- Linux-stable-mirror - lists.linaro.org

FAILED: patch "[PATCH] misc: pci_endpoint_test: Avoid issue of interrupts remaining" failed to apply to 5.15-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 5.15-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. To reproduce the conflict and resubmit, you may use the following commands: git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-5.15.y git checkout FETCH_HEAD git cherry-pick -x f6cb7828c8e17520d4f5afb416515d3fae1af9a9 # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2025041738-tusk-hatchback-f43f@gregkh' --subject-prefix 'PATCH 5.15.y' HEAD^.. Possible dependencies: thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From f6cb7828c8e17520d4f5afb416515d3fae1af9a9 Mon Sep 17 00:00:00 2001 From: Kunihiko Hayashi <hayashi.kunihiko(a)socionext.com> Date: Tue, 25 Feb 2025 20:02:48 +0900 Subject: [PATCH] misc: pci_endpoint_test: Avoid issue of interrupts remaining after request_irq error MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit After devm_request_irq() fails with error in pci_endpoint_test_request_irq(), the pci_endpoint_test_free_irq_vectors() is called assuming that all IRQs have been released. However, some requested IRQs remain unreleased, so there are still /proc/irq/* entries remaining, and this results in WARN() with the following message: remove_proc_entry: removing non-empty directory 'irq/30', leaking at least 'pci-endpoint-test.0' WARNING: CPU: 0 PID: 202 at fs/proc/generic.c:719 remove_proc_entry +0x190/0x19c To solve this issue, set the number of remaining IRQs to test->num_irqs, and release IRQs in advance by calling pci_endpoint_test_release_irq(). Cc: stable(a)vger.kernel.org Fixes: e03327122e2c ("pci_endpoint_test: Add 2 ioctl commands") Reviewed-by: Manivannan Sadhasivam <manivannan.sadhasivam(a)linaro.org> Signed-off-by: Kunihiko Hayashi <hayashi.kunihiko(a)socionext.com> Link: https://lore.kernel.org/r/20250225110252.28866-3-hayashi.kunihiko@socionext… [kwilczynski: commit log] Signed-off-by: Krzysztof Wilczyński <kwilczynski(a)kernel.org> diff --git a/drivers/misc/pci_endpoint_test.c b/drivers/misc/pci_endpoint_test.c index a3d2caa7a6bb..9e56d200d2f0 100644 --- a/drivers/misc/pci_endpoint_test.c +++ b/drivers/misc/pci_endpoint_test.c @@ -259,6 +259,9 @@ static int pci_endpoint_test_request_irq(struct pci_endpoint_test *test) break; } + test->num_irqs = i; + pci_endpoint_test_release_irq(test); + return ret; }

2 months, 2 weeks

3
2
0 0

FAILED: patch "[PATCH] mptcp: sockopt: fix getting freebind & transparent" failed to apply to 6.1-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 6.1-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. To reproduce the conflict and resubmit, you may use the following commands: git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-6.1.y git checkout FETCH_HEAD git cherry-pick -x e2f4ac7bab2205d3c4dd9464e6ffd82502177c51 # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2025041757-treble-debatable-db2c@gregkh' --subject-prefix 'PATCH 6.1.y' HEAD^.. Possible dependencies: thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From e2f4ac7bab2205d3c4dd9464e6ffd82502177c51 Mon Sep 17 00:00:00 2001 From: "Matthieu Baerts (NGI0)" <matttbe(a)kernel.org> Date: Fri, 14 Mar 2025 21:11:33 +0100 Subject: [PATCH] mptcp: sockopt: fix getting freebind & transparent When adding a socket option support in MPTCP, both the get and set parts are supposed to be implemented. IP(V6)_FREEBIND and IP(V6)_TRANSPARENT support for the setsockopt part has been added a while ago, but it looks like the get part got forgotten. It should have been present as a way to verify a setting has been set as expected, and not to act differently from TCP or any other socket types. Everything was in place to expose it, just the last step was missing. Only new code is added to cover these specific getsockopt(), that seems safe. Fixes: c9406a23c116 ("mptcp: sockopt: add SOL_IP freebind & transparent options") Cc: stable(a)vger.kernel.org Reviewed-by: Mat Martineau <martineau(a)kernel.org> Signed-off-by: Matthieu Baerts (NGI0) <matttbe(a)kernel.org> Reviewed-by: Simon Horman <horms(a)kernel.org> Link: https://patch.msgid.link/20250314-net-mptcp-fix-data-stream-corr-sockopt-v1… Signed-off-by: Paolo Abeni <pabeni(a)redhat.com> diff --git a/net/mptcp/sockopt.c b/net/mptcp/sockopt.c index 4b99eb796855..3caa0a9d3b38 100644 --- a/net/mptcp/sockopt.c +++ b/net/mptcp/sockopt.c @@ -1419,6 +1419,12 @@ static int mptcp_getsockopt_v4(struct mptcp_sock *msk, int optname, switch (optname) { case IP_TOS: return mptcp_put_int_option(msk, optval, optlen, READ_ONCE(inet_sk(sk)->tos)); + case IP_FREEBIND: + return mptcp_put_int_option(msk, optval, optlen, + inet_test_bit(FREEBIND, sk)); + case IP_TRANSPARENT: + return mptcp_put_int_option(msk, optval, optlen, + inet_test_bit(TRANSPARENT, sk)); case IP_BIND_ADDRESS_NO_PORT: return mptcp_put_int_option(msk, optval, optlen, inet_test_bit(BIND_ADDRESS_NO_PORT, sk)); @@ -1439,6 +1445,12 @@ static int mptcp_getsockopt_v6(struct mptcp_sock *msk, int optname, case IPV6_V6ONLY: return mptcp_put_int_option(msk, optval, optlen, sk->sk_ipv6only); + case IPV6_TRANSPARENT: + return mptcp_put_int_option(msk, optval, optlen, + inet_test_bit(TRANSPARENT, sk)); + case IPV6_FREEBIND: + return mptcp_put_int_option(msk, optval, optlen, + inet_test_bit(FREEBIND, sk)); } return -EOPNOTSUPP;

