From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: stable@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
patches@lists.linux.dev, Will Deacon <will@kernel.org>,
Mark Rutland <mark.rutland@arm.com>,
Catalin Marinas <catalin.marinas@arm.com>,
Marc Zyngier <maz@kernel.org>, Mark Brown <broonie@kernel.org>,
Sasha Levin <sashal@kernel.org>
Subject: [PATCH 6.6 45/72] arm64/fpsimd: signal: Consistently read FPSIMD context
Date: Wed, 4 Feb 2026 15:40:48 +0100 [thread overview]
Message-ID: <20260204143847.260310999@linuxfoundation.org> (raw)
In-Reply-To: <20260204143845.603454952@linuxfoundation.org>
6.6-stable review patch. If anyone has any objections, please let me know.
------------------
From: Mark Rutland <mark.rutland@arm.com>
[ Upstream commit be625d803c3bbfa9652697eb57589fe6f2f24b89 ]
For historical reasons, restore_sve_fpsimd_context() has an open-coded
copy of the logic from read_fpsimd_context(), which is used to either
restore an FPSIMD-only context, or to merge FPSIMD state into an
SVE state when restoring an SVE+FPSIMD context. The logic is *almost*
identical.
Refactor the logic to avoid duplication and make this clearer.
This comes with two functional changes that I do not believe will be
problematic in practice:
* The user_fpsimd_state::size field will be checked in all restore paths
that consume it user_fpsimd_state. The kernel always populates this
field when delivering a signal, and so this should contain the
expected value unless it has been corrupted.
* If a read of user_fpsimd_state fails, we will return early without
modifying TIF_SVE, the saved SVCR, or the save fp_type. This will
leave the task in a consistent state, without potentially resurrecting
stale FPSIMD state. A read of user_fpsimd_state should never fail
unless the structure has been corrupted or the stack has been
unmapped.
Suggested-by: Will Deacon <will@kernel.org>
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Marc Zyngier <maz@kernel.org>
Cc: Mark Brown <broonie@kernel.org>
Cc: Will Deacon <will@kernel.org>
Link: https://lore.kernel.org/r/20250508132644.1395904-5-mark.rutland@arm.com
[will: Ensure read_fpsimd_context() returns negative error code or zero]
Signed-off-by: Will Deacon <will@kernel.org>
Stable-dep-of: d2907cbe9ea0 ("arm64/fpsimd: signal: Fix restoration of SVE context")
Signed-off-by: Sasha Levin <sashal@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
arch/arm64/kernel/signal.c | 57 ++++++++++++++++++++++-----------------------
1 file changed, 29 insertions(+), 28 deletions(-)
--- a/arch/arm64/kernel/signal.c
+++ b/arch/arm64/kernel/signal.c
@@ -202,29 +202,39 @@ static int preserve_fpsimd_context(struc
return err ? -EFAULT : 0;
}
-static int restore_fpsimd_context(struct user_ctxs *user)
+static int read_fpsimd_context(struct user_fpsimd_state *fpsimd,
+ struct user_ctxs *user)
{
- struct user_fpsimd_state fpsimd;
- int err = 0;
+ int err;
/* check the size information */
if (user->fpsimd_size != sizeof(struct fpsimd_context))
return -EINVAL;
/* copy the FP and status/control registers */
- err = __copy_from_user(fpsimd.vregs, &(user->fpsimd->vregs),
- sizeof(fpsimd.vregs));
- __get_user_error(fpsimd.fpsr, &(user->fpsimd->fpsr), err);
- __get_user_error(fpsimd.fpcr, &(user->fpsimd->fpcr), err);
+ err = __copy_from_user(fpsimd->vregs, &(user->fpsimd->vregs),
+ sizeof(fpsimd->vregs));
+ __get_user_error(fpsimd->fpsr, &(user->fpsimd->fpsr), err);
+ __get_user_error(fpsimd->fpcr, &(user->fpsimd->fpcr), err);
+
+ return err ? -EFAULT : 0;
+}
+
+static int restore_fpsimd_context(struct user_ctxs *user)
+{
+ struct user_fpsimd_state fpsimd;
+ int err;
+
+ err = read_fpsimd_context(&fpsimd, user);
+ if (err)
+ return err;
clear_thread_flag(TIF_SVE);
current->thread.fp_type = FP_STATE_FPSIMD;
/* load the hardware registers from the fpsimd_state structure */
- if (!err)
- fpsimd_update_current_state(&fpsimd);
-
- return err ? -EFAULT : 0;
+ fpsimd_update_current_state(&fpsimd);
+ return 0;
}
@@ -316,12 +326,8 @@ static int restore_sve_fpsimd_context(st
* consistency and robustness, reject restoring streaming SVE state
* without an SVE payload.
*/
- if (!sm && user->sve_size == sizeof(*user->sve)) {
- clear_thread_flag(TIF_SVE);
- current->thread.svcr &= ~SVCR_SM_MASK;
- current->thread.fp_type = FP_STATE_FPSIMD;
- goto fpsimd_only;
- }
+ if (!sm && user->sve_size == sizeof(*user->sve))
+ return restore_fpsimd_context(user);
vq = sve_vq_from_vl(vl);
@@ -357,19 +363,14 @@ static int restore_sve_fpsimd_context(st
set_thread_flag(TIF_SVE);
current->thread.fp_type = FP_STATE_SVE;
-fpsimd_only:
- /* copy the FP and status/control registers */
- /* restore_sigframe() already checked that user->fpsimd != NULL. */
- err = __copy_from_user(fpsimd.vregs, user->fpsimd->vregs,
- sizeof(fpsimd.vregs));
- __get_user_error(fpsimd.fpsr, &user->fpsimd->fpsr, err);
- __get_user_error(fpsimd.fpcr, &user->fpsimd->fpcr, err);
+ err = read_fpsimd_context(&fpsimd, user);
+ if (err)
+ return err;
- /* load the hardware registers from the fpsimd_state structure */
- if (!err)
- fpsimd_update_current_state(&fpsimd);
+ /* Merge the FPSIMD registers into the SVE state */
+ fpsimd_update_current_state(&fpsimd);
- return err ? -EFAULT : 0;
+ return 0;
}
#else /* ! CONFIG_ARM64_SVE */
next prev parent reply other threads:[~2026-02-04 15:23 UTC|newest]
Thread overview: 82+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-02-04 14:40 [PATCH 6.6 00/72] 6.6.123-rc1 review Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 01/72] Bluetooth: hci_uart: fix null-ptr-deref in hci_uart_write_work Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 02/72] net/mlx5: Fix memory leak in esw_acl_ingress_lgcy_setup() Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 03/72] can: gs_usb: gs_usb_receive_bulk_callback(): fix error message Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 04/72] net: bcmasp: fix early exit leak with fixed phy Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 05/72] octeon_ep: Fix memory leak in octep_device_setup() Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 06/72] bonding: annotate data-races around slave->last_rx Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 07/72] net: mvpp2: cls: Fix memory leak in mvpp2_ethtool_cls_rule_ins() Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 08/72] ipv6: use the right ifindex when replying to icmpv6 from localhost Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 09/72] net: wwan: t7xx: fix potential skb->frags overflow in RX path Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 10/72] rocker: fix memory leak in rocker_world_port_post_fini() Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 11/72] nfc: llcp: Fix memleak in nfc_llcp_send_ui_frame() Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 12/72] ice: stop counting UDP csum mismatch as rx_errors Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 13/72] net/mlx5e: TC, delete flows only for existing peers Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 14/72] net/mlx5e: Report rx_discards_phy via rx_dropped Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 15/72] net/mlx5e: Account for netdev stats in ndo_get_stats64 Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 16/72] nfc: nci: Fix race between rfkill and nci_unregister_device() Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 17/72] net: bridge: fix static key check Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 18/72] net/mlx5e: Skip ESN replay window setup for IPsec crypto offload Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 19/72] scsi: firewire: sbp-target: Fix overflow in sbp_make_tpg() Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 20/72] ASoC: Intel: sof_es8336: fix headphone GPIO logic inversion Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 21/72] gpiolib: acpi: use BIT_ULL() for u64 mask in address space handler Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 22/72] dma/pool: distinguish between missing and exhausted atomic pools Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 23/72] pinctrl: meson: mark the GPIO controller as sleeping Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 24/72] riscv: compat: fix COMPAT_UTS_MACHINE definition Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 25/72] rust: kbuild: give `--config-path` to `rustfmt` in `.rsi` target Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 26/72] ASoC: fsl: imx-card: Do not force slot width to sample width Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 27/72] scsi: be2iscsi: Fix a memory leak in beiscsi_boot_get_sinfo() Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 28/72] ASoC: amd: yc: Add DMI quirk for Acer TravelMate P216-41-TCO Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 29/72] gpio: pca953x: mask interrupts in irq shutdown Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 30/72] scsi: qla2xxx: edif: Fix dma_free_coherent() size Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 31/72] efivarfs: fix error propagation in efivar_entry_get() Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 32/72] mptcp: only reset subflow errors when propagated Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 33/72] selftests: mptcp: check no dup close events after error Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 34/72] selftests: mptcp: check subflow errors in close events Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 35/72] selftests: mptcp: join: fix local endp not being tracked Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 36/72] flex_proportions: make fprop_new_period() hardirq safe Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 37/72] scripts: generate_rust_analyzer: Add compiler_builtins -> core dep Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 38/72] drm/imx/tve: fix probe device leak Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 39/72] drm/amdgpu/soc21: fix xclk for APUs Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 40/72] drm/amdgpu/gfx10: fix wptr reset in KGQ init Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 41/72] drm/amdgpu/gfx11: " Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 42/72] gpio: rockchip: Stop calling pinctrl for set_direction Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 43/72] mm/kfence: randomize the freelist on initialization Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 44/72] arm64/fpsimd: signal: Mandate SVE payload for streaming-mode state Greg Kroah-Hartman
2026-02-04 14:40 ` Greg Kroah-Hartman [this message]
2026-02-04 14:40 ` [PATCH 6.6 46/72] arm64/fpsimd: signal: Fix restoration of SVE context Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 47/72] mei: trace: treat reg parameter as string Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 48/72] ksmbd: smbd: fix dma_unmap_sg() nents Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 49/72] drm/amdgpu: Replace Mutex with Spinlock for RLCG register access to avoid Priority Inversion in SRIOV Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 50/72] ksmbd: Fix race condition in RPC handle list access Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 51/72] wifi: mac80211: move TDLS work to wiphy work Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 52/72] wifi: ath11k: add srng->lock for ath11k_hal_srng_* in monitor mode Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 53/72] team: Move team device type change at the end of team_port_add Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 54/72] drm/radeon: delete radeon_fence_process in is_signaled, no deadlock Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 55/72] btrfs: prevent use-after-free on page private data in btrfs_subpage_clear_uptodate() Greg Kroah-Hartman
2026-02-04 14:40 ` [PATCH 6.6 56/72] net/sched: act_ife: convert comma to semicolon Greg Kroah-Hartman
2026-02-04 14:41 ` [PATCH 6.6 57/72] ALSA: usb-audio: Fix missing unlock at error path of maxpacksize check Greg Kroah-Hartman
2026-02-04 14:41 ` [PATCH 6.6 58/72] pinctrl: lpass-lpi: implement .get_direction() for the GPIO driver Greg Kroah-Hartman
2026-02-04 14:41 ` [PATCH 6.6 59/72] drm/msm/a6xx: fix bogus hwcg register updates Greg Kroah-Hartman
2026-02-04 14:41 ` [PATCH 6.6 60/72] perf: sched: Fix perf crash with new is_user_task() helper Greg Kroah-Hartman
2026-02-04 14:41 ` [PATCH 6.6 61/72] writeback: fix 100% CPU usage when dirtytime_expire_interval is 0 Greg Kroah-Hartman
2026-02-04 14:41 ` [PATCH 6.6 62/72] mptcp: avoid dup SUB_CLOSED events after disconnect Greg Kroah-Hartman
2026-02-04 14:41 ` [PATCH 6.6 63/72] pinctrl: qcom: sm8350-lpass-lpi: Merge with SC7280 to fix I2S2 and SWR TX pins Greg Kroah-Hartman
2026-02-04 14:41 ` [PATCH 6.6 64/72] drm/amdkfd: Dont use sw fault filter if retry cam enabled Greg Kroah-Hartman
2026-02-04 14:41 ` [PATCH 6.6 65/72] drm/amdgpu: fix NULL pointer dereference in amdgpu_gmc_filter_faults_remove Greg Kroah-Hartman
2026-02-04 14:41 ` [PATCH 6.6 66/72] xsk: Fix race condition in AF_XDP generic RX path Greg Kroah-Hartman
2026-02-04 14:41 ` [PATCH 6.6 67/72] ksmbd: fix recursive locking in RPC handle list access Greg Kroah-Hartman
2026-02-04 14:41 ` [PATCH 6.6 68/72] ptr_ring: do not block hard interrupts in ptr_ring_resize_multiple() Greg Kroah-Hartman
2026-02-04 14:41 ` [PATCH 6.6 69/72] drm/amd/display: use udelay rather than fsleep Greg Kroah-Hartman
2026-02-04 14:41 ` [PATCH 6.6 70/72] Revert "net: Allow to use SMP threads for backlog NAPI." Greg Kroah-Hartman
2026-02-04 14:41 ` [PATCH 6.6 71/72] Revert "net: Remove conditional threaded-NAPI wakeup based on task state." Greg Kroah-Hartman
2026-02-04 14:41 ` [PATCH 6.6 72/72] bpf/selftests: test_select_reuseport_kern: Remove unused header Greg Kroah-Hartman
2026-02-04 19:52 ` [PATCH 6.6 00/72] 6.6.123-rc1 review Brett A C Sheffield
2026-02-04 20:01 ` Florian Fainelli
2026-02-04 20:27 ` Jon Hunter
2026-02-04 22:49 ` Peter Schneider
2026-02-05 7:46 ` Ron Economos
2026-02-05 8:28 ` Francesco Dolcini
2026-02-05 11:39 ` Mark Brown
2026-02-06 5:54 ` Shung-Hsi Yu
2026-02-06 9:25 ` Miguel Ojeda
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20260204143847.260310999@linuxfoundation.org \
--to=gregkh@linuxfoundation.org \
--cc=broonie@kernel.org \
--cc=catalin.marinas@arm.com \
--cc=mark.rutland@arm.com \
--cc=maz@kernel.org \
--cc=patches@lists.linux.dev \
--cc=sashal@kernel.org \
--cc=stable@vger.kernel.org \
--cc=will@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox