From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
stable@vger.kernel.org, Andy Lutomirski <luto@kernel.org>,
Rik van Riel <riel@redhat.com>, Borislav Petkov <bp@alien8.de>,
Brian Gerst <brgerst@gmail.com>,
Dave Hansen <dave.hansen@linux.intel.com>,
Denys Vlasenko <dvlasenk@redhat.com>,
Fenghua Yu <fenghua.yu@intel.com>,
"H. Peter Anvin" <hpa@zytor.com>,
Josh Poimboeuf <jpoimboe@redhat.com>,
Linus Torvalds <torvalds@linux-foundation.org>,
Oleg Nesterov <oleg@redhat.com>,
Peter Zijlstra <peterz@infradead.org>,
Quentin Casasnovas <quentin.casasnovas@oracle.com>,
Thomas Gleixner <tglx@linutronix.de>,
pbonzini@redhat.com, Ingo Molnar <mingo@kernel.org>,
Daniel Sangorrin <daniel.sangorrin@toshiba.co.jp>
Subject: [PATCH 4.4 32/48] x86/fpu: Remove use_eager_fpu()
Date: Thu, 18 Oct 2018 19:55:07 +0200 [thread overview]
Message-ID: <20181018175429.672632710@linuxfoundation.org> (raw)
In-Reply-To: <20181018175427.133690306@linuxfoundation.org>
4.4-stable review patch. If anyone has any objections, please let me know.
------------------
From: Andy Lutomirski <luto@kernel.org>
commit c592b57347069abfc0dcad3b3a302cf882602597 upstream.
This removes all the obvious code paths that depend on lazy FPU mode.
It shouldn't change the generated code at all.
Signed-off-by: Andy Lutomirski <luto@kernel.org>
Signed-off-by: Rik van Riel <riel@redhat.com>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Brian Gerst <brgerst@gmail.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Denys Vlasenko <dvlasenk@redhat.com>
Cc: Fenghua Yu <fenghua.yu@intel.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Quentin Casasnovas <quentin.casasnovas@oracle.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: pbonzini@redhat.com
Link: http://lkml.kernel.org/r/1475627678-20788-5-git-send-email-riel@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Daniel Sangorrin <daniel.sangorrin@toshiba.co.jp>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
arch/x86/crypto/crc32c-intel_glue.c | 17 +++-------------
arch/x86/include/asm/fpu/internal.h | 34 --------------------------------
arch/x86/kernel/fpu/core.c | 38 ++++--------------------------------
arch/x86/kernel/fpu/signal.c | 8 ++-----
arch/x86/kvm/cpuid.c | 4 ---
arch/x86/kvm/x86.c | 10 ---------
6 files changed, 14 insertions(+), 97 deletions(-)
--- a/arch/x86/crypto/crc32c-intel_glue.c
+++ b/arch/x86/crypto/crc32c-intel_glue.c
@@ -48,21 +48,13 @@
#ifdef CONFIG_X86_64
/*
* use carryless multiply version of crc32c when buffer
- * size is >= 512 (when eager fpu is enabled) or
- * >= 1024 (when eager fpu is disabled) to account
+ * size is >= 512 to account
* for fpu state save/restore overhead.
*/
-#define CRC32C_PCL_BREAKEVEN_EAGERFPU 512
-#define CRC32C_PCL_BREAKEVEN_NOEAGERFPU 1024
+#define CRC32C_PCL_BREAKEVEN 512
asmlinkage unsigned int crc_pcl(const u8 *buffer, int len,
unsigned int crc_init);
-static int crc32c_pcl_breakeven = CRC32C_PCL_BREAKEVEN_EAGERFPU;
-#define set_pcl_breakeven_point() \
-do { \
- if (!use_eager_fpu()) \
- crc32c_pcl_breakeven = CRC32C_PCL_BREAKEVEN_NOEAGERFPU; \
-} while (0)
#endif /* CONFIG_X86_64 */
static u32 crc32c_intel_le_hw_byte(u32 crc, unsigned char const *data, size_t length)
@@ -185,7 +177,7 @@ static int crc32c_pcl_intel_update(struc
* use faster PCL version if datasize is large enough to
* overcome kernel fpu state save/restore overhead
*/
- if (len >= crc32c_pcl_breakeven && irq_fpu_usable()) {
+ if (len >= CRC32C_PCL_BREAKEVEN && irq_fpu_usable()) {
kernel_fpu_begin();
