Move the SEV startup code into arch/x86/boot/startup/, where it will
reside along with other code that executes extremely early, and
therefore needs to be built in a special manner.
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cc: Brian Gerst <brgerst@gmail.com>
Cc: David Woodhouse <dwmw@amazon.co.uk>
Cc: Dionna Amalie Glaze <dionnaglaze@google.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Juergen Gross <jgross@suse.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Kevin Loughlin <kevinloughlin@google.com>
Cc: Len Brown <len.brown@intel.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Cc: Tom Lendacky <thomas.lendacky@amd.com>
Link: https://lore.kernel.org/r/20250418141253.2601348-12-ardb+git@google.com
Disentangle the SEV core code and the SEV code that is called during
early boot. The latter piece will be moved into startup/ in a subsequent
patch.
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cc: Brian Gerst <brgerst@gmail.com>
Cc: David Woodhouse <dwmw@amazon.co.uk>
Cc: Dionna Amalie Glaze <dionnaglaze@google.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Juergen Gross <jgross@suse.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Kevin Loughlin <kevinloughlin@google.com>
Cc: Len Brown <len.brown@intel.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Cc: Tom Lendacky <thomas.lendacky@amd.com>
Link: https://lore.kernel.org/r/20250418141253.2601348-11-ardb+git@google.com
GCC may ignore the __no_sanitize_address function attribute when
inlining, resulting in KASAN instrumentation in code tagged as
'noinstr'.
Move the SEV NMI handling code, which is noinstr, into a separate source
file so KASAN can be disabled on the whole file without losing coverage
of other SEV core code, once the startup code is split off from it too.
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cc: Brian Gerst <brgerst@gmail.com>
Cc: David Woodhouse <dwmw@amazon.co.uk>
Cc: Dionna Amalie Glaze <dionnaglaze@google.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Juergen Gross <jgross@suse.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Kevin Loughlin <kevinloughlin@google.com>
Cc: Len Brown <len.brown@intel.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Cc: Tom Lendacky <thomas.lendacky@amd.com>
Link: https://lore.kernel.org/r/20250418141253.2601348-10-ardb+git@google.com
When running in -fPIC mode, the compiler may decide to emit statically
initialized data objects into .data.rel or .data.rel.local if they
contain absolute references to global or local objects, respectively,
which require fixing up at load time.
This distinction is irrelevant for the kernel, so fold .data.rel and
.data.rel.local into .data.
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cc: Brian Gerst <brgerst@gmail.com>
Cc: David Woodhouse <dwmw@amazon.co.uk>
Cc: Dionna Amalie Glaze <dionnaglaze@google.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Juergen Gross <jgross@suse.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Kevin Loughlin <kevinloughlin@google.com>
Cc: Len Brown <len.brown@intel.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Cc: Tom Lendacky <thomas.lendacky@amd.com>
Link: https://lore.kernel.org/r/20250418141253.2601348-9-ardb+git@google.com
In particular we need this fix before applying subsequent changes:
d54d610243 ("x86/boot/sev: Avoid shared GHCB page for early memory acceptance")
Signed-off-by: Ingo Molnar <mingo@kernel.org>
With ATA devices supporting the CDL feature, using CDL requires that the
feature be enabled with a SET FEATURES command. This command is issued
as the translated command for the MODE SELECT command issued by
scsi_cdl_enable() when the user enables CDL through the device
cdl_enable sysfs attribute.
However, the implementation of scsi_cdl_enable() always issues a MODE
SELECT command for ATA devices when the enable argument is true, even if
CDL is already enabled on the device. While this does not cause any
issue with using CDL descriptors with read/write commands (the CDL
feature will be enabled on the drive), issuing the MODE SELECT command
even when the device CDL feature is already enabled will cause a reset
of the ATA device CDL statistics log page (as defined in ACS, any CDL
enable action must reset the device statistics).
Avoid this needless actions (and the implied statistics log page reset)
by modifying scsi_cdl_enable() to issue the MODE SELECT command to
enable CDL if and only if CDL is not reported as already enabled on the
device.
And while at it, simplify the initialization of the is_ata boolean
variable and move the declaration of the scsi mode data and sense header
variables to within the scope of ATA device handling.
