* Re: [PATCH v8 1/2] powerpc/pseries: Interface to represent PAPR firmware attributes
From: Michael Ellerman @ 2021-09-29 13:49 UTC (permalink / raw)
To: Pratik R. Sampat, benh, paulus, shuah, farosas, kjain,
linuxppc-dev, kvm-ppc, linux-kselftest, linux-kernel, psampat,
pratik.r.sampat
In-Reply-To: <20210928115102.57117-2-psampat@linux.ibm.com>
Hi Pratik,
Some comments inline below ...
"Pratik R. Sampat" <psampat@linux.ibm.com> writes:
> Adds a generic interface to represent the energy and frequency related
> PAPR attributes on the system using the new H_CALL
> "H_GET_ENERGY_SCALE_INFO".
In Linux "generic" would usually mean something that is not architecture
specific, but this is architecture specific.
It's also not generic across different types of attributes, it's only
for information from this particular hcall (right?).
I think you mean generic in contrast to lparcfg, which I guess makes
some sense, this is more generic than lparcfg :)
But I think it'd be better to just say "a sysfs interface".
> H_GET_EM_PARMS H_CALL was previously responsible for exporting this
> information in the lparcfg, however the H_GET_EM_PARMS H_CALL
> will be deprecated P10 onwards.
>
> The H_GET_ENERGY_SCALE_INFO H_CALL is of the following call format:
> hcall(
> uint64 H_GET_ENERGY_SCALE_INFO, // Get energy scale info
> uint64 flags, // Per the flag request
> uint64 firstAttributeId,// The attribute id
> uint64 bufferAddress, // Guest physical address of the output buffer
> uint64 bufferSize // The size in bytes of the output buffer
> );
As specified in PAPR+ v2.11, section 14.14.3.
> This H_CALL can query either all the attributes at once with
> firstAttributeId = 0, flags = 0 as well as query only one attribute
> at a time with firstAttributeId = id, flags = 1.
>
> The output buffer consists of the following
> 1. number of attributes - 8 bytes
> 2. array offset to the data location - 8 bytes
> 3. version info - 1 byte
(possible future expansion here)
> 4. A data array of size num attributes, which contains the following:
> a. attribute ID - 8 bytes
> b. attribute value in number - 8 bytes
> c. attribute name in string - 64 bytes
> d. attribute value in string - 64 bytes
>
> The new H_CALL exports information in direct string value format, hence
> a new interface has been introduced in
> /sys/firmware/papr/energy_scale_info to export this information to
> userspace in an extensible pass-through format.
I'm not really sure what "extensible pass-through format" means? Do you
mean the kernel doesn't interpret the values, so firmware can add new
values and they will just appear without any kernel changes required?
> The H_CALL returns the name, numeric value and string value (if exists)
>
> The format of exposing the sysfs information is as follows:
> /sys/firmware/papr/energy_scale_info/
> |-- <id>/
> |-- desc
> |-- value
> |-- value_desc (if exists)
> |-- <id>/
> |-- desc
> |-- value
> |-- value_desc (if exists)
> ...
>
> The energy information that is exported is useful for userspace tools
> such as powerpc-utils. Currently these tools infer the
> "power_mode_data" value in the lparcfg, which in turn is obtained from
> the to be deprecated H_GET_EM_PARMS H_CALL.
> On future platforms, such userspace utilities will have to look at the
> data returned from the new H_CALL being populated in this new sysfs
> interface and report this information directly without the need of
> interpretation.
I don't see anything here or in PAPR about how/if the values change over
time.
Are we expected to always read the values from the hypervisor, or are we
allowed to cache them for any period of time?
The reason I ask is currently we're doing 3 hcalls for each attribute,
once for desc, value and value_desc. Is there any issue with the
consistency of the values given we're doing 3 separate calls?
Or are the values static? Or static except for LPM? In which case we
could read them once at boot/LPM. Or maybe we not know if/how the values
will change, especially in future?
> diff --git a/Documentation/ABI/testing/sysfs-firmware-papr-energy-scale-info b/Documentation/ABI/testing/sysfs-firmware-papr-energy-scale-info
> new file mode 100644
> index 000000000000..139a576c7c9d
> --- /dev/null
> +++ b/Documentation/ABI/testing/sysfs-firmware-papr-energy-scale-info
> @@ -0,0 +1,26 @@
> +What: /sys/firmware/papr/energy_scale_info
> +Date: June 2021
> +Contact: Linux for PowerPC mailing list <linuxppc-dev@ozlabs.org>
> +Description: Directory hosting a set of platform attributes like
> + energy/frequency on Linux running as a PAPR guest.
> +
> + Each file in a directory contains a platform
> + attribute hierarchy pertaining to performance/
> + energy-savings mode and processor frequency.
> +
> +What: /sys/firmware/papr/energy_scale_info/<id>
> + /sys/firmware/papr/energy_scale_info/<id>/desc
> + /sys/firmware/papr/energy_scale_info/<id>/value
> + /sys/firmware/papr/energy_scale_info/<id>/value_desc
> +Date: June 2021
> +Contact: Linux for PowerPC mailing list <linuxppc-dev@ozlabs.org>
> +Description: Energy, frequency attributes directory for POWERVM servers
> +
> + This directory provides energy, frequency, folding information. It
> + contains below sysfs attributes:
> +
> + - desc: String description of the attribute <id>
> +
> + - value: Numeric value of attribute <id>
> +
> + - value_desc: String value of attribute <id>
> diff --git a/arch/powerpc/include/asm/hvcall.h b/arch/powerpc/include/asm/hvcall.h
> index 9bcf345cb208..38980fef7a3d 100644
> --- a/arch/powerpc/include/asm/hvcall.h
> +++ b/arch/powerpc/include/asm/hvcall.h
> @@ -323,7 +323,8 @@
> #define H_SCM_PERFORMANCE_STATS 0x418
> #define H_RPT_INVALIDATE 0x448
> #define H_SCM_FLUSH 0x44C
> -#define MAX_HCALL_OPCODE H_SCM_FLUSH
> +#define H_GET_ENERGY_SCALE_INFO 0x450
> +#define MAX_HCALL_OPCODE H_GET_ENERGY_SCALE_INFO
>
> /* Scope args for H_SCM_UNBIND_ALL */
> #define H_UNBIND_SCOPE_ALL (0x1)
> @@ -641,6 +642,27 @@ struct hv_gpci_request_buffer {
> uint8_t bytes[HGPCI_MAX_DATA_BYTES];
> } __packed;
>
None of the following needs to be in this header AFAICS. It can all go
in the C file.
> +#define ESI_VERSION 0x1
> +#define MAX_ESI_ATTRS 10
Where does that max come from? I don't see it mentioned in PAPR.
PAPR implies any number of attributes up to 2^64 can be returned, and if
the buffer is too small H_PARTIAL is returned.
We really mustn't hard code a limit of 10 attributes. We need a loop that
reads N attributes, and if we receive H_PARTIAL, we repeat the hcall,
until we've received all the attributes.
To test that is working on an existing system, which presumably only has
a small number of attributes, you should shrink the buffer to the size
of 1 attribute, to make sure the loop logic works.
> +#define MAX_BUF_SZ (sizeof(struct h_energy_scale_info_hdr) + \
> + (sizeof(struct energy_scale_attribute) * MAX_ESI_ATTRS))
>
> +struct energy_scale_attribute {
> + __be64 id;
> + __be64 value;
> + unsigned char desc[64];
> + unsigned char value_desc[64];
I prefer u8.
> +} __packed;
> +
> +struct h_energy_scale_info_hdr {
> + __be64 num_attrs;
> + __be64 array_offset;
> + __u8 data_header_version;
You can use u8 here, __u8 is only needed in uapi headers.
> +} __packed;
So because of array_offset, we don't actually know what the size of the
header plus a single attribute will be.
Because the header tells us the offset to the first attribute, when the
header expands in a future version of the hcall, sizeof(hdr) +
sizeof(energy_scale_attribute) will not necessarily be large enough to
fit a single attribute. I guess maybe that's why you're always
allocating space for 10 of them (MAX_BUF_SZ).
To be truly forward compatible we should do the hcall at startup with
the current known size of the structs, and if we receive H_P4 try again
with increasing sizes until it succeeds.
That's probably overly complex though, given we'd probably only expect
the header to grow by a few 10s of bytes in future.
Maybe we should just allocate 2KB for the buffer, which is still small
as far as allocations go, but gives us a huge amount of headroom for the
header to grow.
> +/* /sys/firmware/papr */
> +extern struct kobject *papr_kobj;
> +
> #endif /* __ASSEMBLY__ */
> #endif /* __KERNEL__ */
> #endif /* _ASM_POWERPC_HVCALL_H */
> diff --git a/arch/powerpc/kvm/trace_hv.h b/arch/powerpc/kvm/trace_hv.h
> index 830a126e095d..38cd0ed0a617 100644
> --- a/arch/powerpc/kvm/trace_hv.h
> +++ b/arch/powerpc/kvm/trace_hv.h
> @@ -115,6 +115,7 @@
> {H_VASI_STATE, "H_VASI_STATE"}, \
> {H_ENABLE_CRQ, "H_ENABLE_CRQ"}, \
> {H_GET_EM_PARMS, "H_GET_EM_PARMS"}, \
> + {H_GET_ENERGY_SCALE_INFO, "H_GET_ENERGY_SCALE_INFO"}, \
> {H_SET_MPP, "H_SET_MPP"}, \
> {H_GET_MPP, "H_GET_MPP"}, \
> {H_HOME_NODE_ASSOCIATIVITY, "H_HOME_NODE_ASSOCIATIVITY"}, \
> diff --git a/arch/powerpc/platforms/pseries/Makefile b/arch/powerpc/platforms/pseries/Makefile
> index 4cda0ef87be0..c4c19f6a5975 100644
> --- a/arch/powerpc/platforms/pseries/Makefile
> +++ b/arch/powerpc/platforms/pseries/Makefile
> @@ -6,7 +6,8 @@ obj-y := lpar.o hvCall.o nvram.o reconfig.o \
> of_helpers.o \
> setup.o iommu.o event_sources.o ras.o \
> firmware.o power.o dlpar.o mobility.o rng.o \
> - pci.o pci_dlpar.o eeh_pseries.o msi.o
> + pci.o pci_dlpar.o eeh_pseries.o msi.o \
> + papr_platform_attributes.o
I don't really mind, but is there a reason we're calling it
papr_platform_attributes.c and not energy_info.c or something?
Is the plan to support more attributes in future?
> obj-$(CONFIG_SMP) += smp.o
> obj-$(CONFIG_SCANLOG) += scanlog.o
> obj-$(CONFIG_KEXEC_CORE) += kexec.o
> diff --git a/arch/powerpc/platforms/pseries/papr_platform_attributes.c b/arch/powerpc/platforms/pseries/papr_platform_attributes.c
> new file mode 100644
> index 000000000000..84ddce52e519
> --- /dev/null
> +++ b/arch/powerpc/platforms/pseries/papr_platform_attributes.c
> @@ -0,0 +1,312 @@
> +// SPDX-License-Identifier: GPL-2.0-or-later
> +/*
> + * Platform energy and frequency attributes driver
> + *
> + * This driver creates a sys file at /sys/firmware/papr/ which encapsulates a
> + * directory structure containing files in keyword - value pairs that specify
> + * energy and frequency configuration of the system.
> + *
> + * The format of exposing the sysfs information is as follows:
> + * /sys/firmware/papr/energy_scale_info/
> + * |-- <id>/
> + * |-- desc
> + * |-- value
> + * |-- value_desc (if exists)
> + * |-- <id>/
> + * |-- desc
> + * |-- value
> + * |-- value_desc (if exists)
> + *
> + * Copyright 2021 IBM Corp.
> + */
> +
> +#include <asm/hvcall.h>
> +#include <asm/machdep.h>
> +
> +#include "pseries.h"
> +
> +/*
> + * Flag attributes to fetch either all or one attribute from the HCALL
> + * flag = BE(0) => fetch all attributes with firstAttributeId = 0
> + * flag = BE(1) => fetch a single attribute with firstAttributeId = id
> + */
> +#define ESI_FLAGS_ALL 0
> +#define ESI_FLAGS_SINGLE PPC_BIT(0)
I dislike PPC_BIT(). It's obscure and not helpful for people reading the
code compared to a simple (1ull << 63).
> +
> +#define MAX_ATTRS 3
> +
> +struct papr_attr {
> + u64 id;
> + struct kobj_attribute kobj_attr;
> +};
> +struct papr_group {
> + struct attribute_group pg;
> + struct papr_attr pgattrs[MAX_ATTRS];
> +} *pgs;
pgs is not a great name for a global.
Can it be static?
Also would be more normal to define the structure separately from
declaring *pgs.
> +
> +/* /sys/firmware/papr */
> +struct kobject *papr_kobj;
static?
> +/* /sys/firmware/papr/energy_scale_info */
> +struct kobject *esi_kobj;
static?
The following three functions are almost exactly the same, except they
print a different field with a different format string. Seems like there
could be some shared code.
> +/*
> + * Extract and export the description of the energy scale attributes
> + */
> +static ssize_t papr_show_desc(struct kobject *kobj,
> + struct kobj_attribute *kobj_attr,
> + char *buf)
> +{
> + struct papr_attr *pattr = container_of(kobj_attr, struct papr_attr,
> + kobj_attr);
> + struct h_energy_scale_info_hdr *t_hdr;
> + struct energy_scale_attribute *t_esi;
What's the "t_", temp? I don't really like it. hdr and esi would be fine IMHO.
> + char *t_buf;
> + int ret = 0;
That's initialisation is pointless, you override it below.
> + t_buf = kmalloc(MAX_BUF_SZ, GFP_KERNEL);
> + if (t_buf == NULL)
> + return -ENOMEM;
> +
> + ret = plpar_hcall_norets(H_GET_ENERGY_SCALE_INFO, ESI_FLAGS_SINGLE,
> + pattr->id, virt_to_phys(t_buf),
> + MAX_BUF_SZ);
> +
> + if (ret != H_SUCCESS) {
> + pr_warn("hcall failed: H_GET_ENERGY_SCALE_INFO");
> + goto out;
Here you're going to return ret back to userspace. But ret currently
holds a hcall return value, hcall return values are not necessarily
valid Linux error codes.
eg. H_PARTIAL is 5, H_BUSY is 1, etc. Those are potentially read as
success by your caller.
You must never return a hcall error back to Linux code that's not
expecting it.
Here it should probably just return -EIO I guess.
> + }
> +
> + t_hdr = (struct h_energy_scale_info_hdr *) t_buf;
> + t_esi = (struct energy_scale_attribute *)
> + (t_buf + be64_to_cpu(t_hdr->array_offset));
You should check that array_offset + sizeof(attribute) doesn't point
past the end of your buffer.
> +
> + ret = snprintf(buf, sizeof(t_esi->desc), "%s\n", t_esi->desc);
> + if (ret < 0)
> + ret = -EIO;
> +out:
> + kfree(t_buf);
> +
> + return ret;
> +}
> +
> +/*
> + * Extract and export the numeric value of the energy scale attributes
> + */
> +static ssize_t papr_show_value(struct kobject *kobj,
> + struct kobj_attribute *kobj_attr,
> + char *buf)
> +{
> + struct papr_attr *pattr = container_of(kobj_attr, struct papr_attr,
> + kobj_attr);
> + struct h_energy_scale_info_hdr *t_hdr;
> + struct energy_scale_attribute *t_esi;
> + char *t_buf;
> + int ret = 0;
> +
> + t_buf = kmalloc(MAX_BUF_SZ, GFP_KERNEL);
> + if (t_buf == NULL)
> + return -ENOMEM;
> +
> + ret = plpar_hcall_norets(H_GET_ENERGY_SCALE_INFO, ESI_FLAGS_SINGLE,
> + pattr->id, virt_to_phys(t_buf),
> + MAX_BUF_SZ);
> +
> + if (ret != H_SUCCESS) {
> + pr_warn("hcall failed: H_GET_ENERGY_SCALE_INFO");
> + goto out;
> + }
> +
> + t_hdr = (struct h_energy_scale_info_hdr *) t_buf;
> + t_esi = (struct energy_scale_attribute *)
> + (t_buf + be64_to_cpu(t_hdr->array_offset));
> +
> + ret = snprintf(buf, sizeof(t_esi->value), "%llu\n",
> + be64_to_cpu(t_esi->value));
> + if (ret < 0)
> + ret = -EIO;
> +out:
> + kfree(t_buf);
> +
> + return ret;
> +}
> +
> +/*
> + * Extract and export the value description in string format of the energy
> + * scale attributes
> + */
> +static ssize_t papr_show_value_desc(struct kobject *kobj,
> + struct kobj_attribute *kobj_attr,
> + char *buf)
> +{
> + struct papr_attr *pattr = container_of(kobj_attr, struct papr_attr,
> + kobj_attr);
> + struct h_energy_scale_info_hdr *t_hdr;
> + struct energy_scale_attribute *t_esi;
> + char *t_buf;
> + int ret = 0;
> +
> + t_buf = kmalloc(MAX_BUF_SZ, GFP_KERNEL);
> + if (t_buf == NULL)
> + return -ENOMEM;
> +
> + ret = plpar_hcall_norets(H_GET_ENERGY_SCALE_INFO, ESI_FLAGS_SINGLE,
> + pattr->id, virt_to_phys(t_buf),
> + MAX_BUF_SZ);
> +
> + if (ret != H_SUCCESS) {
> + pr_warn("hcall failed: H_GET_ENERGY_SCALE_INFO");
> + goto out;
> + }
> +
> + t_hdr = (struct h_energy_scale_info_hdr *) t_buf;
> + t_esi = (struct energy_scale_attribute *)
> + (t_buf + be64_to_cpu(t_hdr->array_offset));
> +
> + ret = snprintf(buf, sizeof(t_esi->value_desc), "%s\n",
> + t_esi->value_desc);
> + if (ret < 0)
> + ret = -EIO;
> +out:
> + kfree(t_buf);
> +
> + return ret;
> +}
> +
> +static struct papr_ops_info {
> + const char *attr_name;
> + ssize_t (*show)(struct kobject *kobj, struct kobj_attribute *kobj_attr,
> + char *buf);
> +} ops_info[MAX_ATTRS] = {
> + { "desc", papr_show_desc },
> + { "value", papr_show_value },
> + { "value_desc", papr_show_value_desc },
> +};
> +
> +static void add_attr(u64 id, int index, struct papr_attr *attr)
> +{
> + attr->id = id;
> + sysfs_attr_init(&attr->kobj_attr.attr);
> + attr->kobj_attr.attr.name = ops_info[index].attr_name;
> + attr->kobj_attr.attr.mode = 0444;
> + attr->kobj_attr.show = ops_info[index].show;
> +}
> +
> +static int add_attr_group(u64 id, struct papr_group *pg, bool show_val_desc)
> +{
> + int i;
> +
> + for (i = 0; i < MAX_ATTRS; i++) {
> + if (!strcmp(ops_info[i].attr_name, "value_desc") &&
> + !show_val_desc) {
> + continue;
> + }
> + add_attr(id, i, &pg->pgattrs[i]);
> + pg->pg.attrs[i] = &pg->pgattrs[i].kobj_attr.attr;
> + }
> +
> + return sysfs_create_group(esi_kobj, &pg->pg);
> +}
> +
> +static int __init papr_init(void)
> +{
> + struct h_energy_scale_info_hdr *esi_hdr;
> + struct energy_scale_attribute *esi_attrs;
> + uint64_t num_attrs;
> + int ret, idx, i;
> + char *esi_buf;
> +
> + if (!firmware_has_feature(FW_FEATURE_LPAR))
> + return -ENXIO;
> +
> + esi_buf = kmalloc(MAX_BUF_SZ, GFP_KERNEL);
> + if (esi_buf == NULL)
> + return -ENOMEM;
> + /*
> + * hcall(
> + * uint64 H_GET_ENERGY_SCALE_INFO, // Get energy scale info
> + * uint64 flags, // Per the flag request
> + * uint64 firstAttributeId, // The attribute id
> + * uint64 bufferAddress, // Guest physical address of the output buffer
> + * uint64 bufferSize); // The size in bytes of the output buffer
> + */
> + ret = plpar_hcall_norets(H_GET_ENERGY_SCALE_INFO, ESI_FLAGS_ALL, 0,
> + virt_to_phys(esi_buf), MAX_BUF_SZ);
> + if (ret != H_SUCCESS) {
> + pr_warn("hcall failed: H_GET_ENERGY_SCALE_INFO");
> + goto out;
This prints a warning on all existing systems (that don't support the
new hcall), which is not ideal :)
The proper way to handle discovery of the hcall is by looking at
ibm,hypertas-functions and setting a FW_FEATURE flag based on that.
