Re: [PATCH 5.4 only] powerpc: boot: Remove unnecessary zero in label in udelay()

Hi! On Tue, Sep 02, 2025 at 04:52:34PM -0700, Nathan Chancellor wrote: > When building powerpc configurations in linux-5.4.y with binutils 2.43 > or newer, there is an assembler error in arch/powerpc/boot/util.S: > > arch/powerpc/boot/util.S: Assembler messages: > arch/powerpc/boot/util.S:44:

Re: [PATCH v3 12/12] PM: EM: Use scope-based cleanup helper

On 03/09/2025 04:12, Zihuan Zhang wrote: >> You are not improving the source code here. This is not how to use >> __free() and you clearly do not understand the source code. > > Sorry for the problem, policy should be assigned after cpumask_test_cpu(). > > I actually realized earlier that __free

Re: [PATCH v5 6/7] dt-bindings: soc: fsl: qe: Convert QE GPIO to DT schema

Hi Rob, Le 02/09/2025 à 23:24, Rob Herring a écrit : On Mon, Sep 01, 2025 at 02:05:13PM +0200, Christophe Leroy wrote: Convert QE QPIO devicetree binding to DT schema. Signed-off-by: Christophe Leroy --- v5: New --- + +properties: + compatible: +items: + - enum: + - fsl,

Re: [PATCH] pasemi: fix PCI device reference leaks in pas_setup_mce_regs

On Tue, Sep 02, 2025 at 12:24:11PM +0200, Markus Elfring wrote: > … > > Add missing pci_dev_put() calls to ensure all device references > > are properly released. > > * See also: > > https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/Documentation/process/submitting-patches

[PATCH 5.4 only] powerpc: boot: Remove unnecessary zero in label in udelay()

When building powerpc configurations in linux-5.4.y with binutils 2.43 or newer, there is an assembler error in arch/powerpc/boot/util.S: arch/powerpc/boot/util.S: Assembler messages: arch/powerpc/boot/util.S:44: Error: junk at end of line, first unrecognized character is `0' arch/powerpc/b

Re: [PATCH v2 3/4] arch: copy_thread: pass clone_flags as u64

Hi Andreas, On Tue, 2025-09-02 at 09:02 +0200, Andreas Larsson wrote: > On 2025-09-01 15:09, Simon Schuster via B4 Relay wrote: > > From: Simon Schuster > > > > With the introduction of clone3 in commit 7f192e3cd316 ("fork: add > > clone3") the effective bit width of clone_flags on all architect

[PATCH v2 3/4] arch: copy_thread: pass clone_flags as u64

From: Simon Schuster With the introduction of clone3 in commit 7f192e3cd316 ("fork: add clone3") the effective bit width of clone_flags on all architectures was increased from 32-bit to 64-bit, with a new type of u64 for the flags. However, for most consumers of clone_flags the interface was not

Re: [PATCH v4 14/16] block-dma: migrate to dma_map_phys instead of map_page

On Tue, Sep 02, 2025 at 03:59:37PM -0600, Keith Busch wrote: > On Tue, Sep 02, 2025 at 10:49:48PM +0200, Marek Szyprowski wrote: > > On 19.08.2025 19:36, Leon Romanovsky wrote: > > > @@ -87,8 +87,8 @@ static bool blk_dma_map_bus(struct blk_dma_iter *iter, > > > struct phys_vec *vec) > > > static

[PATCH v3 01/12] arm64: topology: Use scope-based cleanup helper

Replace the manual cpufreq_cpu_put() with __free(put_cpufreq_policy) annotation for policy references. This reduces the risk of reference counting mistakes and aligns the code with the latest kernel style. No functional change intended. Signed-off-by: Zihuan Zhang --- arch/arm64/kernel/topology

[PATCH 0/6] Curve25519 cleanup

This series is targeting libcrypto-next. It can also be retrieved from: git fetch https://git.kernel.org/pub/scm/linux/kernel/git/ebiggers/linux.git curve25519-v1 This series removes the unused crypto_kpp support for Curve25519, migrates curve25519-selftest to a KUnit test suite, and reorga

[PATCH v3 04/12] cpufreq: longhaul: Use scope-based cleanup helper

Replace the manual cpufreq_cpu_put() with __free(put_cpufreq_policy) annotation for policy references. This reduces the risk of reference counting mistakes and aligns the code with the latest kernel style. No functional change intended. Signed-off-by: Zihuan Zhang --- drivers/cpufreq/longhaul.c