Fixes: 1b22cfb141 ("scsi: core: Allow enabling and disabling command duration limits")
Cc: stable@vger.kernel.org
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Niklas Cassel <cassel@kernel.org>
Reviewed-by: Igor Pylypiv <ipylypiv@google.com>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
With ATA devices supporting the CDL feature, using CDL requires that the
feature be enabled with a SET FEATURES command. This command is issued
as the translated command for the MODE SELECT command issued by
scsi_cdl_enable() when the user enables CDL through the device
cdl_enable sysfs attribute.
Currently, ata_mselect_control_ata_feature() always translates a MODE
SELECT command for the ATA features subpage of the control mode page to
a SET FEATURES command to enable or disable CDL based on the cdl_ctrl
field. However, there is no need to issue the SET FEATURES command if:
1) The MODE SELECT command requests disabling CDL and CDL is already
disabled.
2) The MODE SELECT command requests enabling CDL and CDL is already
enabled.
Fix ata_mselect_control_ata_feature() to issue the SET FEATURES command
only when necessary. Since enabling CDL also implies a reset of the CDL
statistics log page, avoiding useless CDL enable operations also avoids
clearing the CDL statistics log.
Also add debug messages to clearly signal when CDL is being enabled or
disabled using a SET FEATURES command.
Fixes: df60f9c645 ("scsi: ata: libata: Add ATA feature control sub-page translation")
Cc: stable@vger.kernel.org
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Niklas Cassel <cassel@kernel.org>
Reviewed-by: Igor Pylypiv <ipylypiv@google.com>
For the ATA features subpage of the control mode page, the T10 SAT-6
specifications state that:
For a MODE SENSE command, the SATL shall return the CDL_CTRL field value
that was last set by an application client.
However, the function ata_msense_control_ata_feature() always sets the
CDL_CTRL field to the 0x02 value to indicate support for the CDL T2A and
T2B pages. This is thus incorrect and the value 0x02 must be reported
only after the user enables the CDL feature, which is indicated with the
ATA_DFLAG_CDL_ENABLED device flag. When this flag is not set, the
CDL_CTRL field of the ATA feature subpage of the control mode page must
report a value of 0x00.
Fix ata_msense_control_ata_feature() to report the correct values for
the CDL_CTRL field, according to the enable/disable state of the device
CDL feature.
Fixes: df60f9c645 ("scsi: ata: libata: Add ATA feature control sub-page translation")
Cc: stable@vger.kernel.org
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Niklas Cassel <cassel@kernel.org>
Reviewed-by: Igor Pylypiv <ipylypiv@google.com>
The function ata_mselect_control_ata_feature() has a return type defined
as unsigned int but this function may return negative error codes, which
are correctly propagated up the call chain as integers.
Fix ata_mselect_control_ata_feature() to have the correct int return
type.
While at it, also fix a typo in this function description comment.
Fixes: df60f9c645 ("scsi: ata: libata: Add ATA feature control sub-page translation")
Cc: stable@vger.kernel.org
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Niklas Cassel <cassel@kernel.org>
Reviewed-by: Igor Pylypiv <ipylypiv@google.com>
Device-tree expects absent clocks to be specified as <0> (instead of
using <>). This fixes using the FCLK4/FCLK3 clocks as they are now
seen at their correct index (while before they were recognized, but at
the correct index - resulting in the hardware using a different clock
than what the kernel sees).
Fixes: e6884f2e41 ("arm64: dts: amlogic: g12: switch to the new PWM controller binding")
Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20250420164801.330505-5-martin.blumenstingl@googlemail.com
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Device-tree expects absent clocks to be specified as <0> (instead of
using <>). This fixes using the FCLK4/FCLK3 clocks as they are now
seen at their correct index (while before they were recognized, but at
the correct index - resulting in the hardware using a different clock
than what the kernel sees).
Fixes: a526eeef9a ("arm64: dts: amlogic: gx: switch to the new PWM controller binding")
Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20250420164801.330505-4-martin.blumenstingl@googlemail.com
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Device-tree expects absent clocks to be specified as <0> (instead of
using <>). This fixes using the FCLK4/FCLK3 clocks as they are now
seen at their correct index (while before they were recognized, but at
the correct index - resulting in the hardware using a different clock
than what the kernel sees).
Fixes: dbf9218619 ("ARM: dts: amlogic: meson8b: switch to the new PWM controller binding")
Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20250420164801.330505-3-martin.blumenstingl@googlemail.com
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Device-tree expects absent clocks to be specified as <0> (instead of
using <>). This fixes using the FCLK4/FCLK3 clocks as they are now
seen at their correct index (while before they were recognized, but at
the correct index - resulting in the hardware using a different clock
than what the kernel sees).
