| 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: Sun, 27 Sep 2020 16:28:30 +0800
From: Jisheng Zhang <Jisheng.Zhang@...aptics.com>
To: Jon Hunter <jonathanh@...dia.com>
Cc: Kunihiko Hayashi <hayashi.kunihiko@...ionext.com>,
Neil Armstrong <narmstrong@...libre.com>,
<linux-pci@...r.kernel.org>,
Binghui Wang <wangbinghui@...ilicon.com>,
Bjorn Andersson <bjorn.andersson@...aro.org>,
Masahiro Yamada <yamada.masahiro@...ionext.com>,
Thierry Reding <thierry.reding@...il.com>,
<linux-arm-kernel@...s.com>, Vidya Sagar <vidyas@...dia.com>,
Fabio Estevam <festevam@...il.com>,
Jerome Brunet <jbrunet@...libre.com>,
Rob Herring <robh@...nel.org>,
Jesper Nilsson <jesper.nilsson@...s.com>,
Lorenzo Pieralisi <lorenzo.pieralisi@....com>,
Kevin Hilman <khilman@...libre.com>,
Pratyush Anand <pratyush.anand@...il.com>,
<linux-tegra@...r.kernel.org>,
"Krzysztof Kozlowski" <krzk@...nel.org>,
Kishon Vijay Abraham I <kishon@...com>,
"Kukjin Kim" <kgene@...nel.org>,
NXP Linux Team <linux-imx@....com>,
Xiaowei Song <songxiaowei@...ilicon.com>,
Richard Zhu <hongxing.zhu@....com>,
"Martin Blumenstingl" <martin.blumenstingl@...glemail.com>,
<linux-arm-msm@...r.kernel.org>,
Sascha Hauer <s.hauer@...gutronix.de>,
"Yue Wang" <yue.wang@...ogic.com>,
<linux-samsung-soc@...r.kernel.org>,
"Bjorn Helgaas" <bhelgaas@...gle.com>,
<linux-amlogic@...ts.infradead.org>, <linux-omap@...r.kernel.org>,
<linux-arm-kernel@...ts.infradead.org>,
"Jingoo Han" <jingoohan1@...il.com>,
Andy Gross <agross@...nel.org>, <linux-kernel@...r.kernel.org>,
Stanimir Varbanov <svarbanov@...sol.com>,
Pengutronix Kernel Team <kernel@...gutronix.de>,
Gustavo Pimentel <gustavo.pimentel@...opsys.com>,
Shawn Guo <shawnguo@...nel.org>,
Lucas Stach <l.stach@...gutronix.de>
Subject: Re: [PATCH v2 0/5] PCI: dwc: improve msi handling
Hi,
On Fri, 25 Sep 2020 16:13:02 +0100 Jon Hunter wrote:
>
> Hi Jisheng,
>
> On 25/09/2020 10:27, Jisheng Zhang wrote:
>
> ...
>
> >> Could you please try below patch?
> >>
> >>
> >> diff --git a/drivers/pci/controller/dwc/pcie-designware-host.c b/drivers/pci/controller/dwc/pcie-designware-host.c
> >> index bf25d783b5c5..7e5dc54d060e 100644
> >> --- a/drivers/pci/controller/dwc/pcie-designware-host.c
> >> +++ b/drivers/pci/controller/dwc/pcie-designware-host.c
> >> @@ -197,7 +197,6 @@ static struct irq_chip dw_pci_msi_bottom_irq_chip = {
> >> .name = "DWPCI-MSI",
> >> .irq_ack = dw_pci_bottom_ack,
> >> .irq_compose_msi_msg = dw_pci_setup_msi_msg,
> >> - .irq_set_affinity = dw_pci_msi_set_affinity,
> >> .irq_mask = dw_pci_bottom_mask,
> >> .irq_unmask = dw_pci_bottom_unmask,
> >> };
> >
> > A complete patch w/o compiler warning:
> >
> > diff --git a/drivers/pci/controller/dwc/pcie-designware-host.c b/drivers/pci/controller/dwc/pcie-designware-host.c
> > index bf25d783b5c5..18f719cfed0b 100644
> > --- a/drivers/pci/controller/dwc/pcie-designware-host.c
> > +++ b/drivers/pci/controller/dwc/pcie-designware-host.c
> > @@ -137,12 +137,6 @@ static void dw_pci_setup_msi_msg(struct irq_data *d, struct msi_msg *msg)
> > (int)d->hwirq, msg->address_hi, msg->address_lo);
> > }
> >
> > -static int dw_pci_msi_set_affinity(struct irq_data *d,
> > - const struct cpumask *mask, bool force)
> > -{
> > - return -EINVAL;
> > -}
> > -
> > static void dw_pci_bottom_mask(struct irq_data *d)
> > {
> > struct pcie_port *pp = irq_data_get_irq_chip_data(d);
> > @@ -197,7 +191,6 @@ static struct irq_chip dw_pci_msi_bottom_irq_chip = {
> > .name = "DWPCI-MSI",
> > .irq_ack = dw_pci_bottom_ack,
> > .irq_compose_msi_msg = dw_pci_setup_msi_msg,
> > - .irq_set_affinity = dw_pci_msi_set_affinity,
> > .irq_mask = dw_pci_bottom_mask,
> > .irq_unmask = dw_pci_bottom_unmask,
> > };
> >
>
>
> Thanks I was not expecting this to work because ...