2 months, 2 weeks

3
2
0 0

FAILED: patch "[PATCH] landlock: Add the errata interface" failed to apply to 6.1-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 6.1-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. To reproduce the conflict and resubmit, you may use the following commands: git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-6.1.y git checkout FETCH_HEAD git cherry-pick -x 15383a0d63dbcd63dc7e8d9ec1bf3a0f7ebf64ac # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2025041712-remark-lyricist-a86b@gregkh' --subject-prefix 'PATCH 6.1.y' HEAD^.. Possible dependencies: thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 15383a0d63dbcd63dc7e8d9ec1bf3a0f7ebf64ac Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Micka=C3=ABl=20Sala=C3=BCn?= <mic(a)digikod.net> Date: Tue, 18 Mar 2025 17:14:37 +0100 Subject: [PATCH] landlock: Add the errata interface MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Some fixes may require user space to check if they are applied on the running kernel before using a specific feature. For instance, this applies when a restriction was previously too restrictive and is now getting relaxed (e.g. for compatibility reasons). However, non-visible changes for legitimate use (e.g. security fixes) do not require an erratum. Because fixes are backported down to a specific Landlock ABI, we need a way to avoid cherry-pick conflicts. The solution is to only update a file related to the lower ABI impacted by this issue. All the ABI files are then used to create a bitmask of fixes. The new errata interface is similar to the one used to get the supported Landlock ABI version, but it returns a bitmask instead because the order of fixes may not match the order of versions, and not all fixes may apply to all versions. The actual errata will come with dedicated commits. The description is not actually used in the code but serves as documentation. Create the landlock_abi_version symbol and use its value to check errata consistency. Update test_base's create_ruleset_checks_ordering tests and add errata tests. This commit is backportable down to the first version of Landlock. Fixes: 3532b0b4352c ("landlock: Enable user space to infer supported features") Cc: Günther Noack <gnoack(a)google.com> Cc: stable(a)vger.kernel.org Link: https://lore.kernel.org/r/20250318161443.279194-3-mic@digikod.net Signed-off-by: Mickaël Salaün <mic(a)digikod.net> diff --git a/include/uapi/linux/landlock.h b/include/uapi/linux/landlock.h index e1d2c27533b4..8806a132d7b8 100644 --- a/include/uapi/linux/landlock.h +++ b/include/uapi/linux/landlock.h @@ -57,9 +57,11 @@ struct landlock_ruleset_attr { * * - %LANDLOCK_CREATE_RULESET_VERSION: Get the highest supported Landlock ABI * version. + * - %LANDLOCK_CREATE_RULESET_ERRATA: Get a bitmask of fixed issues. */ /* clang-format off */ #define LANDLOCK_CREATE_RULESET_VERSION (1U << 0) +#define LANDLOCK_CREATE_RULESET_ERRATA (1U << 1) /* clang-format on */ /** diff --git a/security/landlock/errata.h b/security/landlock/errata.h new file mode 100644 index 000000000000..f26b28b9873d --- /dev/null +++ b/security/landlock/errata.h @@ -0,0 +1,87 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Landlock - Errata information + * + * Copyright © 2025 Microsoft Corporation + */ + +#ifndef _SECURITY_LANDLOCK_ERRATA_H +#define _SECURITY_LANDLOCK_ERRATA_H + +#include <linux/init.h> + +struct landlock_erratum { + const int abi; + const u8 number; +}; + +/* clang-format off */ +#define LANDLOCK_ERRATUM(NUMBER) \ + { \ + .abi = LANDLOCK_ERRATA_ABI, \ + .number = NUMBER, \ + }, +/* clang-format on */ + +/* + * Some fixes may require user space to