[PATCH v2 2/4] copy_process: pass clone_flags as u64 across calltree

From: Simon Schuster With the introduction of clone3 in commit 7f192e3cd316 ("fork: add clone3") the effective bit width of clone_flags on all architectures was increased from 32-bit to 64-bit, with a new type of u64 for the flags. However, for most consumers of clone_flags the interface was not

[PATCH v5 08/16] kmsan: convert kmsan_handle_dma to use physical addresses

From: Leon Romanovsky Convert the KMSAN DMA handling function from page-based to physical address-based interface. The refactoring renames kmsan_handle_dma() parameters from accepting (struct page *page, size_t offset, size_t size) to (phys_addr_t phys, size_t size). The existing semantics where

Re: pasemi: fix PCI device reference leaks in pas_setup_mce_regs

>> * Would you like to increase the application of scope-based resource  >> management? >>    >> https://elixir.bootlin.com/linux/v6.17-rc4/source/include/linux/device.h#L1180 > > That won't work here at all. Why do you think in this direction here? Regards, Markus

[PATCH 6/6] wireguard: kconfig: Simplify crypto kconfig selections

Simplify the kconfig entry for WIREGUARD: - Drop the selections of the arch-optimized ChaCha20, Poly1305, BLAKE2s, and Curve25519 code. These options no longer exist, as lib/crypto/ now enables the arch-optimized code automatically. - Drop the selection of CRYPTO. This was needed only to ma

Re: [PATCH v2 4/4] nios2: implement architecture-specific portion of sys_clone3

On Mon, Sep 1, 2025, at 15:09, Simon Schuster via B4 Relay wrote: > From: Simon Schuster > > This commit adds the sys_clone3 entry point for nios2. An > architecture-specific wrapper (__sys_clone3) is required to save and > restore additional registers to the kernel stack via SAVE_SWITCH_STACK > a

Re: [PATCH v3 03/12] cpufreq: intel_pstate: Use scope-based cleanup helper

在 2025/9/2 19:47, Rafael J. Wysocki 写道: On Tue, Sep 2, 2025 at 12:33 PM Zihuan Zhang wrote: 在 2025/9/1 23:17, Rafael J. Wysocki 写道: On Mon, Sep 1, 2025 at 10:58 AM Zihuan Zhang wrote: Replace the manual cpufreq_cpu_put() with __free(put_cpufreq_policy) annotation for policy references. Th

Re: [PATCH v3 12/12] PM: EM: Use scope-based cleanup helper

> You are not improving the source code here. This is not how to use > __free() and you clearly do not understand the source code. Sorry for the problem, policy should be assigned after cpumask_test_cpu(). I actually realized earlier that __free() only frees at the end of the variable???s lifet

Re: [PATCH 5/6] lib/crypto: curve25519: Consolidate into single module

On Mon, Sep 01, 2025 at 01:18:14PM -0700, Eric Biggers wrote: > diff --git a/lib/crypto/tests/Kconfig b/lib/crypto/tests/Kconfig > index 7e4e66f30a7a6..c0f3b64489dd6 100644 > --- a/lib/crypto/tests/Kconfig > +++ b/lib/crypto/tests/Kconfig > @@ -11,13 +11,14 @@ config CRYPTO_LIB_BLAKE2S_KUNIT_TEST >

Re: [PATCH v5 6/7] dt-bindings: soc: fsl: qe: Convert QE GPIO to DT schema

On Mon, Sep 01, 2025 at 02:05:13PM +0200, Christophe Leroy wrote: > Convert QE QPIO devicetree binding to DT schema. > > Signed-off-by: Christophe Leroy > --- > v5: New > --- > .../fsl/cpm_qe/fsl,mpc8323-qe-pario-bank.yaml | 53 +++ This should move to bindings/gpio/ > .../bind

Re: next-20250901 drivers bluetooth bpa10x.c:77:33: error: array type has incomplete element type 'struct h4_recv_pkt'

On Tuesday 09/02 at 15:00 +0530, Naresh Kamboju wrote: > The following build warnings / errors are noticed on powerpc ppc6xx_defconfig > with gcc-13 toolchain running on Linux next-20250901 and next-20250902. > > Regression Analysis: > - New regression? yes > - Reproducibili

Re: [PATCH v4 14/16] block-dma: migrate to dma_map_phys instead of map_page

On Tue, Sep 02, 2025 at 10:49:48PM +0200, Marek Szyprowski wrote: > On 19.08.2025 19:36, Leon Romanovsky wrote: > > @@ -87,8 +87,8 @@ static bool blk_dma_map_bus(struct blk_dma_iter *iter, > > struct phys_vec *vec) > > static bool blk_dma_map_direct(struct request *req, struct device > > *dma_d