See fw_hypertas_feature_init().
> + }
Also the hcall can return H_PARTIAL if the buffer isn't big enough.
> + esi_hdr = (struct h_energy_scale_info_hdr *) esi_buf;
> + if (esi_hdr->data_header_version != ESI_VERSION) {
> + pr_warn("H_GET_ENERGY_SCALE_INFO VER MISMATCH - EXP: 0x%x, REC: 0x%x",
> + ESI_VERSION, esi_hdr->data_header_version);
That shouldn't be a warning. PAPR says any changes will be backward
compatible. If we need to check the version at all, it would be to make
sure the hypervisor version is >= the version we expect. But I'm not
sure it's necessary to check at all?
> + }
> +
> + num_attrs = be64_to_cpu(esi_hdr->num_attrs);
> + esi_attrs = (struct energy_scale_attribute *)
> + (esi_buf + be64_to_cpu(esi_hdr->array_offset));
> +
> + pgs = kcalloc(num_attrs, sizeof(*pgs), GFP_KERNEL);
> + if (!pgs)
> + goto out;
> +
> + papr_kobj = kobject_create_and_add("papr", firmware_kobj);
> + if (!papr_kobj) {
> + pr_warn("kobject_create_and_add papr failed\n");
> + goto out_pgs;
> + }
> +
> + esi_kobj = kobject_create_and_add("energy_scale_info", papr_kobj);
> + if (!esi_kobj) {
> + pr_warn("kobject_create_and_add energy_scale_info failed\n");
> + goto out_kobj;
> + }
> +
> + for (idx = 0; idx < num_attrs; idx++) {
> + bool show_val_desc = true;
> +
> + pgs[idx].pg.attrs = kcalloc(MAX_ATTRS + 1,
> + sizeof(*pgs[idx].pg.attrs),
> + GFP_KERNEL);
> + if (!pgs[idx].pg.attrs) {
> + goto out_pgattrs;
> + }
> +
> + pgs[idx].pg.name = kasprintf(GFP_KERNEL, "%lld",
> + be64_to_cpu(esi_attrs[idx].id));
> + if (pgs[idx].pg.name == NULL) {
> + goto out_pgattrs;
> + }
> + /* Do not add the value description if it does not exist */
> + if (strnlen(esi_attrs[idx].value_desc,
> + sizeof(esi_attrs[idx].value_desc)) == 0)
> + show_val_desc = false;
> +
> + if (add_attr_group(be64_to_cpu(esi_attrs[idx].id), &pgs[idx],
> + show_val_desc)) {
> + pr_warn("Failed to create papr attribute group %s\n",
> + pgs[idx].pg.name);
> + goto out_pgattrs;
> + }
> + }
> +
> + kfree(esi_buf);
> + return 0;
> +
> +out_pgattrs:
Is this right? If you failed the allocation for i = 1, then the attrs
for i = 0 would already be registered with sysfs, and then you free them
while they're still in use?
To simplify it you can probably do all the allocations first before
registering anything.
> + for (i = 0; i < idx ; i++) {
> + kfree(pgs[i].pg.attrs);
> + kfree(pgs[i].pg.name);
> + }
> + kobject_put(esi_kobj);
> +out_kobj:
> + kobject_put(papr_kobj);
> +out_pgs:
> + kfree(pgs);
> +out:
If you're freeing something the label should be "out_free_thing". So in
this case out_free_esi_buf.
> + kfree(esi_buf);
> +
> + return -ENOMEM;
> +}
> +
> +machine_device_initcall(pseries, papr_init);
> --
> 2.31.1
cheers
^ permalink raw reply
* Re: [PATCH v5 10/11] PCI: Replace pci_dev::driver usage by pci_dev::dev.driver
From: Uwe Kleine-König @ 2021-09-29 13:43 UTC (permalink / raw)
To: Andrew Donnellan
Cc: Mark Rutland, Peter Zijlstra, H. Peter Anvin,
Oliver O'Halloran, Jiri Olsa, Christoph Hellwig,
Stefano Stabellini, Arnd Bergmann, Boris Ostrovsky, x86,
Alexander Shishkin, Ingo Molnar, Bjorn Helgaas, linux-pci,
xen-devel, Mathias Nyman, Konrad Rzeszutek Wilk,
Arnaldo Carvalho de Melo, Borislav Petkov, Bjorn Helgaas,
Namhyung Kim, Thomas Gleixner, Juergen Gross, Greg Kroah-Hartman,
linux-usb, linux-perf-users, kernel, Frederic Barrat,
Paul Mackerras, linuxppc-dev
In-Reply-To: <75dd6d60-08b9-fa68-352c-3a0c5a04c0ab@linux.ibm.com>
[-- Attachment #1: Type: text/plain, Size: 1935 bytes --]
On Wed, Sep 29, 2021 at 11:15:37PM +1000, Andrew Donnellan wrote:
> On 29/9/21 6:53 pm, Uwe Kleine-König wrote:>
> list_for_each_entry(afu_dev, &afu->phb->bus->devices, bus_list) {
> > - if (afu_dev->driver && afu_dev->driver->err_handler &&
> > - afu_dev->driver->err_handler->resume)
> > - afu_dev->driver->err_handler->resume(afu_dev);
> > + struct pci_driver *afu_drv;
> > + if (afu_dev->dev.driver &&
> > + (afu_drv = to_pci_driver(afu_dev->dev.driver))->err_handler &&
>
> I'm not a huge fan of assignments in if statements and if you send a v6 I'd
> prefer you break it up.
I'm not a huge fan either, I used it to keep the control flow as is and
without introducing several calls to to_pci_driver.
The whole code looks as follows:
list_for_each_entry(afu_dev, &afu->phb->bus->devices, bus_list) {
struct pci_driver *afu_drv;
if (afu_dev->dev.driver &&
(afu_drv = to_pci_driver(afu_dev->dev.driver))->err_handler &&
afu_drv->err_handler->resume)
afu_drv->err_handler->resume(afu_dev);
}
Without assignment in the if it could look as follows:
list_for_each_entry(afu_dev, &afu->phb->bus->devices, bus_list) {
struct pci_driver *afu_drv;
if (!afu_dev->dev.driver)
continue;
afu_drv = to_pci_driver(afu_dev->dev.driver));
if (afu_drv->err_handler && afu_drv->err_handler->resume)
afu_drv->err_handler->resume(afu_dev);
}
Fine for me.
(Sidenote: What happens if the device is unbound directly after the
check for afu_dev->dev.driver? This is a problem the old code had, too
(assuming it is a real problem, didn't check deeply).)
> Apart from that everything in the powerpc and cxl sections looks good to me.
Thanks for checking.
Best regards
Uwe
--
Pengutronix e.K. | Uwe Kleine-König |
Industrial Linux Solutions | https://www.pengutronix.de/ |
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 484 bytes --]
^ permalink raw reply
* Re: [PATCH v5 10/11] PCI: Replace pci_dev::driver usage by pci_dev::dev.driver
From: Andrew Donnellan @ 2021-09-29 13:15 UTC (permalink / raw)
To: Uwe Kleine-König, Bjorn Helgaas
Cc: Mark Rutland, Peter Zijlstra, H. Peter Anvin,
Oliver O'Halloran, Jiri Olsa, Christoph Hellwig,
Stefano Stabellini, Mathias Nyman, x86, Alexander Shishkin,
Ingo Molnar, linux-pci, xen-devel, Juergen Gross, Arnd Bergmann,
Konrad Rzeszutek Wilk, Arnaldo Carvalho de Melo, Borislav Petkov,
Bjorn Helgaas, Namhyung Kim, Thomas Gleixner, Boris Ostrovsky,
Greg Kroah-Hartman, linux-usb, linux-perf-users, kernel,
Frederic Barrat, Paul Mackerras, linuxppc-dev
In-Reply-To: <20210929085306.2203850-11-u.kleine-koenig@pengutronix.de>
On 29/9/21 6:53 pm, Uwe Kleine-König wrote:>
list_for_each_entry(afu_dev, &afu->phb->bus->devices, bus_list) {
> - if (afu_dev->driver && afu_dev->driver->err_handler &&
> - afu_dev->driver->err_handler->resume)
> - afu_dev->driver->err_handler->resume(afu_dev);
> + struct pci_driver *afu_drv;
> + if (afu_dev->dev.driver &&
> + (afu_drv = to_pci_driver(afu_dev->dev.driver))->err_handler &&
I'm not a huge fan of assignments in if statements and if you send a v6
I'd prefer you break it up.
Apart from that everything in the powerpc and cxl sections looks good to me.
--
Andrew Donnellan OzLabs, ADL Canberra
ajd@linux.ibm.com IBM Australia Limited
^ permalink raw reply
* [PATCH v5 00/11] PCI: Drop duplicated tracking of a pci_dev's bound driver
From: Uwe Kleine-König @ 2021-09-29 8:52 UTC (permalink / raw)
To: Bjorn Helgaas
Cc: Giovanni Cabiddu, Mark Rutland, Sathya Prakash,
Alexander Shishkin, linux-pci, Alexander Duyck, x86, qat-linux,
oss-drivers, Oliver O'Halloran, H. Peter Anvin, Jiri Olsa,
Christoph Hellwig, Marco Chiappero, Stefano Stabellini,
Herbert Xu, linux-scsi, Rafał Miłecki, Jesse Brandeburg,
Peter Zijlstra, Ingo Molnar, linux-wireless, Jakub Kicinski,
Yisen Zhuang, Suganath Prabu Subramani, Fiona Trahe,
Andrew Donnellan, Arnd Bergmann, Konrad Rzeszutek Wilk,
Ido Schimmel, Simon Horman, linuxppc-dev,
Arnaldo Carvalho de Melo, Jack Xu, Borislav Petkov,
Michael Buesch, Jiri Pirko, Bjorn Helgaas, Namhyung Kim,
Boris Ostrovsky, Andy Shevchenko, Juergen Gross, Salil Mehta,
Sreekanth Reddy, xen-devel, Vadym Kochan, MPT-FusionLinux.pdl,
Greg Kroah-Hartman, linux-usb, Wojciech Ziemba, linux-kernel,
Mathias Nyman, Zhou Wang, Thomas Gleixner, linux-crypto, kernel,
netdev, Frederic Barrat, Paul Mackerras, Tomaszx Kowalik,
Taras Chornyi, David S. Miller, linux-perf-users
Hello,
this is v5 of the quest to drop the "driver" member from struct pci_dev
which tracks the same data (apart from a constant offset) as dev.driver.
Changes since v4:
- split some changes out of "PCI: replace pci_dev::driver usage that
gets the driver name" and reworked them a bit as suggested by Bjorn.
- Add a line break in
drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c to please
checkpatch --strict and Simon Horman.
- Fixed a build problem in "crypto: qat - simplify adf_enable_aer()",
thanks to Giovanni Cabiddu for spotting and a suggested fix.
I didn't replace :: by . as suggested by Bjorn as I'm unsure if his
preference is stronger than mine.
This patch stack survived an allmodconfig build on arm64, m68k, powerpc,
riscv, s390, sparc64 and x86_64 on top of v5.15-rc3.
Best regards
Uwe
Uwe Kleine-König (11):
PCI: Simplify pci_device_remove()
PCI: Drop useless check from pci_device_probe()
xen/pci: Drop some checks that are always true
bcma: simplify reference to the driver's name
powerpc/eeh: Don't use driver member of struct pci_dev and further
cleanups
ssb: Simplify determination of driver name
PCI: Replace pci_dev::driver usage that gets the driver name
scsi: message: fusion: Remove unused parameter of mpt_pci driver's
probe()
crypto: qat - simplify adf_enable_aer()
PCI: Replace pci_dev::driver usage by pci_dev::dev.driver
PCI: Drop duplicated tracking of a pci_dev's bound driver
arch/powerpc/include/asm/ppc-pci.h | 5 --
arch/powerpc/kernel/eeh.c | 8 +++
arch/powerpc/kernel/eeh_driver.c | 10 +--
arch/x86/events/intel/uncore.c | 2 +-
arch/x86/kernel/probe_roms.c | 2 +-
drivers/bcma/host_pci.c | 6 +-
drivers/crypto/hisilicon/qm.c | 2 +-
drivers/crypto/qat/qat_4xxx/adf_drv.c | 7 +--
drivers/crypto/qat/qat_c3xxx/adf_drv.c | 7 +--
drivers/crypto/qat/qat_c62x/adf_drv.c | 7 +--
drivers/crypto/qat/qat_common/adf_aer.c | 10 +--
.../crypto/qat/qat_common/adf_common_drv.h | 3 +-
drivers/crypto/qat/qat_dh895xcc/adf_drv.c | 7 +--
drivers/message/fusion/mptbase.c | 7 +--
drivers/message/fusion/mptbase.h | 2 +-
drivers/message/fusion/mptctl.c | 4 +-
drivers/message/fusion/mptlan.c | 2 +-
drivers/misc/cxl/guest.c | 24 ++++---
drivers/misc/cxl/pci.c | 30 +++++----
.../ethernet/hisilicon/hns3/hns3_ethtool.c | 2 +-
.../ethernet/marvell/prestera/prestera_pci.c | 2 +-
drivers/net/ethernet/mellanox/mlxsw/pci.c | 2 +-
.../ethernet/netronome/nfp/nfp_net_ethtool.c | 3 +-
drivers/pci/iov.c | 25 +++++---
drivers/pci/pci-driver.c | 45 ++++++-------
drivers/pci/pci.c | 4 +-
drivers/pci/pcie/err.c | 36 ++++++-----
drivers/pci/xen-pcifront.c | 63 +++++++++----------
drivers/ssb/pcihost_wrapper.c | 6 +-
drivers/usb/host/xhci-pci.c | 2 +-
include/linux/pci.h | 1 -
31 files changed, 161 insertions(+), 175 deletions(-)
Range-diff against v4:
1: 8d064bbc74b0 = 1: c2b53ab26a6b PCI: Simplify pci_device_remove()
2: 966b49c308b4 = 2: 2c733e1d5186 PCI: Drop useless check from pci_device_probe()
3: ce710d6e8a1b = 3: 547ca5a7aa16 xen/pci: Drop some checks that are always true
-: ------------ > 4: 40eb07353844 bcma: simplify reference to the driver's name
-: ------------ > 5: bab59c1dff6d powerpc/eeh: Don't use driver member of struct pci_dev and further cleanups
-: ------------ > 6: abd70de9782d ssb: Simplify determination of driver name
4: 3e4e6994a59d ! 7: 735845bd26b9 PCI: replace pci_dev::driver usage that gets the driver name
@@ Metadata
Author: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
## Commit message ##
- PCI: replace pci_dev::driver usage that gets the driver name
+ PCI: Replace pci_dev::driver usage that gets the driver name
struct pci_dev::driver holds (apart from a constant offset) the same
data as struct pci_dev::dev->driver. With the goal to remove struct
@@ Commit message
Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
- ## arch/powerpc/include/asm/ppc-pci.h ##
-@@ arch/powerpc/include/asm/ppc-pci.h: void eeh_sysfs_remove_device(struct pci_dev *pdev);
-
- static inline const char *eeh_driver_name(struct pci_dev *pdev)
- {
-- return (pdev && pdev->driver) ? pdev->driver->name : "<null>";
-+ if (pdev) {
-+ const char *drvstr = dev_driver_string(&pdev->dev);
-+
-+ if (strcmp(drvstr, ""))
-+ return drvstr;
-+ }
-+
-+ return "<null>";
- }
-
- #endif /* CONFIG_EEH */
-
- ## drivers/bcma/host_pci.c ##
-@@ drivers/bcma/host_pci.c: static int bcma_host_pci_probe(struct pci_dev *dev,
- if (err)
- goto err_kfree_bus;
-
-- name = dev_name(&dev->dev);
-- if (dev->driver && dev->driver->name)
-- name = dev->driver->name;
-+ name = dev_driver_string(&dev->dev);
-+ if (!strcmp(name, ""))
-+ name = dev_name(&dev->dev);
-+
- err = pci_request_regions(dev, name);
- if (err)
- goto err_pci_disable;
-
## drivers/crypto/hisilicon/qm.c ##
@@ drivers/crypto/hisilicon/qm.c: static int qm_alloc_uacce(struct hisi_qm *qm)
};
@@ drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c: nfp_get_drvinfo(struct nfp
char nsp_version[ETHTOOL_FWVERS_LEN] = {};
- strlcpy(drvinfo->driver, pdev->driver->name, sizeof(drvinfo->driver));
-+ strlcpy(drvinfo->driver, dev_driver_string(&pdev->dev), sizeof(drvinfo->driver));
++ strlcpy(drvinfo->driver, dev_driver_string(&pdev->dev),
++ sizeof(drvinfo->driver));
nfp_net_get_nspinfo(app, nsp_version);
snprintf(drvinfo->fw_version, sizeof(drvinfo->fw_version),
"%s %s %s %s", vnic_version, nsp_version,
-
- ## drivers/ssb/pcihost_wrapper.c ##
-@@ drivers/ssb/pcihost_wrapper.c: static int ssb_pcihost_probe(struct pci_dev *dev,
- err = pci_enable_device(dev);
- if (err)
- goto err_kfree_ssb;
-- name = dev_name(&dev->dev);
-- if (dev->driver && dev->driver->name)
-- name = dev->driver->name;
-+
-+ name = dev_driver_string(&dev->dev);
-+ if (*name == '\0')
-+ name = dev_name(&dev->dev);
-+
- err = pci_request_regions(dev, name);
- if (err)
- goto err_pci_disable;
5: 574b743327b8 = 8: 1e58019165b9 scsi: message: fusion: Remove unused parameter of mpt_pci driver's probe()
6: 674c6efd3dab ! 9: dea72a470141 crypto: qat - simplify adf_enable_aer()
@@ drivers/crypto/qat/qat_4xxx/adf_drv.c: static struct pci_driver adf_driver = {
.probe = adf_probe,
.remove = adf_remove,
.sriov_configure = adf_sriov_configure,
-+ .err_handler = adf_err_handler,
++ .err_handler = &adf_err_handler,
};
module_pci_driver(adf_driver);
@@ drivers/crypto/qat/qat_c3xxx/adf_drv.c: static struct pci_driver adf_driver = {
.probe = adf_probe,
.remove = adf_remove,
.sriov_configure = adf_sriov_configure,
-+ .err_handler = adf_err_handler,
++ .err_handler = &adf_err_handler,
};
static void adf_cleanup_pci_dev(struct adf_accel_dev *accel_dev)
@@ drivers/crypto/qat/qat_c62x/adf_drv.c: static struct pci_driver adf_driver = {
.probe = adf_probe,
.remove = adf_remove,
.sriov_configure = adf_sriov_configure,
-+ .err_handler = adf_err_handler,
++ .err_handler = &adf_err_handler,
};
static void adf_cleanup_pci_dev(struct adf_accel_dev *accel_dev)
@@ drivers/crypto/qat/qat_common/adf_common_drv.h: void adf_ae_fw_release(struct ad
int adf_ae_stop(struct adf_accel_dev *accel_dev);
-int adf_enable_aer(struct adf_accel_dev *accel_dev);
++extern const struct pci_error_handlers adf_err_handler;
+void adf_enable_aer(struct adf_accel_dev *accel_dev);
void adf_disable_aer(struct adf_accel_dev *accel_dev);
void adf_reset_sbr(struct adf_accel_dev *accel_dev);
@@ drivers/crypto/qat/qat_dh895xcc/adf_drv.c: static struct pci_driver adf_driver =
.probe = adf_probe,
.remove = adf_remove,
.sriov_configure = adf_sriov_configure,
-+ .err_handler = adf_err_handler,
++ .err_handler = &adf_err_handler,
};
static void adf_cleanup_pci_dev(struct adf_accel_dev *accel_dev)
7: 0e022deb0f75 = 10: b4165dda38ea PCI: Replace pci_dev::driver usage by pci_dev::dev.driver
8: edd9d24df02a = 11: d93a138bd7ab PCI: Drop duplicated tracking of a pci_dev's bound driver
base-commit: 5816b3e6577eaa676ceb00a848f0fd65fe2adc29
--
2.30.2
^ permalink raw reply
* Re: [PATCH v2 4/4] powerpc/pseries/cpuhp: remove obsolete comment from pseries_cpu_die
From: Nathan Lynch @ 2021-09-29 12:06 UTC (permalink / raw)
To: Michael Ellerman, Daniel Henrique Barboza, linuxppc-dev
Cc: tyreld, ldufour, aneesh.kumar
In-Reply-To: <87h7e40wac.fsf@mpe.ellerman.id.au>
Michael Ellerman <mpe@ellerman.id.au> writes:
> Daniel Henrique Barboza <danielhb413@gmail.com> writes:
>> This is enough to say that we can't easily see the history behind this comment.
>> I also believe that we're better of without it since it doesn't make sense
>> with the current codebase.