check if they are applied on the running + * kernel before using a specific feature. For instance, this applies when a + * restriction was previously too restrictive and is now getting relaxed (for + * compatibility or semantic reasons). However, non-visible changes for + * legitimate use (e.g. security fixes) do not require an erratum. + */ +static const struct landlock_erratum landlock_errata_init[] __initconst = { + +/* + * Only Sparse may not implement __has_include. If a compiler does not + * implement __has_include, a warning will be printed at boot time (see + * setup.c). + */ +#ifdef __has_include + +#define LANDLOCK_ERRATA_ABI 1 +#if __has_include("errata/abi-1.h") +#include "errata/abi-1.h" +#endif +#undef LANDLOCK_ERRATA_ABI + +#define LANDLOCK_ERRATA_ABI 2 +#if __has_include("errata/abi-2.h") +#include "errata/abi-2.h" +#endif +#undef LANDLOCK_ERRATA_ABI + +#define LANDLOCK_ERRATA_ABI 3 +#if __has_include("errata/abi-3.h") +#include "errata/abi-3.h" +#endif +#undef LANDLOCK_ERRATA_ABI + +#define LANDLOCK_ERRATA_ABI 4 +#if __has_include("errata/abi-4.h") +#include "errata/abi-4.h" +#endif +#undef LANDLOCK_ERRATA_ABI + +/* + * For each new erratum, we need to include all the ABI files up to the impacted + * ABI to make all potential future intermediate errata easy to backport. + * + * If such change involves more than one ABI addition, then it must be in a + * dedicated commit with the same Fixes tag as used for the actual fix. + * + * Each commit creating a new security/landlock/errata/abi-*.h file must have a + * Depends-on tag to reference the commit that previously added the line to + * include this new file, except if the original Fixes tag is enough. + * + * Each erratum must be documented in its related ABI file, and a dedicated + * commit must update Documentation/userspace-api/landlock.rst to include this + * erratum. This commit will not be backported. + */ + +#endif + + {} +}; + +#endif /* _SECURITY_LANDLOCK_ERRATA_H */ diff --git a/security/landlock/setup.c b/security/landlock/setup.c index c71832a8e369..0c85ea27e409 100644 --- a/security/landlock/setup.c +++ b/security/landlock/setup.c @@ -6,12 +6,14 @@ * Copyright © 2018-2020 ANSSI */ +#include <linux/bits.h> #include <linux/init.h> #include <linux/lsm_hooks.h> #include <uapi/linux/lsm.h> #include "common.h" #include "cred.h" +#include "errata.h" #include "fs.h" #include "net.h" #include "setup.h" @@ -31,8 +33,36 @@ struct lsm_blob_sizes landlock_blob_sizes __ro_after_init = { .lbs_superblock = sizeof(struct landlock_superblock_security), }; +int landlock_errata __ro_after_init; + +static void __init compute_errata(void) +{ + size_t i; + +#ifndef __has_include + /* + * This is a safeguard to make sure the compiler implements + * __has_include (see errata.h). + */ + WARN_ON_ONCE(1); + return; +#endif + + for (i = 0; landlock_errata_init[i].number; i++) { + const int prev_errata = landlock_errata; + + if (WARN_ON_ONCE(landlock_errata_init[i].abi > + landlock_abi_version)) + continue; + + landlock_errata |= BIT(landlock_errata_init[i].number - 1); + WARN_ON_ONCE(prev_errata == landlock_errata); + } +} + static int __init landlock_init(void) { + compute_errata(); landlock_add_cred_hooks(); landlock_add_task_hooks(); landlock_add_fs_hooks(); diff --git a/security/landlock/setup.h b/security/landlock/setup.h index c4252d46d49d..fca307c35fee 100644 --- a/security/landlock/setup.h +++ b/security/landlock/setup.h @@ -11,7 +11,10 @@ #include <linux/lsm_hooks.h> +extern const int landlock_abi_version; + extern bool landlock_initialized; +extern int landlock_errata; extern struct lsm_blob_sizes landlock_blob_sizes; extern const struct