Le 26/08/2025 à 06:13, K Prateek Nayak a écrit :
Unify the tl_cls_mask() used by both the scheduler core and x86.
No functional changes intended.
Signed-off-by: K Prateek Nayak
---
arch/x86/kernel/smpboot.c | 4
include/linux/sched/topology.h | 8 +++-
kernel/sched/topology
Le 26/08/2025 à 06:13, K Prateek Nayak a écrit :
Unify the tl_pkg_mask() wrapper around cpu_nod_mask() across core, x86,
powerpc, and s390.
No functional changes intended.
Signed-off-by: K Prateek Nayak
---
arch/powerpc/kernel/smp.c | 7 +--
arch/s390/kernel/topology.c| 7 +-
On Thu, Aug 14, 2025 at 04:02:01PM -0500, Bjorn Helgaas wrote:
> On Thu, Aug 07, 2025 at 03:55:37PM +0200, Niklas Schnelle wrote:
> > Niklas Schnelle (3):
> > PCI/AER: Fix missing uevent on recovery when a reset is requested
> > PCI/ERR: s390/pci: Use pci_uevent_ers() in PCI recovery
>
Le 26/08/2025 à 06:13, K Prateek Nayak a écrit :
PowerPC enables the MC scheduling domain by default on systems with
coregroup support without having a SCHED_MC config in Kconfig.
The scheduler uses CONFIG_SCHED_MC to introduce the MC domain in the
default topology (core) and to optimize the
All users of the generic vDSO library also use the generic vDSO datastore.
Remove the now unnecessary kconfig symbol.
Signed-off-by: Thomas Weißschuh
---
arch/Kconfig| 2 +-
arch/arm/mm/Kconfig | 1 -
arch/arm64/Kconfig | 1 -
arch/loonga
This configuration is never used.
Remove it.
Signed-off-by: Thomas Weißschuh
---
arch/arm64/Kconfig | 1 -
lib/vdso/Kconfig | 5 -
2 files changed, 6 deletions(-)
diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
index
e9bbfacc35a64d7ef1793a5d7f7ff8db138f2814..5c61b19ea9c80559ca3c4d1
The generic vDSO can be used without the time-related functionality.
In that case the generic update_vsyscall() from kernel/time/vsyscall.c
should not be built.
Signed-off-by: Thomas Weißschuh
---
kernel/time/Makefile | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/kernel/tim
All architectures implementing time-related functionality in the vDSO are
using the generic vDSO library which handles time namespaces properly.
Remove the now unnecessary kconfig symbol.
Enables the use of time namespaces on architectures, which use the
generic vDSO but did not enable GENERIC_VD
This configuration is never used.
Remove it.
Signed-off-by: Thomas Weißschuh
---
arch/arm/mm/Kconfig | 1 -
arch/x86/Kconfig| 1 -
lib/vdso/Kconfig| 7 ---
3 files changed, 9 deletions(-)
diff --git a/arch/arm/mm/Kconfig b/arch/arm/mm/Kconfig
index
5c1023a6d78c1b4db67b2d62b71af5a7
All architectures which want to implement getrandom() in the vDSO need to
use the generic vDSO library.
Signed-off-by: Thomas Weißschuh
---
lib/vdso/Kconfig | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/lib/vdso/Kconfig b/lib/vdso/Kconfig
index
3d2c2b90d193f984e3a202e7
| 1 -
24 files changed, 31 insertions(+), 110 deletions(-)
---
base-commit: 3cd1f6bc6aa056cfd32946f9ce0aa3eb3db180e8
change-id: 20250825-vdso-cleanups-fd6fbb32fc19
Best regards,
--
Thomas Weißschuh
From: Rasmus Villemoes
The cntvct_ok variable has not had any external user since commit
c7a18100bdff ("lib/vdso: Avoid highres update if clocksource is not
VDSO capable").
It also only has one user in vdso.c, once during init, so rather than
having the caller of patch_vdso() initialize cntvct_o
The ENABLE_COMAPT_VDSO symbol is only used by arm64 and only for the
time-related functionality. There should be no new users, so it doesn't
need to be in the generic vDSO code.
Move the logic into arm64 architecture-specific code and replace the
explicit define by the standard '#ifdef __aarch64__
On riscv32 the generic vDSO infrastructure is used but without its
time-related functionality. The kconfig logic to implement this
treats HAVE_GENERIC_VDSO as a synonym for GENERIC_GETTIMEOFDAY.