>
> It was added by the original CPU hotplug commit for ppc64::
>
> https://github.com/mpe/linux-fullhistory/commit/0e9fd9441cd2113b67b14e739267c9e69761489b
>
>
> The code was fairly similar:
>
> void __cpu_die(unsigned int cpu)
> {
> int tries;
> int cpu_status;
> unsigned int pcpu = get_hard_smp_processor_id(cpu);
>
> for (tries = 0; tries < 5; tries++) {
> cpu_status = query_cpu_stopped(pcpu);
>
> if (cpu_status == 0)
> break;
> set_current_state(TASK_UNINTERRUPTIBLE);
> schedule_timeout(HZ);
> }
> if (cpu_status != 0) {
> printk("Querying DEAD? cpu %i (%i) shows %i\n",
> cpu, pcpu, cpu_status);
> }
>
> /* Isolation and deallocation are definatly done by
> * drslot_chrp_cpu. If they were not they would be
> * done here. Change isolate state to Isolate and
> * change allocation-state to Unusable.
> */
> paca[cpu].xProcStart = 0;
>
> /* So we can recognize if it fails to come up next time. */
> cpu_callin_map[cpu] = 0;
> }
>
>
> drslot_chrp_cpu() still exists in drmgr:
>
> https://github.com/ibm-power-utilities/powerpc-utils/blob/e798c4a09fbf0fa0f421e624cfa366a6c405c9fe/src/drmgr/drslot_chrp_cpu.c#L406
>
>
> I agree the comment is no longer meaningful and can be removed.
Thanks for providing this background.
> It might be good to then add a comment explaining why we need to set
> cpu_start = 0.
Sure, I can take that as a follow-up. Or perhaps it should be moved to
the online path.
> It's not immediately clear why we need to. When we bring a CPU back
> online in smp_pSeries_kick_cpu() we ask RTAS to start it and then
> immediately set cpu_start = 1, ie. there isn't a separate step that sets
> cpu_start = 1 for hotplugged CPUs.
Hmm I'm not following the distinction you seem to be drawing between
bringing a CPU back online and a hotplugged CPU. kick_cpu is used in all
cases AFAIK.
^ permalink raw reply
* Re: [PATCH v4 5/8] bpf ppc64: Add BPF_PROBE_MEM support for JIT
From: Christophe Leroy @ 2021-09-29 11:47 UTC (permalink / raw)
To: Hari Bathini, naveen.n.rao, mpe, ast, daniel
Cc: Ravi Bangoria, songliubraving, netdev, john.fastabend, andrii,
kpsingh, paulus, yhs, bpf, linuxppc-dev, kafai
In-Reply-To: <20210929111855.50254-6-hbathini@linux.ibm.com>
Le 29/09/2021 à 13:18, Hari Bathini a écrit :
> From: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
>
> BPF load instruction with BPF_PROBE_MEM mode can cause a fault
> inside kernel. Append exception table for such instructions
> within BPF program.
>
> Unlike other archs which uses extable 'fixup' field to pass dest_reg
> and nip, BPF exception table on PowerPC follows the generic PowerPC
For my curiosity, can you explain why we don't want and can't do the
same on powerpc as on other archs ?
> exception table design, where it populates both fixup and extable
> sections within BPF program. fixup section contains two instructions,
> first instruction clears dest_reg and 2nd jumps to next instruction
> in the BPF code. extable 'insn' field contains relative offset of
> the instruction and 'fixup' field contains relative offset of the
> fixup entry. Example layout of BPF program with extable present:
>
> +------------------+
> | |
> | |
> 0x4020 -->| ld r27,4(r3) |
> | |
> | |
> 0x40ac -->| lwz r3,0(r4) |
> | |
> | |
> |------------------|
> 0x4280 -->| li r27,0 | \ fixup entry
> | b 0x4024 | /
> 0x4288 -->| li r3,0 |
> | b 0x40b0 |
> |------------------|
> 0x4290 -->| insn=0xfffffd90 | \ extable entry
> | fixup=0xffffffec | /
> 0x4298 -->| insn=0xfffffe14 |
> | fixup=0xffffffec |
> +------------------+
>
> (Addresses shown here are chosen random, not real)
>
^ permalink raw reply
* Re: [PATCH v4 8/8] bpf ppc32: Access only if addr is kernel address
From: Christophe Leroy @ 2021-09-29 11:45 UTC (permalink / raw)
To: Hari Bathini, naveen.n.rao, mpe, ast, daniel
Cc: songliubraving, netdev, john.fastabend, andrii, kpsingh, paulus,
yhs, bpf, linuxppc-dev, kafai
In-Reply-To: <20210929111855.50254-9-hbathini@linux.ibm.com>
Le 29/09/2021 à 13:18, Hari Bathini a écrit :
> With KUAP enabled, any kernel code which wants to access userspace
> needs to be surrounded by disable-enable KUAP. But that is not
> happening for BPF_PROBE_MEM load instruction. Though PPC32 does not
> support read protection, considering the fact that PTR_TO_BTF_ID
> (which uses BPF_PROBE_MEM mode) could either be a valid kernel pointer
> or NULL but should never be a pointer to userspace address, execute
> BPF_PROBE_MEM load only if addr is kernel address, otherwise set
> dst_reg=0 and move on.
>
> This will catch NULL, valid or invalid userspace pointers. Only bad
> kernel pointer will be handled by BPF exception table.
>
> [Alexei suggested for x86]
> Suggested-by: Alexei Starovoitov <ast@kernel.org>
> Signed-off-by: Hari Bathini <hbathini@linux.ibm.com>
Reviewed-by: Christophe Leroy <christophe.leroy@csgroup.eu>
> ---
>
> Changes in v4:
> * Adjusted the emit code to avoid using temporary reg.
>
>
> arch/powerpc/net/bpf_jit_comp32.c | 34 +++++++++++++++++++++++++++++++
> 1 file changed, 34 insertions(+)
>
> diff --git a/arch/powerpc/net/bpf_jit_comp32.c b/arch/powerpc/net/bpf_jit_comp32.c
> index 6ee13a09c70d..2ac81563c78d 100644
> --- a/arch/powerpc/net/bpf_jit_comp32.c
> +++ b/arch/powerpc/net/bpf_jit_comp32.c
> @@ -818,6 +818,40 @@ int bpf_jit_build_body(struct bpf_prog *fp, u32 *image, struct codegen_context *
> case BPF_LDX | BPF_PROBE_MEM | BPF_W:
> case BPF_LDX | BPF_MEM | BPF_DW: /* dst = *(u64 *)(ul) (src + off) */
> case BPF_LDX | BPF_PROBE_MEM | BPF_DW:
> + /*
> + * As PTR_TO_BTF_ID that uses BPF_PROBE_MEM mode could either be a valid
> + * kernel pointer or NULL but not a userspace address, execute BPF_PROBE_MEM
> + * load only if addr is kernel address (see is_kernel_addr()), otherwise
> + * set dst_reg=0 and move on.
> + */
> + if (BPF_MODE(code) == BPF_PROBE_MEM) {
> + PPC_LI32(_R0, TASK_SIZE - off);
> + EMIT(PPC_RAW_CMPLW(src_reg, _R0));
> + PPC_BCC(COND_GT, (ctx->idx + 5) * 4);
> + EMIT(PPC_RAW_LI(dst_reg, 0));
> + /*
> + * For BPF_DW case, "li reg_h,0" would be needed when
> + * !fp->aux->verifier_zext. Emit NOP otherwise.
> + *
> + * Note that "li reg_h,0" is emitted for BPF_B/H/W case,
> + * if necessary. So, jump there insted of emitting an
> + * additional "li reg_h,0" instruction.
> + */
> + if (size == BPF_DW && !fp->aux->verifier_zext)
> + EMIT(PPC_RAW_LI(dst_reg_h, 0));
> + else
> + EMIT(PPC_RAW_NOP());
> + /*
> + * Need to jump two instructions instead of one for BPF_DW case
> + * as there are two load instructions for dst_reg_h & dst_reg
> + * respectively.
> + */
> + if (size == BPF_DW)
> + PPC_JMP((ctx->idx + 3) * 4);
> + else
> + PPC_JMP((ctx->idx + 2) * 4);
> + }
> +
> switch (size) {
> case BPF_B:
> EMIT(PPC_RAW_LBZ(dst_reg, src_reg, off));
>
^ permalink raw reply
* Re: [PATCH v4 7/8] bpf ppc32: Add BPF_PROBE_MEM support for JIT
From: Christophe Leroy @ 2021-09-29 11:44 UTC (permalink / raw)
To: Hari Bathini, naveen.n.rao, mpe, ast, daniel
Cc: songliubraving, netdev, john.fastabend, andrii, kpsingh, paulus,
yhs, bpf, linuxppc-dev, kafai
In-Reply-To: <20210929111855.50254-8-hbathini@linux.ibm.com>
Le 29/09/2021 à 13:18, Hari Bathini a écrit :
> BPF load instruction with BPF_PROBE_MEM mode can cause a fault
> inside kernel. Append exception table for such instructions
> within BPF program.
>
> Unlike other archs which uses extable 'fixup' field to pass dest_reg
> and nip, BPF exception table on PowerPC follows the generic PowerPC
> exception table design, where it populates both fixup and extable
> sections within BPF program. fixup section contains 3 instructions,
> first 2 instructions clear dest_reg (lower & higher 32-bit registers)
> and last instruction jumps to next instruction in the BPF code.
> extable 'insn' field contains relative offset of the instruction and
> 'fixup' field contains relative offset of the fixup entry. Example
> layout of BPF program with extable present:
>
> +------------------+
> | |
> | |
> 0x4020 -->| lwz r28,4(r4) |
> | |
> | |
> 0x40ac -->| lwz r3,0(r24) |
> | lwz r4,4(r24) |
> | |
> | |
> |------------------|
> 0x4278 -->| li r28,0 | \
> | li r27,0 | | fixup entry
> | b 0x4024 | /
> 0x4284 -->| li r4,0 |
> | li r3,0 |
> | b 0x40b4 |
> |------------------|
> 0x4290 -->| insn=0xfffffd90 | \ extable entry
> | fixup=0xffffffe4 | /
> 0x4298 -->| insn=0xfffffe14 |
> | fixup=0xffffffe8 |
> +------------------+
>
> (Addresses shown here are chosen random, not real)
>
> Signed-off-by: Hari Bathini <hbathini@linux.ibm.com>
Reviewed-by: Christophe Leroy <christophe.leroy@csgroup.eu>
> ---
>
> Changes in v4:
> * Dropped explicit fallthrough statement for empty switch cases.
>
>
> arch/powerpc/net/bpf_jit.h | 4 ++++
> arch/powerpc/net/bpf_jit_comp.c | 2 ++
> arch/powerpc/net/bpf_jit_comp32.c | 30 ++++++++++++++++++++++++++++++
> 3 files changed, 36 insertions(+)
>
> diff --git a/arch/powerpc/net/bpf_jit.h b/arch/powerpc/net/bpf_jit.h
> index 561689a2abdf..800734056200 100644
> --- a/arch/powerpc/net/bpf_jit.h
> +++ b/arch/powerpc/net/bpf_jit.h
> @@ -144,7 +144,11 @@ struct codegen_context {
> unsigned int exentry_idx;
> };
>
> +#ifdef CONFIG_PPC32
> +#define BPF_FIXUP_LEN 3 /* Three instructions => 12 bytes */
> +#else
> #define BPF_FIXUP_LEN 2 /* Two instructions => 8 bytes */
> +#endif
>
> static inline void bpf_flush_icache(void *start, void *end)
> {
> diff --git a/arch/powerpc/net/bpf_jit_comp.c b/arch/powerpc/net/bpf_jit_comp.c
> index f02457c6b54f..1a0041997050 100644
> --- a/arch/powerpc/net/bpf_jit_comp.c
> +++ b/arch/powerpc/net/bpf_jit_comp.c
> @@ -297,6 +297,8 @@ int bpf_add_extable_entry(struct bpf_prog *fp, u32 *image, int pass, struct code
> (ctx->exentry_idx * BPF_FIXUP_LEN * 4);
>
> fixup[0] = PPC_RAW_LI(dst_reg, 0);
> + if (IS_ENABLED(CONFIG_PPC32))
> + fixup[1] = PPC_RAW_LI(dst_reg - 1, 0); /* clear higher 32-bit register too */
>
> fixup[BPF_FIXUP_LEN - 1] =
> PPC_RAW_BRANCH((long)(pc + jmp_off) - (long)&fixup[BPF_FIXUP_LEN - 1]);
> diff --git a/arch/powerpc/net/bpf_jit_comp32.c b/arch/powerpc/net/bpf_jit_comp32.c
> index c04291517a7e..6ee13a09c70d 100644
> --- a/arch/powerpc/net/bpf_jit_comp32.c
> +++ b/arch/powerpc/net/bpf_jit_comp32.c
> @@ -811,9 +811,13 @@ int bpf_jit_build_body(struct bpf_prog *fp, u32 *image, struct codegen_context *
> * BPF_LDX
> */
> case BPF_LDX | BPF_MEM | BPF_B: /* dst = *(u8 *)(ul) (src + off) */
> + case BPF_LDX | BPF_PROBE_MEM | BPF_B:
> case BPF_LDX | BPF_MEM | BPF_H: /* dst = *(u16 *)(ul) (src + off) */
> + case BPF_LDX | BPF_PROBE_MEM | BPF_H:
> case BPF_LDX | BPF_MEM | BPF_W: /* dst = *(u32 *)(ul) (src + off) */
> + case BPF_LDX | BPF_PROBE_MEM | BPF_W:
> case BPF_LDX | BPF_MEM | BPF_DW: /* dst = *(u64 *)(ul) (src + off) */
> + case BPF_LDX | BPF_PROBE_MEM | BPF_DW:
> switch (size) {
> case BPF_B:
> EMIT(PPC_RAW_LBZ(dst_reg, src_reg, off));
> @@ -832,6 +836,32 @@ int bpf_jit_build_body(struct bpf_prog *fp, u32 *image, struct codegen_context *
>
> if (size != BPF_DW && !fp->aux->verifier_zext)
> EMIT(PPC_RAW_LI(dst_reg_h, 0));
> +
> + if (BPF_MODE(code) == BPF_PROBE_MEM) {
> + int insn_idx = ctx->idx - 1;
> + int jmp_off = 4;
> +
> + /*
> + * In case of BPF_DW, two lwz instructions are emitted, one
> + * for higher 32-bit and another for lower 32-bit. So, set
> + * ex->insn to the first of the two and jump over both
> + * instructions in fixup.
> + *
> + * Similarly, with !verifier_zext, two instructions are
> + * emitted for BPF_B/H/W case. So, set ex->insn to the
> + * instruction that could fault and skip over both
> + * instructions.
> + */
> + if (size == BPF_DW || !fp->aux->verifier_zext) {
> + insn_idx -= 1;
> + jmp_off += 4;
> + }
> +
> + ret = bpf_add_extable_entry(fp, image, pass, ctx, insn_idx,
> + jmp_off, dst_reg);
> + if (ret)
> + return ret;
> + }
> break;
>
> /*
>
^ permalink raw reply
* Re: [PATCH v4 6/8] bpf ppc64: Access only if addr is kernel address
From: LEROY Christophe @ 2021-09-29 11:44 UTC (permalink / raw)
To: Hari Bathini, naveen.n.rao@linux.ibm.com, mpe@ellerman.id.au,
ast@kernel.org, daniel@iogearbox.net
Cc: Ravi Bangoria, songliubraving@fb.com, netdev@vger.kernel.org,
john.fastabend@gmail.com, andrii@kernel.org, kpsingh@kernel.org,
paulus@samba.org, yhs@fb.com, bpf@vger.kernel.org,
linuxppc-dev@lists.ozlabs.org, kafai@fb.com
In-Reply-To: <20210929111855.50254-7-hbathini@linux.ibm.com>
Le 29/09/2021 à 13:18, Hari Bathini a écrit :
> From: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
>
> On PPC64 with KUAP enabled, any kernel code which wants to
> access userspace needs to be surrounded by disable-enable KUAP.
> But that is not happening for BPF_PROBE_MEM load instruction.
> So, when BPF program tries to access invalid userspace address,
> page-fault handler considers it as bad KUAP fault:
>
> Kernel attempted to read user page (d0000000) - exploit attempt? (uid: 0)
>
> Considering the fact that PTR_TO_BTF_ID (which uses BPF_PROBE_MEM
> mode) could either be a valid kernel pointer or NULL but should
> never be a pointer to userspace address, execute BPF_PROBE_MEM load
> only if addr is kernel address, otherwise set dst_reg=0 and move on.
>
> This will catch NULL, valid or invalid userspace pointers. Only bad
> kernel pointer will be handled by BPF exception table.
>
> [Alexei suggested for x86]
> Suggested-by: Alexei Starovoitov <ast@kernel.org>
> Signed-off-by: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
> Signed-off-by: Hari Bathini <hbathini@linux.ibm.com>
Reviewed-by: Christophe Leroy <christophe.leroy@csgroup.eu>
> ---
>
> Changes in v4:
> * Used IS_ENABLED() instead of #ifdef.
> * Dropped the else case that is not applicable for PPC64.
>
>
> arch/powerpc/net/bpf_jit_comp64.c | 26 ++++++++++++++++++++++++++
> 1 file changed, 26 insertions(+)
>
> diff --git a/arch/powerpc/net/bpf_jit_comp64.c b/arch/powerpc/net/bpf_jit_comp64.c
> index 4170999371ee..e1ea64081ae1 100644
> --- a/arch/powerpc/net/bpf_jit_comp64.c
> +++ b/arch/powerpc/net/bpf_jit_comp64.c
> @@ -727,6 +727,32 @@ int bpf_jit_build_body(struct bpf_prog *fp, u32 *image, struct codegen_context *
> /* dst = *(u64 *)(ul) (src + off) */
> case BPF_LDX | BPF_MEM | BPF_DW:
> case BPF_LDX | BPF_PROBE_MEM | BPF_DW:
> + /*
> + * As PTR_TO_BTF_ID that uses BPF_PROBE_MEM mode could either be a valid
> + * kernel pointer or NULL but not a userspace address, execute BPF_PROBE_MEM
> + * load only if addr is kernel address (see is_kernel_addr()), otherwise
> + * set dst_reg=0 and move on.
> + */
> + if (BPF_MODE(code) == BPF_PROBE_MEM) {
> + EMIT(PPC_RAW_ADDI(b2p[TMP_REG_1], src_reg, off));
> + if (IS_ENABLED(CONFIG_PPC_BOOK3E_64))
> + PPC_LI64(b2p[TMP_REG_2], 0x8000000000000000ul);
> + else /* BOOK3S_64 */
> + PPC_LI64(b2p[TMP_REG_2], PAGE_OFFSET);
> + EMIT(PPC_RAW_CMPLD(b2p[TMP_REG_1], b2p[TMP_REG_2]));
> + PPC_BCC(COND_GT, (ctx->idx + 4) * 4);
> + EMIT(PPC_RAW_LI(dst_reg, 0));
> + /*
> + * Check if 'off' is word aligned because PPC_BPF_LL()
> + * (BPF_DW case) generates two instructions if 'off' is not
> + * word-aligned and one instruction otherwise.
> + */
> + if (BPF_SIZE(code) == BPF_DW && (off & 3))
> + PPC_JMP((ctx->idx + 3) * 4);
> + else
> + PPC_JMP((ctx->idx + 2) * 4);
> + }
> +
> switch (size) {
> case BPF_B:
> EMIT(PPC_RAW_LBZ(dst_reg, src_reg, off));
>
^ permalink raw reply
* Re: [PATCH v4 5/8] bpf ppc64: Add BPF_PROBE_MEM support for JIT
From: Christophe Leroy @ 2021-09-29 11:43 UTC (permalink / raw)
To: Hari Bathini, naveen.n.rao, mpe, ast, daniel
Cc: Ravi Bangoria, songliubraving, netdev, john.fastabend, andrii,
kpsingh, paulus, yhs, bpf, linuxppc-dev, kafai
In-Reply-To: <20210929111855.50254-6-hbathini@linux.ibm.com>
Le 29/09/2021 à 13:18, Hari Bathini a écrit :
> From: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
>
> BPF load instruction with BPF_PROBE_MEM mode can cause a fault
> inside kernel. Append exception table for such instructions
> within BPF program.