lsm_id landlock_lsmid; diff --git a/security/landlock/syscalls.c b/security/landlock/syscalls.c index a9760d252fc2..cf9e0483e542 100644 --- a/security/landlock/syscalls.c +++ b/security/landlock/syscalls.c @@ -160,7 +160,9 @@ static const struct file_operations ruleset_fops = { * the new ruleset. * @size: Size of the pointed &struct landlock_ruleset_attr (needed for * backward and forward compatibility). - * @flags: Supported value: %LANDLOCK_CREATE_RULESET_VERSION. + * @flags: Supported value: + * - %LANDLOCK_CREATE_RULESET_VERSION + * - %LANDLOCK_CREATE_RULESET_ERRATA * * This system call enables to create a new Landlock ruleset, and returns the * related file descriptor on success. @@ -169,6 +171,10 @@ static const struct file_operations ruleset_fops = { * 0, then the returned value is the highest supported Landlock ABI version * (starting at 1). * + * If @flags is %LANDLOCK_CREATE_RULESET_ERRATA and @attr is NULL and @size is + * 0, then the returned value is a bitmask of fixed issues for the current + * Landlock ABI version. + * * Possible returned errors are: * * - %EOPNOTSUPP: Landlock is supported by the kernel but disabled at boot time; @@ -192,9 +198,15 @@ SYSCALL_DEFINE3(landlock_create_ruleset, return -EOPNOTSUPP; if (flags) { - if ((flags == LANDLOCK_CREATE_RULESET_VERSION) && !attr && - !size) - return LANDLOCK_ABI_VERSION; + if (attr || size) + return -EINVAL; + + if (flags == LANDLOCK_CREATE_RULESET_VERSION) + return landlock_abi_version; + + if (flags == LANDLOCK_CREATE_RULESET_ERRATA) + return landlock_errata; + return -EINVAL; } @@ -235,6 +247,8 @@ SYSCALL_DEFINE3(landlock_create_ruleset, return ruleset_fd; } +const int landlock_abi_version = LANDLOCK_ABI_VERSION; + /* * Returns an owned ruleset from a FD. It is thus needed to call * landlock_put_ruleset() on the return value. diff --git a/tools/testing/selftests/landlock/base_test.c b/tools/testing/selftests/landlock/base_test.c index 1bc16fde2e8a..4766f8fec9f6 100644 --- a/tools/testing/selftests/landlock/base_test.c +++ b/tools/testing/selftests/landlock/base_test.c @@ -98,10 +98,54 @@ TEST(abi_version) ASSERT_EQ(EINVAL, errno); } +/* + * Old source trees might not have the set of Kselftest fixes related to kernel + * UAPI headers. + */ +#ifndef LANDLOCK_CREATE_RULESET_ERRATA +#define LANDLOCK_CREATE_RULESET_ERRATA (1U << 1) +#endif + +TEST(errata) +{ + const struct landlock_ruleset_attr ruleset_attr = { + .handled_access_fs = LANDLOCK_ACCESS_FS_READ_FILE, + }; + int errata; + + errata = landlock_create_ruleset(NULL, 0, + LANDLOCK_CREATE_RULESET_ERRATA); + /* The errata bitmask will not be backported to tests. */ + ASSERT_LE(0, errata); + TH_LOG("errata: 0x%x", errata); + + ASSERT_EQ(-1, landlock_create_ruleset(&ruleset_attr, 0, + LANDLOCK_CREATE_RULESET_ERRATA)); + ASSERT_EQ(EINVAL, errno); + + ASSERT_EQ(-1, landlock_create_ruleset(NULL, sizeof(ruleset_attr), + LANDLOCK_CREATE_RULESET_ERRATA)); + ASSERT_EQ(EINVAL, errno); + + ASSERT_EQ(-1, + landlock_create_ruleset(&ruleset_attr, sizeof(ruleset_attr), + LANDLOCK_CREATE_RULESET_ERRATA)); + ASSERT_EQ(EINVAL, errno); + + ASSERT_EQ(-1, landlock_create_ruleset( + NULL, 0, + LANDLOCK_CREATE_RULESET_VERSION | + LANDLOCK_CREATE_RULESET_ERRATA)); + ASSERT_EQ(-1, landlock_create_ruleset(NULL, 0, + LANDLOCK_CREATE_RULESET_ERRATA | + 1 << 31)); + ASSERT_EQ(EINVAL, errno); +} + /* Tests ordering of syscall argument checks. */ TEST(create_ruleset_checks_ordering) { - const int last_flag = LANDLOCK_CREATE_RULESET_VERSION; + const int last_flag = LANDLOCK_CREATE_RULESET_ERRATA; const int invalid_flag = last_flag << 1; int ruleset_fd; const struct landlock_ruleset_attr ruleset_attr = {