This works today due to some underlying issues in how the generic vDSO
library works. Some future cleanu
All calls of these functions are already gated behind CONFIG_TIME_NS. The
compiler will already optimize them away if time namespaces are disabled.
Drop the unnecessary stubs.
Signed-off-by: Thomas Weißschuh
---
lib/vdso/gettimeofday.c | 25 -
1 file changed, 25 deletion
When the generic vDSO does not provide time functions, as for example on
riscv32, then the time data store is not necessary.
Avoid allocating these time data pages when not used.
Fixes: df7fcbefa710 ("vdso: Add generic time data storage")
Signed-off-by: Thomas Weißschuh
---
lib/vdso/datastore.c
On 26.08.25 07:25, Lorenzo Stoakes wrote:
On Mon, Aug 25, 2025 at 02:31:00PM +0200, David Hildenbrand wrote:
On 11.08.25 13:26, David Hildenbrand wrote:
print_bad_pte() looks like something that should actually be a WARN
or similar, but historically it apparently has proven to be useful to
dete
On Mon, Aug 25, 2025 at 02:31:00PM +0200, David Hildenbrand wrote:
> On 11.08.25 13:26, David Hildenbrand wrote:
> > print_bad_pte() looks like something that should actually be a WARN
> > or similar, but historically it apparently has proven to be useful to
> > detect corruption of page tables eve
Unify the tl_pkg_mask() wrapper around cpu_nod_mask() across core, x86,
powerpc, and s390.
No functional changes intended.
Signed-off-by: K Prateek Nayak
---
arch/powerpc/kernel/smp.c | 7 +--
arch/s390/kernel/topology.c| 7 +--
arch/x86/kernel/smpboot.c | 5 -
include
Unify the tl_mc_mask() wrapper around cpu_coregroup_mask() used by core,
x86, powerpc, and s390.
No functional changes intended.
Signed-off-by: K Prateek Nayak
---
arch/powerpc/kernel/smp.c | 7 +--
arch/s390/kernel/topology.c| 7 +--
arch/x86/kernel/smpboot.c | 7 ---
Unify the tl_cls_mask() used by both the scheduler core and x86.
No functional changes intended.
Signed-off-by: K Prateek Nayak
---
arch/x86/kernel/smpboot.c | 4
include/linux/sched/topology.h | 8 +++-
kernel/sched/topology.c| 7 ---
3 files changed, 7 insertions(+),
Unify the tl_smt_mask() wrapper around cpu_smt_mask() across core, x86,
ppc, and s390.
On s390, include/linux/topology.c defines an explicit cpu_smt_mask()
wrapper around topology_sibling_cpumask() when cpu_smt_mask() is not
defined by the arch/ bits and topology_sibling_cpumask() on s390 returns
PowerPC enables the MC scheduling domain by default on systems with
coregroup support without having a SCHED_MC config in Kconfig.
The scheduler uses CONFIG_SCHED_MC to introduce the MC domain in the
default topology (core) and to optimize the default CPU selection
routine (sched-ext).
Introduce
Deine cpu_coregroup_mask() to export the per-cpu cpu_corgrp_map when
coregroups are supported(). When has_coregroup_support() returns false,
cpu_coregroup_mask() returns the mask used by the PKG domain.
Since this will only be used after CONFIG_SCHED_MC is added for PowerPC,
no functional changes
Rename cpu_corgroup_{map,mask} to cpu_corgrp_{map,mask} to free up the
cpu_corgroup_* namespace. cpu_corgroup_mask() will be added back in the
subsequent commit for CONFIG_SCHED_MC enablement.
No functional changes intended.
Signed-off-by: K Prateek Nayak
---
arch/powerpc/kernel/smp.c | 26
From: Peter Zijlstra
Leon [1] and Vinicius [2] noted a topology_span_sane() warning during
their testing starting from v6.16-rc1. Debug that followed pointed to
the tl->mask() for the NODE domain being incorrectly resolved to that of
the highest NUMA domain.
tl->mask() for NODE is set to the sd_
This version uses Peter's suggestion from [1] as if and incrementally
adds cleanup on top to the arch/ bits. I've tested the x86 side but the
PowerPC and the s390 bits are only build tested. Review and feedback is
greatly appreciated.
