From: Nadav Amit <nadav.amit@gmail.com>
To: Masami Hiramatsu <mhiramat@kernel.org>,
Andy Lutomirski <luto@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>,
Kees Cook <keescook@chromium.org>,
Linus Torvalds <torvalds@linux-foundation.org>,
Paolo Bonzini <pbonzini@redhat.com>,
Jiri Kosina <jkosina@suse.cz>, Will Deacon <will.deacon@arm.com>,
Benjamin Herrenschmidt <benh@au1.ibm.com>,
Nick Piggin <npiggin@gmail.com>,
the arch/x86 maintainers <x86@kernel.org>,
Borislav Petkov <bp@alien8.de>, Rik van Riel <riel@surriel.com>,
Jann Horn <jannh@google.com>,
Adin Scannell <ascannell@google.com>,
Dave Hansen <dave.hansen@intel.com>,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
linux-mm <linux-mm@kvack.org>, David Miller <davem@davemloft.net>,
Martin Schwidefsky <schwidefsky@de.ibm.com>,
Michael Ellerman <mpe@ellerman.id.au>
Subject: Re: TLB flushes on fixmap changes
Date: Tue, 28 Aug 2018 10:33:02 -0700 [thread overview]
Message-ID: <7F3235AF-5300-49B6-A9A4-C8BA8EF1D87B@gmail.com> (raw)
In-Reply-To: <20180828174948.dc0474009be1ec7a0392cfcd@kernel.org>
at 1:49 AM, Masami Hiramatsu <mhiramat@kernel.org> wrote:
> On Mon, 27 Aug 2018 16:01:32 -0700
> Andy Lutomirski <luto@kernel.org> wrote:
>
>> On Mon, Aug 27, 2018 at 3:54 PM, Nadav Amit <nadav.amit@gmail.com> wrote:
>>> at 3:32 PM, Andy Lutomirski <luto@kernel.org> wrote:
>>>
>>>> On Mon, Aug 27, 2018 at 2:55 PM, Nadav Amit <nadav.amit@gmail.com> wrote:
>>>>> at 1:16 PM, Nadav Amit <nadav.amit@gmail.com> wrote:
>>>>>
>>>>>> at 12:58 PM, Andy Lutomirski <luto@kernel.org> wrote:
>>>>>>
>>>>>>> On Mon, Aug 27, 2018 at 12:43 PM, Nadav Amit <nadav.amit@gmail.com> wrote:
>>>>>>>> at 12:10 PM, Nadav Amit <nadav.amit@gmail.com> wrote:
>>>>>>>>
>>>>>>>>> at 11:58 AM, Andy Lutomirski <luto@kernel.org> wrote:
>>>>>>>>>
>>>>>>>>>> On Mon, Aug 27, 2018 at 11:54 AM, Nadav Amit <nadav.amit@gmail.com> wrote:
>>>>>>>>>>>> On Mon, Aug 27, 2018 at 10:34 AM, Nadav Amit <nadav.amit@gmail.com> wrote:
>>>>>>>>>>>> What do you all think?
>>>>>>>>>>>
>>>>>>>>>>> I agree in general. But I think that current->mm would need to be loaded, as
>>>>>>>>>>> otherwise I am afraid it would break switch_mm_irqs_off().
>>>>>>>>>>
>>>>>>>>>> What breaks?
>>>>>>>>>
>>>>>>>>> Actually nothing. I just saw the IBPB stuff regarding tsk, but it should not
>>>>>>>>> matter.
>>>>>>>>
>>>>>>>> So here is what I got. It certainly needs some cleanup, but it boots.
>>>>>>>>
>>>>>>>> Let me know how crappy you find it...