>
> Unlike other archs which uses extable 'fixup' field to pass dest_reg
> and nip, BPF exception table on PowerPC follows the generic PowerPC
> exception table design, where it populates both fixup and extable
> sections within BPF program. fixup section contains two instructions,
> first instruction clears dest_reg and 2nd jumps to next instruction
> in the BPF code. extable 'insn' field contains relative offset of
> the instruction and 'fixup' field contains relative offset of the
> fixup entry. Example layout of BPF program with extable present:
>
> +------------------+
> | |
> | |
> 0x4020 -->| ld r27,4(r3) |
> | |
> | |
> 0x40ac -->| lwz r3,0(r4) |
> | |
> | |
> |------------------|
> 0x4280 -->| li r27,0 | \ fixup entry
> | b 0x4024 | /
> 0x4288 -->| li r3,0 |
> | b 0x40b0 |
> |------------------|
> 0x4290 -->| insn=0xfffffd90 | \ extable entry
> | fixup=0xffffffec | /
> 0x4298 -->| insn=0xfffffe14 |
> | fixup=0xffffffec |
> +------------------+
>
> (Addresses shown here are chosen random, not real)
>
> Signed-off-by: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
> Signed-off-by: Hari Bathini <hbathini@linux.ibm.com>
Reviewed-by: Christophe Leroy <christophe.leroy@csgroup.eu>
> ---
>
> Changes in v4:
> * Dropped explicit fallthrough statement for empty switch cases.
>
>
> arch/powerpc/net/bpf_jit.h | 8 +++-
> arch/powerpc/net/bpf_jit_comp.c | 70 ++++++++++++++++++++++++++++---
> arch/powerpc/net/bpf_jit_comp32.c | 2 +-
> arch/powerpc/net/bpf_jit_comp64.c | 13 +++++-
> 4 files changed, 84 insertions(+), 9 deletions(-)
>
> diff --git a/arch/powerpc/net/bpf_jit.h b/arch/powerpc/net/bpf_jit.h
> index 0c8f885b8f48..561689a2abdf 100644
> --- a/arch/powerpc/net/bpf_jit.h
> +++ b/arch/powerpc/net/bpf_jit.h
> @@ -141,8 +141,11 @@ struct codegen_context {
> unsigned int idx;
> unsigned int stack_size;
> int b2p[ARRAY_SIZE(b2p)];
> + unsigned int exentry_idx;
> };
>
> +#define BPF_FIXUP_LEN 2 /* Two instructions => 8 bytes */
> +
> static inline void bpf_flush_icache(void *start, void *end)
> {
> smp_wmb(); /* smp write barrier */
> @@ -166,11 +169,14 @@ static inline void bpf_clear_seen_register(struct codegen_context *ctx, int i)
>
> void bpf_jit_emit_func_call_rel(u32 *image, struct codegen_context *ctx, u64 func);
> int bpf_jit_build_body(struct bpf_prog *fp, u32 *image, struct codegen_context *ctx,
> - u32 *addrs);
> + u32 *addrs, int pass);
> void bpf_jit_build_prologue(u32 *image, struct codegen_context *ctx);
> void bpf_jit_build_epilogue(u32 *image, struct codegen_context *ctx);
> void bpf_jit_realloc_regs(struct codegen_context *ctx);
>
> +int bpf_add_extable_entry(struct bpf_prog *fp, u32 *image, int pass, struct codegen_context *ctx,
> + int insn_idx, int jmp_off, int dst_reg);
> +
> #endif
>
> #endif
> diff --git a/arch/powerpc/net/bpf_jit_comp.c b/arch/powerpc/net/bpf_jit_comp.c
> index c5c9e8ad1de7..f02457c6b54f 100644
> --- a/arch/powerpc/net/bpf_jit_comp.c
> +++ b/arch/powerpc/net/bpf_jit_comp.c
> @@ -101,6 +101,8 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *fp)
> struct bpf_prog *tmp_fp;
> bool bpf_blinded = false;
> bool extra_pass = false;
> + u32 extable_len;
> + u32 fixup_len;
>
> if (!fp->jit_requested)
> return org_fp;
> @@ -131,7 +133,6 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *fp)
> image = jit_data->image;
> bpf_hdr = jit_data->header;
> proglen = jit_data->proglen;
> - alloclen = proglen + FUNCTION_DESCR_SIZE;
> extra_pass = true;
> goto skip_init_ctx;
> }
> @@ -149,7 +150,7 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *fp)
> cgctx.stack_size = round_up(fp->aux->stack_depth, 16);
>
> /* Scouting faux-generate pass 0 */
> - if (bpf_jit_build_body(fp, 0, &cgctx, addrs)) {
> + if (bpf_jit_build_body(fp, 0, &cgctx, addrs, 0)) {
> /* We hit something illegal or unsupported. */
> fp = org_fp;
> goto out_addrs;
> @@ -162,7 +163,7 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *fp)
> */
> if (cgctx.seen & SEEN_TAILCALL) {
> cgctx.idx = 0;
> - if (bpf_jit_build_body(fp, 0, &cgctx, addrs)) {
> + if (bpf_jit_build_body(fp, 0, &cgctx, addrs, 0)) {
> fp = org_fp;
> goto out_addrs;
> }
> @@ -177,8 +178,11 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *fp)
> bpf_jit_build_prologue(0, &cgctx);
> bpf_jit_build_epilogue(0, &cgctx);
>
> + fixup_len = fp->aux->num_exentries * BPF_FIXUP_LEN * 4;
> + extable_len = fp->aux->num_exentries * sizeof(struct exception_table_entry);
> +
> proglen = cgctx.idx * 4;
> - alloclen = proglen + FUNCTION_DESCR_SIZE;
> + alloclen = proglen + FUNCTION_DESCR_SIZE + fixup_len + extable_len;
>
> bpf_hdr = bpf_jit_binary_alloc(alloclen, &image, 4, bpf_jit_fill_ill_insns);
> if (!bpf_hdr) {
> @@ -186,6 +190,9 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *fp)
> goto out_addrs;
> }
>
> + if (extable_len)
> + fp->aux->extable = (void *)image + FUNCTION_DESCR_SIZE + proglen + fixup_len;
> +
> skip_init_ctx:
> code_base = (u32 *)(image + FUNCTION_DESCR_SIZE);
>
> @@ -210,7 +217,11 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *fp)
> /* Now build the prologue, body code & epilogue for real. */
> cgctx.idx = 0;
> bpf_jit_build_prologue(code_base, &cgctx);
> - bpf_jit_build_body(fp, code_base, &cgctx, addrs);
> + if (bpf_jit_build_body(fp, code_base, &cgctx, addrs, pass)) {
> + bpf_jit_binary_free(bpf_hdr);
> + fp = org_fp;
> + goto out_addrs;
> + }
> bpf_jit_build_epilogue(code_base, &cgctx);
>
> if (bpf_jit_enable > 1)
> @@ -234,7 +245,7 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *fp)
>
> fp->bpf_func = (void *)image;
> fp->jited = 1;
> - fp->jited_len = alloclen;
> + fp->jited_len = proglen + FUNCTION_DESCR_SIZE;
>
> bpf_flush_icache(bpf_hdr, (u8 *)bpf_hdr + (bpf_hdr->pages * PAGE_SIZE));
> bpf_jit_binary_lock_ro(bpf_hdr);
> @@ -258,3 +269,50 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *fp)
>
> return fp;
> }
> +
> +/*
> + * The caller should check for (BPF_MODE(code) == BPF_PROBE_MEM) before calling
> + * this function, as this only applies to BPF_PROBE_MEM, for now.
> + */
> +int bpf_add_extable_entry(struct bpf_prog *fp, u32 *image, int pass, struct codegen_context *ctx,
> + int insn_idx, int jmp_off, int dst_reg)
> +{
> + off_t offset;
> + unsigned long pc;
> + struct exception_table_entry *ex;
> + u32 *fixup;
> +
> + /* Populate extable entries only in the last pass */
> + if (pass != 2)
> + return 0;
> +
> + if (!fp->aux->extable ||
> + WARN_ON_ONCE(ctx->exentry_idx >= fp->aux->num_exentries))
> + return -EINVAL;
> +
> + pc = (unsigned long)&image[insn_idx];
> +
> + fixup = (void *)fp->aux->extable -
> + (fp->aux->num_exentries * BPF_FIXUP_LEN * 4) +
> + (ctx->exentry_idx * BPF_FIXUP_LEN * 4);
> +
> + fixup[0] = PPC_RAW_LI(dst_reg, 0);
> +
> + fixup[BPF_FIXUP_LEN - 1] =
> + PPC_RAW_BRANCH((long)(pc + jmp_off) - (long)&fixup[BPF_FIXUP_LEN - 1]);
> +
> + ex = &fp->aux->extable[ctx->exentry_idx];
> +
> + offset = pc - (long)&ex->insn;
> + if (WARN_ON_ONCE(offset >= 0 || offset < INT_MIN))
> + return -ERANGE;
> + ex->insn = offset;
> +
> + offset = (long)fixup - (long)&ex->fixup;
> + if (WARN_ON_ONCE(offset >= 0 || offset < INT_MIN))
> + return -ERANGE;
> + ex->fixup = offset;
> +
> + ctx->exentry_idx++;
> + return 0;
> +}
> diff --git a/arch/powerpc/net/bpf_jit_comp32.c b/arch/powerpc/net/bpf_jit_comp32.c
> index 65a4d1ed97bf..c04291517a7e 100644
> --- a/arch/powerpc/net/bpf_jit_comp32.c
> +++ b/arch/powerpc/net/bpf_jit_comp32.c
> @@ -266,7 +266,7 @@ static void bpf_jit_emit_tail_call(u32 *image, struct codegen_context *ctx, u32
>
> /* Assemble the body code between the prologue & epilogue */
> int bpf_jit_build_body(struct bpf_prog *fp, u32 *image, struct codegen_context *ctx,
> - u32 *addrs)
> + u32 *addrs, int pass)
> {
> const struct bpf_insn *insn = fp->insnsi;
> int flen = fp->len;
> diff --git a/arch/powerpc/net/bpf_jit_comp64.c b/arch/powerpc/net/bpf_jit_comp64.c
> index 49e6e0b6e4d2..4170999371ee 100644
> --- a/arch/powerpc/net/bpf_jit_comp64.c
> +++ b/arch/powerpc/net/bpf_jit_comp64.c
> @@ -272,7 +272,7 @@ static void bpf_jit_emit_tail_call(u32 *image, struct codegen_context *ctx, u32
>
> /* Assemble the body code between the prologue & epilogue */
> int bpf_jit_build_body(struct bpf_prog *fp, u32 *image, struct codegen_context *ctx,
> - u32 *addrs)
> + u32 *addrs, int pass)
> {
> const struct bpf_insn *insn = fp->insnsi;
> int flen = fp->len;
> @@ -717,12 +717,16 @@ int bpf_jit_build_body(struct bpf_prog *fp, u32 *image, struct codegen_context *
> */
> /* dst = *(u8 *)(ul) (src + off) */
> case BPF_LDX | BPF_MEM | BPF_B:
> + case BPF_LDX | BPF_PROBE_MEM | BPF_B:
> /* dst = *(u16 *)(ul) (src + off) */
> case BPF_LDX | BPF_MEM | BPF_H:
> + case BPF_LDX | BPF_PROBE_MEM | BPF_H:
> /* dst = *(u32 *)(ul) (src + off) */
> case BPF_LDX | BPF_MEM | BPF_W:
> + case BPF_LDX | BPF_PROBE_MEM | BPF_W:
> /* dst = *(u64 *)(ul) (src + off) */
> case BPF_LDX | BPF_MEM | BPF_DW:
> + case BPF_LDX | BPF_PROBE_MEM | BPF_DW:
> switch (size) {
> case BPF_B:
> EMIT(PPC_RAW_LBZ(dst_reg, src_reg, off));
> @@ -740,6 +744,13 @@ int bpf_jit_build_body(struct bpf_prog *fp, u32 *image, struct codegen_context *
>
> if (size != BPF_DW && insn_is_zext(&insn[i + 1]))
> addrs[++i] = ctx->idx * 4;
> +
> + if (BPF_MODE(code) == BPF_PROBE_MEM) {
> + ret = bpf_add_extable_entry(fp, image, pass, ctx, ctx->idx - 1,
> + 4, dst_reg);
> + if (ret)
> + return ret;
> + }
> break;
>
> /*
>
^ permalink raw reply
* Re: [PATCH v4 3/8] bpf powerpc: refactor JIT compiler code
From: Christophe Leroy @ 2021-09-29 11:42 UTC (permalink / raw)
To: Hari Bathini, naveen.n.rao, mpe, ast, daniel
Cc: songliubraving, netdev, john.fastabend, andrii, kpsingh, paulus,
yhs, bpf, linuxppc-dev, kafai
In-Reply-To: <20210929111855.50254-4-hbathini@linux.ibm.com>
Le 29/09/2021 à 13:18, Hari Bathini a écrit :
> Refactor powerpc LDX JITing code to simplify adding BPF_PROBE_MEM
> support.
>
> Signed-off-by: Hari Bathini <hbathini@linux.ibm.com>
Reviewed-by: Christophe Leroy <christophe.leroy@csgroup.eu>
> ---
>
> Changes in v4:
> * Dropped the default case in the switch statement for bpf size.
> * Dropped explicit fallthrough statement for empty switch cases.
>
>
> arch/powerpc/net/bpf_jit_comp32.c | 33 ++++++++++++++++++-------------
> arch/powerpc/net/bpf_jit_comp64.c | 31 +++++++++++++++++------------
> 2 files changed, 37 insertions(+), 27 deletions(-)
>
> diff --git a/arch/powerpc/net/bpf_jit_comp32.c b/arch/powerpc/net/bpf_jit_comp32.c
> index b60b59426a24..65a4d1ed97bf 100644
> --- a/arch/powerpc/net/bpf_jit_comp32.c
> +++ b/arch/powerpc/net/bpf_jit_comp32.c
> @@ -282,6 +282,7 @@ int bpf_jit_build_body(struct bpf_prog *fp, u32 *image, struct codegen_context *
> u32 src_reg = bpf_to_ppc(ctx, insn[i].src_reg);
> u32 src_reg_h = src_reg - 1;
> u32 tmp_reg = bpf_to_ppc(ctx, TMP_REG);
> + u32 size = BPF_SIZE(code);
> s16 off = insn[i].off;
> s32 imm = insn[i].imm;
> bool func_addr_fixed;
> @@ -810,23 +811,27 @@ int bpf_jit_build_body(struct bpf_prog *fp, u32 *image, struct codegen_context *
> * BPF_LDX
> */
> case BPF_LDX | BPF_MEM | BPF_B: /* dst = *(u8 *)(ul) (src + off) */
> - EMIT(PPC_RAW_LBZ(dst_reg, src_reg, off));
> - if (!fp->aux->verifier_zext)
> - EMIT(PPC_RAW_LI(dst_reg_h, 0));
> - break;
> case BPF_LDX | BPF_MEM | BPF_H: /* dst = *(u16 *)(ul) (src + off) */
> - EMIT(PPC_RAW_LHZ(dst_reg, src_reg, off));
> - if (!fp->aux->verifier_zext)
> - EMIT(PPC_RAW_LI(dst_reg_h, 0));
> - break;
> case BPF_LDX | BPF_MEM | BPF_W: /* dst = *(u32 *)(ul) (src + off) */
> - EMIT(PPC_RAW_LWZ(dst_reg, src_reg, off));
> - if (!fp->aux->verifier_zext)
> - EMIT(PPC_RAW_LI(dst_reg_h, 0));
> - break;
> case BPF_LDX | BPF_MEM | BPF_DW: /* dst = *(u64 *)(ul) (src + off) */
> - EMIT(PPC_RAW_LWZ(dst_reg_h, src_reg, off));
> - EMIT(PPC_RAW_LWZ(dst_reg, src_reg, off + 4));
> + switch (size) {
> + case BPF_B:
> + EMIT(PPC_RAW_LBZ(dst_reg, src_reg, off));
> + break;
> + case BPF_H:
> + EMIT(PPC_RAW_LHZ(dst_reg, src_reg, off));
> + break;
> + case BPF_W:
> + EMIT(PPC_RAW_LWZ(dst_reg, src_reg, off));
> + break;
> + case BPF_DW:
> + EMIT(PPC_RAW_LWZ(dst_reg_h, src_reg, off));
> + EMIT(PPC_RAW_LWZ(dst_reg, src_reg, off + 4));
> + break;
> + }
> +
> + if (size != BPF_DW && !fp->aux->verifier_zext)
> + EMIT(PPC_RAW_LI(dst_reg_h, 0));
> break;
>
> /*
> diff --git a/arch/powerpc/net/bpf_jit_comp64.c b/arch/powerpc/net/bpf_jit_comp64.c
> index 2a87da50d9a4..49e6e0b6e4d2 100644
> --- a/arch/powerpc/net/bpf_jit_comp64.c
> +++ b/arch/powerpc/net/bpf_jit_comp64.c
> @@ -285,6 +285,7 @@ int bpf_jit_build_body(struct bpf_prog *fp, u32 *image, struct codegen_context *
> u32 code = insn[i].code;
> u32 dst_reg = b2p[insn[i].dst_reg];
> u32 src_reg = b2p[insn[i].src_reg];
> + u32 size = BPF_SIZE(code);
> s16 off = insn[i].off;
> s32 imm = insn[i].imm;
> bool func_addr_fixed;
> @@ -716,25 +717,29 @@ int bpf_jit_build_body(struct bpf_prog *fp, u32 *image, struct codegen_context *
> */
> /* dst = *(u8 *)(ul) (src + off) */
> case BPF_LDX | BPF_MEM | BPF_B:
> - EMIT(PPC_RAW_LBZ(dst_reg, src_reg, off));
> - if (insn_is_zext(&insn[i + 1]))
> - addrs[++i] = ctx->idx * 4;
> - break;
> /* dst = *(u16 *)(ul) (src + off) */
> case BPF_LDX | BPF_MEM | BPF_H:
> - EMIT(PPC_RAW_LHZ(dst_reg, src_reg, off));
> - if (insn_is_zext(&insn[i + 1]))
> - addrs[++i] = ctx->idx * 4;
> - break;
> /* dst = *(u32 *)(ul) (src + off) */
> case BPF_LDX | BPF_MEM | BPF_W:
> - EMIT(PPC_RAW_LWZ(dst_reg, src_reg, off));
> - if (insn_is_zext(&insn[i + 1]))
> - addrs[++i] = ctx->idx * 4;
> - break;
> /* dst = *(u64 *)(ul) (src + off) */
> case BPF_LDX | BPF_MEM | BPF_DW:
> - PPC_BPF_LL(dst_reg, src_reg, off);
> + switch (size) {
> + case BPF_B:
> + EMIT(PPC_RAW_LBZ(dst_reg, src_reg, off));
> + break;
> + case BPF_H:
> + EMIT(PPC_RAW_LHZ(dst_reg, src_reg, off));
> + break;
> + case BPF_W:
> + EMIT(PPC_RAW_LWZ(dst_reg, src_reg, off));
> + break;
> + case BPF_DW:
> + PPC_BPF_LL(dst_reg, src_reg, off);
> + break;
> + }
> +
> + if (size != BPF_DW && insn_is_zext(&insn[i + 1]))
> + addrs[++i] = ctx->idx * 4;
> break;
>
> /*
>
^ permalink raw reply
* [PATCH v4 8/8] bpf ppc32: Access only if addr is kernel address
From: Hari Bathini @ 2021-09-29 11:18 UTC (permalink / raw)
To: naveen.n.rao, christophe.leroy, mpe, ast, daniel
Cc: songliubraving, netdev, john.fastabend, andrii, kpsingh, paulus,
yhs, bpf, linuxppc-dev, kafai, Hari Bathini
In-Reply-To: <20210929111855.50254-1-hbathini@linux.ibm.com>
With KUAP enabled, any kernel code which wants to access userspace
needs to be surrounded by disable-enable KUAP. But that is not
happening for BPF_PROBE_MEM load instruction. Though PPC32 does not
support read protection, considering the fact that PTR_TO_BTF_ID
(which uses BPF_PROBE_MEM mode) could either be a valid kernel pointer
or NULL but should never be a pointer to userspace address, execute
BPF_PROBE_MEM load only if addr is kernel address, otherwise set
dst_reg=0 and move on.
This will catch NULL, valid or invalid userspace pointers. Only bad
kernel pointer will be handled by BPF exception table.
[Alexei suggested for x86]
Suggested-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Hari Bathini <hbathini@linux.ibm.com>
---
Changes in v4:
* Adjusted the emit code to avoid using temporary reg.
arch/powerpc/net/bpf_jit_comp32.c | 34 +++++++++++++++++++++++++++++++
1 file changed, 34 insertions(+)
diff --git a/arch/powerpc/net/bpf_jit_comp32.c b/arch/powerpc/net/bpf_jit_comp32.c
index 6ee13a09c70d..2ac81563c78d 100644
--- a/arch/powerpc/net/bpf_jit_comp32.c
+++ b/arch/powerpc/net/bpf_jit_comp32.c
@@ -818,6 +818,40 @@ int bpf_jit_build_body(struct bpf_prog *fp, u32 *image, struct codegen_context *
case BPF_LDX | BPF_PROBE_MEM | BPF_W:
case BPF_LDX | BPF_MEM | BPF_DW: /* dst = *(u64 *)(ul) (src + off) */
case BPF_LDX | BPF_PROBE_MEM | BPF_DW:
+ /*
+ * As PTR_TO_BTF_ID that uses BPF_PROBE_MEM mode could either be a valid
+ * kernel pointer or NULL but not a userspace address, execute BPF_PROBE_MEM
+ * load only if addr is kernel address (see is_kernel_addr()), otherwise
+ * set dst_reg=0 and move on.
