Received-SPF: pass (google.com: domain of linux-kernel+bounces-157188-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:40f1:3f00::1 as permitted sender) client-ip=2604:1380:40f1:3f00::1;
Message-ID: <f972ce5a-0351-450c-98a2-38188eae5001@redhat.com>
Date: Wed, 24 Apr 2024 17:26:39 +0200
Precedence: bulk
MIME-Version: 1.0
User-Agent: Mozilla Thunderbird
Subject: Re: [PATCH v21 2/5] ring-buffer: Introducing ring-buffer mapping
 functions
To: Vincent Donnefort <vdonnefort@google.com>, rostedt@goodmis.org,
 mhiramat@kernel.org, linux-kernel@vger.kernel.org,
 linux-trace-kernel@vger.kernel.org
Cc: mathieu.desnoyers@efficios.com, kernel-team@android.com,
 rdunlap@infradead.org, rppt@kernel.org, linux-mm@kvack.org
References: <20240423232728.1492340-1-vdonnefort@google.com>
 <20240423232728.1492340-3-vdonnefort@google.com>
From: David Hildenbrand <david@redhat.com>
Content-Language: en-US
Autocrypt: addr=david@redhat.com; keydata=
 xsFNBFXLn5EBEAC+zYvAFJxCBY9Tr1xZgcESmxVNI/0ffzE/ZQOiHJl6mGkmA1R7/uUpiCjJ
 dBrn+lhhOYjjNefFQou6478faXE6o2AhmebqT4KiQoUQFV4R7y1KMEKoSyy8hQaK1umALTdL
 QZLQMzNE74ap+GDK0wnacPQFpcG1AE9RMq3aeErY5tujekBS32jfC/7AnH7I0v1v1TbbK3Gp
 XNeiN4QroO+5qaSr0ID2sz5jtBLRb15RMre27E1ImpaIv2Jw8NJgW0k/D1RyKCwaTsgRdwuK
 Kx/Y91XuSBdz0uOyU/S8kM1+ag0wvsGlpBVxRR/xw/E8M7TEwuCZQArqqTCmkG6HGcXFT0V9
 PXFNNgV5jXMQRwU0O/ztJIQqsE5LsUomE//bLwzj9IVsaQpKDqW6TAPjcdBDPLHvriq7kGjt
 WhVhdl0qEYB8lkBEU7V2Yb+SYhmhpDrti9Fq1EsmhiHSkxJcGREoMK/63r9WLZYI3+4W2rAc
 UucZa4OT27U5ZISjNg3Ev0rxU5UH2/pT4wJCfxwocmqaRr6UYmrtZmND89X0KigoFD/XSeVv
 jwBRNjPAubK9/k5NoRrYqztM9W6sJqrH8+UWZ1Idd/DdmogJh0gNC0+N42Za9yBRURfIdKSb
 B3JfpUqcWwE7vUaYrHG1nw54pLUoPG6sAA7Mehl3nd4pZUALHwARAQABzSREYXZpZCBIaWxk
 ZW5icmFuZCA8ZGF2aWRAcmVkaGF0LmNvbT7CwZgEEwEIAEICGwMGCwkIBwMCBhUIAgkKCwQW
 AgMBAh4BAheAAhkBFiEEG9nKrXNcTDpGDfzKTd4Q9wD/g1oFAl8Ox4kFCRKpKXgACgkQTd4Q
 9wD/g1oHcA//a6Tj7SBNjFNM1iNhWUo1lxAja0lpSodSnB2g4FCZ4R61SBR4l/psBL73xktp
 rDHrx4aSpwkRP6Epu6mLvhlfjmkRG4OynJ5HG1gfv7RJJfnUdUM1z5kdS8JBrOhMJS2c/gPf
 wv1TGRq2XdMPnfY2o0CxRqpcLkx4vBODvJGl2mQyJF/gPepdDfcT8/PY9BJ7FL6Hrq1gnAo4
 3Iv9qV0JiT2wmZciNyYQhmA1V6dyTRiQ4YAc31zOo2IM+xisPzeSHgw3ONY/XhYvfZ9r7W1l