Fixes: 802cff460a ("ARM: dts: amlogic: meson8: switch to the new PWM controller binding")
Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20250420164801.330505-2-martin.blumenstingl@googlemail.com
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Old microcode is bad for users and for kernel developers.
For users, it exposes them to known fixed security and/or functional
issues. These obviously rarely result in instant dumpster fires in
every environment. But it is as important to keep your microcode up
to date as it is to keep your kernel up to date.
Old microcode also makes kernels harder to debug. A developer looking
at an oops need to consider kernel bugs, known CPU issues and unknown
CPU issues as possible causes. If they know the microcode is up to
date, they can mostly eliminate known CPU issues as the cause.
Make it easier to tell if CPU microcode is out of date. Add a list
of released microcode. If the loaded microcode is older than the
release, tell users in a place that folks can find it:
/sys/devices/system/cpu/vulnerabilities/old_microcode
Tell kernel kernel developers about it with the existing taint
flag:
TAINT_CPU_OUT_OF_SPEC
== Discussion ==
When a user reports a potential kernel issue, it is very common
to ask them to reproduce the issue on mainline. Running mainline,
they will (independently from the distro) acquire a more up-to-date
microcode version list. If their microcode is old, they will
get a warning about the taint and kernel developers can take that
into consideration when debugging.
Just like any other entry in "vulnerabilities/", users are free to
make their own assessment of their exposure.
== Microcode Revision Discussion ==
The microcode versions in the table were generated from the Intel
microcode git repo:
8ac9378a8487 ("microcode-20241112 Release")
which as of this writing lags behind the latest microcode-20250211.
It can be argued that the versions that the kernel picks to call "old"
should be a revision or two old. Which specific version is picked is
less important to me than picking *a* version and enforcing it.
This repository contains only microcode versions that Intel has deemed
to be OS-loadable. It is quite possible that the BIOS has loaded a
newer microcode than the latest in this repo. If this happens, the
system is considered to have new microcode, not old.
Specifically, the sysfs file and taint flag answer the question:
Is the CPU running on the latest OS-loadable microcode,
or something even later that the BIOS loaded?
In other words, Intel never publishes an authoritative list of CPUs
and latest microcode revisions. Until it does, this is the best that
Linux can do.
Also note that the "intel-ucode-defs.h" file is simple, ugly and
has lots of magic numbers. That's on purpose and should allow a
single file to be shared across lots of stable kernel regardless of if
they have the new "VFM" infrastructure or not. It was generated with
a dumb script.
== FAQ ==
Q: Does this tell me if my system is secure or insecure?
A: No. It only tells you if your microcode was old when the
system booted.
Q: Should the kernel warn if the microcode list itself is too old?
A: No. New kernels will get new microcode lists, both mainline
and stable. The only way to have an old list is to be running
an old kernel in which case you have bigger problems.
Q: Is this for security or functional issues?
A: Both.
Q: If a given microcode update only has functional problems but
no security issues, will it be considered old?
A: Yes. All microcode image versions within a microcode release
are treated identically. Intel appears to make security
updates without disclosing them in the release notes. Thus,
all updates are considered to be security-relevant.
Q: Who runs old microcode?
A: Anybody with an old distro. This happens all the time inside
of Intel where there are lots of weird systems in labs that
might not be getting regular distro updates and might also
be running rather exotic microcode images.
Q: If I update my microcode after booting will it stop saying
"Vulnerable"?
A: No. Just like all the other vulnerabilies, you need to
reboot before the kernel will reassess your vulnerability.
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cc: "Ahmed S. Darwish" <darwi@linutronix.de>
Cc: Andrew Cooper <andrew.cooper3@citrix.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Brian Gerst <brgerst@gmail.com>
Cc: John Ogness <john.ogness@linutronix.de>
Cc: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Juergen Gross <jgross@suse.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Link: https://lore.kernel.org/all/20250421195659.CF426C07%40davehans-spike.ostc.intel.com
(cherry picked from commit 9127865b15eb0a1bd05ad7efe29489c44394bdc1)
DSP can be used with various frameworks (e.g audio firmware, rproc).
Currently 'dsp' configuration is intended for audio firmware but it
doesn't work well with board level DTs (e.g imx8mp-evk) because
board level DT enables audio related IPs (e.g SAI) while audio firmware
needs this IPs disabled (because firmware will configure them).