+ */
+ if (BPF_MODE(code) == BPF_PROBE_MEM) {
+ PPC_LI32(_R0, TASK_SIZE - off);
+ EMIT(PPC_RAW_CMPLW(src_reg, _R0));
+ PPC_BCC(COND_GT, (ctx->idx + 5) * 4);
+ EMIT(PPC_RAW_LI(dst_reg, 0));
+ /*
+ * For BPF_DW case, "li reg_h,0" would be needed when
+ * !fp->aux->verifier_zext. Emit NOP otherwise.
+ *
+ * Note that "li reg_h,0" is emitted for BPF_B/H/W case,
+ * if necessary. So, jump there insted of emitting an
+ * additional "li reg_h,0" instruction.
+ */
+ if (size == BPF_DW && !fp->aux->verifier_zext)
+ EMIT(PPC_RAW_LI(dst_reg_h, 0));
+ else
+ EMIT(PPC_RAW_NOP());
+ /*
+ * Need to jump two instructions instead of one for BPF_DW case
+ * as there are two load instructions for dst_reg_h & dst_reg
+ * respectively.
+ */
+ if (size == BPF_DW)
+ PPC_JMP((ctx->idx + 3) * 4);
+ else
+ PPC_JMP((ctx->idx + 2) * 4);
+ }
+
switch (size) {
case BPF_B:
EMIT(PPC_RAW_LBZ(dst_reg, src_reg, off));
--
2.31.1
^ permalink raw reply related
* [PATCH v4 3/8] bpf powerpc: refactor JIT compiler code
From: Hari Bathini @ 2021-09-29 11:18 UTC (permalink / raw)
To: naveen.n.rao, christophe.leroy, mpe, ast, daniel
Cc: songliubraving, netdev, john.fastabend, andrii, kpsingh, paulus,
yhs, bpf, linuxppc-dev, kafai, Hari Bathini
In-Reply-To: <20210929111855.50254-1-hbathini@linux.ibm.com>
Refactor powerpc LDX JITing code to simplify adding BPF_PROBE_MEM
support.
Signed-off-by: Hari Bathini <hbathini@linux.ibm.com>
---
Changes in v4:
* Dropped the default case in the switch statement for bpf size.
* Dropped explicit fallthrough statement for empty switch cases.
arch/powerpc/net/bpf_jit_comp32.c | 33 ++++++++++++++++++-------------
arch/powerpc/net/bpf_jit_comp64.c | 31 +++++++++++++++++------------
2 files changed, 37 insertions(+), 27 deletions(-)
diff --git a/arch/powerpc/net/bpf_jit_comp32.c b/arch/powerpc/net/bpf_jit_comp32.c
index b60b59426a24..65a4d1ed97bf 100644
--- a/arch/powerpc/net/bpf_jit_comp32.c
+++ b/arch/powerpc/net/bpf_jit_comp32.c
@@ -282,6 +282,7 @@ int bpf_jit_build_body(struct bpf_prog *fp, u32 *image, struct codegen_context *
u32 src_reg = bpf_to_ppc(ctx, insn[i].src_reg);
u32 src_reg_h = src_reg - 1;
u32 tmp_reg = bpf_to_ppc(ctx, TMP_REG);
+ u32 size = BPF_SIZE(code);
s16 off = insn[i].off;
s32 imm = insn[i].imm;
bool func_addr_fixed;
@@ -810,23 +811,27 @@ int bpf_jit_build_body(struct bpf_prog *fp, u32 *image, struct codegen_context *
* BPF_LDX
*/
case BPF_LDX | BPF_MEM | BPF_B: /* dst = *(u8 *)(ul) (src + off) */
- EMIT(PPC_RAW_LBZ(dst_reg, src_reg, off));
- if (!fp->aux->verifier_zext)
- EMIT(PPC_RAW_LI(dst_reg_h, 0));
- break;
case BPF_LDX | BPF_MEM | BPF_H: /* dst = *(u16 *)(ul) (src + off) */
- EMIT(PPC_RAW_LHZ(dst_reg, src_reg, off));
- if (!fp->aux->verifier_zext)
- EMIT(PPC_RAW_LI(dst_reg_h, 0));
- break;
case BPF_LDX | BPF_MEM | BPF_W: /* dst = *(u32 *)(ul) (src + off) */
- EMIT(PPC_RAW_LWZ(dst_reg, src_reg, off));
- if (!fp->aux->verifier_zext)
- EMIT(PPC_RAW_LI(dst_reg_h, 0));
- break;
case BPF_LDX | BPF_MEM | BPF_DW: /* dst = *(u64 *)(ul) (src + off) */
- EMIT(PPC_RAW_LWZ(dst_reg_h, src_reg, off));
- EMIT(PPC_RAW_LWZ(dst_reg, src_reg, off + 4));
+ switch (size) {
+ case BPF_B:
+ EMIT(PPC_RAW_LBZ(dst_reg, src_reg, off));
+ break;
+ case BPF_H:
+ EMIT(PPC_RAW_LHZ(dst_reg, src_reg, off));
+ break;
+ case BPF_W:
+ EMIT(PPC_RAW_LWZ(dst_reg, src_reg, off));
+ break;
+ case BPF_DW:
+ EMIT(PPC_RAW_LWZ(dst_reg_h, src_reg, off));
+ EMIT(PPC_RAW_LWZ(dst_reg, src_reg, off + 4));
+ break;
+ }
+
+ if (size != BPF_DW && !fp->aux->verifier_zext)
+ EMIT(PPC_RAW_LI(dst_reg_h, 0));
break;
/*
diff --git a/arch/powerpc/net/bpf_jit_comp64.c b/arch/powerpc/net/bpf_jit_comp64.c
index 2a87da50d9a4..49e6e0b6e4d2 100644
--- a/arch/powerpc/net/bpf_jit_comp64.c
+++ b/arch/powerpc/net/bpf_jit_comp64.c
@@ -285,6 +285,7 @@ int bpf_jit_build_body(struct bpf_prog *fp, u32 *image, struct codegen_context *
u32 code = insn[i].code;
u32 dst_reg = b2p[insn[i].dst_reg];
u32 src_reg = b2p[insn[i].src_reg];
+ u32 size = BPF_SIZE(code);
s16 off = insn[i].off;
s32 imm = insn[i].imm;
bool func_addr_fixed;
@@ -716,25 +717,29 @@ int bpf_jit_build_body(struct bpf_prog *fp, u32 *image, struct codegen_context *
*/
/* dst = *(u8 *)(ul) (src + off) */
case BPF_LDX | BPF_MEM | BPF_B:
- EMIT(PPC_RAW_LBZ(dst_reg, src_reg, off));
- if (insn_is_zext(&insn[i + 1]))
- addrs[++i] = ctx->idx * 4;
- break;
/* dst = *(u16 *)(ul) (src + off) */
case BPF_LDX | BPF_MEM | BPF_H:
- EMIT(PPC_RAW_LHZ(dst_reg, src_reg, off));
- if (insn_is_zext(&insn[i + 1]))
- addrs[++i] = ctx->idx * 4;
- break;
/* dst = *(u32 *)(ul) (src + off) */
case BPF_LDX | BPF_MEM | BPF_W:
- EMIT(PPC_RAW_LWZ(dst_reg, src_reg, off));
- if (insn_is_zext(&insn[i + 1]))
- addrs[++i] = ctx->idx * 4;
- break;
/* dst = *(u64 *)(ul) (src + off) */
case BPF_LDX | BPF_MEM | BPF_DW:
- PPC_BPF_LL(dst_reg, src_reg, off);
+ switch (size) {
+ case BPF_B:
+ EMIT(PPC_RAW_LBZ(dst_reg, src_reg, off));
+ break;
+ case BPF_H:
+ EMIT(PPC_RAW_LHZ(dst_reg, src_reg, off));
+ break;
+ case BPF_W:
+ EMIT(PPC_RAW_LWZ(dst_reg, src_reg, off));
+ break;
+ case BPF_DW:
+ PPC_BPF_LL(dst_reg, src_reg, off);
+ break;
+ }
+
+ if (size != BPF_DW && insn_is_zext(&insn[i + 1]))
+ addrs[++i] = ctx->idx * 4;
break;
/*
--
2.31.1
^ permalink raw reply related
* [PATCH v4 7/8] bpf ppc32: Add BPF_PROBE_MEM support for JIT
From: Hari Bathini @ 2021-09-29 11:18 UTC (permalink / raw)
To: naveen.n.rao, christophe.leroy, mpe, ast, daniel
Cc: songliubraving, netdev, john.fastabend, andrii, kpsingh, paulus,
yhs, bpf, linuxppc-dev, kafai, Hari Bathini
In-Reply-To: <20210929111855.50254-1-hbathini@linux.ibm.com>
BPF load instruction with BPF_PROBE_MEM mode can cause a fault
inside kernel. Append exception table for such instructions
within BPF program.
Unlike other archs which uses extable 'fixup' field to pass dest_reg
and nip, BPF exception table on PowerPC follows the generic PowerPC
exception table design, where it populates both fixup and extable
sections within BPF program. fixup section contains 3 instructions,
first 2 instructions clear dest_reg (lower & higher 32-bit registers)
and last instruction jumps to next instruction in the BPF code.
extable 'insn' field contains relative offset of the instruction and
'fixup' field contains relative offset of the fixup entry. Example
layout of BPF program with extable present:
+------------------+
| |
| |
0x4020 -->| lwz r28,4(r4) |
| |
| |
0x40ac -->| lwz r3,0(r24) |
| lwz r4,4(r24) |
| |
| |
|------------------|
0x4278 -->| li r28,0 | \
| li r27,0 | | fixup entry
| b 0x4024 | /
0x4284 -->| li r4,0 |
| li r3,0 |
| b 0x40b4 |
|------------------|
0x4290 -->| insn=0xfffffd90 | \ extable entry
| fixup=0xffffffe4 | /
0x4298 -->| insn=0xfffffe14 |
| fixup=0xffffffe8 |
+------------------+
(Addresses shown here are chosen random, not real)
Signed-off-by: Hari Bathini <hbathini@linux.ibm.com>
---
Changes in v4:
* Dropped explicit fallthrough statement for empty switch cases.
arch/powerpc/net/bpf_jit.h | 4 ++++
arch/powerpc/net/bpf_jit_comp.c | 2 ++
arch/powerpc/net/bpf_jit_comp32.c | 30 ++++++++++++++++++++++++++++++
3 files changed, 36 insertions(+)
diff --git a/arch/powerpc/net/bpf_jit.h b/arch/powerpc/net/bpf_jit.h
index 561689a2abdf..800734056200 100644
--- a/arch/powerpc/net/bpf_jit.h
+++ b/arch/powerpc/net/bpf_jit.h
@@ -144,7 +144,11 @@ struct codegen_context {
unsigned int exentry_idx;
};
+#ifdef CONFIG_PPC32
+#define BPF_FIXUP_LEN 3 /* Three instructions => 12 bytes */
+#else
#define BPF_FIXUP_LEN 2 /* Two instructions => 8 bytes */
+#endif
static inline void bpf_flush_icache(void *start, void *end)
{
diff --git a/arch/powerpc/net/bpf_jit_comp.c b/arch/powerpc/net/bpf_jit_comp.c
index f02457c6b54f..1a0041997050 100644
--- a/arch/powerpc/net/bpf_jit_comp.c
+++ b/arch/powerpc/net/bpf_jit_comp.c
@@ -297,6 +297,8 @@ int bpf_add_extable_entry(struct bpf_prog *fp, u32 *image, int pass, struct code
(ctx->exentry_idx * BPF_FIXUP_LEN * 4);
fixup[0] = PPC_RAW_LI(dst_reg, 0);
+ if (IS_ENABLED(CONFIG_PPC32))
+ fixup[1] = PPC_RAW_LI(dst_reg - 1, 0); /* clear higher 32-bit register too */
fixup[BPF_FIXUP_LEN - 1] =
PPC_RAW_BRANCH((long)(pc + jmp_off) - (long)&fixup[BPF_FIXUP_LEN - 1]);
diff --git a/arch/powerpc/net/bpf_jit_comp32.c b/arch/powerpc/net/bpf_jit_comp32.c
index c04291517a7e..6ee13a09c70d 100644
--- a/arch/powerpc/net/bpf_jit_comp32.c
+++ b/arch/powerpc/net/bpf_jit_comp32.c
@@ -811,9 +811,13 @@ int bpf_jit_build_body(struct bpf_prog *fp, u32 *image, struct codegen_context *
* BPF_LDX
*/
case BPF_LDX | BPF_MEM | BPF_B: /* dst = *(u8 *)(ul) (src + off) */
+ case BPF_LDX | BPF_PROBE_MEM | BPF_B:
case BPF_LDX | BPF_MEM | BPF_H: /* dst = *(u16 *)(ul) (src + off) */
+ case BPF_LDX | BPF_PROBE_MEM | BPF_H:
case BPF_LDX | BPF_MEM | BPF_W: /* dst = *(u32 *)(ul) (src + off) */
+ case BPF_LDX | BPF_PROBE_MEM | BPF_W:
case BPF_LDX | BPF_MEM | BPF_DW: /* dst = *(u64 *)(ul) (src + off) */
+ case BPF_LDX | BPF_PROBE_MEM | BPF_DW:
switch (size) {
case BPF_B:
EMIT(PPC_RAW_LBZ(dst_reg, src_reg, off));
@@ -832,6 +836,32 @@ int bpf_jit_build_body(struct bpf_prog *fp, u32 *image, struct codegen_context *
if (size != BPF_DW && !fp->aux->verifier_zext)
EMIT(PPC_RAW_LI(dst_reg_h, 0));
+
+ if (BPF_MODE(code) == BPF_PROBE_MEM) {
+ int insn_idx = ctx->idx - 1;
+ int jmp_off = 4;
+
+ /*
+ * In case of BPF_DW, two lwz instructions are emitted, one
+ * for higher 32-bit and another for lower 32-bit. So, set
+ * ex->insn to the first of the two and jump over both
+ * instructions in fixup.
+ *
+ * Similarly, with !verifier_zext, two instructions are
+ * emitted for BPF_B/H/W case. So, set ex->insn to the
+ * instruction that could fault and skip over both
+ * instructions.
+ */
+ if (size == BPF_DW || !fp->aux->verifier_zext) {
+ insn_idx -= 1;
+ jmp_off += 4;
+ }
+
+ ret = bpf_add_extable_entry(fp, image, pass, ctx, insn_idx,
+ jmp_off, dst_reg);
+ if (ret)
+ return ret;
+ }
break;
/*
--
2.31.1
^ permalink raw reply related
* [PATCH v4 6/8] bpf ppc64: Access only if addr is kernel address
From: Hari Bathini @ 2021-09-29 11:18 UTC (permalink / raw)
To: naveen.n.rao, christophe.leroy, mpe, ast, daniel
Cc: Ravi Bangoria, songliubraving, netdev, john.fastabend, andrii,
kpsingh, paulus, yhs, bpf, linuxppc-dev, kafai, Hari Bathini
In-Reply-To: <20210929111855.50254-1-hbathini@linux.ibm.com>
From: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
On PPC64 with KUAP enabled, any kernel code which wants to
access userspace needs to be surrounded by disable-enable KUAP.
But that is not happening for BPF_PROBE_MEM load instruction.
So, when BPF program tries to access invalid userspace address,
page-fault handler considers it as bad KUAP fault:
Kernel attempted to read user page (d0000000) - exploit attempt? (uid: 0)
Considering the fact that PTR_TO_BTF_ID (which uses BPF_PROBE_MEM
mode) could either be a valid kernel pointer or NULL but should
never be a pointer to userspace address, execute BPF_PROBE_MEM load
only if addr is kernel address, otherwise set dst_reg=0 and move on.
This will catch NULL, valid or invalid userspace pointers. Only bad
kernel pointer will be handled by BPF exception table.
[Alexei suggested for x86]
Suggested-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
Signed-off-by: Hari Bathini <hbathini@linux.ibm.com>
---
Changes in v4:
* Used IS_ENABLED() instead of #ifdef.
* Dropped the else case that is not applicable for PPC64.
arch/powerpc/net/bpf_jit_comp64.c | 26 ++++++++++++++++++++++++++
1 file changed, 26 insertions(+)
diff --git a/arch/powerpc/net/bpf_jit_comp64.c b/arch/powerpc/net/bpf_jit_comp64.c
index 4170999371ee..e1ea64081ae1 100644
--- a/arch/powerpc/net/bpf_jit_comp64.c
+++ b/arch/powerpc/net/bpf_jit_comp64.c
@@ -727,6 +727,32 @@ int bpf_jit_build_body(struct bpf_prog *fp, u32 *image, struct codegen_context *
/* dst = *(u64 *)(ul) (src + off) */
case BPF_LDX | BPF_MEM | BPF_DW:
case BPF_LDX | BPF_PROBE_MEM | BPF_DW:
+ /*
+ * As PTR_TO_BTF_ID that uses BPF_PROBE_MEM mode could either be a valid
+ * kernel pointer or NULL but not a userspace address, execute BPF_PROBE_MEM
+ * load only if addr is kernel address (see is_kernel_addr()), otherwise
+ * set dst_reg=0 and move on.
+ */
+ if (BPF_MODE(code) == BPF_PROBE_MEM) {
+ EMIT(PPC_RAW_ADDI(b2p[TMP_REG_1], src_reg, off));
+ if (IS_ENABLED(CONFIG_PPC_BOOK3E_64))
+ PPC_LI64(b2p[TMP_REG_2], 0x8000000000000000ul);
+ else /* BOOK3S_64 */
+ PPC_LI64(b2p[TMP_REG_2], PAGE_OFFSET);
+ EMIT(PPC_RAW_CMPLD(b2p[TMP_REG_1], b2p[TMP_REG_2]));
+ PPC_BCC(COND_GT, (ctx->idx + 4) * 4);
+ EMIT(PPC_RAW_LI(dst_reg, 0));
+ /*
+ * Check if 'off' is word aligned because PPC_BPF_LL()
+ * (BPF_DW case) generates two instructions if 'off' is not
+ * word-aligned and one instruction otherwise.
+ */
+ if (BPF_SIZE(code) == BPF_DW && (off & 3))
+ PPC_JMP((ctx->idx + 3) * 4);
+ else
+ PPC_JMP((ctx->idx + 2) * 4);
+ }
+
switch (size) {
case BPF_B:
EMIT(PPC_RAW_LBZ(dst_reg, src_reg, off));
--
2.31.1
^ permalink raw reply related
* [PATCH v4 5/8] bpf ppc64: Add BPF_PROBE_MEM support for JIT
From: Hari Bathini @ 2021-09-29 11:18 UTC (permalink / raw)
To: naveen.n.rao, christophe.leroy, mpe, ast, daniel
Cc: Ravi Bangoria, songliubraving, netdev, john.fastabend, andrii,
kpsingh, paulus, yhs, bpf, linuxppc-dev, kafai, Hari Bathini
In-Reply-To: <20210929111855.50254-1-hbathini@linux.ibm.com>
From: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
BPF load instruction with BPF_PROBE_MEM mode can cause a fault
inside kernel. Append exception table for such instructions
within BPF program.
