[PATCH v6] um: Enable preemption in UML

[Peter Lafreniere]

On Fri, Sept 22, 2023 at 06:56, anton.ivanov at cambridgegreys.com wrote:
> 
> From: Anton Ivanov anton.ivanov at cambridgegreys.com
> 
> 
> 1. Preemption requires saving/restoring FPU state. This patch
> adds support for it using GCC intrinsics as well as appropriate
> storage space in the thread structure.
> 
> 2. irq critical sections need preempt_disable()/preempt_enable().
> 
> 3. TLB critical sections need preempt_disable()/preempt_enable().
> 
> 4. UML TLB flush is also invoked during a fork. This happens
> with interrupts and preempt disabled which disagrees with the
> standard mm locking via rwsem. The mm lock for this code path
> had to be replaced with an rcu.

"Had to be"? Why? It's a very unconventional locking method
and should have some justification.

> 
> 5. The FPU state area is statically allocated depending on
> the enabled PREEMPT options. PREEMPT_DYNAMIC and chosing the
> preemption model at start time is disabled for the UM arch.
> 
> Signed-off-by: Anton Ivanov anton.ivanov at cambridgegreys.com
> 
> ---
> arch/um/Kconfig | 2 +-
> arch/um/include/asm/fpu/api.h | 9 ++-
> arch/um/include/asm/processor-generic.h | 4 ++
> arch/um/kernel/Makefile | 4 ++
> arch/um/kernel/fpu.c | 75 +++++++++++++++++++++++++
> arch/um/kernel/irq.c | 2 +
> arch/um/kernel/tlb.c | 21 ++++++-
> 7 files changed, 111 insertions(+), 6 deletions(-)
> create mode 100644 arch/um/kernel/fpu.c
> 
> diff --git a/arch/um/Kconfig b/arch/um/Kconfig
> index b5e179360534..19176fde82f3 100644
> --- a/arch/um/Kconfig
> +++ b/arch/um/Kconfig
> @@ -11,7 +11,7 @@ config UML
> select ARCH_HAS_KCOV
> select ARCH_HAS_STRNCPY_FROM_USER
> select ARCH_HAS_STRNLEN_USER
> - select ARCH_NO_PREEMPT
> + select ARCH_NO_PREEMPT_DYNAMIC
> select HAVE_ARCH_AUDITSYSCALL
> select HAVE_ARCH_KASAN if X86_64
> select HAVE_ARCH_KASAN_VMALLOC if HAVE_ARCH_KASAN
> diff --git a/arch/um/include/asm/fpu/api.h b/arch/um/include/asm/fpu/api.h
> index 71bfd9ef3938..9e7680bf48f0 100644
> --- a/arch/um/include/asm/fpu/api.h
> +++ b/arch/um/include/asm/fpu/api.h
> @@ -4,12 +4,15 @@
> 
> /* Copyright (c) 2020 Cambridge Greys Ltd
> * Copyright (c) 2020 Red Hat Inc.
> - * A set of "dummy" defines to allow the direct inclusion
> - * of x86 optimized copy, xor, etc routines into the
> - * UML code tree. */
> + /
> 
> +#if defined(CONFIG_PREEMPT) || defined(CONFIG_PREEMPT_VOLUNTARY)
> +extern void kernel_fpu_begin(void);
> +extern void kernel_fpu_end(void);
> +#else
> #define kernel_fpu_begin() (void)0
> #define kernel_fpu_end() (void)0
> +#endif
> 
> static inline bool irq_fpu_usable(void)
> {
> diff --git a/arch/um/include/asm/processor-generic.h b/arch/um/include/asm/processor-generic.h
> index 7414154b8e9a..9970e70be1e4 100644
> --- a/arch/um/include/asm/processor-generic.h
> +++ b/arch/um/include/asm/processor-generic.h
> @@ -44,6 +44,10 @@ struct thread_struct {
> } cb;
> } u;
> } request;
> +#if defined(CONFIG_PREEMPT) || defined(CONFIG_PREEMPT_VOLUNTARY)
> +/ Intel docs require xsave/xrestore area to be aligned to 64 bytes /
> + u8 fpu[2048] __aligned(64);
> +#endif
> };
> 
> #define INIT_THREAD \
> diff --git a/arch/um/kernel/Makefile b/arch/um/kernel/Makefile
> index 811188be954c..c616e884a488 100644
> --- a/arch/um/kernel/Makefile
> +++ b/arch/um/kernel/Makefile
> @@ -26,9 +26,13 @@ obj-$(CONFIG_OF) += dtb.o
> obj-$(CONFIG_EARLY_PRINTK) += early_printk.o
> obj-$(CONFIG_STACKTRACE) += stacktrace.o
> obj-$(CONFIG_GENERIC_PCI_IOMAP) += ioport.o
> +obj-$(CONFIG_PREEMPT) += fpu.o
> +obj-$(CONFIG_PREEMPT_VOLUNTARY) += fpu.o
> 
> USER_OBJS := config.o
> 
> +CFLAGS_fpu.o += -mxsave -mxsaveopt
> +
> include $(srctree)/arch/um/scripts/Makefile.rules
> 
> targets := config.c config.tmp capflags.c
> diff --git a/arch/um/kernel/fpu.c b/arch/um/kernel/fpu.c
> new file mode 100644
> index 000000000000..4817276b2a26
> --- /dev/null
> +++ b/arch/um/kernel/fpu.c
> @@ -0,0 +1,75 @@
> +// SPDX-License-Identifier: GPL-2.0-only
> +/
> + * Copyright (C) 2023 Cambridge Greys Ltd
> + * Copyright (C) 2023 Red Hat Inc
> + */
> +
> +#include <linux/cpu.h>
> 
> +#include <linux/init.h>
> 
> +#include <asm/fpu/api.h>
> 
> +#include <asm/cpufeature.h>
> 
> +
> +/*
> + * The critical section between kernel_fpu_begin() and kernel_fpu_end()
> + * is non-reentrant. It is the caller's responsibility to avoid reentrance.
> + /
> +
> +static DEFINE_PER_CPU(bool, in_kernel_fpu);
> +
> +/ UML and driver code it pulls out of the x86 tree knows about 387 features
> + * up to and including AVX512. TILE, etc are not yet supported.
> + */
> +
> +#define KNOWN_387_FEATURES 0xFF
> +
> +void kernel_fpu_begin(void)
> +{
> + preempt_disable();
> +
> + WARN_ON(this_cpu_read(in_kernel_fpu));
> +
> + this_cpu_write(in_kernel_fpu, true);
> +
> +#ifdef CONFIG_64BIT
> + if (likely(cpu_has(&boot_cpu_data, X86_FEATURE_XSAVEOPT)))
> + __builtin_ia32_xsaveopt64(&current->thread.fpu, KNOWN_387_FEATURES);
> 
> + else {
> + if (likely(cpu_has(&boot_cpu_data, X86_FEATURE_XSAVE)))
> + __builtin_ia32_xsave64(&current->thread.fpu, KNOWN_387_FEATURES);
> 
> + else
> + __builtin_ia32_fxsave64(&current->thread.fpu);
> 
> + }
> +#else
> + if (likely(cpu_has(&boot_cpu_data, X86_FEATURE_XSAVEOPT)))
> + __builtin_ia32_xsaveopt(&current->thread.fpu, KNOWN_387_FEATURES);
> 
> + else {
> + if (likely(cpu_has(&boot_cpu_data, X86_FEATURE_XSAVE)))
> + __builtin_ia32_xsave(&current->thread.fpu, KNOWN_387_FEATURES);
> 
> + else
> + __builtin_ia32_fxsave(&current->thread.fpu);
> 
> + }
> +#endif
> +}
> +EXPORT_SYMBOL_GPL(kernel_fpu_begin);
> +

