv_netvsc: Sync offloading features to VF NIC
From: Haiyang Zhang @ 2019-08-30 3:45 UTC (permalink / raw)
To: sashal@kernel.org, linux-hyperv@vger.kernel.org,
netdev@vger.kernel.org
Cc: Haiyang Zhang, KY Srinivasan, Stephen Hemminger, olaf@aepfle.de,
vkuznets, davem@davemloft.net, linux-kernel@vger.kernel.org,
Mark Bloch
In-Reply-To: <1567136656-49288-1-git-send-email-haiyangz@microsoft.com>
VF NIC may go down then come up during host servicing events. This
causes the VF NIC offloading feature settings to roll back to the
defaults. This patch can synchronize features from synthetic NIC to
the VF NIC during ndo_set_features (ethtool -K),
and netvsc_register_vf when VF comes back after host events.
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Cc: Mark Bloch <markb@mellanox.com>
---
drivers/net/hyperv/netvsc_drv.c | 22 ++++++++++++++++++++--
1 file changed, 20 insertions(+), 2 deletions(-)
diff --git a/drivers/net/hyperv/netvsc_drv.c b/drivers/net/hyperv/netvsc_drv.c
index 1f1192e..39dddcd 100644
--- a/drivers/net/hyperv/netvsc_drv.c
+++ b/drivers/net/hyperv/netvsc_drv.c
@@ -1785,13 +1785,15 @@ static int netvsc_set_features(struct net_device *ndev,
netdev_features_t change = features ^ ndev->features;
struct net_device_context *ndevctx = netdev_priv(ndev);
struct netvsc_device *nvdev = rtnl_dereference(ndevctx->nvdev);
+ struct net_device *vf_netdev = rtnl_dereference(ndevctx->vf_netdev);
struct ndis_offload_params offloads;
+ int ret = 0;
if (!nvdev || nvdev->destroy)
return -ENODEV;
if (!(change & NETIF_F_LRO))
- return 0;
+ goto syncvf;
memset(&offloads, 0, sizeof(struct ndis_offload_params));
@@ -1803,7 +1805,19 @@ static int netvsc_set_features(struct net_device *ndev,
offloads.rsc_ip_v6 = NDIS_OFFLOAD_PARAMETERS_RSC_DISABLED;
}
- return rndis_filter_set_offload_params(ndev, nvdev, &offloads);
+ ret = rndis_filter_set_offload_params(ndev, nvdev, &offloads);
+
+ if (ret)
+ features ^= NETIF_F_LRO;
+
+syncvf:
+ if (!vf_netdev)
+ return ret;
+
+ vf_netdev->wanted_features = features;
+ netdev_update_features(vf_netdev);
+
+ return ret;
}
static u32 netvsc_get_msglevel(struct net_device *ndev)
@@ -2181,6 +2195,10 @@ static int netvsc_register_vf(struct net_device *vf_netdev)
dev_hold(vf_netdev);
rcu_assign_pointer(net_device_ctx->vf_netdev, vf_netdev);
+
+ vf_netdev->wanted_features = ndev->features;
+ netdev_update_features(vf_netdev);
+
return NOTIFY_OK;
}
--
1.8.3.1
^ permalink raw reply related
* [PATCH net-next, 1/2] hv_netvsc: Allow scatter-gather feature to be tunable
From: Haiyang Zhang @ 2019-08-30 3:45 UTC (permalink / raw)
To: sashal@kernel.org, linux-hyperv@vger.kernel.org,
netdev@vger.kernel.org
Cc: Haiyang Zhang, KY Srinivasan, Stephen Hemminger, olaf@aepfle.de,
vkuznets, davem@davemloft.net, linux-kernel@vger.kernel.org
In-Reply-To: <1567136656-49288-1-git-send-email-haiyangz@microsoft.com>
In a previous patch, the NETIF_F_SG was missing after the code changes.
That caused the SG feature to be "fixed". This patch includes it into
hw_features, so it is tunable again.
Fixes: 23312a3be999 ("netvsc: negotiate checksum and segmentation parameters")
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
---
drivers/net/hyperv/hyperv_net.h | 2 +-
drivers/net/hyperv/netvsc_drv.c | 4 ++--
drivers/net/hyperv/rndis_filter.c | 1 +
3 files changed, 4 insertions(+), 3 deletions(-)
diff --git a/drivers/net/hyperv/hyperv_net.h b/drivers/net/hyperv/hyperv_net.h
index ecc9af0..670ef68 100644
--- a/drivers/net/hyperv/hyperv_net.h
+++ b/drivers/net/hyperv/hyperv_net.h
@@ -822,7 +822,7 @@ struct nvsp_message {
#define NETVSC_SUPPORTED_HW_FEATURES (NETIF_F_RXCSUM | NETIF_F_IP_CSUM | \
NETIF_F_TSO | NETIF_F_IPV6_CSUM | \
- NETIF_F_TSO6 | NETIF_F_LRO)
+ NETIF_F_TSO6 | NETIF_F_LRO | NETIF_F_SG)
#define VRSS_SEND_TAB_SIZE 16 /* must be power of 2 */
#define VRSS_CHANNEL_MAX 64
diff --git a/drivers/net/hyperv/netvsc_drv.c b/drivers/net/hyperv/netvsc_drv.c
index 0a6cd2f..1f1192e 100644
--- a/drivers/net/hyperv/netvsc_drv.c
+++ b/drivers/net/hyperv/netvsc_drv.c
@@ -2313,8 +2313,8 @@ static int netvsc_probe(struct hv_device *dev,
/* hw_features computed in rndis_netdev_set_hwcaps() */
net->features = net->hw_features |
- NETIF_F_HIGHDMA | NETIF_F_SG |
- NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_CTAG_RX;
+ NETIF_F_HIGHDMA | NETIF_F_HW_VLAN_CTAG_TX |
+ NETIF_F_HW_VLAN_CTAG_RX;
net->vlan_features = net->features;
netdev_lockdep_set_classes(net);
diff --git a/drivers/net/hyperv/rndis_filter.c b/drivers/net/hyperv/rndis_filter.c
index 317dbe9..abaf815 100644
--- a/drivers/net/hyperv/rndis_filter.c
+++ b/drivers/net/hyperv/rndis_filter.c
@@ -1207,6 +1207,7 @@ static int rndis_netdev_set_hwcaps(struct rndis_device *rndis_device,
/* Compute tx offload settings based on hw capabilities */
net->hw_features |= NETIF_F_RXCSUM;
+ net->hw_features |= NETIF_F_SG;
if ((hwcaps.csum.ip4_txcsum & NDIS_TXCSUM_ALL_TCP4) == NDIS_TXCSUM_ALL_TCP4) {
/* Can checksum TCP */
--
1.8.3.1
^ permalink raw reply related
* [PATCH net-next, 0/2] Enable sg as tunable, sync offload settings to VF NIC
From: Haiyang Zhang @ 2019-08-30 3:45 UTC (permalink / raw)
To: sashal@kernel.org, linux-hyperv@vger.kernel.org,
netdev@vger.kernel.org
Cc: Haiyang Zhang, KY Srinivasan, Stephen Hemminger, olaf@aepfle.de,
vkuznets, davem@davemloft.net, linux-kernel@vger.kernel.org
This patch set fixes an issue in SG tuning, and sync
offload settings from synthetic NIC to VF NIC.
Haiyang Zhang (2):
hv_netvsc: hv_netvsc: Allow scatter-gather feature to be tunable
hv_netvsc: Sync offloading features to VF NIC
drivers/net/hyperv/hyperv_net.h | 2 +-
drivers/net/hyperv/netvsc_drv.c | 26 ++++++++++++++++++++++----
drivers/net/hyperv/rndis_filter.c | 1 +
3 files changed, 24 insertions(+), 5 deletions(-)
--
1.8.3.1
^ permalink raw reply
* Re: [PATCH v4 1/2] drivers: hv: vmbus: Introduce latency testing
From: kbuild test robot @ 2019-08-30 3:23 UTC (permalink / raw)
To: Branden Bonaby
Cc: kbuild-all, kys, haiyangz, sthemmin, sashal, Branden Bonaby,
linux-hyperv, linux-kernel
In-Reply-To: <da1ab5c98ce902ec181a799d8d1f040e8cc39af6.1567047549.git.brandonbonaby94@gmail.com>
[-- Attachment #1: Type: text/plain, Size: 1179 bytes --]
Hi Branden,
Thank you for the patch! Yet something to improve:
[auto build test ERROR on linus/master]
[cannot apply to v5.3-rc6 next-20190829]
[if your patch is applied to the wrong git tree, please drop us a note to help improve the system]
url: https://github.com/0day-ci/linux/commits/Branden-Bonaby/drivers-hv-vmbus-Introduce-latency-testing/20190830-032123
config: i386-allmodconfig (attached as .config)
compiler: gcc-7 (Debian 7.4.0-11) 7.4.0
reproduce:
# save the attached .config to linux build tree
make ARCH=i386
If you fix the issue, kindly add following tag
Reported-by: kbuild test robot <lkp@intel.com>
All errors (new ones prefixed by >>):
>> ERROR: "hv_debug_add_dev_dir" [drivers/hv/hv_vmbus.ko] undefined!