>>>>>>>>
>>>>>>>>
>>>>>>>> diff --git a/arch/x86/include/asm/mmu_context.h b/arch/x86/include/asm/mmu_context.h
>>>>>>>> index bbc796eb0a3b..336779650a41 100644
>>>>>>>> --- a/arch/x86/include/asm/mmu_context.h
>>>>>>>> +++ b/arch/x86/include/asm/mmu_context.h
>>>>>>>> @@ -343,4 +343,24 @@ static inline unsigned long __get_current_cr3_fast(void)
>>>>>>>> return cr3;
>>>>>>>> }
>>>>>>>>
>>>>>>>> +typedef struct {
>>>>>>>> + struct mm_struct *prev;
>>>>>>>> +} temporary_mm_state_t;
>>>>>>>> +
>>>>>>>> +static inline temporary_mm_state_t use_temporary_mm(struct mm_struct *mm)
>>>>>>>> +{
>>>>>>>> + temporary_mm_state_t state;
>>>>>>>> +
>>>>>>>> + lockdep_assert_irqs_disabled();
>>>>>>>> + state.prev = this_cpu_read(cpu_tlbstate.loaded_mm);
>>>>>>>> + switch_mm_irqs_off(NULL, mm, current);
>>>>>>>> + return state;
>>>>>>>> +}
>>>>>>>> +
>>>>>>>> +static inline void unuse_temporary_mm(temporary_mm_state_t prev)
>>>>>>>> +{
>>>>>>>> + lockdep_assert_irqs_disabled();
>>>>>>>> + switch_mm_irqs_off(NULL, prev.prev, current);
>>>>>>>> +}
>>>>>>>> +
>>>>>>>> #endif /* _ASM_X86_MMU_CONTEXT_H */
>>>>>>>> diff --git a/arch/x86/include/asm/pgtable.h b/arch/x86/include/asm/pgtable.h
>>>>>>>> index 5715647fc4fe..ef62af9a0ef7 100644
>>>>>>>> --- a/arch/x86/include/asm/pgtable.h
>>>>>>>> +++ b/arch/x86/include/asm/pgtable.h
>>>>>>>> @@ -976,6 +976,10 @@ static inline void __meminit init_trampoline_default(void)
>>>>>>>> /* Default trampoline pgd value */
>>>>>>>> trampoline_pgd_entry = init_top_pgt[pgd_index(__PAGE_OFFSET)];
>>>>>>>> }
>>>>>>>> +
>>>>>>>> +void __init patching_mm_init(void);
>>>>>>>> +#define patching_mm_init patching_mm_init
>>>>>>>> +
>>>>>>>> # ifdef CONFIG_RANDOMIZE_MEMORY
>>>>>>>> void __meminit init_trampoline(void);
>>>>>>>> # else
>>>>>>>> diff --git a/arch/x86/include/asm/pgtable_64_types.h b/arch/x86/include/asm/pgtable_64_types.h
>>>>>>>> index 054765ab2da2..9f44262abde0 100644
>>>>>>>> --- a/arch/x86/include/asm/pgtable_64_types.h
>>>>>>>> +++ b/arch/x86/include/asm/pgtable_64_types.h
>>>>>>>> @@ -116,6 +116,9 @@ extern unsigned int ptrs_per_p4d;
>>>>>>>> #define LDT_PGD_ENTRY (pgtable_l5_enabled() ? LDT_PGD_ENTRY_L5 : LDT_PGD_ENTRY_L4)
>>>>>>>> #define LDT_BASE_ADDR (LDT_PGD_ENTRY << PGDIR_SHIFT)
>>>>>>>>
>>>>>>>> +#define TEXT_POKE_PGD_ENTRY -5UL
>>>>>>>> +#define TEXT_POKE_ADDR (TEXT_POKE_PGD_ENTRY << PGDIR_SHIFT)
>>>>>>>> +
>>>>>>>> #define __VMALLOC_BASE_L4 0xffffc90000000000UL
>>>>>>>> #define __VMALLOC_BASE_L5 0xffa0000000000000UL
>>>>>>>>
>>>>>>>> diff --git a/arch/x86/include/asm/pgtable_types.h b/arch/x86/include/asm/pgtable_types.h
>>>>>>>> index 99fff853c944..840c72ec8c4f 100644
>>>>>>>> --- a/arch/x86/include/asm/pgtable_types.h
>>>>>>>> +++ b/arch/x86/include/asm/pgtable_types.h
>>>>>>>> @@ -505,6 +505,9 @@ pgprot_t phys_mem_access_prot(struct file *file, unsigned long pfn,
>>>>>>>> /* Install a pte for a particular vaddr in kernel space. */