So, configure 'dsp' node to be used with rproc. This way users will be
able to use board DT to use the DSP as long as they don't clash with
Audio IP configurations.
More comples usage of 'dsp' node (e.g by audio firmware) will need to
create a separate dts file (or an overlay).
This change follows the approach taken for other i.MX8 boards
in commit 391a319c81 ("arm64: dts: imx8-ss-audio: configure dsp node
for rproc usage")
Reviewed-by: Iuliana Prodan <iuliana.prodan@nxp.com>
Reviewed-by: Peng Fan <peng.fan@nxp.com>
Signed-off-by: Daniel Baluta <daniel.baluta@nxp.com>
Reviewed-by: Laurentiu Mihalcea <laurentiu.mihalcea@nxp.com>
Signed-off-by: Shawn Guo <shawnguo@kernel.org>
Add resets property to dsp node in order to be able to control the dsp
run/stall bit from audio block control.
Reviewed-by: Peng Fan <peng.fan@nxp.com>
Reviewed-by: Frank Li <Frank.Li@nxp.com>
Signed-off-by: Daniel Baluta <daniel.baluta@nxp.com>
Reviewed-by: Laurentiu Mihalcea <laurentiu.mihalcea@nxp.com>
Signed-off-by: Shawn Guo <shawnguo@kernel.org>
The "fsl,mma7455l" compatible string is not documented anywhere.
MMA7455L is the exact same device as the MMA7455, with the exception that
it is lead-free. Use the documented compatible string.
Signed-off-by: Fabio Estevam <festevam@denx.de>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Shawn Guo <shawnguo@kernel.org>
Commit c7e73b5051 ("ARM: imx: mach-imx6ul: remove 14x14 EVK specific
PHY fixup") removed a PHY fixup that setted the clock mode and the LED
mode.
Make the Ethernet interface work again by doing as advised in the
commit's log, set clock mode and the LED mode in the device tree.
Fixes: c7e73b5051 ("ARM: imx: mach-imx6ul: remove 14x14 EVK specific PHY fixup")
Signed-off-by: Sébastien Szymanski <sebastien.szymanski@armadeus.com>
Reviewed-by: Oleksij Rempel <o.rempel@pengutronix.de>
Signed-off-by: Shawn Guo <shawnguo@kernel.org>
Correct the range of PCIe app-reg region from 0x2000 to 0x4000 refer to
SerDes_SS memory map of i.MX95 Rerference Manual.
Fixes: 3b1d5deb29 ("arm64: dts: imx95: add pcie[0,1] and pcie-ep[0,1] support")
Signed-off-by: Richard Zhu <hongxing.zhu@nxp.com>
Reviewed-by: Frank Li <Frank.Li@nxp.com>
Signed-off-by: Shawn Guo <shawnguo@kernel.org>
Per imx-iim.yaml, the compatible string should only contain a single
entry.
Change it accordingly to fix the following dt-schema warnings:
efuse@83f98000: compatible: ['fsl,imx51-iim', 'fsl,imx27-iim', 'syscon'] is too long
Signed-off-by: Fabio Estevam <festevam@gmail.com>
Signed-off-by: Shawn Guo <shawnguo@kernel.org>
According to mtd-physmap.yaml, 'nor' is not a valid node name.
Change it to 'flash' to fix the following dt-schema warning:
nor@0,0: $nodename:0: 'nor@0,0' does not match '^(flash|.*sram|nand)(@.*)?$'
Signed-off-by: Fabio Estevam <festevam@denx.de>
Signed-off-by: Shawn Guo <shawnguo@kernel.org>
Commit 255fbd9eab ("arm64: dts: imx8mp: Add optional nominal drive
mode DTSI") added imx8mp-nominal.dtsi, which overrides all overdrive
clock rates in imx8mp.dtsi to the nominal rates.
At the same time, commit 9f7595b3e5 ("arm64: dts: imx8mp: configure
GPU and NPU clocks to overdrive rate") went in, which changed some
clock rates away from the nominal values.
Resolve the discrepancy by effectively reverting the changes in the
latter commit inside imx8mp-nominal.dtsi. This is required for proper
operation of the imx8mp-skov boards, which are currently
imx8mp-nominal.dtsi's only users and lets all other boards that don't
include it benefit from the new higher frequencies.
Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
Signed-off-by: Shawn Guo <shawnguo@kernel.org>
- Use kvzalloc() so that large exit_dump buffer allocations don't fail
easily.
- Remove cpu.weight / cpu.idle unimplemented warnings which are more
annoying than helpful. This makes SCX_OPS_HAS_CGROUP_WEIGHT unnecessary.
Mark it for deprecation.
-----BEGIN PGP SIGNATURE-----
iIQEABYKACwWIQTfIjM1kS57o3GsC/uxYfJx3gVYGQUCaAaeKw4cdGpAa2VybmVs
Lm9yZwAKCRCxYfJx3gVYGcDjAQDM14FReObIKOOuqzlNaYXQqcSW2///ZVf/FR8j
HpWtyAD/Tsqg6CzBpTxKkpMRLsE2iKI1t770vkUnDbjcnR0Rxgc=
=rIV5
-----END PGP SIGNATURE-----
Merge tag 'sched_ext-for-6.15-rc3-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/sched_ext
Pull sched_ext fixes from Tejun Heo:
- Use kvzalloc() so that large exit_dump buffer allocations don't fail
easily
- Remove cpu.weight / cpu.idle unimplemented warnings which are more
annoying than helpful.
This makes SCX_OPS_HAS_CGROUP_WEIGHT unnecessary. Mark it for
deprecation
* tag 'sched_ext-for-6.15-rc3-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/sched_ext:
sched_ext: Mark SCX_OPS_HAS_CGROUP_WEIGHT for deprecation
sched_ext: Remove cpu.weight / cpu.idle unimplemented warnings
sched_ext: Use kvzalloc for large exit_dump allocation
- Fix compilation in CONFIG_LOCKDEP && !CONFIG_PROVE_RCU configurations.
- Allow "cpuset_v2_mode" mount option for "cpuset" filesystem type to make
life easier for android.
-----BEGIN PGP SIGNATURE-----
iIQEABYKACwWIQTfIjM1kS57o3GsC/uxYfJx3gVYGQUCaAac+Q4cdGpAa2VybmVs
Lm9yZwAKCRCxYfJx3gVYGUiAAQCbw6eOFAE+sjI6GgAeMVORbLqyufGDNPvBwgzJ
xPxgcwD/ZLlsJWRG6BzQ/KHeFZnGWSJEiqSSFHGCCr0l4QkIdgA=
=ay1E
-----END PGP SIGNATURE-----
Merge tag 'cgroup-for-6.15-rc3-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup
Pull cgroup fixes from Tejun Heo:
- Fix compilation in CONFIG_LOCKDEP && !CONFIG_PROVE_RCU configurations
- Allow "cpuset_v2_mode" mount option for "cpuset" filesystem type to
make life easier for android
* tag 'cgroup-for-6.15-rc3-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup:
cgroup/cpuset-v1: Add missing support for cpuset_v2_mode
cgroup: Fix compilation issue due to cgroup_mutex not being exported
Vladimir Oltean says:
====================
ENETC bug fixes for bpf_xdp_adjust_head() and bpf_xdp_adjust_tail()
It has been reported that on the ENETC driver, bpf_xdp_adjust_head()
and bpf_xdp_adjust_tail() are broken in combination with the XDP_PASS
verdict. I have constructed a series a simple XDP programs and tested
with various packet sizes and confirmed that this is the case.
Patch 3/3 fixes the core issue, which is that the sk_buff created on
XDP_PASS is created by the driver as if XDP never ran, but in fact the
geometry needs to be adjusted according to the delta applied by the
program on the original xdp_buff. It depends on commit 539c1fba1a
("xdp: add generic xdp_build_skb_from_buff()") which is not available in
"stable" but perhaps should be.
Patch 2/3 is a small refactor necessary for 3/3.
Patch 1/3 fixes a related issue I noticed, which is that
bpf_xdp_adjust_tail() with a positive offset works for linear XDP
buffers, but returns an error for non-linear ones, even if there is
plenty of space in the final page fragment.
====================
Link: https://patch.msgid.link/20250417120005.3288549-1-vladimir.oltean@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Vlatko Markovikj reported that XDP programs attached to ENETC do not
work well if they use bpf_xdp_adjust_head() or bpf_xdp_adjust_tail(),
combined with the XDP_PASS verdict. A typical use case is to add or
remove a VLAN tag.