>> ERROR: "hv_debug_init" [drivers/hv/hv_vmbus.ko] undefined!
>> ERROR: "hv_debug_delay_test" [drivers/hv/hv_vmbus.ko] undefined!
>> ERROR: "hv_debug_rm_all_dir" [drivers/hv/hv_vmbus.ko] undefined!
>> ERROR: "hv_debug_rm_dev_dir" [drivers/hv/hv_vmbus.ko] undefined!
---
0-DAY kernel test infrastructure Open Source Technology Center
https://lists.01.org/pipermail/kbuild-all Intel Corporation
[-- Attachment #2: .config.gz --]
[-- Type: application/gzip, Size: 69479 bytes --]
^ permalink raw reply
* Re: [PATCH v3 1/3] drivers: hv: vmbus: Introduce latency testing
From: Stephen Hemminger @ 2019-08-29 21:57 UTC (permalink / raw)
To: Branden Bonaby
Cc: KY Srinivasan, Haiyang Zhang, Stephen Hemminger, sashal,
linux-hyperv
In-Reply-To: <ebca54bf70d2af53de419c1b7ac8db5b77b888cb.1566340843.git.brandonbonaby94@gmail.com>
On Tue, 20 Aug 2019 16:39:02 -0700
"Branden Bonaby" <brandonbonaby94@gmail.com> wrote:
> diff --git a/drivers/hv/Kconfig b/drivers/hv/Kconfig
> index 9a59957922d4..d97437ba0626 100644
> --- a/drivers/hv/Kconfig
> +++ b/drivers/hv/Kconfig
> @@ -29,4 +29,11 @@ config HYPERV_BALLOON
> help
> Select this option to enable Hyper-V Balloon driver.
>
> +config HYPERV_TESTING
> + bool "Hyper-V testing"
> + default n
> + depends on HYPERV && DEBUG_FS
> + help
> + Select this option to enable Hyper-V vmbus testing.
> +
> endmenu
Maybe this should go under the Kernel hacking
section in lib/Kconfig.debug?
^ permalink raw reply
* [PATCH AUTOSEL 5.2 54/76] Tools: hv: kvp: eliminate 'may be used uninitialized' warning
From: Sasha Levin @ 2019-08-29 18:12 UTC (permalink / raw)
To: linux-kernel, stable; +Cc: Vitaly Kuznetsov, Sasha Levin, linux-hyperv
In-Reply-To: <20190829181311.7562-1-sashal@kernel.org>
From: Vitaly Kuznetsov <vkuznets@redhat.com>
[ Upstream commit 89eb4d8d25722a0a0194cf7fa47ba602e32a6da7 ]
When building hv_kvp_daemon GCC-8.3 complains:
hv_kvp_daemon.c: In function ‘kvp_get_ip_info.constprop’:
hv_kvp_daemon.c:812:30: warning: ‘ip_buffer’ may be used uninitialized in this function [-Wmaybe-uninitialized]
struct hv_kvp_ipaddr_value *ip_buffer;
this seems to be a false positive: we only use ip_buffer when
op == KVP_OP_GET_IP_INFO and it is only unset when op == KVP_OP_ENUMERATE.
Silence the warning by initializing ip_buffer to NULL.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
tools/hv/hv_kvp_daemon.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools/hv/hv_kvp_daemon.c b/tools/hv/hv_kvp_daemon.c
index d7e06fe0270ee..6d809abaf338a 100644
--- a/tools/hv/hv_kvp_daemon.c
+++ b/tools/hv/hv_kvp_daemon.c
@@ -809,7 +809,7 @@ kvp_get_ip_info(int family, char *if_name, int op,
int sn_offset = 0;
int error = 0;
char *buffer;
- struct hv_kvp_ipaddr_value *ip_buffer;
+ struct hv_kvp_ipaddr_value *ip_buffer = NULL;
char cidr_mask[5]; /* /xyz */
int weight;
int i;
--
2.20.1
^ permalink raw reply related
* [PATCH AUTOSEL 4.19 33/45] Tools: hv: kvp: eliminate 'may be used uninitialized' warning
From: Sasha Levin @ 2019-08-29 18:15 UTC (permalink / raw)
To: linux-kernel, stable; +Cc: Vitaly Kuznetsov, Sasha Levin, linux-hyperv
In-Reply-To: <20190829181547.8280-1-sashal@kernel.org>
From: Vitaly Kuznetsov <vkuznets@redhat.com>
[ Upstream commit 89eb4d8d25722a0a0194cf7fa47ba602e32a6da7 ]
When building hv_kvp_daemon GCC-8.3 complains:
hv_kvp_daemon.c: In function ‘kvp_get_ip_info.constprop’:
hv_kvp_daemon.c:812:30: warning: ‘ip_buffer’ may be used uninitialized in this function [-Wmaybe-uninitialized]
struct hv_kvp_ipaddr_value *ip_buffer;
this seems to be a false positive: we only use ip_buffer when
op == KVP_OP_GET_IP_INFO and it is only unset when op == KVP_OP_ENUMERATE.
Silence the warning by initializing ip_buffer to NULL.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
tools/hv/hv_kvp_daemon.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools/hv/hv_kvp_daemon.c b/tools/hv/hv_kvp_daemon.c
index d7e06fe0270ee..6d809abaf338a 100644
--- a/tools/hv/hv_kvp_daemon.c
+++ b/tools/hv/hv_kvp_daemon.c
@@ -809,7 +809,7 @@ kvp_get_ip_info(int family, char *if_name, int op,
int sn_offset = 0;
int error = 0;
char *buffer;
- struct hv_kvp_ipaddr_value *ip_buffer;
+ struct hv_kvp_ipaddr_value *ip_buffer = NULL;
char cidr_mask[5]; /* /xyz */
int weight;
int i;
--
2.20.1
^ permalink raw reply related
* [PATCH AUTOSEL 4.19 32/45] Input: hyperv-keyboard: Use in-place iterator API in the channel callback
From: Sasha Levin @ 2019-08-29 18:15 UTC (permalink / raw)
To: linux-kernel, stable
Cc: Dexuan Cui, Dmitry Torokhov, Sasha Levin, linux-hyperv,
linux-input
In-Reply-To: <20190829181547.8280-1-sashal@kernel.org>
From: Dexuan Cui <decui@microsoft.com>
[ Upstream commit d09bc83640d524b8467a660db7b1d15e6562a1de ]
Simplify the ring buffer handling with the in-place API.
Also avoid the dynamic allocation and the memory leak in the channel
callback function.
Signed-off-by: Dexuan Cui <decui@microsoft.com>
Acked-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
drivers/input/serio/hyperv-keyboard.c | 35 +++++----------------------
1 file changed, 6 insertions(+), 29 deletions(-)
diff --git a/drivers/input/serio/hyperv-keyboard.c b/drivers/input/serio/hyperv-keyboard.c
index a8b9be3e28db7..7d0a5ccf57751 100644
--- a/drivers/input/serio/hyperv-keyboard.c
+++ b/drivers/input/serio/hyperv-keyboard.c
@@ -245,40 +245,17 @@ static void hv_kbd_handle_received_packet(struct hv_device *hv_dev,
static void hv_kbd_on_channel_callback(void *context)
{
+ struct vmpacket_descriptor *desc;
struct hv_device *hv_dev = context;
- void *buffer;
- int bufferlen = 0x100; /* Start with sensible size */
u32 bytes_recvd;
u64 req_id;
- int error;
- buffer = kmalloc(bufferlen, GFP_ATOMIC);
- if (!buffer)
- return;
-
- while (1) {
- error = vmbus_recvpacket_raw(hv_dev->channel, buffer, bufferlen,
- &bytes_recvd, &req_id);
- switch (error) {
- case 0:
- if (bytes_recvd == 0) {
- kfree(buffer);
- return;
- }
-
- hv_kbd_handle_received_packet(hv_dev, buffer,
- bytes_recvd, req_id);
- break;
+ foreach_vmbus_pkt(desc, hv_dev->channel) {
+ bytes_recvd = desc->len8 * 8;
+ req_id = desc->trans_id;
- case -ENOBUFS:
- kfree(buffer);
- /* Handle large packet */
- bufferlen = bytes_recvd;
- buffer = kmalloc(bytes_recvd, GFP_ATOMIC);
- if (!buffer)
- return;
- break;
- }
+ hv_kbd_handle_received_packet(hv_dev, desc, bytes_recvd,
+ req_id);
}
}
--
2.20.1
^ permalink raw reply related
* [PATCH AUTOSEL 4.14 02/27] hv_netvsc: Fix a warning of suspicious RCU usage
From: Sasha Levin @ 2019-08-29 18:16 UTC (permalink / raw)
To: linux-kernel, stable
Cc: Dexuan Cui, David S . Miller, Sasha Levin, linux-hyperv, netdev
In-Reply-To: <20190829181655.8741-1-sashal@kernel.org>
From: Dexuan Cui <decui@microsoft.com>
[ Upstream commit 6d0d779dca73cd5acb649c54f81401f93098b298 ]
This fixes a warning of "suspicious rcu_dereference_check() usage"
when nload runs.