>>>>>>>> void set_pte_vaddr(unsigned long vaddr, pte_t pte);
>>>>>>>>
>>>>>>>> +struct mm_struct;
>>>>>>>> +void set_mm_pte_vaddr(struct mm_struct *mm, unsigned long vaddr, pte_t pte);
>>>>>>>> +
>>>>>>>> #ifdef CONFIG_X86_32
>>>>>>>> extern void native_pagetable_init(void);
>>>>>>>> #else
>>>>>>>> diff --git a/arch/x86/include/asm/text-patching.h b/arch/x86/include/asm/text-patching.h
>>>>>>>> index 2ecd34e2d46c..cb364ea5b19d 100644
>>>>>>>> --- a/arch/x86/include/asm/text-patching.h
>>>>>>>> +++ b/arch/x86/include/asm/text-patching.h
>>>>>>>> @@ -38,4 +38,6 @@ extern void *text_poke(void *addr, const void *opcode, size_t len);
>>>>>>>> extern int poke_int3_handler(struct pt_regs *regs);
>>>>>>>> extern void *text_poke_bp(void *addr, const void *opcode, size_t len, void *handler);
>>>>>>>>
>>>>>>>> +extern struct mm_struct *patching_mm;
>>>>>>>> +
>>>>>>>> #endif /* _ASM_X86_TEXT_PATCHING_H */
>>>>>>>> diff --git a/arch/x86/kernel/alternative.c b/arch/x86/kernel/alternative.c
>>>>>>>> index a481763a3776..fd8a950b0d62 100644
>>>>>>>> --- a/arch/x86/kernel/alternative.c
>>>>>>>> +++ b/arch/x86/kernel/alternative.c
>>>>>>>> @@ -11,6 +11,7 @@
>>>>>>>> #include <linux/stop_machine.h>
>>>>>>>> #include <linux/slab.h>
>>>>>>>> #include <linux/kdebug.h>
>>>>>>>> +#include <linux/mmu_context.h>
>>>>>>>> #include <asm/text-patching.h>
>>>>>>>> #include <asm/alternative.h>
>>>>>>>> #include <asm/sections.h>
>>>>>>>> @@ -701,8 +702,36 @@ void *text_poke(void *addr, const void *opcode, size_t len)
>>>>>>>> WARN_ON(!PageReserved(pages[0]));
>>>>>>>> pages[1] = virt_to_page(addr + PAGE_SIZE);
>>>>>>>> }
>>>>>>>> - BUG_ON(!pages[0]);
>>>>>>>> +
>>>>>>>> local_irq_save(flags);
>>>>>>>> + BUG_ON(!pages[0]);
>>>>>>>> +
>>>>>>>> + /*
>>>>>>>> + * During initial boot, it is hard to initialize patching_mm due to
>>>>>>>> + * dependencies in boot order.
>>>>>>>> + */
>>>>>>>> + if (patching_mm) {
>>>>>>>> + pte_t pte;
>>>>>>>> + temporary_mm_state_t prev;
>>>>>>>> +
>>>>>>>> + prev = use_temporary_mm(patching_mm);
>>>>>>>> + pte = mk_pte(pages[0], PAGE_KERNEL);
>>>>>>>> + set_mm_pte_vaddr(patching_mm, TEXT_POKE_ADDR, pte);
>>>>>>>> + pte = mk_pte(pages[1], PAGE_KERNEL);
>>>>>>>> + set_mm_pte_vaddr(patching_mm, TEXT_POKE_ADDR + PAGE_SIZE, pte);
>>>>>>>> +
>>>>>>>> + memcpy((void *)(TEXT_POKE_ADDR | ((unsigned long)addr & ~PAGE_MASK)),
>>>>>>>> + opcode, len);
>>>>>>>> +
>>>>>>>> + set_mm_pte_vaddr(patching_mm, TEXT_POKE_ADDR, __pte(0));
>>>>>>>> + set_mm_pte_vaddr(patching_mm, TEXT_POKE_ADDR + PAGE_SIZE, __pte(0));
>>>>>>>> + local_flush_tlb();
>>>>>>>
>>>>>>> Hmm. This is stuff busted on SMP, and it's IMO more complicated than
>>>>>>> needed. How about getting rid of all the weird TLB flushing stuff and
>>>>>>> instead putting the mapping at vaddr - __START_KERNEL_map or whatever
>>>>>>> it is? You *might* need to flush_tlb_mm_range() on module unload, but
>>>>>>> that's it.
>>>>>>
>>>>>> I don’t see what’s wrong in SMP, since this entire piece of code should be
>>>>>> running under text_mutex.
>>>>>>
>>>>>> I don’t quite understand your proposal. I really don’t want to have any
>>>>>> chance in which the page-tables for the poked address is not preallocated.
>>>>>>
>>>>>> It is more complicated than needed, and there are redundant TLB flushes. The
>>>>>> reason I preferred to do it this way, is in order not to use other functions
>>>>>> that take locks during the software page-walk and not to duplicate existing
>>>>>> code. Yet, duplication might be the way to go.
>>>>>>
>>>>>>>> + sync_core();
>>>>>>>
>>>>>>> I can't think of any case where sync_core() is needed. The mm switch
>>>>>>> serializes.
>>>>>>
>>>>>> Good point!
>>>>>>
>>>>>>> Also, is there any circumstance in which any of this is used before at
>>>>>>> least jump table init? All the early stuff is text_poke_early(),
>>>>>>> right?
>>>>>>
>>>>>> Not before jump_label_init. However, I did not manage to get rid of the two
>>>>>> code-patches in text_poke(), since text_poke is used relatively early by
>>>>>> x86_late_time_init(), and at this stage kmem_cache_alloc() - which is needed
>>>>>> to duplicate init_mm - still fails.
>>>>>
>>>>> Another correction: the populate_extra_pte() is not needed.
>>>>>
>>>>> Anyhow, if you want to do this whole thing differently, I obviously will not
>>>>> object, but I think it will end up more complicated.
>>>>>
>>>>> I think I finally understood your comment about "vaddr -
>>>>> __START_KERNEL_map”. I did something like that before, and it is not
>>>>> super-simple. You need not only to conditionally flush the TLB, but also
>>>>> to synchronize the PUD/PMD on changes. Don’t forget that module memory
>>>>> is installed even when BPF programs are installed.
>>>>>
>>>>> Let me know if you want me to submit cleaner patches or you want to carry on
>>>>> yourself.
>>>>
>>>> I think your approach is a good start and should be good enough (with
>>>> cleanups) as a fix for the bug. But I think your code has the same
>>>> bug that we have now! You're reusing the same address on multiple
>>>> CPUs without flushing. You can easily fix it by forcing a flush
>>>> before loading the mm, which should be as simple as adding
>>>> flush_tlb_mm() before you load the mm. (It won't actually flush
>>>> anything by itself, since the mm isn't loaded, but it will update the
>>>> bookkeeping so that switch_mm_irqs_off() flushes the mm.)
>>>
>>> What am I missing?
>>>
>>> We have a lock (text_mutex) which prevents the use of the new page-table
>>> hierarchy on multiple CPUs. In addition we have __set_pte_vaddr() which does
>>> a local TLB flush before the lock is released and before IRQs are enabled.
>>> So how can the PTE be cached on multiple CPUs?
>>>
>>> Yes, __set_pte_vaddr() is ugly and flushes too much. I’ll try to remove some
>>> redundant TLB flushes, but these flushes are already there.