[1]
https://lore.kernel.org/lkml/20250825091910.gt3245...@nois
On 8/26/25 3:28 AM, Namhyung Kim wrote:
> To pick up the changes in this cset:
>
> 69bf2053608423cb powerpc: Drop GPL boilerplate text with obsolete FSF
> address
>
> This addresses these perf build warnings:
>
> Warning: Kernel ABI header differences:
> diff -u tools/arch/powerpc/in
On 08/08/25 1:52 PM, Ganesh Goudar wrote:
parse AER uncorrectable and correctable error status
registers to print error type and severity.
output looks like
EEH:AER severity=Uncorrected (Fatal), Error Type: Data Link Protocol Error
Signed-off-by: Ganesh Goudar
---
v2:
* Remove unnecessary ch
Hi Christophe,
> I still don't receive anything it seems. I have modified the status
> of several patches over the last two weeks and haven't received any
> notification from patchwork.
In your case, it looks like you have previously opted-out of email
notifications from patchwork (for this sende
在 2025/8/25 22:13, Rafael J. Wysocki 写道:
On Mon, Aug 25, 2025 at 11:29 AM Zihuan Zhang wrote:
This patch replaces all remaining uses of cpufreq_cpu_get() with
the __free(cpufreq_cpu_put) annotation.
Motivation:
- Ensures automatic cleanup of policy references when they go out of scope,
re
在 2025/8/25 19:28, Gautam Menghani 写道:
On Mon, Aug 25, 2025 at 05:28:33PM +0800, Zihuan Zhang wrote:
This patch replaces all remaining uses of cpufreq_cpu_get() with
the __free(cpufreq_cpu_put) annotation.
Motivation:
- Ensures automatic cleanup of policy references when they go out of scope,
To pick up the changes in this cset:
69bf2053608423cb powerpc: Drop GPL boilerplate text with obsolete FSF address
This addresses these perf build warnings:
Warning: Kernel ABI header differences:
diff -u tools/arch/powerpc/include/uapi/asm/kvm.h
arch/powerpc/include/uapi/asm/kvm.h
Ple
On Tue, Aug 05, 2025 at 06:13:56PM +0200, Uladzislau Rezki wrote:
> I agree. Also it can be even moved under vmalloc.c. There is only one
> user which needs it globally, it is usercopy.c. It uses find_vmap_area()
> which is wrong. See:
>
>
> if (is_vmalloc_addr(ptr) && !pagefault_disabled()
On Mon, 25 Aug 2025 18:32:54 +0200, Andy Shevchenko wrote:
> There are a few places in architecture code for the floppy driver
> that may be cleaned up. Do it so.
>
> Assumed to route via Andrew Morton's tree as floppy is basically orphaned.
>
> Changelog v2:
> - combined separate patches sent
Remove legacy-of-mm-gpiochip.h header file. The above mentioned
file provides an OF API that's deprecated. There is no agnostic
alternatives to it and we have to open code the logic which was
hidden behind of_mm_gpiochip_add_data(). Note, most of the GPIO
drivers are using their own labeling schema
On Wed, 20 Aug 2025 11:16:41 +0800, Shengjiu Wang wrote:
> If a defer probe happens for the cpu device, the return value for
> cpu_pdev "ret = -EINVAL" breaks defer probe. As the cpus->dai_name
> can be replaced by cpus->of_node, to simplify the code, remove related
> code for cpu_pdev.
>
>
Appl
Since the commit 3d86739c6343 ("floppy: always use the track buffer")
the CROSS_64KB() is not used by the driver, remove the leftovers.
Acked-by: Helge Deller #parisc
Acked-by: Geert Uytterhoeven # m68k
Signed-off-by: Andy Shevchenko
---
arch/alpha/include/asm/floppy.h| 19
There are only two headers using the K_64 custom constant. Moreover,
its usage tangles a code because the constant is defined in the C
file, while users are in the headers. Replace it with well defined
SZ_64K from sizes.h.
Acked-by: Helge Deller
Signed-off-by: Andy Shevchenko
---
arch/parisc/in
Sorting headers alphabetically helps locating duplicates, and makes it
easier to figure out where to insert new headers.
Signed-off-by: Andy Shevchenko
---
drivers/block/floppy.c | 54 +-
1 file changed, 27 insertions(+), 27 deletions(-)
diff --git a/driv
There are a few places in architecture code for the floppy driver
that may be cleaned up. Do it so.