Fixes: 776e726bfb34 ("netvsc: fix RCU warning in get_stats")
Signed-off-by: Dexuan Cui <decui@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
drivers/net/hyperv/netvsc_drv.c | 9 +++++++--
1 file changed, 7 insertions(+), 2 deletions(-)
diff --git a/drivers/net/hyperv/netvsc_drv.c b/drivers/net/hyperv/netvsc_drv.c
index eb92720dd1c4a..33c1f6548fb79 100644
--- a/drivers/net/hyperv/netvsc_drv.c
+++ b/drivers/net/hyperv/netvsc_drv.c
@@ -1170,12 +1170,15 @@ static void netvsc_get_stats64(struct net_device *net,
struct rtnl_link_stats64 *t)
{
struct net_device_context *ndev_ctx = netdev_priv(net);
- struct netvsc_device *nvdev = rcu_dereference_rtnl(ndev_ctx->nvdev);
+ struct netvsc_device *nvdev;
struct netvsc_vf_pcpu_stats vf_tot;
int i;
+ rcu_read_lock();
+
+ nvdev = rcu_dereference(ndev_ctx->nvdev);
if (!nvdev)
- return;
+ goto out;
netdev_stats_to_stats64(t, &net->stats);
@@ -1214,6 +1217,8 @@ static void netvsc_get_stats64(struct net_device *net,
t->rx_packets += packets;
t->multicast += multicast;
}
+out:
+ rcu_read_unlock();
}
static int netvsc_set_mac_addr(struct net_device *ndev, void *p)
--
2.20.1
^ permalink raw reply related
* [PATCH AUTOSEL 4.14 22/27] Tools: hv: kvp: eliminate 'may be used uninitialized' warning
From: Sasha Levin @ 2019-08-29 18:16 UTC (permalink / raw)
To: linux-kernel, stable; +Cc: Vitaly Kuznetsov, Sasha Levin, linux-hyperv
In-Reply-To: <20190829181655.8741-1-sashal@kernel.org>
From: Vitaly Kuznetsov <vkuznets@redhat.com>
[ Upstream commit 89eb4d8d25722a0a0194cf7fa47ba602e32a6da7 ]
When building hv_kvp_daemon GCC-8.3 complains:
hv_kvp_daemon.c: In function ‘kvp_get_ip_info.constprop’:
hv_kvp_daemon.c:812:30: warning: ‘ip_buffer’ may be used uninitialized in this function [-Wmaybe-uninitialized]
struct hv_kvp_ipaddr_value *ip_buffer;
this seems to be a false positive: we only use ip_buffer when
op == KVP_OP_GET_IP_INFO and it is only unset when op == KVP_OP_ENUMERATE.
Silence the warning by initializing ip_buffer to NULL.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
tools/hv/hv_kvp_daemon.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools/hv/hv_kvp_daemon.c b/tools/hv/hv_kvp_daemon.c
index 62c9a503ae052..bb245d4afc0cf 100644
--- a/tools/hv/hv_kvp_daemon.c
+++ b/tools/hv/hv_kvp_daemon.c
@@ -867,7 +867,7 @@ kvp_get_ip_info(int family, char *if_name, int op,
int sn_offset = 0;
int error = 0;
char *buffer;
- struct hv_kvp_ipaddr_value *ip_buffer;
+ struct hv_kvp_ipaddr_value *ip_buffer = NULL;
char cidr_mask[5]; /* /xyz */
int weight;
int i;
--
2.20.1
^ permalink raw reply related
* [PATCH AUTOSEL 4.14 21/27] Input: hyperv-keyboard: Use in-place iterator API in the channel callback
From: Sasha Levin @ 2019-08-29 18:16 UTC (permalink / raw)
To: linux-kernel, stable
Cc: Dexuan Cui, Dmitry Torokhov, Sasha Levin, linux-hyperv,
linux-input
In-Reply-To: <20190829181655.8741-1-sashal@kernel.org>
From: Dexuan Cui <decui@microsoft.com>
[ Upstream commit d09bc83640d524b8467a660db7b1d15e6562a1de ]
Simplify the ring buffer handling with the in-place API.
Also avoid the dynamic allocation and the memory leak in the channel
callback function.
Signed-off-by: Dexuan Cui <decui@microsoft.com>
Acked-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
drivers/input/serio/hyperv-keyboard.c | 35 +++++----------------------
1 file changed, 6 insertions(+), 29 deletions(-)
diff --git a/drivers/input/serio/hyperv-keyboard.c b/drivers/input/serio/hyperv-keyboard.c
index 55288a026e4e2..c137ffa6fdec8 100644
--- a/drivers/input/serio/hyperv-keyboard.c
+++ b/drivers/input/serio/hyperv-keyboard.c
@@ -245,40 +245,17 @@ static void hv_kbd_handle_received_packet(struct hv_device *hv_dev,
static void hv_kbd_on_channel_callback(void *context)
{
+ struct vmpacket_descriptor *desc;
struct hv_device *hv_dev = context;
- void *buffer;
- int bufferlen = 0x100; /* Start with sensible size */
u32 bytes_recvd;
u64 req_id;
- int error;
- buffer = kmalloc(bufferlen, GFP_ATOMIC);
- if (!buffer)
- return;
-
- while (1) {
- error = vmbus_recvpacket_raw(hv_dev->channel, buffer, bufferlen,
- &bytes_recvd, &req_id);
- switch (error) {
- case 0:
- if (bytes_recvd == 0) {
- kfree(buffer);
- return;
- }
-
- hv_kbd_handle_received_packet(hv_dev, buffer,
- bytes_recvd, req_id);
- break;
+ foreach_vmbus_pkt(desc, hv_dev->channel) {
+ bytes_recvd = desc->len8 * 8;
+ req_id = desc->trans_id;
- case -ENOBUFS:
- kfree(buffer);
- /* Handle large packet */
- bufferlen = bytes_recvd;
- buffer = kmalloc(bytes_recvd, GFP_ATOMIC);
- if (!buffer)
- return;
- break;
- }
+ hv_kbd_handle_received_packet(hv_dev, desc, bytes_recvd,
+ req_id);
}
}
--
2.20.1
^ permalink raw reply related
* [PATCH AUTOSEL 4.19 05/45] hv_netvsc: Fix a warning of suspicious RCU usage
From: Sasha Levin @ 2019-08-29 18:15 UTC (permalink / raw)
To: linux-kernel, stable
Cc: Dexuan Cui, David S . Miller, Sasha Levin, linux-hyperv, netdev
In-Reply-To: <20190829181547.8280-1-sashal@kernel.org>
From: Dexuan Cui <decui@microsoft.com>
[ Upstream commit 6d0d779dca73cd5acb649c54f81401f93098b298 ]
This fixes a warning of "suspicious rcu_dereference_check() usage"
when nload runs.
Fixes: 776e726bfb34 ("netvsc: fix RCU warning in get_stats")
Signed-off-by: Dexuan Cui <decui@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
drivers/net/hyperv/netvsc_drv.c | 9 +++++++--
1 file changed, 7 insertions(+), 2 deletions(-)
diff --git a/drivers/net/hyperv/netvsc_drv.c b/drivers/net/hyperv/netvsc_drv.c
index cc60ef9634db2..6f6c0dbd91fc8 100644
--- a/drivers/net/hyperv/netvsc_drv.c
+++ b/drivers/net/hyperv/netvsc_drv.c
@@ -1248,12 +1248,15 @@ static void netvsc_get_stats64(struct net_device *net,
struct rtnl_link_stats64 *t)
{
struct net_device_context *ndev_ctx = netdev_priv(net);
- struct netvsc_device *nvdev = rcu_dereference_rtnl(ndev_ctx->nvdev);
+ struct netvsc_device *nvdev;
struct netvsc_vf_pcpu_stats vf_tot;
int i;
+ rcu_read_lock();
+
+ nvdev = rcu_dereference(ndev_ctx->nvdev);
if (!nvdev)
- return;
+ goto out;
netdev_stats_to_stats64(t, &net->stats);
@@ -1292,6 +1295,8 @@ static void netvsc_get_stats64(struct net_device *net,
t->rx_packets += packets;
t->multicast += multicast;
}
+out:
+ rcu_read_unlock();
}
static int netvsc_set_mac_addr(struct net_device *ndev, void *p)
--
2.20.1
^ permalink raw reply related
* [PATCH AUTOSEL 5.2 53/76] Input: hyperv-keyboard: Use in-place iterator API in the channel callback
From: Sasha Levin @ 2019-08-29 18:12 UTC (permalink / raw)
To: linux-kernel, stable
Cc: Dexuan Cui, Dmitry Torokhov, Sasha Levin, linux-hyperv,
linux-input
In-Reply-To: <20190829181311.7562-1-sashal@kernel.org>
From: Dexuan Cui <decui@microsoft.com>
[ Upstream commit d09bc83640d524b8467a660db7b1d15e6562a1de ]
Simplify the ring buffer handling with the in-place API.
Also avoid the dynamic allocation and the memory leak in the channel
callback function.