>>
>> I missed that set_pte_vaddr() contained a flush. It's probably nicer
>> to use one of the APIs that doesn't imply a flush. If you end up with
>> vm_insert_pfn() or similar, you can use can use copy_to_user() if
>> needed :)
>>
>>>> Also, please at least get rid of TEXT_POKE_ADDR. If you don't want to
>>>> do the vaddr - __START_KERNEL_map thing, then at least pick an address
>>>> in the low half of the address space such as 0 :) Ideally you'd only
>>>> use this thing late enough that you could even use the normal
>>>> insert_pfn (or similar) API for it, but that doesn't really matter.
>>>
>>> Perhaps the vaddr - __START_KERNEL_map actually makes sense. I was
>>> misunderstanding it (again) before, thinking you want me to use vaddr (and
>>> not the delta). I’ll give it a try.
>>
>> It does prevent easy preallocation of the intermediate tables.
>
> Would we have to change the mapping address? If so, how about using hash
> of vaddr, which allow us to map it in one pte?
We do need two PTEs, and I will reuse the same two PTEs in order not to
waste memory on page-tables. I will try to randomize their address to be on
the safe side.
next prev parent reply other threads:[~2018-08-28 17:33 UTC|newest]
Thread overview: 96+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-08-22 15:30 [PATCH 0/4] x86: TLB invalidate fixes Peter Zijlstra
2018-08-22 15:30 ` [PATCH 1/4] x86/mm/tlb: Revert the recent lazy TLB patches Peter Zijlstra
2018-08-22 21:37 ` Linus Torvalds
2018-08-22 22:11 ` Rik van Riel
2018-08-22 15:30 ` [PATCH 2/4] mm/tlb: Remove tlb_remove_table() non-concurrent condition Peter Zijlstra
2018-08-23 3:31 ` Nicholas Piggin
2018-08-23 3:35 ` Linus Torvalds
2018-08-23 3:44 ` Linus Torvalds
2018-08-23 4:16 ` Nicholas Piggin
2018-08-23 4:54 ` Linus Torvalds
2018-08-23 5:15 ` Nicholas Piggin
2018-08-24 8:42 ` Peter Zijlstra
2018-08-23 13:40 ` Will Deacon
2018-08-22 15:30 ` [PATCH 3/4] mm/tlb, x86/mm: Support invalidating TLB caches for RCU_TABLE_FREE Peter Zijlstra
2018-08-22 15:55 ` Peter Zijlstra
2018-08-23 3:45 ` Nicholas Piggin
2018-08-23 3:59 ` Linus Torvalds
2018-08-23 4:33 ` Nicholas Piggin
2018-08-23 5:03 ` Linus Torvalds
2018-08-23 5:58 ` Nicholas Piggin
2018-08-23 4:54 ` Benjamin Herrenschmidt
2018-08-23 5:11 ` Linus Torvalds
2018-08-23 5:20 ` Linus Torvalds
2018-08-23 6:48 ` Martin Schwidefsky
2018-08-23 5:21 ` Benjamin Herrenschmidt
2018-08-23 6:15 ` Nicholas Piggin
2018-08-23 13:39 ` Will Deacon
2018-08-24 8:47 ` Peter Zijlstra
2018-08-24 11:32 ` Peter Zijlstra
2018-08-24 11:39 ` Peter Zijlstra
2018-08-27 5:00 ` Nicholas Piggin
2018-08-27 7:47 ` Peter Zijlstra
2018-08-27 8:04 ` Nicholas Piggin
2018-08-27 8:09 ` Benjamin Herrenschmidt
2018-08-27 8:20 ` Peter Zijlstra
2018-08-27 8:54 ` Nicholas Piggin
2018-08-27 9:02 ` Nicholas Piggin
2018-08-27 22:13 ` Benjamin Herrenschmidt