*crcp = crc_pcl(data, len, *crcp);
kernel_fpu_end();
@@ -197,7 +189,7 @@ static int crc32c_pcl_intel_update(struc
static int __crc32c_pcl_intel_finup(u32 *crcp, const u8 *data, unsigned int len,
u8 *out)
{
- if (len >= crc32c_pcl_breakeven && irq_fpu_usable()) {
+ if (len >= CRC32C_PCL_BREAKEVEN && irq_fpu_usable()) {
kernel_fpu_begin();
*(__le32 *)out = ~cpu_to_le32(crc_pcl(data, len, *crcp));
kernel_fpu_end();
@@ -256,7 +248,6 @@ static int __init crc32c_intel_mod_init(
alg.update = crc32c_pcl_intel_update;
alg.finup = crc32c_pcl_intel_finup;
alg.digest = crc32c_pcl_intel_digest;
- set_pcl_breakeven_point();
}
#endif
return crypto_register_shash(&alg);
--- a/arch/x86/include/asm/fpu/internal.h
+++ b/arch/x86/include/asm/fpu/internal.h
@@ -57,11 +57,6 @@ extern u64 fpu__get_supported_xfeatures_
/*
* FPU related CPU feature flag helper routines:
*/
-static __always_inline __pure bool use_eager_fpu(void)
-{
- return true;
-}
-
static __always_inline __pure bool use_xsaveopt(void)
{
return static_cpu_has(X86_FEATURE_XSAVEOPT);
@@ -498,24 +493,6 @@ static inline int fpu_want_lazy_restore(
}
-/*
- * Wrap lazy FPU TS handling in a 'hw fpregs activation/deactivation'
- * idiom, which is then paired with the sw-flag (fpregs_active) later on:
- */
-
-static inline void __fpregs_activate_hw(void)
-{
- if (!use_eager_fpu())
- clts();
-}
-
-static inline void __fpregs_deactivate_hw(void)
-{
- if (!use_eager_fpu())
- stts();
-}
-
-/* Must be paired with an 'stts' (fpregs_deactivate_hw()) after! */
static inline void __fpregs_deactivate(struct fpu *fpu)
{
WARN_ON_FPU(!fpu->fpregs_active);
@@ -524,7 +501,6 @@ static inline void __fpregs_deactivate(s
this_cpu_write(fpu_fpregs_owner_ctx, NULL);
}
-/* Must be paired with a 'clts' (fpregs_activate_hw()) before! */
static inline void __fpregs_activate(struct fpu *fpu)
{
WARN_ON_FPU(fpu->fpregs_active);
@@ -549,22 +525,17 @@ static inline int fpregs_active(void)
}
/*
- * Encapsulate the CR0.TS handling together with the
- * software flag.
- *
* These generally need preemption protection to work,
* do try to avoid using these on their own.
*/
static inline void fpregs_activate(struct fpu *fpu)
{
- __fpregs_activate_hw();
__fpregs_activate(fpu);
}
static inline void fpregs_deactivate(struct fpu *fpu)
{
__fpregs_deactivate(fpu);
- __fpregs_deactivate_hw();
}
/*
@@ -591,8 +562,7 @@ switch_fpu_prepare(struct fpu *old_fpu,
* or if the past 5 consecutive context-switches used math.
*/
fpu.preload = static_cpu_has(X86_FEATURE_FPU) &&
- new_fpu->fpstate_active &&
- (use_eager_fpu() || new_fpu->counter > 5);
+ new_fpu->fpstate_active;
if (old_fpu->fpregs_active) {
if (!copy_fpregs_to_fpstate(old_fpu))
@@ -608,8 +578,6 @@ switch_fpu_prepare(struct fpu *old_fpu,
new_fpu->counter++;
__fpregs_activate(new_fpu);
prefetch(&new_fpu->state);
- } else {
- __fpregs_deactivate_hw();
}
} else {
old_fpu->counter = 0;
--- a/arch/x86/kernel/fpu/core.c
+++ b/arch/x86/kernel/fpu/core.c
@@ -53,27 +53,9 @@ static bool kernel_fpu_disabled(void)
return this_cpu_read(in_kernel_fpu);
}
-/*
- * Were we in an interrupt that interrupted kernel mode?
- *
- * On others, we can do a kernel_fpu_begin/end() pair *ONLY* if that
- * pair does nothing at all: the thread must not have fpu (so
- * that we don't try to save the FPU state), and TS must
- * be set (so that the clts/stts pair does nothing that is
- * visible in the interrupted kernel thread).
- *
- * Except for the eagerfpu case when we return true; in the likely case
- * the thread has FPU but we are not going to set/clear TS.