2 months, 2 weeks

3
2
0 0

[PATCH] x86/pvh: Call C code via the kernel virtual mapping

by Jason Andryuk

From: Ard Biesheuvel <ardb(a)kernel.org> [ Upstream commit e8fbc0d9cab6c1ee6403f42c0991b0c1d5dbc092 ] Calling C code via a different mapping than it was linked at is problematic, because the compiler assumes that RIP-relative and absolute symbol references are interchangeable. GCC in particular may use RIP-relative per-CPU variable references even when not using -fpic. So call xen_prepare_pvh() via its kernel virtual mapping on x86_64, so that those RIP-relative references produce the correct values. This matches the pre-existing behavior for i386, which also invokes xen_prepare_pvh() via the kernel virtual mapping before invoking startup_32 with paging disabled again. Fixes: 7243b93345f7 ("xen/pvh: Bootstrap PVH guest") Tested-by: Jason Andryuk <jason.andryuk(a)amd.com> Reviewed-by: Jason Andryuk <jason.andryuk(a)amd.com> Signed-off-by: Ard Biesheuvel <ardb(a)kernel.org> Message-ID: <20241009160438.3884381-8-ardb+git(a)google.com> Signed-off-by: Juergen Gross <jgross(a)suse.com> [ Stable context update ] Signed-off-by: Jason Andryuk <jason.andryuk(a)amd.com> --- Stable backport for 6.6 .. 5.10. Direct cherry-pick needed context fixups, which are made here. This upstream commit was previously included in stable, but with the pre-req of b464b461d27d ("x86/pvh: Set phys_base when calling xen_prepare_pvh()"). Both were subsequently reverted as b464b461d27d caused regressions. This backport, e8fbc0d9cab6, in isolation is correct. This fixes a regression introduced by the backport of upstream commit b4845bb6383821a9516ce30af3a27dc873e37fd4 ("x86/xen: add central hypercall functions") b4845bb63838 adds a comparison between rip-relative xen_hypercall_amd() and kernel virtual address of xen_hypercall_amd() to determine whether to use the AMD or Intel variant. When running from the identity mapped address, the comparison always fail. The leads to calling xen_hypercall_intel(), even on AMD processors, which faults and halts boot. This affects PVH dom0 - domU doesn't seem to be affected. This patch performs the rip-relative mapping from the kernel virtual mapping, so the values can be properly compared. --- arch/x86/platform/pvh/head.S | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/arch/x86/platform/pvh/head.S b/arch/x86/platform/pvh/head.S index c4365a05ab83..fc46b4dfbd74 100644 --- a/arch/x86/platform/pvh/head.S +++ b/arch/x86/platform/pvh/head.S @@ -100,7 +100,12 @@ SYM_CODE_START_LOCAL(pvh_start_xen) xor %edx, %edx wrmsr - call xen_prepare_pvh + /* Call xen_prepare_pvh() via the kernel virtual mapping */ + leaq xen_prepare_pvh(%rip), %rax + subq phys_base(%rip), %rax + addq $__START_KERNEL_map, %rax + ANNOTATE_RETPOLINE_SAFE + call *%rax /* startup_64 expects boot_params in %rsi. */ mov $_pa(pvh_bootparams), %rsi -- 2.49.0

2 months, 2 weeks

2
2
0 0

[PATCH v2] Input: cyttsp5 - ensure minimum reset pulse width

by Hugo Villeneuve

From: Hugo Villeneuve <hvilleneuve(a)dimonoff.com> The current reset pulse width is measured to be 5us on a Renesas RZ/G2L SOM. The manufacturer's minimum reset pulse width is specified as 10us. Extend reset pulse width to make sure it is long enough on all platforms. Also reword confusing comments about reset pin assertion. Fixes: 5b0c03e24a06 ("Input: Add driver for Cypress Generation 5 touchscreen") Cc: <stable(a)vger.kernel.org> Acked-by: Alistair Francis <alistair(a)alistair23.me> Signed-off-by: Hugo Villeneuve <hvilleneuve(a)dimonoff.com> --- Changes for v2: - change delay from 1000us to 10us, to eliminate confusion - Add Alistair acked-by tag --- drivers/input/touchscreen/cyttsp5.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/input/touchscreen/cyttsp5.c b/drivers/input/touchscreen/cyttsp5.c index eafe5a9b89648..14c43f0a6c217 100644 --- a/drivers/input/touchscreen/cyttsp5.c +++ b/drivers/input/touchscreen/cyttsp5.c @@ -870,13 +870,16 @@ static int cyttsp5_probe(struct device *dev, struct regmap *regmap, int irq, ts->input->phys = ts->phys; input_set_drvdata(ts->input, ts); - /* Reset the gpio to be in a reset state */ + /* Assert gpio to be in a reset state */ ts->reset_gpio = devm_gpiod_get_optional(dev, "reset", GPIOD_OUT_HIGH); if (IS_ERR(ts->reset_gpio)) { error = PTR_ERR(ts->reset_gpio); dev_err(dev, "Failed to request reset gpio, error %d\n", error); return error; } + + fsleep(10); /* Ensure long-enough reset pulse (minimum 10us). */ + gpiod_set_value_cansleep(ts->reset_gpio, 0); /* Need a delay to have device up */ base-commit: 3b07108ada81a8ebcebf1fe61367b4e436c895bd -- 2.39.5

2 months, 2 weeks

2
1
0 0

[PATCH] staging: axis-fifo: Correct handling of tx_fifo_depth for size validation

by Gabriel Shahrouzi

Remove erroneous subtraction of 4 from the total FIFO depth read from device tree. The stored depth is for checking against total capacity, not initial vacancy. This prevented writes near the FIFO's full size. The check performed just before data transfer, which uses live reads of the TDFV register to determine current vacancy, correctly handles the initial Depth - 4 hardware state and subsequent FIFO fullness. Fixes: 4a965c5f89de ("staging: add driver for Xilinx AXI-Stream FIFO v4.1 IP core") Cc: stable(a)vger.kernel.org Signed-off-by: Gabriel Shahrouzi <gshahrouzi(a)gmail.com> --- drivers/staging/axis-fifo/axis-fifo.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/drivers/staging/axis-fifo/axis-fifo.c b/drivers/staging/axis-fifo/axis-fifo.c index 76db29e4d2828..351f983ef9149 100644 --- a/drivers/staging/axis-fifo/axis-fifo.c +++ b/drivers/staging/axis-fifo/axis-fifo.c @@ -770,9 +770,6 @@ static int axis_fifo_parse_dt(struct axis_fifo *fifo) goto end; } - /* IP sets TDFV to fifo depth - 4 so we will do the same */ - fifo->tx_fifo_depth -= 4; - ret = get_dts_property(fifo, "xlnx,use-rx-data", &fifo->has_rx_fifo); if (ret) { dev_err(fifo->dt_device, "missing xlnx,use-rx-data property\n"); -- 2.43.0