Assumed to route via Andrew Morton's tree as floppy is basically orphaned.
Changelog v2:
- combined separate patches sent earlier into a series
- added tags (Helge, Geert)
- fixed typo in the commi
The QUICC Engine provides interrupts for a few I/O ports. This is
handled via a separate interrupt ID and managed via a triplet of
dedicated registers hosted by the SoC.
Implement an interrupt driver for it for that those IRQs can then
be linked to the related GPIOs.
The number of ports for which
In the QE, a few GPIOs are IRQ capable. Similarly to
commit 726bd223105c ("powerpc/8xx: Adding support of IRQ in MPC8xx
GPIO"), add IRQ support to QE GPIO.
Add property 'fsl,qe-gpio-irq-mask' similar to
'fsl,cpm1-gpio-irq-mask' that define which of the GPIOs have IRQs.
Here is an exemple for port
On Fri, Aug 22, 2025 at 05:43:39AM -0700, Chelsy Ratnawat wrote:
> Use sysfs_emit() instead of snprintf()/sprintf() when writing
> to sysfs buffers, as recommended by the kernel documentation.
>
> Signed-off-by: Chelsy Ratnawat
Acked-by: Ioana Ciornei
On Mon, Aug 25, 2025 at 10:34:35AM +0100, Salah Triki wrote:
> platform_get_resource() returns NULL in case of failure, so check its
> return value and propagate the error in order to prevent NULL pointer
> dereference.
>
> Fixes: 6305166c8771 ("bus: fsl-mc: Add ACPI support for fsl-mc")
> Cc:
>
On Mon, Aug 25, 2025 at 11:29 AM Zihuan Zhang wrote:
>
> This patch replaces all remaining uses of cpufreq_cpu_get() with
> the __free(cpufreq_cpu_put) annotation.
>
> Motivation:
> - Ensures automatic cleanup of policy references when they go out of scope,
> reducing the risk of forgetting to c
On 7/29/25 19:42, Sean Christopherson wrote:
> Use the newfangled EXPORT_SYMBOL_GPL_FOR_MODULES() along with some macro
> shenanigans to export KVM-internal symbols if and only if KVM has one or
Note it was renamed to EXPORT_SYMBOL_FOR_MODULES() only in 6.17-rc3 so this
series will need rebasing t
On Mon, Aug 25, 2025 at 8:53 AM Christophe Leroy
wrote:
>
> In the QE, a few GPIOs are IRQ capable. Similarly to
> commit 726bd223105c ("powerpc/8xx: Adding support of IRQ in MPC8xx
> GPIO"), add IRQ support to QE GPIO.
>
> Add property 'fsl,qe-gpio-irq-mask' similar to
> 'fsl,cpm1-gpio-irq-mask'
On Mon, Aug 25, 2025 at 2:56 PM Bartosz Golaszewski wrote:
>
> On Mon, Aug 25, 2025 at 8:53 AM Christophe Leroy
> wrote:
> >
> > In order to be able to add interrupts to the GPIOs, first change the
> > QE GPIO driver to the proper platform driver in order to allow
> > initialisation to be done in
On Mon, Aug 25, 2025 at 8:53 AM Christophe Leroy
wrote:
>
> In order to be able to add interrupts to the GPIOs, first change the
> QE GPIO driver to the proper platform driver in order to allow
> initialisation to be done in the right order, otherwise the GPIOs
> get added before the interrupts ar
This patch replaces all remaining uses of cpufreq_cpu_get() with
the __free(cpufreq_cpu_put) annotation.
Motivation:
- Ensures automatic cleanup of policy references when they go out of scope,
reducing the risk of forgetting to call cpufreq_cpu_put() on early return
or error paths.
- Brings th
On 11.08.25 13:26, David Hildenbrand wrote:
print_bad_pte() looks like something that should actually be a WARN
or similar, but historically it apparently has proven to be useful to
detect corruption of page tables even on production systems -- report
the issue and keep the system running to make
On Mon, Aug 25, 2025 at 05:28:33PM +0800, Zihuan Zhang wrote:
> This patch replaces all remaining uses of cpufreq_cpu_get() with
> the __free(cpufreq_cpu_put) annotation.
>
> Motivation:
> - Ensures automatic cleanup of policy references when they go out of scope,
> reducing the risk of forgetti
The QUICC Engine provides interrupts for a few I/O ports. This is
handled via a separate interrupt ID and managed via a triplet of
dedicated registers hosted by the SoC.