Unlike other archs which uses extable 'fixup' field to pass dest_reg
and nip, BPF exception table on PowerPC follows the generic PowerPC
exception table design, where it populates both fixup and extable
sections within BPF program. fixup section contains two instructions,
first instruction clears dest_reg and 2nd jumps to next instruction
in the BPF code. extable 'insn' field contains relative offset of
the instruction and 'fixup' field contains relative offset of the
fixup entry. Example layout of BPF program with extable present:
+------------------+
| |
| |
0x4020 -->| ld r27,4(r3) |
| |
| |
0x40ac -->| lwz r3,0(r4) |
| |
| |
|------------------|
0x4280 -->| li r27,0 | \ fixup entry
| b 0x4024 | /
0x4288 -->| li r3,0 |
| b 0x40b0 |
|------------------|
0x4290 -->| insn=0xfffffd90 | \ extable entry
| fixup=0xffffffec | /
0x4298 -->| insn=0xfffffe14 |
| fixup=0xffffffec |
+------------------+
(Addresses shown here are chosen random, not real)
Signed-off-by: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
Signed-off-by: Hari Bathini <hbathini@linux.ibm.com>
---
Changes in v4:
* Dropped explicit fallthrough statement for empty switch cases.
arch/powerpc/net/bpf_jit.h | 8 +++-
arch/powerpc/net/bpf_jit_comp.c | 70 ++++++++++++++++++++++++++++---
arch/powerpc/net/bpf_jit_comp32.c | 2 +-
arch/powerpc/net/bpf_jit_comp64.c | 13 +++++-
4 files changed, 84 insertions(+), 9 deletions(-)
diff --git a/arch/powerpc/net/bpf_jit.h b/arch/powerpc/net/bpf_jit.h
index 0c8f885b8f48..561689a2abdf 100644
--- a/arch/powerpc/net/bpf_jit.h
+++ b/arch/powerpc/net/bpf_jit.h
@@ -141,8 +141,11 @@ struct codegen_context {
unsigned int idx;
unsigned int stack_size;
int b2p[ARRAY_SIZE(b2p)];
+ unsigned int exentry_idx;
};
+#define BPF_FIXUP_LEN 2 /* Two instructions => 8 bytes */
+
static inline void bpf_flush_icache(void *start, void *end)
{
smp_wmb(); /* smp write barrier */
@@ -166,11 +169,14 @@ static inline void bpf_clear_seen_register(struct codegen_context *ctx, int i)
void bpf_jit_emit_func_call_rel(u32 *image, struct codegen_context *ctx, u64 func);
int bpf_jit_build_body(struct bpf_prog *fp, u32 *image, struct codegen_context *ctx,
- u32 *addrs);
+ u32 *addrs, int pass);
void bpf_jit_build_prologue(u32 *image, struct codegen_context *ctx);
void bpf_jit_build_epilogue(u32 *image, struct codegen_context *ctx);
void bpf_jit_realloc_regs(struct codegen_context *ctx);
+int bpf_add_extable_entry(struct bpf_prog *fp, u32 *image, int pass, struct codegen_context *ctx,
+ int insn_idx, int jmp_off, int dst_reg);
+
#endif
#endif
diff --git a/arch/powerpc/net/bpf_jit_comp.c b/arch/powerpc/net/bpf_jit_comp.c
index c5c9e8ad1de7..f02457c6b54f 100644
--- a/arch/powerpc/net/bpf_jit_comp.c
+++ b/arch/powerpc/net/bpf_jit_comp.c
@@ -101,6 +101,8 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *fp)
struct bpf_prog *tmp_fp;
bool bpf_blinded = false;
bool extra_pass = false;
+ u32 extable_len;
+ u32 fixup_len;
if (!fp->jit_requested)
return org_fp;
@@ -131,7 +133,6 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *fp)
image = jit_data->image;
bpf_hdr = jit_data->header;
proglen = jit_data->proglen;
- alloclen = proglen + FUNCTION_DESCR_SIZE;
extra_pass = true;
goto skip_init_ctx;
}
@@ -149,7 +150,7 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *fp)
cgctx.stack_size = round_up(fp->aux->stack_depth, 16);
/* Scouting faux-generate pass 0 */
- if (bpf_jit_build_body(fp, 0, &cgctx, addrs)) {
+ if (bpf_jit_build_body(fp, 0, &cgctx, addrs, 0)) {
/* We hit something illegal or unsupported. */
fp = org_fp;
goto out_addrs;
@@ -162,7 +163,7 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *fp)
*/
if (cgctx.seen & SEEN_TAILCALL) {
cgctx.idx = 0;
- if (bpf_jit_build_body(fp, 0, &cgctx, addrs)) {
+ if (bpf_jit_build_body(fp, 0, &cgctx, addrs, 0)) {
fp = org_fp;
goto out_addrs;
}
@@ -177,8 +178,11 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *fp)
bpf_jit_build_prologue(0, &cgctx);
bpf_jit_build_epilogue(0, &cgctx);
+ fixup_len = fp->aux->num_exentries * BPF_FIXUP_LEN * 4;
+ extable_len = fp->aux->num_exentries * sizeof(struct exception_table_entry);
+
proglen = cgctx.idx * 4;
- alloclen = proglen + FUNCTION_DESCR_SIZE;
+ alloclen = proglen + FUNCTION_DESCR_SIZE + fixup_len + extable_len;
bpf_hdr = bpf_jit_binary_alloc(alloclen, &image, 4, bpf_jit_fill_ill_insns);
if (!bpf_hdr) {
@@ -186,6 +190,9 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *fp)
goto out_addrs;
}
+ if (extable_len)
+ fp->aux->extable = (void *)image + FUNCTION_DESCR_SIZE + proglen + fixup_len;
+
skip_init_ctx:
code_base = (u32 *)(image + FUNCTION_DESCR_SIZE);
@@ -210,7 +217,11 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *fp)
/* Now build the prologue, body code & epilogue for real. */
cgctx.idx = 0;
bpf_jit_build_prologue(code_base, &cgctx);
- bpf_jit_build_body(fp, code_base, &cgctx, addrs);
+ if (bpf_jit_build_body(fp, code_base, &cgctx, addrs, pass)) {
+ bpf_jit_binary_free(bpf_hdr);
+ fp = org_fp;
+ goto out_addrs;
+ }
bpf_jit_build_epilogue(code_base, &cgctx);
if (bpf_jit_enable > 1)
@@ -234,7 +245,7 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *fp)
fp->bpf_func = (void *)image;
fp->jited = 1;
- fp->jited_len = alloclen;
+ fp->jited_len = proglen + FUNCTION_DESCR_SIZE;
bpf_flush_icache(bpf_hdr, (u8 *)bpf_hdr + (bpf_hdr->pages * PAGE_SIZE));
bpf_jit_binary_lock_ro(bpf_hdr);
@@ -258,3 +269,50 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *fp)
return fp;
}
+
+/*
+ * The caller should check for (BPF_MODE(code) == BPF_PROBE_MEM) before calling
+ * this function, as this only applies to BPF_PROBE_MEM, for now.
+ */
+int bpf_add_extable_entry(struct bpf_prog *fp, u32 *image, int pass, struct codegen_context *ctx,
+ int insn_idx, int jmp_off, int dst_reg)
+{
+ off_t offset;
+ unsigned long pc;
+ struct exception_table_entry *ex;
+ u32 *fixup;
+
+ /* Populate extable entries only in the last pass */
+ if (pass != 2)
+ return 0;
+
+ if (!fp->aux->extable ||
+ WARN_ON_ONCE(ctx->exentry_idx >= fp->aux->num_exentries))
+ return -EINVAL;
+
+ pc = (unsigned long)&image[insn_idx];
+
+ fixup = (void *)fp->aux->extable -
+ (fp->aux->num_exentries * BPF_FIXUP_LEN * 4) +
+ (ctx->exentry_idx * BPF_FIXUP_LEN * 4);
+
+ fixup[0] = PPC_RAW_LI(dst_reg, 0);
+
+ fixup[BPF_FIXUP_LEN - 1] =
+ PPC_RAW_BRANCH((long)(pc + jmp_off) - (long)&fixup[BPF_FIXUP_LEN - 1]);
+
+ ex = &fp->aux->extable[ctx->exentry_idx];
+
+ offset = pc - (long)&ex->insn;
+ if (WARN_ON_ONCE(offset >= 0 || offset < INT_MIN))
+ return -ERANGE;
+ ex->insn = offset;
+
+ offset = (long)fixup - (long)&ex->fixup;
+ if (WARN_ON_ONCE(offset >= 0 || offset < INT_MIN))
+ return -ERANGE;
+ ex->fixup = offset;
+
+ ctx->exentry_idx++;
+ return 0;
+}
diff --git a/arch/powerpc/net/bpf_jit_comp32.c b/arch/powerpc/net/bpf_jit_comp32.c
index 65a4d1ed97bf..c04291517a7e 100644
--- a/arch/powerpc/net/bpf_jit_comp32.c
+++ b/arch/powerpc/net/bpf_jit_comp32.c
@@ -266,7 +266,7 @@ static void bpf_jit_emit_tail_call(u32 *image, struct codegen_context *ctx, u32
/* Assemble the body code between the prologue & epilogue */
int bpf_jit_build_body(struct bpf_prog *fp, u32 *image, struct codegen_context *ctx,
- u32 *addrs)
+ u32 *addrs, int pass)
{
const struct bpf_insn *insn = fp->insnsi;
int flen = fp->len;
diff --git a/arch/powerpc/net/bpf_jit_comp64.c b/arch/powerpc/net/bpf_jit_comp64.c
index 49e6e0b6e4d2..4170999371ee 100644
--- a/arch/powerpc/net/bpf_jit_comp64.c
+++ b/arch/powerpc/net/bpf_jit_comp64.c
@@ -272,7 +272,7 @@ static void bpf_jit_emit_tail_call(u32 *image, struct codegen_context *ctx, u32
/* Assemble the body code between the prologue & epilogue */
int bpf_jit_build_body(struct bpf_prog *fp, u32 *image, struct codegen_context *ctx,
- u32 *addrs)
+ u32 *addrs, int pass)
{
const struct bpf_insn *insn = fp->insnsi;
int flen = fp->len;
@@ -717,12 +717,16 @@ int bpf_jit_build_body(struct bpf_prog *fp, u32 *image, struct codegen_context *
*/
/* dst = *(u8 *)(ul) (src + off) */
case BPF_LDX | BPF_MEM | BPF_B:
+ case BPF_LDX | BPF_PROBE_MEM | BPF_B:
/* dst = *(u16 *)(ul) (src + off) */
case BPF_LDX | BPF_MEM | BPF_H:
+ case BPF_LDX | BPF_PROBE_MEM | BPF_H:
/* dst = *(u32 *)(ul) (src + off) */
case BPF_LDX | BPF_MEM | BPF_W:
+ case BPF_LDX | BPF_PROBE_MEM | BPF_W:
/* dst = *(u64 *)(ul) (src + off) */
case BPF_LDX | BPF_MEM | BPF_DW:
+ case BPF_LDX | BPF_PROBE_MEM | BPF_DW:
switch (size) {
case BPF_B:
EMIT(PPC_RAW_LBZ(dst_reg, src_reg, off));
@@ -740,6 +744,13 @@ int bpf_jit_build_body(struct bpf_prog *fp, u32 *image, struct codegen_context *
if (size != BPF_DW && insn_is_zext(&insn[i + 1]))
addrs[++i] = ctx->idx * 4;
+
+ if (BPF_MODE(code) == BPF_PROBE_MEM) {
+ ret = bpf_add_extable_entry(fp, image, pass, ctx, ctx->idx - 1,
+ 4, dst_reg);
+ if (ret)
+ return ret;
+ }
break;
/*
--
2.31.1
^ permalink raw reply related
* [PATCH v4 4/8] powerpc/ppc-opcode: introduce PPC_RAW_BRANCH() macro
From: Hari Bathini @ 2021-09-29 11:18 UTC (permalink / raw)
To: naveen.n.rao, christophe.leroy, mpe, ast, daniel
Cc: songliubraving, netdev, john.fastabend, andrii, kpsingh, paulus,
yhs, bpf, linuxppc-dev, kafai, Hari Bathini
In-Reply-To: <20210929111855.50254-1-hbathini@linux.ibm.com>
Define and use PPC_RAW_BRANCH() macro instead of open coding it. This
macro is used while adding BPF_PROBE_MEM support.
Signed-off-by: Hari Bathini <hbathini@linux.ibm.com>
Reviewed-by: Christophe Leroy <christophe.leroy@csgroup.eu>
---
* No changes in v4.
arch/powerpc/include/asm/ppc-opcode.h | 2 ++
arch/powerpc/net/bpf_jit.h | 4 ++--
2 files changed, 4 insertions(+), 2 deletions(-)
diff --git a/arch/powerpc/include/asm/ppc-opcode.h b/arch/powerpc/include/asm/ppc-opcode.h
index baea657bc868..f50213e2a3e0 100644
--- a/arch/powerpc/include/asm/ppc-opcode.h
+++ b/arch/powerpc/include/asm/ppc-opcode.h
@@ -566,6 +566,8 @@
#define PPC_RAW_MTSPR(spr, d) (0x7c0003a6 | ___PPC_RS(d) | __PPC_SPR(spr))
#define PPC_RAW_EIEIO() (0x7c0006ac)
+#define PPC_RAW_BRANCH(addr) (PPC_INST_BRANCH | ((addr) & 0x03fffffc))
+
/* Deal with instructions that older assemblers aren't aware of */
#define PPC_BCCTR_FLUSH stringify_in_c(.long PPC_INST_BCCTR_FLUSH)
#define PPC_CP_ABORT stringify_in_c(.long PPC_RAW_CP_ABORT)
diff --git a/arch/powerpc/net/bpf_jit.h b/arch/powerpc/net/bpf_jit.h
index 411c63d945c7..0c8f885b8f48 100644
--- a/arch/powerpc/net/bpf_jit.h
+++ b/arch/powerpc/net/bpf_jit.h
@@ -24,8 +24,8 @@
#define EMIT(instr) PLANT_INSTR(image, ctx->idx, instr)
/* Long jump; (unconditional 'branch') */
-#define PPC_JMP(dest) EMIT(PPC_INST_BRANCH | \
- (((dest) - (ctx->idx * 4)) & 0x03fffffc))
+#define PPC_JMP(dest) EMIT(PPC_RAW_BRANCH((dest) - (ctx->idx * 4)))
+
/* blr; (unconditional 'branch' with link) to absolute address */
#define PPC_BL_ABS(dest) EMIT(PPC_INST_BL | \
(((dest) - (unsigned long)(image + ctx->idx)) & 0x03fffffc))
--
2.31.1
^ permalink raw reply related
* [PATCH v4 1/8] bpf powerpc: Remove unused SEEN_STACK
From: Hari Bathini @ 2021-09-29 11:18 UTC (permalink / raw)
To: naveen.n.rao, christophe.leroy, mpe, ast, daniel
Cc: Ravi Bangoria, songliubraving, netdev, john.fastabend, andrii,
kpsingh, paulus, yhs, bpf, linuxppc-dev, kafai
In-Reply-To: <20210929111855.50254-1-hbathini@linux.ibm.com>
From: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
SEEN_STACK is unused on PowerPC. Remove it. Also, have
SEEN_TAILCALL use 0x40000000.
Signed-off-by: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
Reviewed-by: Christophe Leroy <christophe.leroy@csgroup.eu>
---
* No changes in v4.
arch/powerpc/net/bpf_jit.h | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/arch/powerpc/net/bpf_jit.h b/arch/powerpc/net/bpf_jit.h
index 99fad093f43e..d6267e93027a 100644
--- a/arch/powerpc/net/bpf_jit.h
+++ b/arch/powerpc/net/bpf_jit.h
@@ -116,8 +116,7 @@ static inline bool is_nearbranch(int offset)
#define COND_LE (CR0_GT | COND_CMP_FALSE)
#define SEEN_FUNC 0x20000000 /* might call external helpers */
-#define SEEN_STACK 0x40000000 /* uses BPF stack */
-#define SEEN_TAILCALL 0x80000000 /* uses tail calls */
+#define SEEN_TAILCALL 0x40000000 /* uses tail calls */
#define SEEN_VREG_MASK 0x1ff80000 /* Volatile registers r3-r12 */
#define SEEN_NVREG_MASK 0x0003ffff /* Non volatile registers r14-r31 */
--
2.31.1
^ permalink raw reply related
* [PATCH v4 0/8] bpf powerpc: Add BPF_PROBE_MEM support in powerpc JIT compiler
From: Hari Bathini @ 2021-09-29 11:18 UTC (permalink / raw)
To: naveen.n.rao, christophe.leroy, mpe, ast, daniel
Cc: songliubraving, netdev, john.fastabend, andrii, kpsingh, paulus,
yhs, bpf, linuxppc-dev, kafai, Hari Bathini
Patch #1 & #2 are simple cleanup patches. Patch #3 refactors JIT
compiler code with the aim to simplify adding BPF_PROBE_MEM support.
Patch #4 introduces PPC_RAW_BRANCH() macro instead of open coding
branch instruction. Patch #5 & #7 add BPF_PROBE_MEM support for PPC64
& PPC32 JIT compilers respectively. Patch #6 & #8 handle bad userspace
pointers for PPC64 & PPC32 cases respectively.
Changes in v4:
* Addressed all the review comments from Christophe.
Hari Bathini (4):
bpf powerpc: refactor JIT compiler code
powerpc/ppc-opcode: introduce PPC_RAW_BRANCH() macro
bpf ppc32: Add BPF_PROBE_MEM support for JIT
bpf ppc32: Access only if addr is kernel address
Ravi Bangoria (4):
bpf powerpc: Remove unused SEEN_STACK
bpf powerpc: Remove extra_pass from bpf_jit_build_body()
bpf ppc64: Add BPF_PROBE_MEM support for JIT
bpf ppc64: Access only if addr is kernel address
arch/powerpc/include/asm/ppc-opcode.h | 2 +
arch/powerpc/net/bpf_jit.h | 19 +++--
arch/powerpc/net/bpf_jit_comp.c | 72 ++++++++++++++++--
arch/powerpc/net/bpf_jit_comp32.c | 101 ++++++++++++++++++++++----
arch/powerpc/net/bpf_jit_comp64.c | 72 ++++++++++++++----
5 files changed, 224 insertions(+), 42 deletions(-)
--
2.31.1
^ permalink raw reply
* [PATCH v4 2/8] bpf powerpc: Remove extra_pass from bpf_jit_build_body()
From: Hari Bathini @ 2021-09-29 11:18 UTC (permalink / raw)
To: naveen.n.rao, christophe.leroy, mpe, ast, daniel
Cc: Ravi Bangoria, songliubraving, netdev, john.fastabend, andrii,
kpsingh, paulus, yhs, bpf, linuxppc-dev, kafai
In-Reply-To: <20210929111855.50254-1-hbathini@linux.ibm.com>
From: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
In case of extra_pass, usual JIT passes are always skipped. So,
extra_pass is always false while calling bpf_jit_build_body() and
can be removed.