- */
static bool interrupted_kernel_fpu_idle(void)
{
- if (kernel_fpu_disabled())
- return false;
-
- if (use_eager_fpu())
- return true;
-
- return !current->thread.fpu.fpregs_active && (read_cr0() & X86_CR0_TS);
+ return !kernel_fpu_disabled();
}
/*
@@ -121,7 +103,6 @@ void __kernel_fpu_begin(void)
copy_fpregs_to_fpstate(fpu);
} else {
this_cpu_write(fpu_fpregs_owner_ctx, NULL);
- __fpregs_activate_hw();
}
}
EXPORT_SYMBOL(__kernel_fpu_begin);
@@ -132,8 +113,6 @@ void __kernel_fpu_end(void)
if (fpu->fpregs_active)
copy_kernel_to_fpregs(&fpu->state);
- else
- __fpregs_deactivate_hw();
kernel_fpu_enable();
}
@@ -194,10 +173,7 @@ void fpu__save(struct fpu *fpu)
preempt_disable();
if (fpu->fpregs_active) {
if (!copy_fpregs_to_fpstate(fpu)) {
- if (use_eager_fpu())
- copy_kernel_to_fpregs(&fpu->state);
- else
- fpregs_deactivate(fpu);
+ copy_kernel_to_fpregs(&fpu->state);
}
}
preempt_enable();
@@ -245,8 +221,7 @@ static void fpu_copy(struct fpu *dst_fpu
* Don't let 'init optimized' areas of the XSAVE area
* leak into the child task:
*/
- if (use_eager_fpu())
- memset(&dst_fpu->state.xsave, 0, xstate_size);
+ memset(&dst_fpu->state.xsave, 0, xstate_size);
/*
* Save current FPU registers directly into the child
@@ -268,10 +243,7 @@ static void fpu_copy(struct fpu *dst_fpu
if (!copy_fpregs_to_fpstate(dst_fpu)) {
memcpy(&src_fpu->state, &dst_fpu->state, xstate_size);
- if (use_eager_fpu())
- copy_kernel_to_fpregs(&src_fpu->state);
- else
- fpregs_deactivate(src_fpu);
+ copy_kernel_to_fpregs(&src_fpu->state);
}
preempt_enable();
}
@@ -437,7 +409,7 @@ void fpu__clear(struct fpu *fpu)
{
WARN_ON_FPU(fpu != ¤t->thread.fpu); /* Almost certainly an anomaly */
- if (!use_eager_fpu() || !static_cpu_has(X86_FEATURE_FPU)) {
+ if (!static_cpu_has(X86_FEATURE_FPU)) {
/* FPU state will be reallocated lazily at the first use. */
fpu__drop(fpu);
} else {
--- a/arch/x86/kernel/fpu/signal.c
+++ b/arch/x86/kernel/fpu/signal.c
@@ -319,11 +319,9 @@ static int __fpu__restore_sig(void __use
}
fpu->fpstate_active = 1;
- if (use_eager_fpu()) {
- preempt_disable();
- fpu__restore(fpu);
- preempt_enable();
- }
+ preempt_disable();
+ fpu__restore(fpu);
+ preempt_enable();
return err;
} else {
--- a/arch/x86/kvm/cpuid.c
+++ b/arch/x86/kvm/cpuid.c
@@ -16,7 +16,6 @@
#include <linux/module.h>
#include <linux/vmalloc.h>
#include <linux/uaccess.h>
-#include <asm/fpu/internal.h> /* For use_eager_fpu. Ugh! */
#include <asm/user.h>
#include <asm/fpu/xstate.h>
#include "cpuid.h"
@@ -104,8 +103,7 @@ int kvm_update_cpuid(struct kvm_vcpu *vc
if (best && (best->eax & (F(XSAVES) | F(XSAVEC))))
best->ebx = xstate_required_size(vcpu->arch.xcr0, true);
- if (use_eager_fpu())
- kvm_x86_ops->fpu_activate(vcpu);
+ kvm_x86_ops->fpu_activate(vcpu);
/*
* The existing code assumes virtual address is 48-bit in the canonical
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -7319,16 +7319,6 @@ void kvm_put_guest_fpu(struct kvm_vcpu *
copy_fpregs_to_fpstate(&vcpu->arch.guest_fpu);
__kernel_fpu_end();
++vcpu->stat.fpu_reload;
- /*
- * If using eager FPU mode, or if the guest is a frequent user
- * of the FPU, just leave the FPU active for next time.
- * Every 255 times fpu_counter rolls over to 0; a guest that uses
- * the FPU in bursts will revert to loading it on demand.