Re: [PATCH v4 14/16] block-dma: migrate to dma_map_phys instead of map_page

On 19.08.2025 19:36, Leon Romanovsky wrote: > From: Leon Romanovsky > > After introduction of dma_map_phys(), there is no need to convert > from physical address to struct page in order to map page. So let's > use it directly. > > Signed-off-by: Leon Romanovsky > --- > block/blk-mq-dma.c | 4 ++

[PATCH v5 03/16] dma-debug: refactor to use physical addresses for page mapping

From: Leon Romanovsky Convert the DMA debug infrastructure from page-based to physical address-based mapping as a preparation to rely on physical address for DMA mapping routines. The refactoring renames debug_dma_map_page() to debug_dma_map_phys() and changes its signature to accept a phys_addr

Re: [PATCH v3 08/12] cpufreq: powerpc: macintosh: Use scope-based cleanup helper

On 01/09/2025 10:57, Zihuan Zhang wrote: > + struct cpufreq_policy *policy __free(put_cpufreq_policy) = > cpufreq_cpu_get(0); > struct wf_control *clamp; > struct device *dev; > int ret; > > - policy = cpufreq_cpu_get(0); > if (!policy) { > pr_warn("

Re: [PATCH v3 04/12] cpufreq: longhaul: Use scope-based cleanup helper

On 01/09/2025 10:57, Zihuan Zhang wrote: > static void __exit longhaul_exit(void) > { > - struct cpufreq_policy *policy = cpufreq_cpu_get(0); > + struct cpufreq_policy *policy __free(put_cpufreq_policy) = > cpufreq_cpu_get(0); > int i; > > for (i = 0; i < numscales; i++) {

Re: [PATCH v3 12/12] PM: EM: Use scope-based cleanup helper

On 01/09/2025 10:57, Zihuan Zhang wrote: > Replace the manual cpufreq_cpu_put() with __free(put_cpufreq_policy) > annotation for policy references. This reduces the risk of reference > counting mistakes and aligns the code with the latest kernel style. > > No functional change intended. > > Signe

Re: [bpf-next v2 0/5] powerpc64/bpf: Add support for bpf arena and arena atomics

> On 29 Aug 2025, at 10:21 PM, Saket Kumar Bhaskar wrote: > > This patch series introduces support for the PROBE_MEM32, > bpf_addr_space_cast and PROBE_ATOMIC instructions in the powerpc BPF JIT, > facilitating the implementation of BPF arena and arena atomics. > > The last patch in the serie

[PATCH v5 02/16] iommu/dma: implement DMA_ATTR_MMIO for dma_iova_link().

From: Leon Romanovsky This will replace the hacky use of DMA_ATTR_SKIP_CPU_SYNC to avoid touching the possibly non-KVA MMIO memory. Also correct the incorrect caching attribute for the IOMMU, MMIO memory should not be cachable inside the IOMMU mapping or it can possibly create system problems. S

[PATCH v5 16/16] nvme-pci: unmap MMIO pages with appropriate interface

From: Leon Romanovsky Block layer maps MMIO memory through dma_map_phys() interface with help of DMA_ATTR_MMIO attribute. There is a need to unmap that memory with the appropriate unmap function, something which wasn't possible before adding new REQ attribute to block layer in previous patch. Re

[PATCH v5 14/16] block-dma: migrate to dma_map_phys instead of map_page

From: Leon Romanovsky After introduction of dma_map_phys(), there is no need to convert from physical address to struct page in order to map page. So let's use it directly. Reviewed-by: Keith Busch Signed-off-by: Leon Romanovsky --- block/blk-mq-dma.c | 4 ++-- 1 file changed, 2 insertions(+)

[PATCH v5 12/16] mm/hmm: migrate to physical address-based DMA mapping API

From: Leon Romanovsky Convert HMM DMA operations from the legacy page-based API to the new physical address-based dma_map_phys() and dma_unmap_phys() functions. This demonstrates the preferred approach for new code that should use physical addresses directly rather than page+offset parameters. T

[PATCH v5 09/16] dma-mapping: implement DMA_ATTR_MMIO for dma_(un)map_page_attrs()

From: Leon Romanovsky Make dma_map_page_attrs() and dma_map_page_attrs() respect DMA_ATTR_MMIO. DMA_ATR_MMIO makes the functions behave the same as dma_(un)map_resource(): - No swiotlb is possible - Legacy dma_ops arches use ops->map_resource() - No kmsan - No arch_dma_map_phys_direct() The

[PATCH v5 06/16] iommu/dma: implement DMA_ATTR_MMIO for iommu_dma_(un)map_phys()

From: Leon Romanovsky Make iommu_dma_map_phys() and iommu_dma_unmap_phys() respect DMA_ATTR_MMIO. DMA_ATTR_MMIO makes the functions behave the same as iommu_dma_(un)map_resource(): - No swiotlb is possible - No cache flushing is done (ATTR_MMIO should not be cached memory) - prot for iommu_ma

[PATCH v5 07/16] dma-mapping: convert dma_direct_*map_page to be phys_addr_t based

From: Leon Romanovsky Convert the DMA direct mapping functions to accept physical addresses directly instead of page+offset parameters. The functions were already operating on physical addresses internally, so this change eliminates the redundant page-to-physical conversion at the API boundary.