The resulting sk_buff passed to the stack is corrupted, because the
algorithm used by the driver for XDP_PASS is to unwind the current
buffer pointer in the RX ring and to re-process the current frame with
enetc_build_skb() as if XDP hadn't run. That is incorrect because XDP
may have modified the geometry of the buffer, which we then are
completely unaware of. We are looking at a modified buffer with the
original geometry.
The initial reaction, both from me and from Vlatko, was to shop around
the kernel for code to steal that would calculate a delta between the
old and the new XDP buffer geometry, and apply that to the sk_buff too.
We noticed that veth and generic xdp have such code.
The headroom adjustment is pretty uncontroversial, but what turned out
severely problematic is the tailroom.
veth has this snippet:
__skb_put(skb, off); /* positive on grow, negative on shrink */
which on first sight looks decent enough, except __skb_put() takes an
"unsigned int" for the second argument, and the arithmetic seems to only
work correctly by coincidence. Second issue, __skb_put() contains a
SKB_LINEAR_ASSERT(). It's not a great pattern to make more widespread.
The skb may still be nonlinear at that point - it only becomes linear
later when resetting skb->data_len to zero.
To avoid the above, bpf_prog_run_generic_xdp() does this instead:
skb_set_tail_pointer(skb, xdp->data_end - xdp->data);
skb->len += off; /* positive on grow, negative on shrink */
which is more open-coded, uses lower-level functions and is in general a
bit too much to spread around in driver code.
Then there is the snippet:
if (xdp_buff_has_frags(xdp))
skb->data_len = skb_shinfo(skb)->xdp_frags_size;
else
skb->data_len = 0;
One would have expected __pskb_trim() to be the function of choice for
this task. But it's not used in veth/xdpgeneric because the extraneous
fragments were _already_ freed by bpf_xdp_adjust_tail() ->
bpf_xdp_frags_shrink_tail() -> ... -> __xdp_return() - the backing
memory for the skb frags and the xdp frags is the same, but they don't
keep individual references.
In fact, that is the biggest reason why this snippet cannot be reused
as-is, because ENETC temporarily constructs an skb with the original len
and the original number of frags. Because the extraneous frags are
already freed by bpf_xdp_adjust_tail() and returned to the page
allocator, it means the entire approach of using enetc_build_skb() is
questionable for XDP_PASS. To avoid that, one would need to elevate the
page refcount of all frags before calling bpf_prog_run_xdp() and drop it
after XDP_PASS.
There are other things that are missing in ENETC's handling of XDP_PASS,
like for example updating skb_shinfo(skb)->meta_len.
These are all handled correctly and cleanly in commit 539c1fba1a
("xdp: add generic xdp_build_skb_from_buff()"), added to net-next in
Dec 2024, and in addition might even be quicker that way. I have a very
strong preference towards backporting that commit for "stable", and that
is what is used to fix the handling bugs. It is way too messy to go
this deep into the guts of an sk_buff from the code of a device driver.
Fixes: d1b15102dd ("net: enetc: add support for XDP_DROP and XDP_PASS")
Reported-by: Vlatko Markovikj <vlatko.markovikj@etas.com>
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Wei Fang <wei.fang@nxp.com>
Link: https://patch.msgid.link/20250417120005.3288549-4-vladimir.oltean@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
This small snippet of code ensures that we do something with the array
of RX software buffer descriptor elements after passing the skb to the
stack. In this case, we see if the other half of the page is reusable,
and if so, we "turn around" the buffers, making them directly usable by
enetc_refill_rx_ring() without going to enetc_new_page().
We will need to perform this kind of buffer flipping from a new code
path, i.e. from XDP_PASS. Currently, enetc_build_skb() does it there
buffer by buffer, but in a subsequent change we will stop using
enetc_build_skb() for XDP_PASS.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Wei Fang <wei.fang@nxp.com>
Link: https://patch.msgid.link/20250417120005.3288549-3-vladimir.oltean@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
At the time when bpf_xdp_adjust_tail() gained support for non-linear
buffers, ENETC was already generating this kind of geometry on RX, due
to its use of 2K half page buffers. Frames larger than 1472 bytes
(without FCS) are stored as multi-buffer, presenting a need for multi
buffer support to work properly even in standard MTU circumstances.
Allow bpf_xdp_frags_increase_tail() to know the allocation size of paged
data, so it can safely permit growing the tailroom of the buffer from
XDP programs.