- */
- if (!use_eager_fpu()) {
- if (++vcpu->fpu_counter < 5)
- kvm_make_request(KVM_REQ_DEACTIVATE_FPU, vcpu);
- }
trace_kvm_fpu(0);
}
next prev parent reply other threads:[~2018-10-18 18:06 UTC|newest]
Thread overview: 57+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-10-18 17:54 [PATCH 4.4 00/48] 4.4.162-stable review Greg Kroah-Hartman
2018-10-18 17:54 ` [PATCH 4.4 01/48] ASoC: wm8804: Add ACPI support Greg Kroah-Hartman
2018-10-18 17:54 ` [PATCH 4.4 02/48] ASoC: sigmadsp: safeload should not have lower byte limit Greg Kroah-Hartman
2018-10-18 17:54 ` [PATCH 4.4 03/48] selftests/efivarfs: add required kernel configs Greg Kroah-Hartman
2018-10-18 17:54 ` [PATCH 4.4 04/48] mfd: omap-usb-host: Fix dts probe of children Greg Kroah-Hartman
2018-10-18 17:54 ` [PATCH 4.4 05/48] sound: enable interrupt after dma buffer initialization Greg Kroah-Hartman
2018-10-18 17:54 ` [PATCH 4.4 06/48] stmmac: fix valid numbers of unicast filter entries Greg Kroah-Hartman
2018-10-18 17:54 ` [PATCH 4.4 07/48] net: macb: disable scatter-gather for macb on sama5d3 Greg Kroah-Hartman
2018-10-18 17:54 ` [PATCH 4.4 08/48] ARM: dts: at91: add new compatibility string " Greg Kroah-Hartman
2018-10-18 17:54 ` [PATCH 4.4 09/48] drm/amdgpu: Fix SDMA HQD destroy error on gfx_v7 Greg Kroah-Hartman
2018-10-18 17:54 ` [PATCH 4.4 10/48] ext4: add corruption check in ext4_xattr_set_entry() Greg Kroah-Hartman
2018-10-18 17:54 ` [PATCH 4.4 11/48] mm/vmstat.c: fix outdated vmstat_text Greg Kroah-Hartman
2018-10-18 17:54 ` [PATCH 4.4 12/48] mach64: detect the dot clock divider correctly on sparc Greg Kroah-Hartman
2018-10-18 17:54 ` [PATCH 4.4 13/48] perf script python: Fix export-to-postgresql.py occasional failure Greg Kroah-Hartman
2018-10-18 17:54 ` [PATCH 4.4 14/48] i2c: i2c-scmi: fix for i2c_smbus_write_block_data Greg Kroah-Hartman
2018-10-18 17:54 ` [PATCH 4.4 15/48] xhci: Dont print a warning when setting link state for disabled ports Greg Kroah-Hartman
2018-10-18 17:54 ` [PATCH 4.4 16/48] jffs2: return -ERANGE when xattr buffer is too small Greg Kroah-Hartman
2018-10-18 17:54 ` [PATCH 4.4 17/48] bnxt_en: Fix TX timeout during netpoll Greg Kroah-Hartman
2018-10-18 17:54 ` [PATCH 4.4 18/48] bonding: avoid possible dead-lock Greg Kroah-Hartman
2018-10-18 17:54 ` [PATCH 4.4 19/48] ip6_tunnel: be careful when accessing the inner header Greg Kroah-Hartman
2018-10-18 17:54 ` [PATCH 4.4 20/48] ip_tunnel: " Greg Kroah-Hartman
2018-10-18 17:54 ` [PATCH 4.4 21/48] ipv4: fix use-after-free in ip_cmsg_recv_dstaddr() Greg Kroah-Hartman
2018-10-18 17:54 ` [PATCH 4.4 22/48] net: ipv4: update fnhe_pmtu when first hops MTU changes Greg Kroah-Hartman
2018-10-18 17:54 ` [PATCH 4.4 23/48] net/ipv6: Display all addresses in output of /proc/net/if_inet6 Greg Kroah-Hartman
2018-10-18 17:54 ` [PATCH 4.4 24/48] netlabel: check for IPV4MASK in addrinfo_get Greg Kroah-Hartman
2018-10-18 17:55 ` [PATCH 4.4 25/48] net/usb: cancel pending work when unbinding smsc75xx Greg Kroah-Hartman
2018-10-18 17:55 ` [PATCH 4.4 26/48] qlcnic: fix Tx descriptor corruption on 82xx devices Greg Kroah-Hartman
2018-10-18 17:55 ` [PATCH 4.4 27/48] team: Forbid enslaving team device to itself Greg Kroah-Hartman