2 months, 2 weeks

1
0
0 0

Please backport 8542870237c3 ("md: fix mddev uaf while iterating all_mddevs list") back down to 6.12.y (and question on older stable series back to 6.1.y)

by Salvatore Bonaccorso

Hi stable mantainers, [note I'm CC'ing here Guillaume, Yu and Christoph] In Debian we were affected for a while by https://bugs.debian.org/1086175 which we found to be reported by Guillaume as https://lore.kernel.org/all/Z7Y0SURoA8xwg7vn@bender.morinfr.org/ . The issue went in fact back to 6.0. The fix was applied upstream and then backported to 6.14.2 already. Can you backport it please as well down to the other stable series, at least back to 6.12.y it goes with applying (and unless I miss something fixes the issue, we cherry-picked the commit for Debian trixie and our 6.12.y kernel in advance already). For going back to 6.1.y it seems it won't apply cleanly, You, Christoph might you be available to look into it to make this possible? Regards, Salvatore

2 months, 2 weeks

2
1
0 0

[PATCH v2] iio: adc: Include valid channel for channel selection

by Gabriel Shahrouzi

According to the datasheet on page 9 under the channel selection table, all devices (AD7816/7/8) are able to use the channel marked as 7. This channel is used for diagnostic purposes by routing the internal 1.23V bandgap source through the MUX to the input of the ADC. Replace checking for string equality with checking for the same chip ID to reduce time complexity. Group invalid channels for all devices together because they are processed the same way. Fixes: 7924425db04a ("staging: iio: adc: new driver for AD7816 devices") Cc: stable(a)vger.kernel.org Signed-off-by: Gabriel Shahrouzi <gshahrouzi(a)gmail.com> --- Changes since v2: - Refactor by adding chip_info struct which simplifies condtional logic. --- drivers/staging/iio/adc/ad7816.c | 68 ++++++++++++++++++-------------- 1 file changed, 38 insertions(+), 30 deletions(-) diff --git a/drivers/staging/iio/adc/ad7816.c b/drivers/staging/iio/adc/ad7816.c index 6c14d7bcdd675..ec955cbf06c17 100644 --- a/drivers/staging/iio/adc/ad7816.c +++ b/drivers/staging/iio/adc/ad7816.c @@ -41,8 +41,20 @@ * struct ad7816_chip_info - chip specific information */ +enum ad7816_type { + ID_AD7816, + ID_AD7817, + ID_AD7818, +}; + struct ad7816_chip_info { - kernel_ulong_t id; + const char *name; + enum ad7816_type type; + u8 max_channels; +}; + +struct ad7816_state { + const struct ad7816_chip_info *chip_info; struct spi_device *spi_dev; struct gpio_desc *rdwr_pin; struct gpio_desc *convert_pin; @@ -52,16 +64,11 @@ struct ad7816_chip_info { u8 mode; }; -enum ad7816_type { - ID_AD7816, - ID_AD7817, - ID_AD7818, -}; /* * ad7816 data access by SPI */ -static int ad7816_spi_read(struct ad7816_chip_info *chip, u16 *data) +static int ad7816_spi_read(struct ad7816_state *chip, u16 *data) { struct spi_device *spi_dev = chip->spi_dev; int ret; @@ -84,7 +91,7 @@ static int ad7816_spi_read(struct ad7816_chip_info *chip, u16 *data) gpiod_set_value(chip->convert_pin, 1); } - if (chip->id == ID_AD7816 || chip->id == ID_AD7817) { + if (chip->chip_info->type == ID_AD7816 || chip->chip_info->type == ID_AD7817) { while (gpiod_get_value(chip->busy_pin)) cpu_relax(); } @@ -102,7 +109,7 @@ static int ad7816_spi_read(struct ad7816_chip_info *chip, u16 *data) return ret; } -static int ad7816_spi_write(struct ad7816_chip_info *chip, u8 data) +static int ad7816_spi_write(struct ad7816_state *chip, u8 data) { struct spi_device *spi_dev = chip->spi_dev; int ret; @@ -121,7 +128,7 @@ static ssize_t ad7816_show_mode(struct device *dev, char *buf) { struct iio_dev *indio_dev = dev_to_iio_dev(dev); - struct ad7816_chip_info *chip = iio_priv(indio_dev); + struct ad7816_state *chip = iio_priv(indio_dev); if (chip->mode) return sprintf(buf, "power-save\n"); @@ -134,7 +141,7 @@ static ssize_t ad7816_store_mode(struct device *dev, size_t len) { struct iio_dev *indio_dev = dev_to_iio_dev(dev); - struct ad7816_chip_info *chip = iio_priv(indio_dev); + struct ad7816_state *chip = iio_priv(indio_dev); if (strcmp(buf, "full")) { gpiod_set_value(chip->rdwr_pin, 