 pNQdc2G+o4Di9NPFHQQhDw3YTRR1opJaTlRDzxYxzU6ZnUUBghxt9cwUWTpfCktkMZiPSDGd
 KgQBjnweV2jw9UOTxjb4LXqDjmSNkjDdQUOU69jGMUXgihvo4zhYcMX8F5gWdRtMR7DzW/YE
 BgVcyxNkMIXoY1aYj6npHYiNQesQlqjU6azjbH70/SXKM5tNRplgW8TNprMDuntdvV9wNkFs
 9TyM02V5aWxFfI42+aivc4KEw69SE9KXwC7FSf5wXzuTot97N9Phj/Z3+jx443jo2NR34XgF
 89cct7wJMjOF7bBefo0fPPZQuIma0Zym71cP61OP/i11ahNye6HGKfxGCOcs5wW9kRQEk8P9
 M/k2wt3mt/fCQnuP/mWutNPt95w9wSsUyATLmtNrwccz63XOwU0EVcufkQEQAOfX3n0g0fZz
 Bgm/S2zF/kxQKCEKP8ID+Vz8sy2GpDvveBq4H2Y34XWsT1zLJdvqPI4af4ZSMxuerWjXbVWb
 T6d4odQIG0fKx4F8NccDqbgHeZRNajXeeJ3R7gAzvWvQNLz4piHrO/B4tf8svmRBL0ZB5P5A
 2uhdwLU3NZuK22zpNn4is87BPWF8HhY0L5fafgDMOqnf4guJVJPYNPhUFzXUbPqOKOkL8ojk
 CXxkOFHAbjstSK5Ca3fKquY3rdX3DNo+EL7FvAiw1mUtS+5GeYE+RMnDCsVFm/C7kY8c2d0G
 NWkB9pJM5+mnIoFNxy7YBcldYATVeOHoY4LyaUWNnAvFYWp08dHWfZo9WCiJMuTfgtH9tc75
 7QanMVdPt6fDK8UUXIBLQ2TWr/sQKE9xtFuEmoQGlE1l6bGaDnnMLcYu+Asp3kDT0w4zYGsx
 5r6XQVRH4+5N6eHZiaeYtFOujp5n+pjBaQK7wUUjDilPQ5QMzIuCL4YjVoylWiBNknvQWBXS
 lQCWmavOT9sttGQXdPCC5ynI+1ymZC1ORZKANLnRAb0NH/UCzcsstw2TAkFnMEbo9Zu9w7Kv
 AxBQXWeXhJI9XQssfrf4Gusdqx8nPEpfOqCtbbwJMATbHyqLt7/oz/5deGuwxgb65pWIzufa
 N7eop7uh+6bezi+rugUI+w6DABEBAAHCwXwEGAEIACYCGwwWIQQb2cqtc1xMOkYN/MpN3hD3
 AP+DWgUCXw7HsgUJEqkpoQAKCRBN3hD3AP+DWrrpD/4qS3dyVRxDcDHIlmguXjC1Q5tZTwNB
 boaBTPHSy/Nksu0eY7x6HfQJ3xajVH32Ms6t1trDQmPx2iP5+7iDsb7OKAb5eOS8h+BEBDeq
 3ecsQDv0fFJOA9ag5O3LLNk+3x3q7e0uo06XMaY7UHS341ozXUUI7wC7iKfoUTv03iO9El5f
 XpNMx/YrIMduZ2+nd9Di7o5+KIwlb2mAB9sTNHdMrXesX8eBL6T9b+MZJk+mZuPxKNVfEQMQ
 a5SxUEADIPQTPNvBewdeI80yeOCrN+Zzwy/Mrx9EPeu59Y5vSJOx/z6OUImD/GhX7Xvkt3kq
 Er5KTrJz3++B6SH9pum9PuoE/k+nntJkNMmQpR4MCBaV/J9gIOPGodDKnjdng+mXliF3Ptu6
 3oxc2RCyGzTlxyMwuc2U5Q7KtUNTdDe8T0uE+9b8BLMVQDDfJjqY0VVqSUwImzTDLX9S4g/8
 kC4HRcclk8hpyhY2jKGluZO0awwTIMgVEzmTyBphDg/Gx7dZU1Xf8HFuE+UZ5UDHDTnwgv7E
 th6RC9+WrhDNspZ9fJjKWRbveQgUFCpe1sa77LAw+XFrKmBHXp9ZVIe90RMe2tRL06BGiRZr
 jPrnvUsUUsjRoRNJjKKA/REq+sAnhkNPPZ/NNMjaZ5b8Tovi8C0tmxiCHaQYqj7G2rgnT0kt
 WNyWQQ==
Organization: Red Hat
In-Reply-To: <20240423232728.1492340-3-vdonnefort@google.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit


I gave it some more thought, and I think we are still missing something 
(I wish PFNMAP/MIXEDMAP wouldn't be that hard).

> +
> +/*
> + *   +--------------+  pgoff == 0
> + *   |   meta page  |
> + *   +--------------+  pgoff == 1
> + *   | subbuffer 0  |
> + *   |              |
> + *   +--------------+  pgoff == (1 + (1 << subbuf_order))
> + *   | subbuffer 1  |
> + *   |              |
> + *         ...
> + */
> +#ifdef CONFIG_MMU
> +static int __rb_map_vma(struct ring_buffer_per_cpu *cpu_buffer,
> +			struct vm_area_struct *vma)
> +{
> +	unsigned long nr_subbufs, nr_pages, vma_pages, pgoff = vma->vm_pgoff;
> +	unsigned int subbuf_pages, subbuf_order;
> +	struct page **pages;
> +	int p = 0, s = 0;
> +	int err;
> +

I'd add some comments here like

/* Refuse any MAP_PRIVATE or writable mappings. */
> +	if (vma->vm_flags & VM_WRITE || vma->vm_flags & VM_EXEC ||
> +	    !(vma->vm_flags & VM_MAYSHARE))
> +		return -EPERM;
> +

/*
  * Make sure the mapping cannot become writable later. Also, tell the VM
  * to not touch these pages pages (VM_DONTCOPY | VM_DONTDUMP) and tell
  * GUP to leave them alone as well (VM_IO).
  */
> +	vm_flags_mod(vma,
> +		     VM_MIXEDMAP | VM_PFNMAP |
> +		     VM_DONTCOPY | VM_DONTDUMP | VM_DONTEXPAND | VM_IO,
> +		     VM_MAYWRITE);

I am still really unsure about VM_PFNMAP ... it's not a PFNMAP at all 
and, as stated, vm_insert_pages() even complains quite a lot when it 
would have to set VM_MIXEDMAP and VM_PFNMAP is already set, likely for a 
very good reason.

Can't we limit ourselves to VM_IO?

But then, I wonder if it really helps much regarding GUP: yes, it blocks 
ordinary GUP (see check_vma_flags()) but as 
insert_page_into_pte_locked() does *not* set pte_special(), GUP-fast 
(gup_fast_pte_range()) will not reject it.

Really, if you want GUP-fast to reject it, remap_pfn_range() and friends 
are the way to go, that will set pte_special() such that also GUP-fast 
will leave it alone, just like vm_normal_page() would.

So ... I know Linus recommended VM_PFNMAP/VM_IO to stop GUP, but it 
alone won't stop all of GUP. We really have to mark the PTE as special, 
which vm_insert_page() must not do (because it is refcounted!).

Which means: do we really have to stop GUP from grabbing that page?

Using vm_insert_page() only with VM_MIXEDMAP (and without 
VM_PFNMAP|VM_IO) would be better.

If we want to stop all of GUP, remap_pfn_range() currently seems 
unavoidable :(


> +
> +	lockdep_assert_held(&cpu_buffer->mapping_lock);
> +
> +	subbuf_order = cpu_buffer->buffer->subbuf_order;
> +	subbuf_pages = 1 << subbuf_order;
> +
> +	nr_subbufs = cpu_buffer->nr_pages + 1; /* + reader-subbuf */
> +	nr_pages = ((nr_subbufs) << subbuf_order) - pgoff + 1; /* + meta-page */
> +
> +	vma_pages = (vma->vm_end - vma->vm_start) >> PAGE_SHIFT;
> +	if (!vma_pages || vma_pages > nr_pages)
> +		return -EINVAL;
> +
> +	nr_pages = vma_pages;
> +
> +	pages = kcalloc(nr_pages, sizeof(*pages), GFP_KERNEL);
> +	if (!pages)
> +		return -ENOMEM;
> +
> +	if (!pgoff) {
> +		pages[p++] = virt_to_page(cpu_buffer->meta_page);
> +
> +		/*
> +		 * TODO: Align sub-buffers on their size, once
> +		 * vm_insert_pages() supports the zero-page.
> +		 */
> +	} else {
> +		/* Skip the meta-page */
> +		pgoff--;
> +
> +		if (pgoff % subbuf_pages) {
> +			err = -EINVAL;
> +			goto out;
> +		}
> +
> +		s += pgoff / subbuf_pages;
> +	}
> +
> +	while (s < nr_subbufs && p < nr_pages) {
> +		struct page *page = virt_to_page(cpu_buffer->subbuf_ids[s]);
> +		int off = 0;
> +
> +		for (; off < (1 << (subbuf_order)); off++, page++) {
> +			if (p >= nr_pages)
> +				break;
> +
> +			pages[p++] = page;
> +		}
> +		s++;
> +	}
> +
> +	err = vm_insert_pages(vma, vma->vm_start, pages, &nr_pages);

vm_insert_pages() documents: "In case of error, we may have mapped a 
subset of the provided pages. It is the caller's responsibility to 
account for this case."

Which could for example happen, when allocating a page table fails.

Would we able to deal with that here?


Again, I wish it would all be easier ...

-- 
Cheers,

David / dhildenb