Fixes: bf25146a55 ("bpf: add frags support to the bpf_xdp_adjust_tail() API")
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Wei Fang <wei.fang@nxp.com>
Link: https://patch.msgid.link/20250417120005.3288549-2-vladimir.oltean@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The function xdp_convert_buff_to_frame() may return NULL if it fails
to correctly convert the XDP buffer into an XDP frame due to memory
constraints, internal errors, or invalid data. Failing to check for NULL
may lead to a NULL pointer dereference if the result is used later in
processing, potentially causing crashes, data corruption, or undefined
behavior.
On XDP redirect failure, the associated page must be released explicitly
if it was previously retained via get_page(). Failing to do so may result
in a memory leak, as the pages reference count is not decremented.
Cc: stable@vger.kernel.org # v5.9+
Fixes: 6c5aa6fc4d ("xen networking: add basic XDP support for xen-netfront")
Signed-off-by: Alexey Nepomnyashih <sdl@nppct.ru>
Link: https://patch.msgid.link/20250417122118.1009824-1-sdl@nppct.ru
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
linux@treblig.org says:
Hi,
This is a batch of deadcoding on the qla2xxx driver.
Note the last patch removes two unused module
parameters, so I guess if anyone has that in some configs
somewhere that might surprise them.
Other than that, it's all simple function deletion.
Build tested only.
Signed-off-by: Dr. David Alan Gilbert <linux@treblig.org>
Link: https://lore.kernel.org/r/20250415002803.135909-1-linux@treblig.org
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Yihang Li <liyihang9@huawei.com> says:
This series contains some minor bugfix and general tidying:
- Ignore the soft reset result by calling I_T_nexus after the SATA disk
is soft reset
- General minor tidying
Link: https://lore.kernel.org/r/20250331123349.99591-1-liyihang9@huawei.com
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
qedi_get_proto_itt() last use was removed in 2021 by commit ed1b86ba0f
("scsi: qedi: Wake up if cmd_cleanup_req is set")
Remove it.
Signed-off-by: Dr. David Alan Gilbert <linux@treblig.org>
Link: https://lore.kernel.org/r/20250416002235.299347-3-linux@treblig.org
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
-----BEGIN PGP SIGNATURE-----
iHUEABYKAB0WIQQ6NaUOruQGUkvPdG4raS+Z+3y5EwUCaAFFvgAKCRAraS+Z+3y5
E/1OAP9SGmTMgHuHLlF8en+MaYdtwgcHy6uurXgbSQAAV/RwwQEAh2oXZE1D9I7a
EtxsaJYqbbhD09RPwWa2Rd8iJrJYXQk=
=qcoU
-----END PGP SIGNATURE-----
Merge tag 'for-netdev' of https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next
Martin KaFai Lau says:
====================
pull-request: bpf-next 2025-04-17
We've added 12 non-merge commits during the last 9 day(s) which contain
a total of 18 files changed, 1748 insertions(+), 19 deletions(-).
The main changes are:
1) bpf qdisc support, from Amery Hung.
A qdisc can be implemented in bpf struct_ops programs and
can be used the same as other existing qdiscs in the
"tc qdisc" command.
2) Add xsk tail adjustment tests, from Tushar Vyavahare.
* tag 'for-netdev' of https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next:
selftests/bpf: Test attaching bpf qdisc to mq and non root
selftests/bpf: Add a bpf fq qdisc to selftest
selftests/bpf: Add a basic fifo qdisc test
libbpf: Support creating and destroying qdisc
bpf: net_sched: Disable attaching bpf qdisc to non root
bpf: net_sched: Support updating bstats
bpf: net_sched: Add a qdisc watchdog timer
bpf: net_sched: Add basic bpf qdisc kfuncs
bpf: net_sched: Support implementation of Qdisc_ops in bpf
bpf: Prepare to reuse get_ctx_arg_idx
selftests/xsk: Add tail adjustment tests and support check
selftests/xsk: Add packet stream replacement function
====================
Link: https://patch.msgid.link/20250417184338.3152168-1-martin.lau@linux.dev
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
qedi_remove_sysfs_attr() and qedi_create_sysfs_attr() were added in 2016
by commit ace7f46ba5 ("scsi: qedi: Add QLogic FastLinQ offload iSCSI
driver framework.") but have remained unused.
Remove them.
Signed-off-by: Dr. David Alan Gilbert <linux@treblig.org>
Link: https://lore.kernel.org/r/20250416002235.299347-2-linux@treblig.org
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
BNXT_ROCE_ULP and BNXT_MAX_ULP are no longer used. Remove them to
clean up the code.