>
> int irq_do_set_affinity(struct irq_data *data, const struct cpumask *mask,
> bool force)
> {
> struct irq_desc *desc = irq_data_to_desc(data);
> struct irq_chip *chip = irq_data_get_irq_chip(data);
> int ret;
>
> if (!chip || !chip->irq_set_affinity)
> return -EINVAL;
>
> However, with your patch Tegra crashes on boot ...
>
> [ 11.613853] Unable to handle kernel NULL pointer dereference at virtual address 0000000000000000
> [ 11.622500] Mem abort info:
> [ 11.622515] ESR = 0x86000004
> [ 11.622524] EC = 0x21: IABT (current EL), IL = 32 bits
> [ 11.622540] SET = 0, FnV = 0
> [ 11.636544] EA = 0, S1PTW = 0
> [ 11.636554] user pgtable: 4k pages, 48-bit VAs, pgdp=000000046a28e000
> [ 11.636559] [0000000000000000] pgd=0000000000000000, p4d=0000000000000000
> [ 11.652652] Internal error: Oops: 86000004 [#1] PREEMPT SMP
> [ 11.652658] Modules linked in: pwm_tegra phy_tegra194_p2u crct10dif_ce lm90 pwm_fan tegra_bpmp_thermal pcie_tegra194 ip_tables x_tables ipv6
> [ 11.670525] CPU: 3 PID: 138 Comm: kworker/3:3 Not tainted 5.9.0-rc4-dirty #12
> [ 11.670534] Hardware name: NVIDIA Jetson AGX Xavier Developer Kit (DT)
> [ 11.683967] Workqueue: events deferred_probe_work_func
> [ 11.683974] pstate: 60c00089 (nZCv daIf +PAN +UAO BTYPE=--)
> [ 11.683985] pc : 0x0
> [ 11.696669] lr : msi_domain_set_affinity+0x44/0xc0
> [ 11.696672] sp : ffff800012bcb390
> [ 11.696680] x29: ffff800012bcb390 x28: ffff0003e3033c20
> [ 11.709891] x27: ffff0003e76cfe58 x26: 0000000000000000
> [ 11.709900] x25: ffff800011d7e850 x24: ffff800011d7e878
> [ 11.709908] x23: 0000000000000000 x22: ffff0003e76cfe00
> [ 11.709914] x21: ffff0003e76cfe58 x20: ffff0003e76cfe58
> [ 11.709921] x19: ffff800011b19000 x18: ffffffffffffffff
> [ 11.709927] x17: 0000000000000000 x16: 0000000000000000
> [ 11.741262] x15: ffff800011b19948 x14: 0000000000000040
> [ 11.741267] x13: 0000000000000228 x12: 0000000000000030
> [ 11.741272] x11: 0101010101010101 x10: 0000000000000040
> [ 11.741277] x9 : 0000000000000000 x8 : 0000000000000004
> [ 11.741281] x7 : ffffffffffffffff x6 : 00000000000000ff
> [ 11.767374] x5 : 0000000000000000 x4 : 0000000000000000
> [ 11.767379] x3 : 0000000000000000 x2 : 0000000000000000
> [ 11.767384] x1 : ffff800011d7e898 x0 : ffff0003e262bf00
> [ 11.767406] Call trace:
> [ 11.767410] 0x0
> [ 11.767424] irq_do_set_affinity+0x4c/0x178
> [ 11.791400] irq_setup_affinity+0x124/0x1b0
> [ 11.791423] irq_startup+0x6c/0x118
> [ 11.791434] __setup_irq+0x810/0x8a0
> [ 11.802510] request_threaded_irq+0xdc/0x188
> [ 11.802517] pcie_pme_probe+0x98/0x110
> [ 11.802536] pcie_port_probe_service+0x34/0x60
> [ 11.814799] really_probe+0x110/0x400
> [ 11.814809] driver_probe_device+0x54/0xb8
> [ 11.822438] __device_attach_driver+0x90/0xc0
> [ 11.822463] bus_for_each_drv+0x70/0xc8
> [ 11.822471] __device_attach+0xec/0x150