[PATCH v5 05/16] iommu/dma: rename iommu_dma_*map_page to iommu_dma_*map_phys

From: Leon Romanovsky Rename the IOMMU DMA mapping functions to better reflect their actual calling convention. The functions iommu_dma_map_page() and iommu_dma_unmap_page() are renamed to iommu_dma_map_phys() and iommu_dma_unmap_phys() respectively, as they already operate on physical addresses

Re: [PATCH v5 3/7] soc: fsl: qe: Drop legacy-of-mm-gpiochip.h header from GPIO driver

On Mon, 1 Sep 2025 14:05:10 +0200, Christophe Leroy said: > 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().

Re: [PATCH v3 03/12] cpufreq: intel_pstate: Use scope-based cleanup helper

On Tue, Sep 2, 2025 at 12:33 PM Zihuan Zhang wrote: > > > 在 2025/9/1 23:17, Rafael J. Wysocki 写道: > > On Mon, Sep 1, 2025 at 10:58 AM Zihuan Zhang wrote: > >> Replace the manual cpufreq_cpu_put() with __free(put_cpufreq_policy) > >> annotation for policy references. This reduces the risk of refer

Re: [PATCH] pasemi: fix PCI device reference leaks in pas_setup_mce_regs

On 02. 09. 25, 12:24, Markus Elfring wrote: … Add missing pci_dev_put() calls to ensure all device references are properly released. * See also: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/Documentation/process/submitting-patches.rst?h=v6.17-rc4#n658 ? * Wou

Re: [PATCH v3 03/12] cpufreq: intel_pstate: Use scope-based cleanup helper

在 2025/9/1 23:17, Rafael J. Wysocki 写道: On Mon, Sep 1, 2025 at 10:58 AM Zihuan Zhang wrote: Replace the manual cpufreq_cpu_put() with __free(put_cpufreq_policy) annotation for policy references. This reduces the risk of reference counting mistakes and aligns the code with the latest kernel st

Re: [PATCH] pasemi: fix PCI device reference leaks in pas_setup_mce_regs

… > Add missing pci_dev_put() calls to ensure all device references > are properly released. * See also: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/Documentation/process/submitting-patches.rst?h=v6.17-rc4#n658 * Would you like to increase the application of scope-b

next-20250901 drivers bluetooth bpa10x.c:77:33: error: array type has incomplete element type 'struct h4_recv_pkt'

The following build warnings / errors are noticed on powerpc ppc6xx_defconfig with gcc-13 toolchain running on Linux next-20250901 and next-20250902. Regression Analysis: - New regression? yes - Reproducibility? yes First seen on next-20250901 Bad: next-20250901 and next-20250902 Good: next

Re: [PATCH v2 3/4] arch: copy_thread: pass clone_flags as u64

On Mon, Sep 1, 2025 at 9:10 PM Simon Schuster via B4 Relay wrote: > > From: Simon Schuster > > With the introduction of clone3 in commit 7f192e3cd316 ("fork: add > clone3") the effective bit width of clone_flags on all architectures was > increased from 32-bit to 64-bit, with a new type of u64 fo

Re: [PATCH v4 00/16] dma-mapping: migrate to physical address-based API

On Mon, Sep 01, 2025 at 07:23:02PM -0300, Jason Gunthorpe wrote: > On Mon, Sep 01, 2025 at 11:47:59PM +0200, Marek Szyprowski wrote: > > I would like to give those patches a try in linux-next, but in meantime > > I tested it on my test farm and found a regression in dma_map_resource() > > handlin

Re: [PATCH v2 3/4] arch: copy_thread: pass clone_flags as u64

On 2025-09-01 15:09, Simon Schuster via B4 Relay wrote: > From: Simon Schuster > > With the introduction of clone3 in commit 7f192e3cd316 ("fork: add > clone3") the effective bit width of clone_flags on all architectures was > increased from 32-bit to 64-bit, with a new type of u64 for the flags.