Signed-off-by: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
---
* No changes in v4.
arch/powerpc/net/bpf_jit.h | 2 +-
arch/powerpc/net/bpf_jit_comp.c | 6 +++---
arch/powerpc/net/bpf_jit_comp32.c | 4 ++--
arch/powerpc/net/bpf_jit_comp64.c | 4 ++--
4 files changed, 8 insertions(+), 8 deletions(-)
diff --git a/arch/powerpc/net/bpf_jit.h b/arch/powerpc/net/bpf_jit.h
index d6267e93027a..411c63d945c7 100644
--- a/arch/powerpc/net/bpf_jit.h
+++ b/arch/powerpc/net/bpf_jit.h
@@ -166,7 +166,7 @@ static inline void bpf_clear_seen_register(struct codegen_context *ctx, int i)
void bpf_jit_emit_func_call_rel(u32 *image, struct codegen_context *ctx, u64 func);
int bpf_jit_build_body(struct bpf_prog *fp, u32 *image, struct codegen_context *ctx,
- u32 *addrs, bool extra_pass);
+ u32 *addrs);
void bpf_jit_build_prologue(u32 *image, struct codegen_context *ctx);
void bpf_jit_build_epilogue(u32 *image, struct codegen_context *ctx);
void bpf_jit_realloc_regs(struct codegen_context *ctx);
diff --git a/arch/powerpc/net/bpf_jit_comp.c b/arch/powerpc/net/bpf_jit_comp.c
index 53aefee3fe70..c5c9e8ad1de7 100644
--- a/arch/powerpc/net/bpf_jit_comp.c
+++ b/arch/powerpc/net/bpf_jit_comp.c
@@ -149,7 +149,7 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *fp)
cgctx.stack_size = round_up(fp->aux->stack_depth, 16);
/* Scouting faux-generate pass 0 */
- if (bpf_jit_build_body(fp, 0, &cgctx, addrs, false)) {
+ if (bpf_jit_build_body(fp, 0, &cgctx, addrs)) {
/* We hit something illegal or unsupported. */
fp = org_fp;
goto out_addrs;
@@ -162,7 +162,7 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *fp)
*/
if (cgctx.seen & SEEN_TAILCALL) {
cgctx.idx = 0;
- if (bpf_jit_build_body(fp, 0, &cgctx, addrs, false)) {
+ if (bpf_jit_build_body(fp, 0, &cgctx, addrs)) {
fp = org_fp;
goto out_addrs;
}
@@ -210,7 +210,7 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *fp)
/* Now build the prologue, body code & epilogue for real. */
cgctx.idx = 0;
bpf_jit_build_prologue(code_base, &cgctx);
- bpf_jit_build_body(fp, code_base, &cgctx, addrs, extra_pass);
+ bpf_jit_build_body(fp, code_base, &cgctx, addrs);
bpf_jit_build_epilogue(code_base, &cgctx);
if (bpf_jit_enable > 1)
diff --git a/arch/powerpc/net/bpf_jit_comp32.c b/arch/powerpc/net/bpf_jit_comp32.c
index beb12cbc8c29..b60b59426a24 100644
--- a/arch/powerpc/net/bpf_jit_comp32.c
+++ b/arch/powerpc/net/bpf_jit_comp32.c
@@ -266,7 +266,7 @@ static void bpf_jit_emit_tail_call(u32 *image, struct codegen_context *ctx, u32
/* Assemble the body code between the prologue & epilogue */
int bpf_jit_build_body(struct bpf_prog *fp, u32 *image, struct codegen_context *ctx,
- u32 *addrs, bool extra_pass)
+ u32 *addrs)
{
const struct bpf_insn *insn = fp->insnsi;
int flen = fp->len;
@@ -860,7 +860,7 @@ int bpf_jit_build_body(struct bpf_prog *fp, u32 *image, struct codegen_context *
case BPF_JMP | BPF_CALL:
ctx->seen |= SEEN_FUNC;
- ret = bpf_jit_get_func_addr(fp, &insn[i], extra_pass,
+ ret = bpf_jit_get_func_addr(fp, &insn[i], false,
&func_addr, &func_addr_fixed);
if (ret < 0)
return ret;
diff --git a/arch/powerpc/net/bpf_jit_comp64.c b/arch/powerpc/net/bpf_jit_comp64.c
index b87a63dba9c8..2a87da50d9a4 100644
--- a/arch/powerpc/net/bpf_jit_comp64.c
+++ b/arch/powerpc/net/bpf_jit_comp64.c
@@ -272,7 +272,7 @@ static void bpf_jit_emit_tail_call(u32 *image, struct codegen_context *ctx, u32
/* Assemble the body code between the prologue & epilogue */
int bpf_jit_build_body(struct bpf_prog *fp, u32 *image, struct codegen_context *ctx,
- u32 *addrs, bool extra_pass)
+ u32 *addrs)
{
const struct bpf_insn *insn = fp->insnsi;
int flen = fp->len;
@@ -769,7 +769,7 @@ int bpf_jit_build_body(struct bpf_prog *fp, u32 *image, struct codegen_context *
case BPF_JMP | BPF_CALL:
ctx->seen |= SEEN_FUNC;
- ret = bpf_jit_get_func_addr(fp, &insn[i], extra_pass,
+ ret = bpf_jit_get_func_addr(fp, &insn[i], false,
&func_addr, &func_addr_fixed);
if (ret < 0)
return ret;
--
2.31.1
^ permalink raw reply related
* Re: [PATCH v5 07/11] PCI: Replace pci_dev::driver usage that gets the driver name
From: Simon Horman @ 2021-09-29 10:17 UTC (permalink / raw)
To: Uwe Kleine-König
Cc: linux-pci, Alexander Duyck, oss-drivers, Paul Mackerras,
Christoph Hellwig, Herbert Xu, Rafał Miłecki,
Jesse Brandeburg, Bjorn Helgaas, Ido Schimmel, Jakub Kicinski,
Yisen Zhuang, Vadym Kochan, Michael Buesch, Jiri Pirko,
Salil Mehta, netdev, linux-wireless, linux-kernel, Taras Chornyi,
Zhou Wang, linux-crypto, kernel, Oliver O'Halloran,
linuxppc-dev, David S. Miller
In-Reply-To: <20210929085306.2203850-8-u.kleine-koenig@pengutronix.de>
On Wed, Sep 29, 2021 at 10:53:02AM +0200, Uwe Kleine-König wrote:
> struct pci_dev::driver holds (apart from a constant offset) the same
> data as struct pci_dev::dev->driver. With the goal to remove struct
> pci_dev::driver to get rid of data duplication replace getting the
> driver name by dev_driver_string() which implicitly makes use of struct
> pci_dev::dev->driver.
>
> Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
> ---
> drivers/crypto/hisilicon/qm.c | 2 +-
> drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.c | 2 +-
> drivers/net/ethernet/marvell/prestera/prestera_pci.c | 2 +-
> drivers/net/ethernet/mellanox/mlxsw/pci.c | 2 +-
> drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c | 3 ++-
> 5 files changed, 6 insertions(+), 5 deletions(-)
Thanks Uwe.
For NFP:
Acked-by: Simon Horman <simon.horman@corigine.com>
^ permalink raw reply
* [powerpc:merge] BUILD SUCCESS 044c2d99d9f43c6d6fde8bed00672517dd9a5a57
From: kernel test robot @ 2021-09-29 10:04 UTC (permalink / raw)
To: Michael Ellerman; +Cc: linuxppc-dev
tree/branch: https://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux.git merge
branch HEAD: 044c2d99d9f43c6d6fde8bed00672517dd9a5a57 Automatic merge of 'fixes' into merge (2021-09-19 22:18)
elapsed time: 730m
configs tested: 118
configs skipped: 3
The following configs have been built successfully.
More configs may be tested in the coming days.
gcc tested configs:
arm defconfig
arm64 allyesconfig
arm64 defconfig
arm allyesconfig
arm allmodconfig
i386 randconfig-c001-20210929
powerpc amigaone_defconfig
powerpc ksi8560_defconfig
sh se7722_defconfig
arm pleb_defconfig
arm ep93xx_defconfig
arm am200epdkit_defconfig
mips maltasmvp_defconfig
powerpc mpc837x_rdb_defconfig
mips loongson1c_defconfig
mips mpc30x_defconfig
riscv nommu_k210_sdcard_defconfig
powerpc tqm8560_defconfig
arm hackkit_defconfig
mips ci20_defconfig
arm mmp2_defconfig
arm assabet_defconfig
m68k mac_defconfig
sh sh2007_defconfig
arm ixp4xx_defconfig
sh urquell_defconfig
mips ip27_defconfig
mips fuloong2e_defconfig
sh r7780mp_defconfig
m68k m5272c3_defconfig
powerpc linkstation_defconfig
sh shx3_defconfig
microblaze mmu_defconfig
x86_64 randconfig-c001-20210929
arm randconfig-c002-20210929
x86_64 randconfig-c001-20210928
arm randconfig-c002-20210928
i386 randconfig-c001-20210928
ia64 allmodconfig
ia64 defconfig
ia64 allyesconfig
m68k allmodconfig
m68k defconfig
m68k allyesconfig
nios2 defconfig
arc allyesconfig
nds32 allnoconfig
nds32 defconfig
nios2 allyesconfig
csky defconfig
alpha defconfig
alpha allyesconfig
xtensa allyesconfig
h8300 allyesconfig
arc defconfig
sh allmodconfig
parisc defconfig
s390 allyesconfig
s390 allmodconfig
parisc allyesconfig
s390 defconfig
i386 allyesconfig
sparc allyesconfig
sparc defconfig
i386 defconfig
mips allyesconfig
mips allmodconfig
powerpc allyesconfig
powerpc allmodconfig
powerpc allnoconfig
x86_64 randconfig-a014-20210928
x86_64 randconfig-a011-20210928
x86_64 randconfig-a013-20210928
x86_64 randconfig-a012-20210928
x86_64 randconfig-a015-20210928
x86_64 randconfig-a016-20210928
i386 randconfig-a014-20210928
i386 randconfig-a013-20210928
i386 randconfig-a016-20210928
i386 randconfig-a011-20210928
i386 randconfig-a015-20210928
i386 randconfig-a012-20210928
arc randconfig-r043-20210928
riscv randconfig-r042-20210928
s390 randconfig-r044-20210928
riscv nommu_k210_defconfig
riscv allyesconfig
riscv nommu_virt_defconfig
riscv allnoconfig
riscv defconfig
riscv rv32_defconfig
riscv allmodconfig
x86_64 rhel-8.3-kselftests
um x86_64_defconfig
um i386_defconfig
x86_64 allyesconfig
x86_64 defconfig
x86_64 rhel-8.3
x86_64 kexec
clang tested configs:
powerpc randconfig-c003-20210929
mips randconfig-c004-20210929
arm randconfig-c002-20210929
x86_64 randconfig-c007-20210929
riscv randconfig-c006-20210929
s390 randconfig-c005-20210929
i386 randconfig-c001-20210929
x86_64 randconfig-a002-20210928
x86_64 randconfig-a005-20210928
x86_64 randconfig-a001-20210928
x86_64 randconfig-a006-20210928
x86_64 randconfig-a003-20210928
x86_64 randconfig-a004-20210928
i386 randconfig-a001-20210928
i386 randconfig-a005-20210928
i386 randconfig-a002-20210928
i386 randconfig-a006-20210928
i386 randconfig-a004-20210928
i386 randconfig-a003-20210928
hexagon randconfig-r045-20210928
hexagon randconfig-r041-20210928
---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org
^ permalink raw reply
* Re: [PATCH v4 4/8] PCI: replace pci_dev::driver usage that gets the driver name
From: Uwe Kleine-König @ 2021-09-29 9:04 UTC (permalink / raw)
To: Simon Horman
Cc: linux-pci, Alexander Duyck, oss-drivers, Paul Mackerras,
Herbert Xu, Ido Schimmel, Rafał Miłecki,
Jesse Brandeburg, Bjorn Helgaas, Jakub Kicinski, Yisen Zhuang,
Uwe Kleine-König, Vadym Kochan, Michael Buesch, Jiri Pirko,
Salil Mehta, netdev, linux-wireless, linux-kernel, Taras Chornyi,
Zhou Wang, linux-crypto, kernel, Oliver O'Halloran,
linuxppc-dev, David S. Miller
In-Reply-To: <20210929080541.GA13506@corigine.com>
[-- Attachment #1: Type: text/plain, Size: 3758 bytes --]
Hello Simon,
On Wed, Sep 29, 2021 at 10:05:42AM +0200, Simon Horman wrote:
> On Tue, Sep 28, 2021 at 12:31:29PM +0200, Uwe Kleine-König wrote:
> > On Tue, Sep 28, 2021 at 12:01:28PM +0200, Simon Horman wrote:
> > > On Mon, Sep 27, 2021 at 10:43:22PM +0200, Uwe Kleine-König wrote:
> > > > From: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
> > > >
> > > > struct pci_dev::driver holds (apart from a constant offset) the same
> > > > data as struct pci_dev::dev->driver. With the goal to remove struct
> > > > pci_dev::driver to get rid of data duplication replace getting the
> > > > driver name by dev_driver_string() which implicitly makes use of struct
> > > > pci_dev::dev->driver.
> > > >
> > > > Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
> > >
> > > ...
> > >
> > > > diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c b/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c
> > > > index 0685ece1f155..23dfb599c828 100644
> > > > --- a/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c
> > > > +++ b/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c
> > > > @@ -202,7 +202,7 @@ nfp_get_drvinfo(struct nfp_app *app, struct pci_dev *pdev,
> > > > {
> > > > char nsp_version[ETHTOOL_FWVERS_LEN] = {};
> > > >
> > > > - strlcpy(drvinfo->driver, pdev->driver->name, sizeof(drvinfo->driver));
> > > > + strlcpy(drvinfo->driver, dev_driver_string(&pdev->dev), sizeof(drvinfo->driver));
> > >
> > > I'd slightly prefer to maintain lines under 80 columns wide.
> > > But not nearly strongly enough to engage in a long debate about it.
> >
> > :-)
> >
> > Looking at the output of
> >
> > git grep strlcpy.\*sizeof
> >
> > I wonder if it would be sensible to introduce something like
> >
> > #define strlcpy_array(arr, src) (strlcpy(arr, src, sizeof(arr)) + __must_be_array(arr))
> >
> > but not sure this is possible without a long debate either (and this
> > line is over 80 chars wide, too :-).
>
> My main motivation for the 80 char limit in nfp_net_ethtool.c is
> not that I think 80 char is universally a good limit (although that is true),
> but rather that I expect that is the prevailing style in nfp_net_ethtool.c.
I sent out v5 with an additional line break now.
> So a macro more than 80 car wide somewhere else is fine by me.
>
> However, when running checkpatch --strict over the patch it told me:
>
> WARNING: Prefer strscpy over strlcpy - see: https://lore.kernel.org/r/CAHk-=wgfRnXz0W3D37d01q3JFkr_i_uTL=V6A6G1oUZcprmknw@mail.gmail.com/
> #276: FILE: drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c:205:
> + strlcpy(drvinfo->driver, dev_driver_string(&pdev->dev), sizeof(drvinfo->driver));
>
> total: 0 errors, 1 warnings, 0 checks, 80 lines checked
>
> (Amusingly, more text wider than 80 column, perhaps suggesting the folly of
> my original comment, but lets move on from that.)
>
> As your patch doesn't introduce the usage of strlcpy() I was considering a
> follow-up patch to change it to strscpy(). And in general the email at the
> link above suggests all usages of strlcpy() should do so. So perhaps
> creating strscpy_array is a better idea?
What I read about strscpy() is that conversions for the sake of the
conversion are not welcome. When such a conversion comes from someone
involved with the driver that is also tested this is probably fine.
> I have not thought about this much, and probably this just leads us to a
> deeper part of the rabbit hole.
I assume so, too.
Best regards
Uwe
--
Pengutronix e.K. | Uwe Kleine-König |
Industrial Linux Solutions | https://www.pengutronix.de/ |
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]
^ permalink raw reply
* [PATCH v5 10/11] PCI: Replace pci_dev::driver usage by pci_dev::dev.driver
From: Uwe Kleine-König @ 2021-09-29 8:53 UTC (permalink / raw)
To: Bjorn Helgaas
Cc: Mark Rutland, Peter Zijlstra, H. Peter Anvin,
Oliver O'Halloran, Jiri Olsa, Christoph Hellwig,
Stefano Stabellini, Mathias Nyman, x86, Alexander Shishkin,
Ingo Molnar, linux-pci, xen-devel, Juergen Gross,
Andrew Donnellan, Arnd Bergmann, Konrad Rzeszutek Wilk,
Arnaldo Carvalho de Melo, Borislav Petkov, Bjorn Helgaas,
Namhyung Kim, Thomas Gleixner, Boris Ostrovsky,
Greg Kroah-Hartman, linux-usb, linux-perf-users, kernel,
Frederic Barrat, Paul Mackerras, linuxppc-dev
In-Reply-To: <20210929085306.2203850-1-u.kleine-koenig@pengutronix.de>
struct pci_dev::driver contains (apart from a constant offset) the same
data as struct pci_dev::dev->driver. Replace all remaining users of the
former pointer by the latter to allow removing the former.
Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
---
arch/powerpc/kernel/eeh_driver.c | 10 ++++-----
arch/x86/events/intel/uncore.c | 2 +-
arch/x86/kernel/probe_roms.c | 2 +-
drivers/misc/cxl/guest.c | 24 ++++++++++++---------
drivers/misc/cxl/pci.c | 30 ++++++++++++++++----------
drivers/pci/iov.c | 25 ++++++++++++++--------
drivers/pci/pci-driver.c | 25 +++++++++++-----------
drivers/pci/pci.c | 4 ++--
drivers/pci/pcie/err.c | 36 ++++++++++++++++++--------------
drivers/pci/xen-pcifront.c | 4 ++--
drivers/usb/host/xhci-pci.c | 2 +-
11 files changed, 93 insertions(+), 71 deletions(-)
diff --git a/arch/powerpc/kernel/eeh_driver.c b/arch/powerpc/kernel/eeh_driver.c
index 3eff6a4888e7..350dab18e137 100644
--- a/arch/powerpc/kernel/eeh_driver.c
+++ b/arch/powerpc/kernel/eeh_driver.c
@@ -104,13 +104,13 @@ static bool eeh_edev_actionable(struct eeh_dev *edev)
*/
static inline struct pci_driver *eeh_pcid_get(struct pci_dev *pdev)
{
- if (!pdev || !pdev->driver)
+ if (!pdev || !pdev->dev.driver)
return NULL;
- if (!try_module_get(pdev->driver->driver.owner))
+ if (!try_module_get(pdev->dev.driver->owner))
return NULL;
- return pdev->driver;
+ return to_pci_driver(pdev->dev.driver);
}
/**
@@ -122,10 +122,10 @@ static inline struct pci_driver *eeh_pcid_get(struct pci_dev *pdev)
*/
static inline void eeh_pcid_put(struct pci_dev *pdev)
{
- if (!pdev || !pdev->driver)
+ if (!pdev || !pdev->dev.driver)
return;
- module_put(pdev->driver->driver.owner);
+ module_put(pdev->dev.driver->owner);
}
/**
diff --git a/arch/x86/events/intel/uncore.c b/arch/x86/events/intel/uncore.c
index c72e368dd164..f1ba6ab2e97e 100644
--- a/arch/x86/events/intel/uncore.c
+++ b/arch/x86/events/intel/uncore.c
@@ -1187,7 +1187,7 @@ static int uncore_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id
* PCI slot and func to indicate the uncore box.
*/
if (id->driver_data & ~0xffff) {
- struct pci_driver *pci_drv = pdev->driver;
+ struct pci_driver *pci_drv = to_pci_driver(pdev->dev.driver);
pmu = uncore_pci_find_dev_pmu(pdev, pci_drv->id_table);
if (pmu == NULL)
diff --git a/arch/x86/kernel/probe_roms.c b/arch/x86/kernel/probe_roms.c
index 9e1def3744f2..36e84d904260 100644
--- a/arch/x86/kernel/probe_roms.c
+++ b/arch/x86/kernel/probe_roms.c
@@ -80,7 +80,7 @@ static struct resource video_rom_resource = {
*/
static bool match_id(struct pci_dev *pdev, unsigned short vendor, unsigned short device)
{
- struct pci_driver *drv = pdev->driver;
+ struct pci_driver *drv = to_pci_driver(pdev->dev.driver);
const struct pci_device_id *id;
if (pdev->vendor == vendor && pdev->device == device)
diff --git a/drivers/misc/cxl/guest.c b/drivers/misc/cxl/guest.c
index 186308f1f8eb..d997c9c3ebb5 100644
--- a/drivers/misc/cxl/guest.c
+++ b/drivers/misc/cxl/guest.c
@@ -25,28 +25,32 @@ static void pci_error_handlers(struct cxl_afu *afu,
return;
list_for_each_entry(afu_dev, &afu->phb->bus->devices, bus_list) {
- if (!afu_dev->driver)
+ struct pci_driver *afu_drv;
+
+ if (!afu_dev->dev.driver)
continue;
+ afu_drv = to_pci_driver(afu_dev->dev.driver);
+
switch (bus_error_event) {
case CXL_ERROR_DETECTED_EVENT:
afu_dev->error_state = state;
- if (afu_dev->driver->err_handler &&
- afu_dev->driver->err_handler->error_detected)
- afu_dev->driver->err_handler->error_detected(afu_dev, state);
+ if (afu_drv->err_handler &&
+ afu_drv->err_handler->error_detected)
+ afu_drv->err_handler->error_detected(afu_dev, state);
break;
case CXL_SLOT_RESET_EVENT:
afu_dev->error_state = state;
- if (afu_dev->driver->err_handler &&
- afu_dev->driver->err_handler->slot_reset)
- afu_dev->driver->err_handler->slot_reset(afu_dev);
+ if (afu_drv->err_handler &&
+ afu_drv->err_handler->slot_reset)
+ afu_drv->err_handler->slot_reset(afu_dev);
break;
case CXL_RESUME_EVENT:
- if (afu_dev->driver->err_handler &&
- afu_dev->driver->err_handler->resume)
- afu_dev->driver->err_handler->resume(afu_dev);
+ if (afu_drv->err_handler &&
+ afu_drv->err_handler->resume)
+ afu_drv->err_handler->resume(afu_dev);
break;
}
}
diff --git a/drivers/misc/cxl/pci.c b/drivers/misc/cxl/pci.c
index 2ba899f5659f..7e7545d01e27 100644
--- a/drivers/misc/cxl/pci.c
+++ b/drivers/misc/cxl/pci.c
@@ -1805,14 +1805,16 @@ static pci_ers_result_t cxl_vphb_error_detected(struct cxl_afu *afu,
return result;
list_for_each_entry(afu_dev, &afu->phb->bus->devices, bus_list) {
- if (!afu_dev->driver)
+ struct pci_driver *afu_drv;
+ if (!afu_dev->dev.driver)
continue;
+ afu_drv = to_pci_driver(afu_dev->dev.driver);
+
afu_dev->error_state = state;
- if (afu_dev->driver->err_handler)
- afu_result = afu_dev->driver->err_handler->error_detected(afu_dev,
- state);
+ if (afu_drv->err_handler)
+ afu_result = afu_drv->err_handler->error_detected(afu_dev, state);
/* Disconnect trumps all, NONE trumps NEED_RESET */
if (afu_result == PCI_ERS_RESULT_DISCONNECT)
result = PCI_ERS_RESULT_DISCONNECT;
@@ -2003,6 +2005,8 @@ static pci_ers_result_t cxl_pci_slot_reset(struct pci_dev *pdev)
continue;
list_for_each_entry(afu_dev, &afu->phb->bus->devices, bus_list) {
+ struct pci_driver *afu_drv;
+
/* Reset the device context.