1); @@ -167,7 +174,7 @@ static ssize_t ad7816_show_channel(struct device *dev, char *buf) { struct iio_dev *indio_dev = dev_to_iio_dev(dev); - struct ad7816_chip_info *chip = iio_priv(indio_dev); + struct ad7816_state *chip = iio_priv(indio_dev); return sprintf(buf, "%d\n", chip->channel_id); } @@ -178,7 +185,7 @@ static ssize_t ad7816_store_channel(struct device *dev, size_t len) { struct iio_dev *indio_dev = dev_to_iio_dev(dev); - struct ad7816_chip_info *chip = iio_priv(indio_dev); + struct ad7816_state *chip = iio_priv(indio_dev); unsigned long data; int ret; @@ -186,17 +193,10 @@ static ssize_t ad7816_store_channel(struct device *dev, if (ret) return ret; - if (data > AD7816_CS_MAX && data != AD7816_CS_MASK) { - dev_err(&chip->spi_dev->dev, "Invalid channel id %lu for %s.\n", - data, indio_dev->name); - return -EINVAL; - } else if (strcmp(indio_dev->name, "ad7818") == 0 && data > 1) { + if (data > chip->chip_info->max_channels && data != AD7816_CS_MASK) { dev_err(&chip->spi_dev->dev, - "Invalid channel id %lu for ad7818.\n", data); - return -EINVAL; - } else if (strcmp(indio_dev->name, "ad7816") == 0 && data > 0) { - dev_err(&chip->spi_dev->dev, - "Invalid channel id %lu for ad7816.\n", data); + "Invalid channel id %lu for %s (max regular: %u).\n", data, + chip->chip_info->name, chip->chip_info->max_channels); return -EINVAL; } @@ -215,7 +215,7 @@ static ssize_t ad7816_show_value(struct device *dev, char *buf) { struct iio_dev *indio_dev = dev_to_iio_dev(dev); - struct ad7816_chip_info *chip = iio_priv(indio_dev); + struct ad7816_state *chip = iio_priv(indio_dev); u16 data; s8 value; int ret; @@ -271,7 +271,7 @@ static ssize_t ad7816_show_oti(struct device *dev, char *buf) { struct iio_dev *indio_dev = dev_to_iio_dev(dev); - struct ad7816_chip_info *chip = iio_priv(indio_dev); + struct ad7816_state *chip = iio_priv(indio_dev); int value; if (chip->channel_id > AD7816_CS_MAX) { @@ -292,7 +292,7 @@ static inline ssize_t ad7816_set_oti(struct device *dev, size_t len) { struct iio_dev *indio_dev = dev_to_iio_dev(dev); - struct ad7816_chip_info *chip = iio_priv(indio_dev); + struct ad7816_state *chip = iio_priv(indio_dev); long value; u8 data; int ret; @@ -345,14 +345,22 @@ static const struct iio_info ad7816_info = { .event_attrs = &ad7816_event_attribute_group, }; +static const struct ad7816_chip_info ad7816_chip_infos[] = { + [ID_AD7816] = { .name = "ad7816", .max_channels = 0, .type = ID_AD7816 }, + [ID_AD7817] = { .name = "ad7817", .max_channels = 3, .type = ID_AD7817 }, + [ID_AD7818] = { .name = "ad7818", .max_channels = 1, .type = ID_AD7818 }, +}; + /* * device probe and remove */ static int ad7816_probe(struct spi_device *spi_dev) { - struct ad7816_chip_info *chip; + struct ad7816_state *chip; struct iio_dev *indio_dev; + const struct spi_device_id *id = spi_get_device_id(spi_dev); + enum ad7816_type chip_type = (enum ad7816_type)id->driver_data; int i, ret; indio_dev = devm_iio_device_alloc(&spi_dev->dev, sizeof(*chip)); @@ -361,12 +369,12 @@ static int ad7816_probe(struct spi_device *spi_dev) chip = iio_priv(indio_dev); /* this is only used for device removal purposes */ dev_set_drvdata(&spi_dev->dev, indio_dev); + chip->chip_info = &ad7816_chip_infos[chip_type]; chip->spi_dev = spi_dev; for (i = 0; i <= AD7816_CS_MAX; i++) chip->oti_data[i] = 203; - chip->id = spi_get_device_id(spi_dev)->driver_data; chip->rdwr_pin = devm_gpiod_get(&spi_dev->dev, "rdwr", GPIOD_OUT_HIGH); if (IS_ERR(chip->rdwr_pin)) { ret = PTR_ERR(chip->rdwr_pin); @@ -382,7 +390,7 @@ static int ad7816_probe(struct spi_device *spi_dev) ret); return ret; } - if (chip->id == ID_AD7816 || chip->id == ID_AD7817) { + if (chip->chip_info->type == ID_AD7816 || chip->chip_info->type == ID_AD7817) { chip->busy_pin = devm_gpiod_get(&spi_dev->dev, "busy", GPIOD_IN); if (IS_ERR(chip->busy_pin)) { @@ -393,7 +401,7 @@ static int ad7816_probe(struct spi_device *spi_dev) } } - indio_dev->name = spi_get_device_id(spi_dev)->name; + indio_dev->name = chip->chip_info->name; indio_dev->info = &ad7816_info; indio_dev->modes = INDIO_DIRECT_MODE; -- 2.43.0