> [ 11.834307] device_initial_probe+0x10/0x18
> [ 11.834311] bus_probe_device+0x94/0xa0
> [ 11.834315] device_add+0x464/0x730
> [ 11.834338] device_register+0x1c/0x28
> [ 11.834349] pcie_port_device_register+0x2d0/0x3e8
> [ 11.854056] pcie_portdrv_probe+0x34/0xd8
> [ 11.854063] local_pci_probe+0x3c/0xa0
> [ 11.854088] pci_device_probe+0x128/0x1c8
> [ 11.854103] really_probe+0x110/0x400
> [ 11.869283] driver_probe_device+0x54/0xb8
> [ 11.869311] __device_attach_driver+0x90/0xc0
> [ 11.877638] bus_for_each_drv+0x70/0xc8
> [ 11.877645] __device_attach+0xec/0x150
> [ 11.877669] device_attach+0x10/0x18
> [ 11.877680] pci_bus_add_device+0x4c/0xb0
> [ 11.892642] pci_bus_add_devices+0x44/0x90
> [ 11.892646] dw_pcie_host_init+0x370/0x4f8
> [ 11.892653] tegra_pcie_dw_probe+0x5e8/0xb50 [pcie_tegra194]
> [ 11.892661] platform_drv_probe+0x50/0xa8
> [ 11.910179] really_probe+0x110/0x400
> [ 11.910183] driver_probe_device+0x54/0xb8
> [ 11.910186] __device_attach_driver+0x90/0xc0
> [ 11.910213] bus_for_each_drv+0x70/0xc8
> [ 11.910240] __device_attach+0xec/0x150
> [ 11.929689] device_initial_probe+0x10/0x18
> [ 11.929694] bus_probe_device+0x94/0xa0
> [ 11.929719] deferred_probe_work_func+0x6c/0xa0
> [ 11.929730] process_one_work+0x1cc/0x360
> [ 11.946008] worker_thread+0x48/0x450
> [ 11.949602] kthread+0x120/0x150
> [ 11.952803] ret_from_fork+0x10/0x1c
> [ 11.956332] Code: bad PC value
> [ 11.959360] ---[ end trace 03c30e252fe4e40b ]---
>
> To be honest, I am not sure I completely understand why it crashes here.
>
I see, the msi_domain_set_affinity() calls parent->chip->irq_set_affinity
without checking, grepping the irqchip and pci dir, I found that
if the MSI is based on some cascaded interrupt mechanism, they all
point the irq_set_affinity to irq_chip_set_affinity_parent(), so I believe
below patch works:
diff --git a/drivers/pci/controller/dwc/pcie-designware-host.c b/drivers/pci/controller/dwc/pcie-designware-host.c
index bf25d783b5c5..093fba616736 100644
--- a/drivers/pci/controller/dwc/pcie-designware-host.c
+++ b/drivers/pci/controller/dwc/pcie-designware-host.c
@@ -137,12 +137,6 @@ static void dw_pci_setup_msi_msg(struct irq_data *d, struct msi_msg *msg)
(int)d->hwirq, msg->address_hi, msg->address_lo);
}
-static int dw_pci_msi_set_affinity(struct irq_data *d,
- const struct cpumask *mask, bool force)
-{
- return -EINVAL;
-}
-
static void dw_pci_bottom_mask(struct irq_data *d)
{
struct pcie_port *pp = irq_data_get_irq_chip_data(d);
@@ -197,7 +191,7 @@ static struct irq_chip dw_pci_msi_bottom_irq_chip = {
.name = "DWPCI-MSI",
.irq_ack = dw_pci_bottom_ack,
.irq_compose_msi_msg = dw_pci_setup_msi_msg,
- .irq_set_affinity = dw_pci_msi_set_affinity,
+ .irq_set_affinity = irq_chip_set_affinity_parent,
.irq_mask = dw_pci_bottom_mask,
.irq_unmask = dw_pci_bottom_unmask,
};
Powered by blists - more mailing lists