2018-10-18 17:55 ` [PATCH 4.4 28/48] net: mvpp2: Extract the correct ethtype from the skb for tx csum offload Greg Kroah-Hartman
2018-10-18 17:55 ` [PATCH 4.4 29/48] net: systemport: Fix wake-up interrupt race during resume Greg Kroah-Hartman
2018-10-18 17:55 ` [PATCH 4.4 30/48] rtnl: limit IFLA_NUM_TX_QUEUES and IFLA_NUM_RX_QUEUES to 4096 Greg Kroah-Hartman
2018-10-18 17:55 ` [PATCH 4.4 31/48] KVM: x86: remove eager_fpu field of struct kvm_vcpu_arch Greg Kroah-Hartman
2018-10-18 17:55 ` Greg Kroah-Hartman [this message]
2018-10-18 17:55 ` [PATCH 4.4 33/48] x86/fpu: Remove struct fpu::counter Greg Kroah-Hartman
2018-10-18 17:55 ` [PATCH 4.4 34/48] x86/fpu: Finish excising eagerfpu Greg Kroah-Hartman
2018-10-18 17:55 ` [PATCH 4.4 35/48] media: af9035: prevent buffer overflow on write Greg Kroah-Hartman
2018-10-18 17:55 ` [PATCH 4.4 36/48] clocksource/drivers/ti-32k: Add CLOCK_SOURCE_SUSPEND_NONSTOP flag for non-am43 SoCs Greg Kroah-Hartman
2018-10-18 17:55 ` [PATCH 4.4 37/48] Input: atakbd - fix Atari keymap Greg Kroah-Hartman
2018-10-18 17:55 ` [PATCH 4.4 38/48] Input: atakbd - fix Atari CapsLock behaviour Greg Kroah-Hartman
2018-10-18 17:55 ` [PATCH 4.4 39/48] net/mlx4: Use cpumask_available for eq->affinity_mask Greg Kroah-Hartman
2018-10-18 17:55 ` [PATCH 4.4 40/48] RISC-V: include linux/ftrace.h in asm-prototypes.h Greg Kroah-Hartman
2018-10-18 17:55 ` [PATCH 4.4 41/48] powerpc/tm: Fix userspace r13 corruption Greg Kroah-Hartman
2018-10-18 17:55 ` [PATCH 4.4 42/48] powerpc/tm: Avoid possible userspace r1 corruption on reclaim Greg Kroah-Hartman
2018-10-18 17:55 ` [PATCH 4.4 43/48] ARC: build: Get rid of toolchain check Greg Kroah-Hartman
2018-10-18 17:55 ` [PATCH 4.4 44/48] usb: gadget: serial: fix oops when data rxd after close Greg Kroah-Hartman
2018-10-18 17:55 ` [PATCH 4.4 45/48] Drivers: hv: utils: Invoke the poll function after handshake Greg Kroah-Hartman
2018-10-18 17:55 ` [PATCH 4.4 46/48] Drivers: hv: util: Pass the channel information during the init call Greg Kroah-Hartman
2018-10-18 17:55 ` [PATCH 4.4 47/48] Drivers: hv: kvp: fix IP Failover Greg Kroah-Hartman
2018-10-18 17:55 ` [PATCH 4.4 48/48] HV: properly delay KVP packets when negotiation is in progress Greg Kroah-Hartman
2018-10-19 1:40 ` [PATCH 4.4 00/48] 4.4.162-stable review Nathan Chancellor
2018-10-19 8:40 ` Sebastian Gottschall
2018-10-19 9:15 ` Greg Kroah-Hartman
2018-10-19 13:58 ` Rafael David Tinoco
2018-10-19 19:50 ` Guenter Roeck
2018-10-19 20:47 ` Shuah Khan
2018-10-19 22:30 ` Shuah Khan
2018-10-22 13:05 ` Jon Hunter
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=20181018175429.672632710@linuxfoundation.org \
--to=gregkh@linuxfoundation.org \
--cc=bp@alien8.de \
--cc=brgerst@gmail.com \
--cc=daniel.sangorrin@toshiba.co.jp \
--cc=dave.hansen@linux.intel.com \
--cc=dvlasenk@redhat.com \
--cc=fenghua.yu@intel.com \
--cc=hpa@zytor.com \
--cc=jpoimboe@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=luto@kernel.org \
--cc=mingo@kernel.org \
--cc=oleg@redhat.com \
--cc=pbonzini@redhat.com \
--cc=peterz@infradead.org \
--cc=quentin.casasnovas@oracle.com \
--cc=riel@redhat.com \
--cc=stable@vger.kernel.org \
--cc=tglx@linutronix.de \
--cc=torvalds@linux-foundation.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;
as well as URLs for NNTP newsgroup(s).