Johannes is right about how flattening the else branch makes
it a bit more readable, imo.

> +void kernel_fpu_end(void)
> +{
> + WARN_ON(!this_cpu_read(in_kernel_fpu));
> +
> +#ifdef CONFIG_64BIT
> + if (likely(cpu_has(&boot_cpu_data, X86_FEATURE_XSAVE)))
> + __builtin_ia32_xrstor64(&current->thread.fpu, KNOWN_387_FEATURES);
> 
> + else
> + __builtin_ia32_fxrstor64(&current->thread.fpu);
> 
> +#else
> + if (likely(cpu_has(&boot_cpu_data, X86_FEATURE_XSAVE)))
> + __builtin_ia32_xrstor(&current->thread.fpu, KNOWN_387_FEATURES);
> 
> + else
> + __builtin_ia32_fxrstor(&current->thread.fpu);
> 
> +#endif
> + this_cpu_write(in_kernel_fpu, false);
> +
> + preempt_enable();
> +}
> +EXPORT_SYMBOL_GPL(kernel_fpu_end);

--->8---

> @@ -597,8 +609,13 @@ void force_flush_all(void)
> struct vm_area_struct vma;
> VMA_ITERATOR(vmi, mm, 0);
> 
> - mmap_read_lock(mm);
> + / We use a RCU lock instead of a mm lock, because
> + * this can be invoked out of critical/atomic sections
> + * and that does not agree with the sleepable semantics
> + * of the standard semaphore based mm lock.
> + */

As Johannes said, this comment doesn't explain why it's needed,
just that an rcu lock is needed.

> + rcu_read_lock();
> for_each_vma(vmi, vma)
> fix_range(mm, vma->vm_start, vma->vm_end, 1);
> 
> - mmap_read_unlock(mm);
> + rcu_read_unlock();
> }
> --
> 2.30.2

I didn't audit the assortment of preempt_{enable,disable}() pairs,
but I did test this patch with defconfig + debug.config + CONFIG_PREEMPT=y
for 15 minutes and I didn't see any warnings that used to show up.

I think that we should fix what currently exists, then rebase this
PREEMPT patch on top of the results. Plus that'll give me time to look
into the avx + !xsave edge case :)

Cheers,
Peter