Implement an interrupt driver for those IRQs then add IRQs capability to
the QUICC ENGINE GPIOs.
The number of GPIOs for which
In order to be able to add interrupts to the GPIOs, first change the
QE GPIO driver to the proper platform driver in order to allow
initialisation to be done in the right order, otherwise the GPIOs
get added before the interrupts are registered.
Signed-off-by: Christophe Leroy
Reviewed-by: Bartos
Hi Jeremy,
Le 10/08/2025 à 05:12, Jeremy Kerr a écrit :
Hi Christophe,
I've (finally!) had some time to take a look at this. Looks like
timezone behaviour had changed, which affected notification expiry.
It should be fixed now - I'm seeing test notifications myself. Let
me know how go.
I sti
Hi Gabriel,
Le 25/08/2025 à 11:04, Gabriel Paubert a écrit :
[Vous ne recevez pas souvent de courriers de paub...@iram.es. D?couvrez
pourquoi ceci est important ? https://aka.ms/LearnAboutSenderIdentification ]
Hi Christophe,
On Fri, Aug 22, 2025 at 11:58:06AM +0200, Christophe Leroy wrote:
On Mon, Aug 25, 2025 at 11:40:48AM +0200, Christophe Leroy wrote:
> Hi Gabriel,
>
> Le 25/08/2025 à 11:04, Gabriel Paubert a écrit :
> > [Vous ne recevez pas souvent de courriers de paub...@iram.es. D?couvrez
> > pourquoi ceci est important ?
> > https://urldefense.com/v3/__https://aka.ms/LearnA
platform_get_resource() returns NULL in case of failure, so check its
return value and propagate the error in order to prevent NULL pointer
dereference.
Fixes: 6305166c8771 ("bus: fsl-mc: Add ACPI support for fsl-mc")
Cc:
Signed-off-by: Salah Triki
---
Changes in v2:
- Add Fixes and Cc tags,
On Mon, 2025-08-25 at 14:10 +0530, Mahesh J Salgaonkar wrote:
> On 2025-08-12 15:57:13 Tue, Haren Myneni wrote:
> > The hypervisor assigns one pipe per partition for all sources and
> > assigns new pipe after migration. Also the partition ID that is
> > used by source as its target ID may be change
Hi Christophe,
On Fri, Aug 22, 2025 at 11:58:06AM +0200, Christophe Leroy wrote:
> Masked user access avoids the address/size verification by access_ok().
> Allthough its main purpose is to skip the speculation in the
> verification of user address and size hence avoid the need of spec
> mitigat
In the QE, a few GPIOs are IRQ capable. Similarly to
commit 726bd223105c ("powerpc/8xx: Adding support of IRQ in MPC8xx
GPIO"), add IRQ support to QE GPIO.
Add property 'fsl,qe-gpio-irq-mask' similar to
'fsl,cpm1-gpio-irq-mask' that define which of the GPIOs have IRQs.
Here is an exemple for port
The QUICC Engine provides interrupts for a few I/O ports. This is
handled via a separate interrupt ID and managed via a triplet of
dedicated registers hosted by the SoC.
Implement an interrupt driver for it for that those IRQs can then
be linked to the related GPIOs.
Signed-off-by: Christophe Ler
On 2025-08-24 23:53:45 Sun, Haren Myneni wrote:
> On Mon, 2025-08-25 at 11:28 +0530, Mahesh J Salgaonkar wrote:
> > On 2025-08-12 15:57:08 Tue, Haren Myneni wrote:
> > > ibm,send-hvpipe-msg RTAS call is used to send data to the source
> > > (Ex: Hardware Management Console) over the hypervisor pipe
On 2025-08-12 15:57:13 Tue, Haren Myneni wrote:
> The hypervisor assigns one pipe per partition for all sources and
> assigns new pipe after migration. Also the partition ID that is
> used by source as its target ID may be changed after the migration.
> So disable hvpipe during SUSPEND event with ‘
tree/branch: https://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux.git
merge
branch HEAD: 0bc42b06664d82d3d9c4f5f54a5960b3760c7cd6 Automatic merge of
'fixes' into merge (2025-08-24 18:30)
elapsed time: 1095m
configs tested: 142
configs skipped: 6
The following configs have been built
66 matches
Mail list logo