Reviewed-by: Shruti Parab <shruti.parab@broadcom.com>
Signed-off-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Link: https://patch.msgid.link/20250417172448.1206107-5-michael.chan@broadcom.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The "ref_count" field in struct bnxt_ulp is unused after
commit a43c26fa2e ("RDMA/bnxt_re: Remove the sriov config callback").
So we can just remove it now.
Reviewed-by: Somnath Kotur <somnath.kotur@broadcom.com>
Signed-off-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Link: https://patch.msgid.link/20250417172448.1206107-4-michael.chan@broadcom.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
ethtool first calls .get_dump_flags() to get the dump length. For
coredump, the driver calls the FW to get the coredump length (L1). The
min. of L1 and the user specified length is then passed to
.get_dump_data() (L2) to get the coredump. The actual coredump length
retrieved by the FW (L3) during .get_dump_data() may be smaller than L1.
This length discrepancy will trigger a WARN_ON() in
ethtool_get_dump_data().
ethtool has already vzalloc'ed a buffer with size L1. Just report
the coredump length as L2 even though the actual coredump length L3
may be smaller. The extra zero padding does not matter. This will
prevent the warning that may alarm the user.
For correctness, only do the final length update if there is no error.
Reviewed-by: Andy Gospodarek <andrew.gospodarek@broadcom.com>
Reviewed-by: Damodharam Ammepalli <damodharam.ammepalli@broadcom.com>
Signed-off-by: Shruti Parab <shruti.parab@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Link: https://patch.msgid.link/20250417172448.1206107-3-michael.chan@broadcom.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The firmware advertises a "hwrm_cmd_max_timeout" value to the driver
for NVRAM and coredump related functions that can take tens of seconds
to complete. The driver polls for the operation to complete under
mutex and may trigger hung task watchdog warning if the wait is too long.
To warn the user about this, the driver currently prints a warning if
this advertised value exceeds 40 seconds:
Device requests max timeout of %d seconds, may trigger hung task watchdog
Initially, we chose 40 seconds, well below the kernel's default
CONFIG_DEFAULT_HUNG_TASK_TIMEOUT (120 seconds) to avoid triggering
the hung task watchdog. But 60 seconds is the timeout on most
production FW and cannot be reduced further. Change the driver's warning
threshold to 60 seconds to avoid triggering this warning on all
production devices. We also print the warning if the value exceeds
CONFIG_DEFAULT_HUNG_TASK_TIMEOUT which may be set to architecture
specific defaults as low as 10 seconds.
Reviewed-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Reviewed-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Reviewed-by: Andy Gospodarek <andrew.gospodarek@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Link: https://patch.msgid.link/20250417172448.1206107-2-michael.chan@broadcom.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
ql2xetsenable last use was removed in 2020 by commit 37efd51f75 ("scsi:
qla2xxx: Use FC generic update firmware options routine for ISP27xx")
ql2xiidmaenable last use was removed in 2017 by commit 726b854870
("qla2xxx: Add framework for async fabric discovery")
Remove them.
Signed-off-by: Dr. David Alan Gilbert <linux@treblig.org>
Link: https://lore.kernel.org/r/20250415002803.135909-9-linux@treblig.org
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
qla2x00_gpsc() was added in 2017 as part of commit 726b854870
("qla2xxx: Add framework for async fabric discovery") but has remained
unused.
Remove it.
Signed-off-by: Dr. David Alan Gilbert <linux@treblig.org>
Link: https://lore.kernel.org/r/20250415002803.135909-8-linux@treblig.org
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
ql_log_qp() was added in 2017 as part of commit 22d84726e3 ("scsi:
qla2xxx: Add debug logging routine for qpair") but has remained unused.
Remove it. (That patch also added ql_dbg_qp but that is still used so is
left in).
Signed-off-by: Dr. David Alan Gilbert <linux@treblig.org>
Link: https://lore.kernel.org/r/20250415002803.135909-7-linux@treblig.org
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
qla82xx_wait_for_state_change() was added in 2010 as part of commit
579d12b58a ("[SCSI] qla2xxx: Added support for quiescence mode for
ISP82xx.") but has remained unused.
Remove it.
Signed-off-by: Dr. David Alan Gilbert <linux@treblig.org>
Link: https://lore.kernel.org/r/20250415002803.135909-6-linux@treblig.org
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