2018-08-27 13:36 ` Rik van Riel
2018-08-27 14:29 ` Nicholas Piggin
2018-08-27 8:57 ` removig ia64, was: " Christoph Hellwig
2018-08-27 11:28 ` Peter Zijlstra
2018-08-27 11:45 ` Jason Duerstock
2018-08-27 11:00 ` Peter Zijlstra
2018-08-30 0:13 ` Vineet Gupta
2018-08-30 10:23 ` Peter Zijlstra
2018-08-24 17:26 ` Nadav Amit
2018-08-24 18:04 ` Peter Zijlstra
2018-08-24 18:35 ` TLB flushes on fixmap changes Nadav Amit
2018-08-24 19:31 ` Linus Torvalds
2018-08-24 20:24 ` Nadav Amit
2018-08-25 0:58 ` Linus Torvalds
2018-08-25 2:29 ` nadav.amit
2018-08-25 4:23 ` Andy Lutomirski
2018-08-26 2:23 ` Masami Hiramatsu
2018-08-26 4:21 ` Andy Lutomirski
2018-08-26 4:43 ` Kees Cook
2018-08-26 5:53 ` Nadav Amit
2018-08-26 14:20 ` Andy Lutomirski
2018-08-26 16:47 ` Kees Cook
2018-08-26 17:25 ` Andy Lutomirski
2018-08-26 20:15 ` Thomas Gleixner
2018-08-26 22:03 ` Kees Cook
2018-08-26 22:15 ` Matthew Wilcox
2018-08-26 22:29 ` Jann Horn
2018-08-26 9:09 ` Peter Zijlstra
2018-08-27 3:03 ` Masami Hiramatsu
2018-08-27 3:26 ` Nadav Amit
2018-08-27 8:05 ` Masami Hiramatsu
2018-08-27 17:34 ` Nadav Amit
2018-08-27 18:45 ` Andy Lutomirski
2018-08-27 18:54 ` Nadav Amit
2018-08-27 18:58 ` Andy Lutomirski
2018-08-27 19:10 ` Nadav Amit
2018-08-27 19:43 ` Nadav Amit
2018-08-27 19:58 ` Andy Lutomirski
2018-08-27 20:16 ` Nadav Amit
2018-08-27 21:55 ` Nadav Amit
2018-08-27 22:32 ` Andy Lutomirski
2018-08-27 22:54 ` Nadav Amit
2018-08-27 23:01 ` Andy Lutomirski
2018-08-28 8:49 ` Masami Hiramatsu
2018-08-28 17:33 ` Nadav Amit [this message]
2018-08-27 8:13 ` Peter Zijlstra
2018-08-27 9:39 ` Masami Hiramatsu
2018-08-27 9:55 ` Jann Horn
2018-08-26 22:48 ` Jann Horn
2018-08-24 8:35 ` [PATCH 3/4] mm/tlb, x86/mm: Support invalidating TLB caches for RCU_TABLE_FREE Peter Zijlstra
2018-08-24 13:13 ` Peter Zijlstra
2018-08-24 13:14 ` Peter Zijlstra
2018-08-24 15:49 ` Will Deacon
2018-08-23 23:31 ` Will Deacon
2018-08-22 21:34 ` Linus Torvalds
2018-08-23 8:46 ` Nicholas Piggin
2018-08-22 15:30 ` [PATCH 4/4] x86/mm: Only use tlb_remove_table() for paravirt Peter Zijlstra
2018-08-22 22:12 ` Eduardo Valentin
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=7F3235AF-5300-49B6-A9A4-C8BA8EF1D87B@gmail.com \
--to=nadav.amit@gmail.com \
--cc=ascannell@google.com \
--cc=benh@au1.ibm.com \
--cc=bp@alien8.de \
--cc=dave.hansen@intel.com \
--cc=davem@davemloft.net \
--cc=jannh@google.com \
--cc=jkosina@suse.cz \
--cc=keescook@chromium.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=luto@kernel.org \
--cc=mhiramat@kernel.org \
--cc=mpe@ellerman.id.au \
--cc=npiggin@gmail.com \
--cc=pbonzini@redhat.com \
--cc=peterz@infradead.org \
--cc=riel@surriel.com \
--cc=schwidefsky@de.ibm.com \
--cc=torvalds@linux-foundation.org \
--cc=will.deacon@arm.com \
--cc=x86@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;
as well as URLs for NNTP newsgroup(s).