2 months, 2 weeks

2
2
0 0

[PATCH v3 1/5] staging: iio: adc: ad7816: Allow channel 7 for all devices

by Gabriel Shahrouzi

According to the datasheet on page 9 under the channel selection table, all devices (AD7816/7/8) are able to use the channel marked as 7. This channel is used for diagnostic purposes by routing the internal 1.23V bandgap source through the MUX to the input of the ADC. Modify the channel validation logic to permit channel 7 for all supported device types. Fixes: 7924425db04a ("staging: iio: adc: new driver for AD7816 devices") Cc: stable(a)vger.kernel.org Signed-off-by: Gabriel Shahrouzi <gshahrouzi(a)gmail.com> --- drivers/staging/iio/adc/ad7816.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/staging/iio/adc/ad7816.c b/drivers/staging/iio/adc/ad7816.c index 6c14d7bcdd675..a44b0c8c82b12 100644 --- a/drivers/staging/iio/adc/ad7816.c +++ b/drivers/staging/iio/adc/ad7816.c @@ -190,11 +190,11 @@ static ssize_t ad7816_store_channel(struct device *dev, dev_err(&chip->spi_dev->dev, "Invalid channel id %lu for %s.\n", data, indio_dev->name); return -EINVAL; - } else if (strcmp(indio_dev->name, "ad7818") == 0 && data > 1) { + } else if (strcmp(indio_dev->name, "ad7818") == 0 && data > 1 && data != AD7816_CS_MASK) { dev_err(&chip->spi_dev->dev, "Invalid channel id %lu for ad7818.\n", data); return -EINVAL; - } else if (strcmp(indio_dev->name, "ad7816") == 0 && data > 0) { + } else if (strcmp(indio_dev->name, "ad7816") == 0 && data > 0 && data != AD7816_CS_MASK) { dev_err(&chip->spi_dev->dev, "Invalid channel id %lu for ad7816.\n", data); return -EINVAL; -- 2.43.0

2 months, 2 weeks

1
0
0 0

[PATCH v3 0/4] J722S: Disable WIZ0 and WIZ1 in SoC file

by Siddharth Vadapalli

Hello, This series disables the "serdes_wiz0" and "serdes_wiz1" device-tree nodes in the J722S SoC file and enables them in the board files where they are required along with "serdes0" and "serdes1". There are two reasons behind this change: 1. To follow the existing convention of disabling nodes in the SoC file and enabling them in the board file as required. 2. To address situations where a board file hasn't explicitly disabled "serdes_wiz0" and "serdes_wiz1" (example: am67a-beagley-ai.dts) as a result of which booting the board displays the following errors: wiz bus@f0000:phy@f000000: probe with driver wiz failed with error -12 ... wiz bus@f0000:phy@f010000: probe with driver wiz failed with error -12 Additionally, another series for DT cleanup at: https://lore.kernel.org/r/20250412052712.927626-1-s-vadapalli@ti.com/ has been squashed into this series as patches 3 and 4. This has been done based on Nishanth's suggestion at: https://lore.kernel.org/r/20250414143916.zhskssezbffmvnsz@dragonfly/ Series is based on linux-next tagged next-20250417. NOTE: For patches 1 and 2 of this series which are "Fixes", it has also been verified that this series applies to the following commit cfb2e2c57aef Merge tag 'mm-hotfixes-stable-2025-04-16-19-59' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm of Mainline Linux. v2 of this series is at: https://lore.kernel.org/r/20250408103606.3679505-1-s-vadapalli@ti.com/ Changes since v2: - Collected Reviewed-by tags from Udit Kumar <u-kumar1(a)ti.com>. - Squashed the DT cleanup series at: https://lore.kernel.org/r/20250412052712.927626-1-s-vadapalli@ti.com/ as patches 3 and 4 of this series. v1 of this series is at: https://lore.kernel.org/r/20250408060636.3413856-1-s-vadapalli@ti.com/ Changes since v1: - Added "Fixes" tag and updated commit message accordingly. Regards, Siddharth. Siddharth Vadapalli (4): arm64: dts: ti: k3-j722s-evm: Enable "serdes_wiz0" and "serdes_wiz1" arm64: dts: ti: k3-j722s-main: Disable "serdes_wiz0" and "serdes_wiz1" arm64: dts: ti: k3-j722s-main: don't disable serdes0 and serdes1 arm64: dts: ti: k3-j722s-evm: drop redundant status within serdes0/serdes1 arch/arm64/boot/dts/ti/k3-j722s-evm.dts | 10 ++++++++-- arch/arm64/boot/dts/ti/k3-j722s-main.dtsi | 8 ++++---- 2 files changed, 12 insertions(+), 6 deletions(-) -- 2.34.1

2 months, 2 weeks

3
6
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror