All of lore.kernel.org
 help / color / mirror / Atom feed
From: Thomas Gleixner <tglx@linutronix.de>
To: LKML <linux-kernel@vger.kernel.org>
Cc: x86@kernel.org, "Chang S. Bae" <chang.seok.bae@intel.com>,
	Dave Hansen <dave.hansen@linux.intel.com>,
	Arjan van de Ven <arjan@linux.intel.com>,
	kvm@vger.kernel.org, Paolo Bonzini <pbonzini@redhat.com>,
	"Liu, Jing2" <jing2.liu@intel.com>,
	Sean Christopherson <seanjc@google.com>,
	Xiaoyao Li <xiaoyao.li@intel.com>
Subject: [patch V2 08/30] x86/fpu: Do not inherit FPU context for kernel and IO worker threads
Date: Fri, 15 Oct 2021 03:16:06 +0200 (CEST)	[thread overview]
Message-ID: <20211015011538.839822981@linutronix.de> (raw)
In-Reply-To: 20211015011411.304289784@linutronix.de

There is no reason why kernel and IO worker threads need a full clone of
the parent's FPU state. Both are kernel threads which are not supposed to
use FPU. So copying a large state or doing XSAVE() is pointless. Just clean
out the minimaly required state for those tasks.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>

---
 arch/x86/kernel/fpu/core.c | 37 ++++++++++++++++++++++++++-----------
 1 file changed, 26 insertions(+), 11 deletions(-)
---
diff --git a/arch/x86/kernel/fpu/core.c b/arch/x86/kernel/fpu/core.c
index 191269edac97..9a6b195a8a00 100644
--- a/arch/x86/kernel/fpu/core.c
+++ b/arch/x86/kernel/fpu/core.c
@@ -212,6 +212,15 @@ static inline void fpstate_init_xstate(struct xregs_state *xsave)
 	xsave->header.xcomp_bv = XCOMP_BV_COMPACTED_FORMAT | xfeatures_mask_all;
 }
 
+static inline unsigned int init_fpstate_copy_size(void)
+{
+	if (!use_xsave())
+		return fpu_kernel_xstate_size;
+
+	/* XSAVE(S) just needs the legacy and the xstate header part */
+	return sizeof(init_fpstate.xsave);
+}
+
 static inline void fpstate_init_fxstate(struct fxregs_state *fx)
 {
 	fx->cwd = 0x37f;
@@ -260,6 +269,23 @@ int fpu_clone(struct task_struct *dst)
 		return 0;
 
 	/*
+	 * Enforce reload for user space tasks and prevent kernel threads
+	 * from trying to save the FPU registers on context switch.
+	 */
+	set_tsk_thread_flag(dst, TIF_NEED_FPU_LOAD);
+
+	/*
+	 * No FPU state inheritance for kernel threads and IO
+	 * worker threads.
+	 */
+	if (dst->flags & (PF_KTHREAD | PF_IO_WORKER)) {
+		/* Clear out the minimal state */
+		memcpy(&dst_fpu->state, &init_fpstate,
+		       init_fpstate_copy_size());
+		return 0;
+	}
+
+	/*
 	 * If the FPU registers are not owned by current just memcpy() the
 	 * state.  Otherwise save the FPU registers directly into the
 	 * child's FPU context, without any memory-to-memory copying.
@@ -272,8 +298,6 @@ int fpu_clone(struct task_struct *dst)
 		save_fpregs_to_fpstate(dst_fpu);
 	fpregs_unlock();
 
-	set_tsk_thread_flag(dst, TIF_NEED_FPU_LOAD);
-
 	trace_x86_fpu_copy_src(src_fpu);
 	trace_x86_fpu_copy_dst(dst_fpu);
 
@@ -322,15 +346,6 @@ static inline void restore_fpregs_from_init_fpstate(u64 features_mask)
 	pkru_write_default();
 }
 
-static inline unsigned int init_fpstate_copy_size(void)
-{
-	if (!use_xsave())
-		return fpu_kernel_xstate_size;
-
-	/* XSAVE(S) just needs the legacy and the xstate header part */
-	return sizeof(init_fpstate.xsave);
-}
-
 /*
  * Reset current->fpu memory state to the init values.
  */


  parent reply	other threads:[~2021-10-15  1:16 UTC|newest]

Thread overview: 62+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-10-15  1:15 [patch V2 00/30] x86/fpu: Preparatory cleanups for AMX support (part 1) Thomas Gleixner
2021-10-15  1:15 ` [patch V2 01/30] x86/fpu: Remove pointless argument from switch_fpu_finish() Thomas Gleixner
2021-10-20 13:44   ` [tip: x86/fpu] " tip-bot2 for Thomas Gleixner
2021-10-15  1:15 ` [patch V2 02/30] x86/fpu: Update stale comments Thomas Gleixner
2021-10-20 13:44   ` [tip: x86/fpu] " tip-bot2 for Thomas Gleixner
2021-10-15  1:15 ` [patch V2 03/30] x86/pkru: Remove useless include Thomas Gleixner
2021-10-20 13:44   ` [tip: x86/fpu] " tip-bot2 for Thomas Gleixner
2021-10-15  1:15 ` [patch V2 04/30] x86/fpu: Restrict xsaves()/xrstors() to independent states Thomas Gleixner
2021-10-20 13:44   ` [tip: x86/fpu] " tip-bot2 for Thomas Gleixner
2021-10-15  1:16 ` [patch V2 05/30] x86/fpu: Cleanup the on_boot_cpu clutter Thomas Gleixner
2021-10-20 13:44   ` [tip: x86/fpu] " tip-bot2 for Thomas Gleixner
2021-10-15  1:16 ` [patch V2 06/30] x86/fpu: Remove pointless memset in fpu_clone() Thomas Gleixner
2021-10-20 13:44   ` [tip: x86/fpu] " tip-bot2 for Thomas Gleixner
2021-10-15  1:16 ` [patch V2 07/30] x86/process: Clone FPU in copy_thread() Thomas Gleixner
2021-10-20 13:44   ` [tip: x86/fpu] " tip-bot2 for Thomas Gleixner
2021-10-15  1:16 ` Thomas Gleixner [this message]
2021-10-20 13:44   ` [tip: x86/fpu] x86/fpu: Do not inherit FPU context for kernel and IO worker threads tip-bot2 for Thomas Gleixner
2021-10-15  1:16 ` [patch V2 09/30] x86/fpu: Cleanup xstate xcomp_bv initialization Thomas Gleixner
2021-10-20 13:44   ` [tip: x86/fpu] " tip-bot2 for Thomas Gleixner
2021-10-15  1:16 ` [patch V2 10/30] x86/fpu/xstate: Provide and use for_each_xfeature() Thomas Gleixner
2021-10-20 13:44   ` [tip: x86/fpu] " tip-bot2 for Thomas Gleixner
2021-10-15  1:16 ` [patch V2 11/30] x86/fpu/xstate: Mark all init only functions __init Thomas Gleixner
2021-10-20 13:44   ` [tip: x86/fpu] " tip-bot2 for Thomas Gleixner
2021-10-15  1:16 ` [patch V2 12/30] x86/fpu: Move KVMs FPU swapping to FPU core Thomas Gleixner
2021-10-20 13:44   ` [tip: x86/fpu] " tip-bot2 for Thomas Gleixner
2021-10-15  1:16 ` [patch V2 13/30] x86/fpu: Replace KVMs home brewed FPU copy from user Thomas Gleixner
2021-10-20 13:44   ` [tip: x86/fpu] " tip-bot2 for Thomas Gleixner
2021-10-15  1:16 ` [patch V2 14/30] x86/fpu: Rework copy_xstate_to_uabi_buf() Thomas Gleixner
2021-10-20 13:44   ` [tip: x86/fpu] " tip-bot2 for Thomas Gleixner
2021-10-15  1:16 ` [patch V2 15/30] x86/fpu: Replace KVMs home brewed FPU copy to user Thomas Gleixner
2021-10-21 15:12   ` [tip: x86/fpu] " tip-bot2 for Thomas Gleixner
2021-10-15  1:16 ` [patch V2 16/30] x86/fpu: Mark fpu__init_prepare_fx_sw_frame() as __init Thomas Gleixner
2021-10-20 13:44   ` [tip: x86/fpu] " tip-bot2 for Thomas Gleixner
2021-10-15  1:16 ` [patch V2 17/30] x86/fpu: Move context switch and exit to user inlines into sched.h Thomas Gleixner
2021-10-20 13:44   ` [tip: x86/fpu] " tip-bot2 for Thomas Gleixner
2021-10-15  1:16 ` [patch V2 18/30] x86/fpu: Clean up cpu feature tests Thomas Gleixner
2021-10-20 13:44   ` [tip: x86/fpu] x86/fpu: Clean up CPU " tip-bot2 for Thomas Gleixner
2021-10-15  1:16 ` [patch V2 19/30] x86/fpu: Make os_xrstor_booting() private Thomas Gleixner
2021-10-20 13:44   ` [tip: x86/fpu] " tip-bot2 for Thomas Gleixner
2021-10-15  1:16 ` [patch V2 20/30] x86/fpu: Move os_xsave() and os_xrstor() to core Thomas Gleixner
2021-10-20 13:44   ` [tip: x86/fpu] " tip-bot2 for Thomas Gleixner
2021-10-15  1:16 ` [patch V2 21/30] x86/fpu: Move legacy ASM wrappers " Thomas Gleixner
2021-10-20 13:44   ` [tip: x86/fpu] " tip-bot2 for Thomas Gleixner
2021-10-15  1:16 ` [patch V2 22/30] x86/fpu: Make WARN_ON_FPU() private Thomas Gleixner
2021-10-20 13:44   ` [tip: x86/fpu] " tip-bot2 for Thomas Gleixner
2021-10-15  1:16 ` [patch V2 23/30] x86/fpu: Move fpregs_restore_userregs() to core Thomas Gleixner
2021-10-19 18:11   ` Borislav Petkov
2021-10-20 13:44   ` [tip: x86/fpu] " tip-bot2 for Thomas Gleixner
2021-10-15  1:16 ` [patch V2 24/30] x86/fpu: Move mxcsr related code " Thomas Gleixner
2021-10-20 13:44   ` [tip: x86/fpu] " tip-bot2 for Thomas Gleixner
2021-10-15  1:16 ` [patch V2 25/30] x86/fpu: Move fpstate functions to api.h Thomas Gleixner
2021-10-20 13:44   ` [tip: x86/fpu] " tip-bot2 for Thomas Gleixner
2021-10-15  1:16 ` [patch V2 26/30] x86/fpu: Remove internal.h dependency from fpu/signal.h Thomas Gleixner
2021-10-20 13:44   ` [tip: x86/fpu] " tip-bot2 for Thomas Gleixner
2021-10-15  1:16 ` [patch V2 27/30] x86/sev: Include fpu/xcr.h Thomas Gleixner
2021-10-20 13:44   ` [tip: x86/fpu] " tip-bot2 for Thomas Gleixner
2021-10-15  1:16 ` [patch V2 28/30] x86/fpu: Mop up the internal.h leftovers Thomas Gleixner
2021-10-20 13:44   ` [tip: x86/fpu] " tip-bot2 for Thomas Gleixner
2021-10-15  1:16 ` [patch V2 29/30] x86/fpu: Replace the includes of fpu/internal.h Thomas Gleixner
2021-10-20 13:44   ` [tip: x86/fpu] " tip-bot2 for Thomas Gleixner
2021-10-15  1:16 ` [patch V2 30/30] x86/fpu: Provide a proper function for ex_handler_fprestore() Thomas Gleixner
2021-10-20 13:44   ` [tip: x86/fpu] " tip-bot2 for Thomas Gleixner

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=20211015011538.839822981@linutronix.de \
    --to=tglx@linutronix.de \
    --cc=arjan@linux.intel.com \
    --cc=chang.seok.bae@intel.com \
    --cc=dave.hansen@linux.intel.com \
    --cc=jing2.liu@intel.com \
    --cc=kvm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=pbonzini@redhat.com \
    --cc=seanjc@google.com \
    --cc=x86@kernel.org \
    --cc=xiaoyao.li@intel.com \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.