Signed-off-by: Dexuan Cui <decui@microsoft.com>
Acked-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
drivers/input/serio/hyperv-keyboard.c | 35 +++++----------------------
1 file changed, 6 insertions(+), 29 deletions(-)
diff --git a/drivers/input/serio/hyperv-keyboard.c b/drivers/input/serio/hyperv-keyboard.c
index 8e457e50f837e..770e36d0c66fb 100644
--- a/drivers/input/serio/hyperv-keyboard.c
+++ b/drivers/input/serio/hyperv-keyboard.c
@@ -237,40 +237,17 @@ static void hv_kbd_handle_received_packet(struct hv_device *hv_dev,
static void hv_kbd_on_channel_callback(void *context)
{
+ struct vmpacket_descriptor *desc;
struct hv_device *hv_dev = context;
- void *buffer;
- int bufferlen = 0x100; /* Start with sensible size */
u32 bytes_recvd;
u64 req_id;
- int error;
- buffer = kmalloc(bufferlen, GFP_ATOMIC);
- if (!buffer)
- return;
-
- while (1) {
- error = vmbus_recvpacket_raw(hv_dev->channel, buffer, bufferlen,
- &bytes_recvd, &req_id);
- switch (error) {
- case 0:
- if (bytes_recvd == 0) {
- kfree(buffer);
- return;
- }
-
- hv_kbd_handle_received_packet(hv_dev, buffer,
- bytes_recvd, req_id);
- break;
+ foreach_vmbus_pkt(desc, hv_dev->channel) {
+ bytes_recvd = desc->len8 * 8;
+ req_id = desc->trans_id;
- case -ENOBUFS:
- kfree(buffer);
- /* Handle large packet */
- bufferlen = bytes_recvd;
- buffer = kmalloc(bytes_recvd, GFP_ATOMIC);
- if (!buffer)
- return;
- break;
- }
+ hv_kbd_handle_received_packet(hv_dev, desc, bytes_recvd,
+ req_id);
}
}
--
2.20.1
^ permalink raw reply related
* [PATCH AUTOSEL 5.2 16/76] hv_netvsc: Fix a warning of suspicious RCU usage
From: Sasha Levin @ 2019-08-29 18:12 UTC (permalink / raw)
To: linux-kernel, stable
Cc: Dexuan Cui, David S . Miller, Sasha Levin, linux-hyperv, netdev
In-Reply-To: <20190829181311.7562-1-sashal@kernel.org>
From: Dexuan Cui <decui@microsoft.com>
[ Upstream commit 6d0d779dca73cd5acb649c54f81401f93098b298 ]
This fixes a warning of "suspicious rcu_dereference_check() usage"
when nload runs.
Fixes: 776e726bfb34 ("netvsc: fix RCU warning in get_stats")
Signed-off-by: Dexuan Cui <decui@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
drivers/net/hyperv/netvsc_drv.c | 9 +++++++--
1 file changed, 7 insertions(+), 2 deletions(-)
diff --git a/drivers/net/hyperv/netvsc_drv.c b/drivers/net/hyperv/netvsc_drv.c
index 3544e19915792..e8fce6d715ef0 100644
--- a/drivers/net/hyperv/netvsc_drv.c
+++ b/drivers/net/hyperv/netvsc_drv.c
@@ -1239,12 +1239,15 @@ static void netvsc_get_stats64(struct net_device *net,
struct rtnl_link_stats64 *t)
{
struct net_device_context *ndev_ctx = netdev_priv(net);
- struct netvsc_device *nvdev = rcu_dereference_rtnl(ndev_ctx->nvdev);
+ struct netvsc_device *nvdev;
struct netvsc_vf_pcpu_stats vf_tot;
int i;
+ rcu_read_lock();
+
+ nvdev = rcu_dereference(ndev_ctx->nvdev);
if (!nvdev)
- return;
+ goto out;
netdev_stats_to_stats64(t, &net->stats);
@@ -1283,6 +1286,8 @@ static void netvsc_get_stats64(struct net_device *net,
t->rx_packets += packets;
t->multicast += multicast;
}
+out:
+ rcu_read_unlock();
}
static int netvsc_set_mac_addr(struct net_device *ndev, void *p)
--
2.20.1
^ permalink raw reply related
* RE: [PATCH v3] PCI: hv: Make functions static
From: Haiyang Zhang @ 2019-08-29 15:50 UTC (permalink / raw)
To: Krzysztof Wilczynski, Bjorn Helgaas
Cc: KY Srinivasan, Stephen Hemminger, Sasha Levin, Lorenzo Pieralisi,
linux-pci@vger.kernel.org, linux-kernel@vger.kernel.org,
linux-hyperv@vger.kernel.org
In-Reply-To: <20190829091713.27130-1-kw@linux.com>
> -----Original Message-----
> From: Krzysztof Wilczynski <kswilczynski@gmail.com> On Behalf Of Krzysztof
> Wilczynski
> Sent: Thursday, August 29, 2019 2:17 AM
> To: Bjorn Helgaas <helgaas@kernel.org>
> Cc: KY Srinivasan <kys@microsoft.com>; Haiyang Zhang
> <haiyangz@microsoft.com>; Stephen Hemminger
> <sthemmin@microsoft.com>; Sasha Levin <sashal@kernel.org>; Lorenzo
> Pieralisi <lorenzo.pieralisi@arm.com>; linux-pci@vger.kernel.org; linux-
> kernel@vger.kernel.org; linux-hyperv@vger.kernel.org
> Subject: [PATCH v3] PCI: hv: Make functions static
>
> Functions hv_read_config_block(), hv_write_config_block() and
> hv_register_block_invalidate() are not used anywhere else and are local to
> drivers/pci/controller/pci-hyperv.c,
> and do not need to be in global scope, so make these static.
>
> Resolve following compiler warning that can be seen when building with
> warnings enabled (W=1):
>
> drivers/pci/controller/pci-hyperv.c:933:5: warning:
> no previous prototype for ‘hv_read_config_block’
> [-Wmissing-prototypes]
>
> drivers/pci/controller/pci-hyperv.c:1013:5: warning:
> no previous prototype for ‘hv_write_config_block’
> [-Wmissing-prototypes]
>
> drivers/pci/controller/pci-hyperv.c:1082:5: warning:
> no previous prototype for ‘hv_register_block_invalidate’
> [-Wmissing-prototypes]
>
> Signed-off-by: Krzysztof Wilczynski <kw@linux.com>
Reviewed-by: Haiyang Zhang <haiyangz@microsoft.com>
Thanks!
^ permalink raw reply
* [PATCH v3] PCI: hv: Make functions static
From: Krzysztof Wilczynski @ 2019-08-29 9:17 UTC (permalink / raw)
To: Bjorn Helgaas
Cc: K. Y. Srinivasan, Haiyang Zhang, Stephen Hemminger, Sasha Levin,
Lorenzo Pieralisi, linux-pci, linux-kernel, linux-hyperv
In-Reply-To: <20190828221846.6672-1-kw@linux.com>
Functions hv_read_config_block(), hv_write_config_block()
and hv_register_block_invalidate() are not used anywhere
else and are local to drivers/pci/controller/pci-hyperv.c,
and do not need to be in global scope, so make these static.
Resolve following compiler warning that can be seen when
building with warnings enabled (W=1):
drivers/pci/controller/pci-hyperv.c:933:5: warning:
no previous prototype for ‘hv_read_config_block’
[-Wmissing-prototypes]
drivers/pci/controller/pci-hyperv.c:1013:5: warning:
no previous prototype for ‘hv_write_config_block’
[-Wmissing-prototypes]
drivers/pci/controller/pci-hyperv.c:1082:5: warning:
no previous prototype for ‘hv_register_block_invalidate’
[-Wmissing-prototypes]
Signed-off-by: Krzysztof Wilczynski <kw@linux.com>
---
Changes in v3:
Commit message has been wrapped to fit 75 columns.
Addressed formatting based on feedback from v2.
Changes in v2:
Update commit message to include compiler warning.
drivers/pci/controller/pci-hyperv.c | 16 ++++++++--------
1 file changed, 8 insertions(+), 8 deletions(-)
diff --git a/drivers/pci/controller/pci-hyperv.c b/drivers/pci/controller/pci-hyperv.c
index f1f300218fab..ba988fe033b5 100644
--- a/drivers/pci/controller/pci-hyperv.c
+++ b/drivers/pci/controller/pci-hyperv.c
@@ -930,8 +930,9 @@ static void hv_pci_read_config_compl(void *context, struct pci_response *resp,
*
* Return: 0 on success, -errno on failure
*/
-int hv_read_config_block(struct pci_dev *pdev, void *buf, unsigned int len,
- unsigned int block_id, unsigned int *bytes_returned)
+static int hv_read_config_block(struct pci_dev *pdev, void *buf,
+ unsigned int len, unsigned int block_id,
+ unsigned int *bytes_returned)
{
struct hv_pcibus_device *hbus =
container_of(pdev->bus->sysdata, struct hv_pcibus_device,
@@ -1010,8 +1011,8 @@ static void hv_pci_write_config_compl(void *context, struct pci_response *resp,
*
* Return: 0 on success, -errno on failure
*/
-int hv_write_config_block(struct pci_dev *pdev, void *buf, unsigned int len,
- unsigned int block_id)
+static int hv_write_config_block(struct pci_dev *pdev, void *buf,
+ unsigned int len, unsigned int block_id)
{
struct hv_pcibus_device *hbus =
container_of(pdev->bus->sysdata, struct hv_pcibus_device,
@@ -1079,9 +1080,9 @@ int hv_write_config_block(struct pci_dev *pdev, void *buf, unsigned int len,
*
* Return: 0 on success, -errno on failure
*/
-int hv_register_block_invalidate(struct pci_dev *pdev, void *context,
- void (*block_invalidate)(void *context,
- u64 block_mask))
+static int hv_register_block_invalidate(
+ struct pci_dev *pdev, void *context,
+ void (*block_invalidate)(void *context, u64 block_mask))
{
struct hv_pcibus_device *hbus =
container_of(pdev->bus->sysdata, struct hv_pcibus_device,
@@ -1097,7 +1098,6 @@ int hv_register_block_invalidate(struct pci_dev *pdev, void *context,
put_pcichild(hpdev);
return 0;
-
}
/* Interrupt management hooks */
--
2.22.1
^ permalink raw reply related
* [PATCH v4 2/2] tools: hv: add vmbus testing tool
From: Branden Bonaby @ 2019-08-29 4:24 UTC (permalink / raw)
To: kys, haiyangz, sthemmin, sashal
Cc: Branden Bonaby, linux-hyperv, linux-kernel
In-Reply-To: <cover.1567047549.git.brandonbonaby94@gmail.com>
This is a userspace tool to drive the testing. Currently it supports
introducing user specified delay in the host to guest communication
path on a per-channel basis.
Signed-off-by: Branden Bonaby <brandonbonaby94@gmail.com>
---
Changes in v4:
- Based on Harrys comments, made the tool more
user friendly and added more error checking.
Changes in v3:
- Align python tool to match Linux coding style.
Changes in v2:
- Move testing location to new location in debugfs.
tools/hv/vmbus_testing | 376 +++++++++++++++++++++++++++++++++++++++++
1 file changed, 376 insertions(+)
create mode 100644 tools/hv/vmbus_testing
diff --git a/tools/hv/vmbus_testing b/tools/hv/vmbus_testing
new file mode 100644
index 000000000000..e7212903dd1d
--- /dev/null
+++ b/tools/hv/vmbus_testing
@@ -0,0 +1,376 @@
+#!/usr/bin/env python3
+# SPDX-License-Identifier: GPL-2.0
+#
+# Program to allow users to fuzz test Hyper-V drivers
+# by interfacing with Hyper-V debugfs attributes.
+# Current test methods available:
+# 1. delay testing
+#
+# Current file/directory structure of hyper-V debugfs:
+# /sys/kernel/debug/hyperv/UUID
+# /sys/kernel/debug/hyperv/UUID/<test-state filename>
+# /sys/kernel/debug/hyperv/UUID/<test-method sub-directory>
+#
+# author: Branden Bonaby <brandonbonaby94@gmail.com>
+
+import os
+import cmd
+import argparse
+import glob
+from argparse import RawDescriptionHelpFormatter
+from argparse import RawTextHelpFormatter
+from enum import Enum
+
+# Do not change unless, you change the debugfs attributes
+# in /drivers/hv/debugfs.c. All fuzz testing
+# attributes will start with "fuzz_test".
+
+# debugfs path for hyperv must exist before proceeding
+debugfs_hyperv_path = "/sys/kernel/debug/hyperv"
+if not os.path.isdir(debugfs_hyperv_path):
+ print("{} doesn't exist/check permissions".format(debugfs_hyperv_path))
+ exit(-1)
+
+class dev_state(Enum):
+ off = 0
+ on = 1
+
+# File names, that correspond to the files created in
+# /drivers/hv/debugfs.c
+class f_names(Enum):
+ state_f = "fuzz_test_state"
+ buff_f = "fuzz_test_buffer_interrupt_delay"
+ mess_f = "fuzz_test_message_delay"
+
+# Both single_actions and all_actions are used
+# for error checking and to allow for some subparser
+# names to be abbreviated. Do not abbreviate the
+# test method names, as it will become less intuitive
+# as to what the user can do. If you do decide to
+# abbreviate the test method name, make sure the main
+# function reflects this change.
+
+all_actions = [
+ "disable_all",
+ "D",
+ "enable_all",
+ "view_all",
+ "V"
+]
+
+single_actions = [
+ "disable_single",
+ "d",
+ "enable_single",
+ "view_single",
+ "v"
+]
+
+def main():
+
+ file_map = recursive_file_lookup(debugfs_hyperv_path, dict())
+ args = parse_args()
+ if (not args.action):
+ print ("Error, no options selected...exiting")
+ exit(-1)
+ arg_set = { k for (k,v) in vars(args).items() if v and k != "action" }
+ arg_set.add(args.action)
+ path = args.path if "path" in arg_set else None
+ if (path and path[-1] == "/"):
+ path = path[:-1]
+ validate_args_path(path, arg_set, file_map)
+ if (path and "enable_single" in arg_set):
+ state_path = locate_state(path, file_map)
+ set_test_state(state_path, dev_state.on.value, args.quiet)
+
+ # Use subparsers as the key for different actions
+ if ("delay" in arg_set):
+ validate_delay_values(args.delay_time)
+ if (args.enable_all):
+ set_delay_all_devices(file_map, args.delay_time,
+ args.quiet)
+ else:
+ set_delay_values(path, file_map, args.delay_time,
+ args.quiet)
+ elif ("disable_all" in arg_set or "D" in arg_set):
+ disable_all_testing(file_map)
+ elif ("disable_single" in arg_set or "d" in arg_set):
+ disable_testing_single_device(path, file_map)
+ elif ("view_all" in arg_set or "V" in arg_set):
+ get_all_devices_test_status(file_map)
+ elif ("view_single" in arg_set or "v" in arg_set):
+ get_device_test_values(path, file_map)
+
+# Get the state location
+def locate_state(device, file_map):
+ return file_map[device][f_names.state_f.value]
+
+# Validate delay values to make sure they are acceptable to
+# enable delays on a device
+def validate_delay_values(delay):
+
+ if (delay[0] == -1 and delay[1] == -1):
+ print("\nError, At least 1 value must be greater than 0")
+ exit(-1)
+ for i in delay:
+ if (i < -1 or i == 0 or i > 1000):
+ print("\nError, Values must be equal to -1 "
+ "or be > 0 and <= 1000")
+ exit(-1)
+
+# Validate argument path
+def validate_args_path(path, arg_set, file_map):
+
+ if (not path and any(element in arg_set for element in single_actions)):
+ print("Error, path (-p) REQUIRED for the specified option. "
+ "Use (-h) to check usage.")
+ exit(-1)
+ elif (path and any(item in arg_set for item in all_actions)):
+ print("Error, path (-p) NOT REQUIRED for the specified option. "
+ "Use (-h) to check usage." )
+ exit(-1)
+ elif (path not in file_map and any(item in arg_set
+ for item in single_actions)):
+ print("Error, path '{}' not a valid vmbus device".format(path))
+ exit(-1)
+
+# display Testing status of single device
+def get_device_test_values(path, file_map):
+
+ for name in file_map[path]:
+ file_location = file_map[path][name]
+ print( name + " = " + str(read_test_files(file_location)))
+
+# Create a map of the vmbus devices and their associated files
+# [key=device, value = [key = filename, value = file path]]
+def recursive_file_lookup(path, file_map):
+
+ for f_path in glob.iglob(path + '**/*'):
+ if (os.path.isfile(f_path)):
+ if (f_path.rsplit("/",2)[0] == debugfs_hyperv_path):
+ directory = f_path.rsplit("/",1)[0]
+ else:
+ directory = f_path.rsplit("/",2)[0]
+ f_name = f_path.split("/")[-1]
+ if (file_map.get(directory)):
+ file_map[directory].update({f_name:f_path})
+ else:
+ file_map[directory] = {f_name:f_path}
+ elif (os.path.isdir(f_path)):
+ recursive_file_lookup(f_path,file_map)
+ return file_map
+
+# display Testing state of devices
+def get_all_devices_test_status(file_map):
+
+ for device in file_map:
+ if (get_test_state(locate_state(device, file_map)) is 1):
+ print("Testing = ON for: {}"
+ .format(device.split("/")[5]))
+ else:
+ print("Testing = OFF for: {}"
+ .format(device.split("/")[5]))
+
+# read the vmbus device files, path must be absolute path before calling
+def read_test_files(path):
+ try:
+ with open(path,"r") as f:
+ file_value = f.readline().strip()
+ return int(file_value)
+
+ except IOError as e:
+ errno, strerror = e.args
+ print("I/O error({0}): {1} on file {2}"
+ .format(errno, strerror, path))
+ exit(-1)
+ except ValueError:
+ print ("Element to int conversion error in: \n{}".format(path))
+ exit(-1)
+
+# writing to vmbus device files, path must be absolute path before calling
+def write_test_files(path, value):
+
+ try:
+ with open(path,"w") as f:
+ f.write("{}".format(value))
+ except IOError as e:
+ errno, strerror = e.args
+ print("I/O error({0}): {1} on file {2}"
+ .format(errno, strerror, path))
+ exit(-1)
+
+# set testing state of device
+def set_test_state(state_path, state_value, quiet):
+
+ write_test_files(state_path, state_value)
+ if (get_test_state(state_path) is 1):
+ if (not quiet):
+ print("Testing = ON for device: {}"
+ .format(state_path.split("/")[5]))
+ else:
+ if (not quiet):
+ print("Testing = OFF for device: {}"
+ .format(state_path.split("/")[5]))
+
+# get testing state of device
+def get_test_state(state_path):
+ #state == 1 - test = ON
+ #state == 0 - test = OFF
+ return read_test_files(state_path)
+
+# write 1 - 1000 microseconds, into a single device using the
+# fuzz_test_buffer_interrupt_delay and fuzz_test_message_delay
+# debugfs attributes
+def set_delay_values(device, file_map, delay_length, quiet):
+
+ try:
+ interrupt = file_map[device][f_names.buff_f.value]
+ message = file_map[device][f_names.mess_f.value]
+
+ # delay[0]- buffer interrupt delay, delay[1]- message delay
+ if (delay_length[0] >= 0 and delay_length[0] <= 1000):
+ write_test_files(interrupt, delay_length[0])
+ if (delay_length[1] >= 0 and delay_length[1] <= 1000):
+ write_test_files(message, delay_length[1])
+ if (not quiet):
+ print("Buffer delay testing = {} for: {}"
+ .format(read_test_files(interrupt),
+ interrupt.split("/")[5]))
+ print("Message delay testing = {} for: {}"
+ .format(read_test_files(message),
+ message.split("/")[5]))
+ except IOError as e:
+ errno, strerror = e.args
+ print("I/O error({0}): {1} on files {2}{3}"
+ .format(errno, strerror, interrupt, message))
+ exit(-1)
+
+# enabling delay testing on all devices
+def set_delay_all_devices(file_map, delay, quiet):
+
+ for device in (file_map):
+ set_test_state(locate_state(device, file_map),
+ dev_state.on.value,
+ quiet)
+ set_delay_values(device, file_map, delay, quiet)
+
+# disable all testing on a SINGLE device.
+def disable_testing_single_device(device, file_map):
+
+ for name in file_map[device]:
+ file_location = file_map[device][name]
+ write_test_files(file_location, dev_state.off.value)
+ print("ALL testing now OFF for {}".format(device.split("/")[-1]))
+
+# disable all testing on ALL devices
+def disable_all_testing(file_map):
+
+ for device in file_map:
+ disable_testing_single_device(device, file_map)
+
+def parse_args():
+ parser = argparse.ArgumentParser(prog = "vmbus_testing",usage ="\n"
+ "%(prog)s [delay] [-h] [-e|-E] -t [-p]\n"
+ "%(prog)s [view_all | V] [-h]\n"
+ "%(prog)s [disable_all | D] [-h]\n"
+ "%(prog)s [disable_single | d] [-h|-p]\n"
+ "%(prog)s [view_single | v] [-h|-p]\n"
+ "%(prog)s --version\n",
+ description = "\nUse lsvmbus to get vmbus device type "
+ "information.\n" "\nThe debugfs root path is "
+ "/sys/kernel/debug/hyperv",
+ formatter_class = RawDescriptionHelpFormatter)
+ subparsers = parser.add_subparsers(dest = "action")
+ parser.add_argument("--version", action = "version",
+ version = '%(prog)s 0.1.0')
+ parser.add_argument("-q","--quiet", action = "store_true",
+ help = "silence none important test messages."
+ " This will only work when enabling testing"
+ " on a device.")
+ # Use the path parser to hold the --path attribute so it can
+ # be shared between subparsers. Also do the same for the state
+ # parser, as all testing methods will use --enable_all and
+ # enable_single.
+ path_parser = argparse.ArgumentParser(add_help=False)
+ path_parser.add_argument("-p","--path", metavar = "",
+ help = "Debugfs path to a vmbus device. The path "
+ "must be the absolute path to the device.")
+ state_parser = argparse.ArgumentParser(add_help=False)
+ state_group = state_parser.add_mutually_exclusive_group(required = True)
+ state_group.add_argument("-E", "--enable_all", action = "store_const",
+ const = "enable_all",
+ help = "Enable the specified test type "
+ "on ALL vmbus devices.")
+ state_group.add_argument("-e", "--enable_single",
+ action = "store_const",
+ const = "enable_single",
+ help = "Enable the specified test type on a "
+ "SINGLE vmbus device.")
+ parser_delay = subparsers.add_parser("delay",
+ parents = [state_parser, path_parser],
+ help = "Delay the ring buffer interrupt or the "
+ "ring buffer message reads in microseconds.",
+ prog = "vmbus_testing",
+ usage = "%(prog)s [-h]\n"
+ "%(prog)s -E -t [value] [value]\n"
+ "%(prog)s -e -t [value] [value] -p",
+ description = "Delay the ring buffer interrupt for "
+ "vmbus devices, or delay the ring buffer message "
+ "reads for vmbus devices (both in microseconds). This "
+ "is only on the host to guest channel.")
+ parser_delay.add_argument("-t", "--delay_time", metavar = "", nargs = 2,
+ type = check_range, default =[0,0], required = (True),
+ help = "Set [buffer] & [message] delay time. "
+ "Value constraints: -1 == value "
+ "or 0 < value <= 1000.\n"
+ "Use -1 to keep the previous value for that delay "
+ "type, or a value > 0 <= 1000 to change the delay "
+ "time.")
+ parser_dis_all = subparsers.add_parser("disable_all",
+ aliases = ['D'], prog = "vmbus_testing",
+ usage = "%(prog)s [disable_all | D] -h\n"
+ "%(prog)s [disable_all | D]\n",
+ help = "Disable ALL testing on ALL vmbus devices.",
+ description = "Disable ALL testing on ALL vmbus "
+ "devices.")
+ parser_dis_single = subparsers.add_parser("disable_single",
+ aliases = ['d'],
+ parents = [path_parser], prog = "vmbus_testing",
+ usage = "%(prog)s [disable_single | d] -h\n"
+ "%(prog)s [disable_single | d] -p\n",
+ help = "Disable ALL testing on a SINGLE vmbus device.",
+ description = "Disable ALL testing on a SINGLE vmbus "
+ "device.")
+ parser_view_all = subparsers.add_parser("view_all", aliases = ['V'],
+ help = "View the test state for ALL vmbus devices.",
+ prog = "vmbus_testing",
+ usage = "%(prog)s [view_all | V] -h\n"
+ "%(prog)s [view_all | V]\n",
+ description = "This shows the test state for ALL the "
+ "vmbus devices.")
+ parser_view_single = subparsers.add_parser("view_single",
+ aliases = ['v'],parents = [path_parser],
+ help = "View the test values for a SINGLE vmbus "
+ "device.",
+ description = "This shows the test values for a SINGLE "
+ "vmbus device.", prog = "vmbus_testing",
+ usage = "%(prog)s [view_single | v] -h\n"
+ "%(prog)s [view_single | v] -p")
+
+ return parser.parse_args()
+
+# value checking for range checking input in parser
+def check_range(arg1):
+
+ try:
+ val = int(arg1)
+ except ValueError as err:
+ raise argparse.ArgumentTypeError(str(err))
+ if val < -1 or val > 1000:
+ message = ("\n\nvalue must be -1 or 0 < value <= 1000. "
+ "Value program received: {}\n").format(val)
+ raise argparse.ArgumentTypeError(message)
+ return val
+
+if __name__ == "__main__":
+ main()
--
2.17.1
^ permalink raw reply related
* [PATCH v4 1/2] drivers: hv: vmbus: Introduce latency testing
From: Branden Bonaby @ 2019-08-29 4:24 UTC (permalink / raw)
To: kys, haiyangz, sthemmin, sashal
Cc: Branden Bonaby, linux-hyperv, linux-kernel
In-Reply-To: <cover.1567047549.git.brandonbonaby94@gmail.com>
Introduce user specified latency in the packet reception path
By exposing the test parameters as part of the debugfs channel
attributes. We will control the testing state via these attributes.
Signed-off-by: Branden Bonaby <brandonbonaby94@gmail.com>
---
changes in v4:
- Combined v3 patch 2 into this patch, and changed the
commit description to reflect this.
- Moved debugfs code from "vmbus_drv.c" that was in
previous v3 patch 2, into a new file "debugfs.c" in
drivers/hv.
- Updated the Makefile to compile "debugfs.c" if
CONFIG_HYPERV_TESTING is enabled
- As per Michael's comments, added empty implementations
of the new functions, so the compiler will not generate
code when CONFIG_HYPERV_TESTING is not enabled.
- Added microseconds into description for files in
Documentation/ABI/testing/debugfs-hyperv.
Changes in v2:
- Add #ifdef in Kconfig file so test code will not interfere
with non-test code.
- Move test code functions for delay to hyperv_vmbus header
file.
- Wrap test code under #ifdef statement.