* TODO: make this less disruptive
*/
@@ -2028,12 +2032,14 @@ static pci_ers_result_t cxl_pci_slot_reset(struct pci_dev *pdev)
* shouldn't start new work until we call
* their resume function.
*/
- if (!afu_dev->driver)
+ if (!afu_dev->dev.driver)
continue;
- if (afu_dev->driver->err_handler &&
- afu_dev->driver->err_handler->slot_reset)
- afu_result = afu_dev->driver->err_handler->slot_reset(afu_dev);
+ afu_drv = to_pci_driver(afu_dev->dev.driver);
+
+ if (afu_drv->err_handler &&
+ afu_drv->err_handler->slot_reset)
+ afu_result = afu_drv->err_handler->slot_reset(afu_dev);
if (afu_result == PCI_ERS_RESULT_DISCONNECT)
result = PCI_ERS_RESULT_DISCONNECT;
@@ -2074,9 +2080,11 @@ static void cxl_pci_resume(struct pci_dev *pdev)
continue;
list_for_each_entry(afu_dev, &afu->phb->bus->devices, bus_list) {
- if (afu_dev->driver && afu_dev->driver->err_handler &&
- afu_dev->driver->err_handler->resume)
- afu_dev->driver->err_handler->resume(afu_dev);
+ struct pci_driver *afu_drv;
+ if (afu_dev->dev.driver &&
+ (afu_drv = to_pci_driver(afu_dev->dev.driver))->err_handler &&
+ afu_drv->err_handler->resume)
+ afu_drv->err_handler->resume(afu_dev);
}
}
spin_unlock(&adapter->afu_list_lock);
diff --git a/drivers/pci/iov.c b/drivers/pci/iov.c
index dafdc652fcd0..f94660927544 100644
--- a/drivers/pci/iov.c
+++ b/drivers/pci/iov.c
@@ -164,13 +164,15 @@ static ssize_t sriov_vf_total_msix_show(struct device *dev,
char *buf)
{
struct pci_dev *pdev = to_pci_dev(dev);
+ struct pci_driver *pdrv;
u32 vf_total_msix = 0;
device_lock(dev);
- if (!pdev->driver || !pdev->driver->sriov_get_vf_total_msix)
+ pdrv = to_pci_driver(dev->driver);
+ if (!pdrv || !pdrv->sriov_get_vf_total_msix)
goto unlock;
- vf_total_msix = pdev->driver->sriov_get_vf_total_msix(pdev);
+ vf_total_msix = pdrv->sriov_get_vf_total_msix(pdev);
unlock:
device_unlock(dev);
return sysfs_emit(buf, "%u\n", vf_total_msix);
@@ -183,6 +185,7 @@ static ssize_t sriov_vf_msix_count_store(struct device *dev,
{
struct pci_dev *vf_dev = to_pci_dev(dev);
struct pci_dev *pdev = pci_physfn(vf_dev);
+ struct pci_driver *pdrv;
int val, ret;
ret = kstrtoint(buf, 0, &val);
@@ -193,13 +196,14 @@ static ssize_t sriov_vf_msix_count_store(struct device *dev,
return -EINVAL;
device_lock(&pdev->dev);
- if (!pdev->driver || !pdev->driver->sriov_set_msix_vec_count) {
+ pdrv = to_pci_driver(pdev->dev.driver);
+ if (!pdrv || !pdrv->sriov_set_msix_vec_count) {
ret = -EOPNOTSUPP;
goto err_pdev;
}
device_lock(&vf_dev->dev);
- if (vf_dev->driver) {
+ if (vf_dev->dev.driver) {
/*
* A driver is already attached to this VF and has configured
* itself based on the current MSI-X vector count. Changing
@@ -209,7 +213,7 @@ static ssize_t sriov_vf_msix_count_store(struct device *dev,
goto err_dev;
}
- ret = pdev->driver->sriov_set_msix_vec_count(vf_dev, val);
+ ret = pdrv->sriov_set_msix_vec_count(vf_dev, val);
err_dev:
device_unlock(&vf_dev->dev);
@@ -376,6 +380,7 @@ static ssize_t sriov_numvfs_store(struct device *dev,
const char *buf, size_t count)
{
struct pci_dev *pdev = to_pci_dev(dev);
+ struct pci_driver *pdrv;
int ret;
u16 num_vfs;
@@ -392,14 +397,16 @@ static ssize_t sriov_numvfs_store(struct device *dev,
goto exit;
/* is PF driver loaded */
- if (!pdev->driver) {
+ if (!pdev->dev.driver) {
pci_info(pdev, "no driver bound to device; cannot configure SR-IOV\n");
ret = -ENOENT;
goto exit;
}
+ pdrv = to_pci_driver(pdev->dev.driver);
+
/* is PF driver loaded w/callback */
- if (!pdev->driver->sriov_configure) {
+ if (!pdrv->sriov_configure) {
pci_info(pdev, "driver does not support SR-IOV configuration via sysfs\n");
ret = -ENOENT;
goto exit;
@@ -407,7 +414,7 @@ static ssize_t sriov_numvfs_store(struct device *dev,
if (num_vfs == 0) {
/* disable VFs */
- ret = pdev->driver->sriov_configure(pdev, 0);
+ ret = pdrv->sriov_configure(pdev, 0);
goto exit;
}
@@ -419,7 +426,7 @@ static ssize_t sriov_numvfs_store(struct device *dev,
goto exit;
}
- ret = pdev->driver->sriov_configure(pdev, num_vfs);
+ ret = pdrv->sriov_configure(pdev, num_vfs);
if (ret < 0)
goto exit;
diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c
index 50449ec622a3..4d20022b8631 100644
--- a/drivers/pci/pci-driver.c
+++ b/drivers/pci/pci-driver.c
@@ -457,7 +457,7 @@ static int pci_device_probe(struct device *dev)
static void pci_device_remove(struct device *dev)
{
struct pci_dev *pci_dev = to_pci_dev(dev);
- struct pci_driver *drv = pci_dev->driver;
+ struct pci_driver *drv = to_pci_driver(pci_dev->dev.driver);
if (drv->remove) {
pm_runtime_get_sync(dev);
@@ -493,7 +493,7 @@ static void pci_device_remove(struct device *dev)
static void pci_device_shutdown(struct device *dev)
{
struct pci_dev *pci_dev = to_pci_dev(dev);
- struct pci_driver *drv = pci_dev->driver;
+ struct pci_driver *drv = to_pci_driver(pci_dev->dev.driver);
pm_runtime_resume(dev);
@@ -589,7 +589,7 @@ static int pci_pm_reenable_device(struct pci_dev *pci_dev)
static int pci_legacy_suspend(struct device *dev, pm_message_t state)
{
struct pci_dev *pci_dev = to_pci_dev(dev);
- struct pci_driver *drv = pci_dev->driver;
+ struct pci_driver *drv = to_pci_driver(dev->driver);
if (drv && drv->suspend) {
pci_power_t prev = pci_dev->current_state;
@@ -630,7 +630,7 @@ static int pci_legacy_suspend_late(struct device *dev, pm_message_t state)
static int pci_legacy_resume(struct device *dev)
{
struct pci_dev *pci_dev = to_pci_dev(dev);
- struct pci_driver *drv = pci_dev->driver;
+ struct pci_driver *drv = to_pci_driver(pci_dev->dev.driver);
pci_fixup_device(pci_fixup_resume, pci_dev);
@@ -649,7 +649,7 @@ static void pci_pm_default_suspend(struct pci_dev *pci_dev)
static bool pci_has_legacy_pm_support(struct pci_dev *pci_dev)
{
- struct pci_driver *drv = pci_dev->driver;
+ struct pci_driver *drv = to_pci_driver(pci_dev->dev.driver);
bool ret = drv && (drv->suspend || drv->resume);
/*
@@ -1242,11 +1242,11 @@ static int pci_pm_runtime_suspend(struct device *dev)
int error;
/*
- * If pci_dev->driver is not set (unbound), we leave the device in D0,
+ * If pci_dev->dev.driver is not set (unbound), we leave the device in D0,
* but it may go to D3cold when the bridge above it runtime suspends.
* Save its config space in case that happens.
*/
- if (!pci_dev->driver) {
+ if (!pci_dev->dev.driver) {
pci_save_state(pci_dev);
return 0;
}
@@ -1303,7 +1303,7 @@ static int pci_pm_runtime_resume(struct device *dev)
*/
pci_restore_standard_config(pci_dev);
- if (!pci_dev->driver)
+ if (!dev->driver)
return 0;
pci_fixup_device(pci_fixup_resume_early, pci_dev);
@@ -1322,14 +1322,13 @@ static int pci_pm_runtime_resume(struct device *dev)
static int pci_pm_runtime_idle(struct device *dev)
{
- struct pci_dev *pci_dev = to_pci_dev(dev);
const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL;
/*
- * If pci_dev->driver is not set (unbound), the device should
+ * If dev->driver is not set (unbound), the device should
* always remain in D0 regardless of the runtime PM status
*/
- if (!pci_dev->driver)
+ if (!dev->driver)
return 0;
if (!pm)
@@ -1436,8 +1435,8 @@ static struct pci_driver pci_compat_driver = {
*/
struct pci_driver *pci_dev_driver(const struct pci_dev *dev)
{
- if (dev->driver)
- return dev->driver;
+ if (dev->dev.driver)
+ return to_pci_driver(dev->dev.driver);
else {
int i;
for (i = 0; i <= PCI_ROM_RESOURCE; i++)
diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index ce2ab62b64cf..ccecf740de59 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c
@@ -5089,7 +5089,7 @@ EXPORT_SYMBOL_GPL(pci_dev_unlock);
static void pci_dev_save_and_disable(struct pci_dev *dev)
{
const struct pci_error_handlers *err_handler =
- dev->driver ? dev->driver->err_handler : NULL;
+ dev->dev.driver ? to_pci_driver(dev->dev.driver)->err_handler : NULL;
/*
* dev->driver->err_handler->reset_prepare() is protected against
@@ -5120,7 +5120,7 @@ static void pci_dev_save_and_disable(struct pci_dev *dev)
static void pci_dev_restore(struct pci_dev *dev)
{
const struct pci_error_handlers *err_handler =
- dev->driver ? dev->driver->err_handler : NULL;
+ dev->dev.driver ? to_pci_driver(dev->dev.driver)->err_handler : NULL;
pci_restore_state(dev);
diff --git a/drivers/pci/pcie/err.c b/drivers/pci/pcie/err.c
index b576aa890c76..b314b54f7821 100644
--- a/drivers/pci/pcie/err.c
+++ b/drivers/pci/pcie/err.c
@@ -49,14 +49,15 @@ static int report_error_detected(struct pci_dev *dev,
pci_channel_state_t state,
enum pci_ers_result *result)
{
+ struct pci_driver *pdrv;
pci_ers_result_t vote;
const struct pci_error_handlers *err_handler;
device_lock(&dev->dev);
if (!pci_dev_set_io_state(dev, state) ||
- !dev->driver ||
- !dev->driver->err_handler ||
- !dev->driver->err_handler->error_detected) {
+ !dev->dev.driver ||
+ !(pdrv = to_pci_driver(dev->dev.driver))->err_handler ||
+ !pdrv->err_handler->error_detected) {
/*
* If any device in the subtree does not have an error_detected
* callback, PCI_ERS_RESULT_NO_AER_DRIVER prevents subsequent
@@ -70,7 +71,7 @@ static int report_error_detected(struct pci_dev *dev,
vote = PCI_ERS_RESULT_NONE;
}
} else {
- err_handler = dev->driver->err_handler;
+ err_handler = pdrv->err_handler;
vote = err_handler->error_detected(dev, state);
}
pci_uevent_ers(dev, vote);
@@ -92,15 +93,16 @@ static int report_normal_detected(struct pci_dev *dev, void *data)
static int report_mmio_enabled(struct pci_dev *dev, void *data)
{
pci_ers_result_t vote, *result = data;
+ struct pci_driver *pdrv;
const struct pci_error_handlers *err_handler;
device_lock(&dev->dev);
- if (!dev->driver ||
- !dev->driver->err_handler ||
- !dev->driver->err_handler->mmio_enabled)
+ if (!dev->dev.driver ||
+ !(pdrv = to_pci_driver(dev->dev.driver))->err_handler ||
+ !pdrv->err_handler->mmio_enabled)
goto out;
- err_handler = dev->driver->err_handler;
+ err_handler = pdrv->err_handler;
vote = err_handler->mmio_enabled(dev);
*result = merge_result(*result, vote);
out:
@@ -112,14 +114,15 @@ static int report_slot_reset(struct pci_dev *dev, void *data)
{
pci_ers_result_t vote, *result = data;
const struct pci_error_handlers *err_handler;
+ struct pci_driver *pdrv;
device_lock(&dev->dev);
- if (!dev->driver ||
- !dev->driver->err_handler ||
- !dev->driver->err_handler->slot_reset)
+ if (!dev->dev.driver ||
+ !(pdrv = to_pci_driver(dev->dev.driver))->err_handler ||
+ !pdrv->err_handler->slot_reset)
goto out;
- err_handler = dev->driver->err_handler;
+ err_handler = pdrv->err_handler;
vote = err_handler->slot_reset(dev);
*result = merge_result(*result, vote);
out:
@@ -130,15 +133,16 @@ static int report_slot_reset(struct pci_dev *dev, void *data)
static int report_resume(struct pci_dev *dev, void *data)
{
const struct pci_error_handlers *err_handler;
+ struct pci_driver *pdrv;
device_lock(&dev->dev);
if (!pci_dev_set_io_state(dev, pci_channel_io_normal) ||
- !dev->driver ||
- !dev->driver->err_handler ||
- !dev->driver->err_handler->resume)
+ !dev->dev.driver ||
+ !(pdrv = to_pci_driver(dev->dev.driver))->err_handler ||
+ !pdrv->err_handler->resume)
goto out;
- err_handler = dev->driver->err_handler;
+ err_handler = pdrv->err_handler;
err_handler->resume(dev);
out:
pci_uevent_ers(dev, PCI_ERS_RESULT_RECOVERED);
diff --git a/drivers/pci/xen-pcifront.c b/drivers/pci/xen-pcifront.c
index f2d7f70a7a10..73831fb87a1e 100644
--- a/drivers/pci/xen-pcifront.c
+++ b/drivers/pci/xen-pcifront.c
@@ -601,12 +601,12 @@ static pci_ers_result_t pcifront_common_process(int cmd,
result = PCI_ERS_RESULT_NONE;
pcidev = pci_get_domain_bus_and_slot(domain, bus, devfn);
- if (!pcidev || !pcidev->driver) {
+ if (!pcidev || !pcidev->dev.driver) {
dev_err(&pdev->xdev->dev, "device or AER driver is NULL\n");
pci_dev_put(pcidev);
return result;
}
- pdrv = pcidev->driver;
+ pdrv = to_pci_driver(pcidev->dev.driver);
if (pdrv->err_handler && pdrv->err_handler->error_detected) {
pci_dbg(pcidev, "trying to call AER service\n");
diff --git a/drivers/usb/host/xhci-pci.c b/drivers/usb/host/xhci-pci.c
index 2c9f25ca8edd..2f4729f4f1e0 100644
--- a/drivers/usb/host/xhci-pci.c
+++ b/drivers/usb/host/xhci-pci.c
@@ -103,7 +103,7 @@ static void xhci_pci_quirks(struct device *dev, struct xhci_hcd *xhci)
struct xhci_driver_data *driver_data;
const struct pci_device_id *id;
- id = pci_match_id(pdev->driver->id_table, pdev);
+ id = pci_match_id(to_pci_driver(pdev->dev.driver)->id_table, pdev);
if (id && id->driver_data) {
driver_data = (struct xhci_driver_data *)id->driver_data;
--
2.30.2
^ permalink raw reply related
* [PATCH v5 07/11] PCI: Replace pci_dev::driver usage that gets the driver name
From: Uwe Kleine-König @ 2021-09-29 8:53 UTC (permalink / raw)
To: Bjorn Helgaas
Cc: linux-pci, Alexander Duyck, oss-drivers, Paul Mackerras,
Christoph Hellwig, Herbert Xu, Rafał Miłecki,
Jesse Brandeburg, Ido Schimmel, Jakub Kicinski, Yisen Zhuang,
Vadym Kochan, Michael Buesch, Jiri Pirko, Salil Mehta, netdev,
linux-wireless, linux-kernel, Taras Chornyi, Zhou Wang,
linux-crypto, kernel, Simon Horman, Oliver O'Halloran,
linuxppc-dev, David S. Miller
In-Reply-To: <20210929085306.2203850-1-u.kleine-koenig@pengutronix.de>
struct pci_dev::driver holds (apart from a constant offset) the same
data as struct pci_dev::dev->driver. With the goal to remove struct
pci_dev::driver to get rid of data duplication replace getting the
driver name by dev_driver_string() which implicitly makes use of struct
pci_dev::dev->driver.
Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
---
drivers/crypto/hisilicon/qm.c | 2 +-
drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.c | 2 +-
drivers/net/ethernet/marvell/prestera/prestera_pci.c | 2 +-
drivers/net/ethernet/mellanox/mlxsw/pci.c | 2 +-
drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c | 3 ++-
5 files changed, 6 insertions(+), 5 deletions(-)
diff --git a/drivers/crypto/hisilicon/qm.c b/drivers/crypto/hisilicon/qm.c
index 369562d34d66..8f361e54e524 100644
--- a/drivers/crypto/hisilicon/qm.c
+++ b/drivers/crypto/hisilicon/qm.c
@@ -3085,7 +3085,7 @@ static int qm_alloc_uacce(struct hisi_qm *qm)
};
int ret;
- ret = strscpy(interface.name, pdev->driver->name,
+ ret = strscpy(interface.name, dev_driver_string(&pdev->dev),
sizeof(interface.name));
if (ret < 0)
return -ENAMETOOLONG;
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.c b/drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.c
index 7ea511d59e91..f279edfce3f1 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.c
@@ -606,7 +606,7 @@ static void hns3_get_drvinfo(struct net_device *netdev,
return;
}
- strncpy(drvinfo->driver, h->pdev->driver->name,
+ strncpy(drvinfo->driver, dev_driver_string(&h->pdev->dev),
sizeof(drvinfo->driver));
drvinfo->driver[sizeof(drvinfo->driver) - 1] = '\0';
diff --git a/drivers/net/ethernet/marvell/prestera/prestera_pci.c b/drivers/net/ethernet/marvell/prestera/prestera_pci.c
index a250d394da38..a8f007f6dad2 100644
--- a/drivers/net/ethernet/marvell/prestera/prestera_pci.c
+++ b/drivers/net/ethernet/marvell/prestera/prestera_pci.c
@@ -720,7 +720,7 @@ static int prestera_fw_load(struct prestera_fw *fw)
static int prestera_pci_probe(struct pci_dev *pdev,
const struct pci_device_id *id)
{
- const char *driver_name = pdev->driver->name;
+ const char *driver_name = dev_driver_string(&pdev->dev);
struct prestera_fw *fw;
int err;
diff --git a/drivers/net/ethernet/mellanox/mlxsw/pci.c b/drivers/net/ethernet/mellanox/mlxsw/pci.c
index 13b0259f7ea6..8f306364f7bf 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/pci.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/pci.c
@@ -1876,7 +1876,7 @@ static void mlxsw_pci_cmd_fini(struct mlxsw_pci *mlxsw_pci)
static int mlxsw_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id)
{
- const char *driver_name = pdev->driver->name;
+ const char *driver_name = dev_driver_string(&pdev->dev);
struct mlxsw_pci *mlxsw_pci;
int err;
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c b/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c
index 0685ece1f155..1de076f55740 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c
@@ -202,7 +202,8 @@ nfp_get_drvinfo(struct nfp_app *app, struct pci_dev *pdev,
{
char nsp_version[ETHTOOL_FWVERS_LEN] = {};
- strlcpy(drvinfo->driver, pdev->driver->name, sizeof(drvinfo->driver));
+ strlcpy(drvinfo->driver, dev_driver_string(&pdev->dev),
+ sizeof(drvinfo->driver));
nfp_net_get_nspinfo(app, nsp_version);
snprintf(drvinfo->fw_version, sizeof(drvinfo->fw_version),
"%s %s %s %s", vnic_version, nsp_version,
--
2.30.2
^ permalink raw reply related
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