Documentation/ABI/testing/debugfs-hyperv | 23 +++
MAINTAINERS | 1 +
drivers/hv/Kconfig | 7 +
drivers/hv/Makefile | 1 +
drivers/hv/connection.c | 1 +
drivers/hv/hv_debugfs.c | 187 +++++++++++++++++++++++
drivers/hv/hyperv_vmbus.h | 31 ++++
drivers/hv/ring_buffer.c | 2 +
drivers/hv/vmbus_drv.c | 6 +
include/linux/hyperv.h | 19 +++
10 files changed, 278 insertions(+)
create mode 100644 Documentation/ABI/testing/debugfs-hyperv
create mode 100644 drivers/hv/hv_debugfs.c
diff --git a/Documentation/ABI/testing/debugfs-hyperv b/Documentation/ABI/testing/debugfs-hyperv
new file mode 100644
index 000000000000..0903e6427a2d
--- /dev/null
+++ b/Documentation/ABI/testing/debugfs-hyperv
@@ -0,0 +1,23 @@
+What: /sys/kernel/debug/hyperv/<UUID>/fuzz_test_state
+Date: August 2019
+KernelVersion: 5.3
+Contact: Branden Bonaby <brandonbonaby94@gmail.com>
+Description: Fuzz testing status of a vmbus device, whether its in an ON
+ state or a OFF state
+Users: Debugging tools
+
+What: /sys/kernel/debug/hyperv/<UUID>/delay/fuzz_test_buffer_interrupt_delay
+Date: August 2019
+KernelVersion: 5.3
+Contact: Branden Bonaby <brandonbonaby94@gmail.com>
+Description: Fuzz testing buffer interrupt delay value between 0 - 1000
+ microseconds (inclusive).
+Users: Debugging tools
+
+What: /sys/kernel/debug/hyperv/<UUID>/delay/fuzz_test_message_delay
+Date: August 2019
+KernelVersion: 5.3
+Contact: Branden Bonaby <brandonbonaby94@gmail.com>
+Description: Fuzz testing message delay value between 0 - 1000 microseconds
+ (inclusive).
+Users: Debugging tools
diff --git a/MAINTAINERS b/MAINTAINERS
index b1bc9c87838b..6931a4eeaac0 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -7460,6 +7460,7 @@ F: include/uapi/linux/hyperv.h
F: include/asm-generic/mshyperv.h
F: tools/hv/
F: Documentation/ABI/stable/sysfs-bus-vmbus
+F: Documentation/ABI/testing/debugfs-hyperv
HYPERBUS SUPPORT
M: Vignesh Raghavendra <vigneshr@ti.com>
diff --git a/drivers/hv/Kconfig b/drivers/hv/Kconfig
index 9a59957922d4..d97437ba0626 100644
--- a/drivers/hv/Kconfig
+++ b/drivers/hv/Kconfig
@@ -29,4 +29,11 @@ config HYPERV_BALLOON
help
Select this option to enable Hyper-V Balloon driver.
+config HYPERV_TESTING
+ bool "Hyper-V testing"
+ default n
+ depends on HYPERV && DEBUG_FS
+ help
+ Select this option to enable Hyper-V vmbus testing.
+
endmenu
diff --git a/drivers/hv/Makefile b/drivers/hv/Makefile
index a1eec7177c2d..d754bd86ca47 100644
--- a/drivers/hv/Makefile
+++ b/drivers/hv/Makefile
@@ -2,6 +2,7 @@
obj-$(CONFIG_HYPERV) += hv_vmbus.o
obj-$(CONFIG_HYPERV_UTILS) += hv_utils.o
obj-$(CONFIG_HYPERV_BALLOON) += hv_balloon.o
+obj-$(CONFIG_HYPERV_TESTING) += hv_debugfs.o
CFLAGS_hv_trace.o = -I$(src)
CFLAGS_hv_balloon.o = -I$(src)
diff --git a/drivers/hv/connection.c b/drivers/hv/connection.c
index 09829e15d4a0..4d4d40832846 100644
--- a/drivers/hv/connection.c
+++ b/drivers/hv/connection.c
@@ -357,6 +357,7 @@ void vmbus_on_event(unsigned long data)
trace_vmbus_on_event(channel);
+ hv_debug_delay_test(channel, INTERRUPT_DELAY);
do {
void (*callback_fn)(void *);
diff --git a/drivers/hv/hv_debugfs.c b/drivers/hv/hv_debugfs.c
new file mode 100644
index 000000000000..7a3f1ce71ffa
--- /dev/null
+++ b/drivers/hv/hv_debugfs.c
@@ -0,0 +1,187 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Authors:
+ * Branden Bonaby <brandonbonaby94@gmail.com>
+ */
+
+#include <linux/hyperv.h>
+#include <linux/debugfs.h>
+#include <linux/delay.h>
+#include <linux/err.h>
+
+#include "hyperv_vmbus.h"
+
+struct dentry *hv_debug_root;
+
+static int hv_debugfs_delay_get(void *data, u64 *val)
+{
+ *val = *(u32 *)data;
+ return 0;
+}
+
+static int hv_debugfs_delay_set(void *data, u64 val)
+{
+ int ret = 0;
+
+ if (val >= 1 && val <= 1000)
+ *(u32 *)data = val;
+ else if (val == 0)
+ *(u32 *)data = 0;
+ else
+ ret = -EINVAL;
+
+ return ret;
+}
+
+DEFINE_DEBUGFS_ATTRIBUTE(hv_debugfs_delay_fops, hv_debugfs_delay_get,
+ hv_debugfs_delay_set, "%llu\n");
+
+static int hv_debugfs_state_get(void *data, u64 *val)
+{
+ *val = *(bool *)data;
+ return 0;
+}
+
+static int hv_debugfs_state_set(void *data, u64 val)
+{
+ int ret = 0;
+
+ if (val == 1)
+ *(bool *)data = true;
+ else if (val == 0)
+ *(bool *)data = false;
+ else
+ ret = -EINVAL;
+
+ return ret;
+}
+
+DEFINE_DEBUGFS_ATTRIBUTE(hv_debugfs_state_fops, hv_debugfs_state_get,
+ hv_debugfs_state_set, "%llu\n");
+
+/* Setup delay files to store test values */
+static int hv_debug_delay_files(struct hv_device *dev, struct dentry *root)
+{
+ struct vmbus_channel *channel = dev->channel;
+ char *buffer = "fuzz_test_buffer_interrupt_delay";
+ char *message = "fuzz_test_message_delay";
+ int *buffer_val = &channel->fuzz_testing_interrupt_delay;
+ int *message_val = &channel->fuzz_testing_message_delay;
+ struct dentry *buffer_file, *message_file;
+
+ buffer_file = debugfs_create_file(buffer, 0644, root,
+ buffer_val,
+ &hv_debugfs_delay_fops);
+ if (IS_ERR(buffer_file)) {
+ pr_debug("debugfs_hyperv: file %s not created\n", buffer);
+ return PTR_ERR(buffer_file);
+ }
+
+ message_file = debugfs_create_file(message, 0644, root,
+ message_val,
+ &hv_debugfs_delay_fops);
+ if (IS_ERR(message_file)) {
+ pr_debug("debugfs_hyperv: file %s not created\n", message);
+ return PTR_ERR(message_file);
+ }
+
+ return 0;
+}
+
+/* Setup test state value for vmbus device */
+static int hv_debug_set_test_state(struct hv_device *dev, struct dentry *root)
+{
+ struct vmbus_channel *channel = dev->channel;
+ bool *state = &channel->fuzz_testing_state;
+ char *status = "fuzz_test_state";
+ struct dentry *test_state;
+
+ test_state = debugfs_create_file(status, 0644, root,
+ state,
+ &hv_debugfs_state_fops);
+ if (IS_ERR(test_state)) {
+ pr_debug("debugfs_hyperv: file %s not created\n", status);
+ return PTR_ERR(test_state);
+ }
+
+ return 0;
+}
+
+/* Bind hv device to a dentry for debugfs */
+static void hv_debug_set_dir_dentry(struct hv_device *dev, struct dentry *root)
+{
+ if (hv_debug_root)
+ dev->debug_dir = root;
+}
+
+/* Create all test dentry's and names for fuzz testing */
+int hv_debug_add_dev_dir(struct hv_device *dev)
+{
+ const char *device = dev_name(&dev->device);
+ char *delay_name = "delay";
+ struct dentry *delay, *dev_root;
+ int ret;
+
+ if (!IS_ERR(hv_debug_root)) {
+ dev_root = debugfs_create_dir(device, hv_debug_root);
+ if (IS_ERR(dev_root)) {
+ pr_debug("debugfs_hyperv: hyperv/%s/ not created\n",
+ device);
+ return PTR_ERR(dev_root);
+ }
+ hv_debug_set_test_state(dev, dev_root);
+ hv_debug_set_dir_dentry(dev, dev_root);
+ delay = debugfs_create_dir(delay_name, dev_root);
+
+ if (IS_ERR(delay)) {
+ pr_debug("debugfs_hyperv: hyperv/%s/%s/ not created\n",
+ device, delay_name);
+ return PTR_ERR(delay);
+ }
+ ret = hv_debug_delay_files(dev, delay);
+
+ return ret;
+ }
+ pr_debug("debugfs_hyperv: hyperv/ not in root debugfs path\n");
+ return PTR_ERR(hv_debug_root);
+}
+
+/* Remove dentry associated with released hv device */
+void hv_debug_rm_dev_dir(struct hv_device *dev)
+{
+ if (!IS_ERR(hv_debug_root))
+ debugfs_remove_recursive(dev->debug_dir);
+}
+
+/* Remove all dentrys associated with vmbus testing */
+void hv_debug_rm_all_dir(void)
+{
+ debugfs_remove_recursive(hv_debug_root);
+}
+
+/* Delay buffer/message reads on a vmbus channel */
+void hv_debug_delay_test(struct vmbus_channel *channel, enum delay delay_type)
+{
+ struct vmbus_channel *test_channel = channel->primary_channel ?
+ channel->primary_channel :
+ channel;
+ bool state = test_channel->fuzz_testing_state;
+
+ if (state) {
+ if (delay_type == 0)
+ udelay(test_channel->fuzz_testing_interrupt_delay);
+ else
+ udelay(test_channel->fuzz_testing_message_delay);
+ }
+}
+
+/* Initialize top dentry for vmbus testing */
+int hv_debug_init(void)
+{
+ hv_debug_root = debugfs_create_dir("hyperv", NULL);
+ if (IS_ERR(hv_debug_root)) {
+ pr_debug("debugfs_hyperv: hyperv/ not created\n");
+ return PTR_ERR(hv_debug_root);
+ }
+ return 0;
+}
diff --git a/drivers/hv/hyperv_vmbus.h b/drivers/hv/hyperv_vmbus.h
index 362e70e9d145..fee2be064d72 100644
--- a/drivers/hv/hyperv_vmbus.h
+++ b/drivers/hv/hyperv_vmbus.h
@@ -357,4 +357,35 @@ enum hvutil_device_state {
HVUTIL_DEVICE_DYING, /* driver unload is in progress */
};
+enum delay {
+ INTERRUPT_DELAY = 0,
+ MESSAGE_DELAY = 1,
+};
+
+#ifdef CONFIG_HYPERV_TESTING
+
+int hv_debug_add_dev_dir(struct hv_device *dev);
+void hv_debug_rm_dev_dir(struct hv_device *dev);
+void hv_debug_rm_all_dir(void);
+int hv_debug_init(void);
+void hv_debug_delay_test(struct vmbus_channel *channel, enum delay delay_type);
+
+#else /* CONFIG_HYPERV_TESTING */
+
+static inline void hv_debug_rm_dev_dir(struct hv_device *dev) {};
+static inline void hv_debug_rm_all_dir(void) {};
+static inline void hv_debug_delay_test(struct vmbus_channel *channel,
+ enum delay delay_type) {};
+static inline int hv_debug_init(void)
+{
+ return -1;
+}
+
+static inline int hv_debug_add_dev_dir(struct hv_device *dev)
+{
+ return -1;
+}
+
+#endif /* CONFIG_HYPERV_TESTING */
+
#endif /* _HYPERV_VMBUS_H */
diff --git a/drivers/hv/ring_buffer.c b/drivers/hv/ring_buffer.c
index 9a03b163cbbd..356e22159e83 100644
--- a/drivers/hv/ring_buffer.c
+++ b/drivers/hv/ring_buffer.c
@@ -396,6 +396,7 @@ struct vmpacket_descriptor *hv_pkt_iter_first(struct vmbus_channel *channel)
struct hv_ring_buffer_info *rbi = &channel->inbound;
struct vmpacket_descriptor *desc;
+ hv_debug_delay_test(channel, MESSAGE_DELAY);
if (hv_pkt_iter_avail(rbi) < sizeof(struct vmpacket_descriptor))
return NULL;
@@ -421,6 +422,7 @@ __hv_pkt_iter_next(struct vmbus_channel *channel,
u32 packetlen = desc->len8 << 3;
u32 dsize = rbi->ring_datasize;
+ hv_debug_delay_test(channel, MESSAGE_DELAY);
/* bump offset to next potential packet */
rbi->priv_read_index += packetlen + VMBUS_PKT_TRAILER;
if (rbi->priv_read_index >= dsize)
diff --git a/drivers/hv/vmbus_drv.c b/drivers/hv/vmbus_drv.c
index ebd35fc35290..8b2e2352fc6c 100644
--- a/drivers/hv/vmbus_drv.c
+++ b/drivers/hv/vmbus_drv.c
@@ -919,6 +919,8 @@ static void vmbus_device_release(struct device *device)
struct hv_device *hv_dev = device_to_hv_device(device);
struct vmbus_channel *channel = hv_dev->channel;
+ hv_debug_rm_dev_dir(hv_dev);
+
mutex_lock(&vmbus_connection.channel_mutex);
hv_process_channel_removal(channel);
mutex_unlock(&vmbus_connection.channel_mutex);
@@ -1727,6 +1729,7 @@ int vmbus_device_register(struct hv_device *child_device_obj)
pr_err("Unable to register primary channeln");
goto err_kset_unregister;
}
+ hv_debug_add_dev_dir(child_device_obj);
return 0;
@@ -2108,6 +2111,7 @@ static int __init hv_acpi_init(void)
ret = -ETIMEDOUT;
goto cleanup;
}
+ hv_debug_init();
ret = vmbus_bus_init();
if (ret)
@@ -2140,6 +2144,8 @@ static void __exit vmbus_exit(void)
tasklet_kill(&hv_cpu->msg_dpc);
}
+ hv_debug_rm_all_dir();
+
vmbus_free_channels();
if (ms_hyperv.misc_features & HV_FEATURE_GUEST_CRASH_MSR_AVAILABLE) {
diff --git a/include/linux/hyperv.h b/include/linux/hyperv.h
index 6256cc34c4a6..6d815f14d97f 100644
--- a/include/linux/hyperv.h
+++ b/include/linux/hyperv.h
@@ -926,6 +926,21 @@ struct vmbus_channel {
* full outbound ring buffer.
*/
u64 out_full_first;
+
+ /* enabling/disabling fuzz testing on the channel (default is false)*/
+ bool fuzz_testing_state;
+
+ /*
+ * Interrupt delay will delay the guest from emptying the ring buffer
+ * for a specific amount of time. The delay is in microseconds and will
+ * be between 1 to a maximum of 1000, its default is 0 (no delay).
+ * The Message delay will delay guest reading on a per message basis
+ * in microseconds between 1 to 1000 with the default being 0
+ * (no delay).
+ */
+ u32 fuzz_testing_interrupt_delay;
+ u32 fuzz_testing_message_delay;
+
};
static inline bool is_hvsock_channel(const struct vmbus_channel *c)
@@ -1166,6 +1181,10 @@ struct hv_device {
struct vmbus_channel *channel;
struct kset *channels_kset;
+
+ /* place holder to keep track of the dir for hv device in debugfs */
+ struct dentry *debug_dir;
+
};
--
2.17.1
^ permalink raw reply related
* [PATCH v4 0/2] hv: vmbus: add fuzz testing to hv device
From: Branden Bonaby @ 2019-08-29 4:23 UTC (permalink / raw)
To: kys, haiyangz, sthemmin, sashal
Cc: Branden Bonaby, linux-hyperv, linux-kernel
This patchset introduces a testing framework for Hyper-V drivers.
This framework allows us to introduce delays in the packet receive
path on a per-device basis. While the current code only supports
introducing arbitrary delays in the host/guest communication path,
we intend to expand this to support error injection in the future.
changes in v4:
patch 1:
Combined previous v3 patches 1 and 2, into a single patch
which is now patch 1. This was done so that calls to
the new debugfs functions are in the same patch as
the definitions for these functions.
Moved debugfs code from "vmbus_drv.c" that was in
previous v3 patch 2, into a new file "debugfs.c" in
drivers/hv.
Updated the Makefile to compile "debugfs.c" if
CONFIG_HYPERV_TESTING is enabled
As per Michael's comments, added empty implementations
of the new functions, so the compiler will not generate
code when CONFIG_HYPERV_TESTING is not enabled.
patch 2 (was previously v3 patch 3):
Based on Harrys comments, made the tool more
user friendly and added more error checking.
changes in v3:
patch 2: change call to IS_ERR_OR_NULL, to IS_ERR.
patch 3: Align python tool to match Linux coding style.
Changes in v2:
Patch 1: As per Vitaly's suggestion, wrapped the test code under an
#ifdef and updated the Kconfig file, so that the test code
will only be used when the config option is set to true.
(default is false).
Updated hyperv_vmbus header to contain new #ifdef with new
new functions for the test code.
Patch 2: Moved code from under sysfs to debugfs and wrapped it under
the new ifdef.
Updated MAINTAINERS file with new debugfs-hyperv file under
the section for hyperv.
Patch 3: Updated testing tool with new debugfs location.
Branden Bonaby (2):
drivers: hv: vmbus: Introduce latency testing
tools: hv: add vmbus testing tool
Documentation/ABI/testing/debugfs-hyperv | 23 ++
MAINTAINERS | 1 +
drivers/hv/Kconfig | 7 +
drivers/hv/Makefile | 1 +
drivers/hv/connection.c | 1 +
drivers/hv/hv_debugfs.c | 187 +++++++++++
drivers/hv/hyperv_vmbus.h | 31 ++
drivers/hv/ring_buffer.c | 2 +
drivers/hv/vmbus_drv.c | 6 +
include/linux/hyperv.h | 19 ++
tools/hv/vmbus_testing | 376 +++++++++++++++++++++++
11 files changed, 654 insertions(+)
create mode 100644 Documentation/ABI/testing/debugfs-hyperv
create mode 100644 drivers/hv/hv_debugfs.c
create mode 100644 tools/hv/vmbus_testing
--
2.17.1
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox