linux-kernel - Re: [PATCH 19/22] mm: always return EBUSY for invalid ranges in hmm_range

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Tue, 2 Jul 2019 21:43:58 +0000
From:   "Kuehling, Felix" <Felix.Kuehling@....com>
To:     Christoph Hellwig <hch@....de>,
        Dan Williams <dan.j.williams@...el.com>,
        Jérôme Glisse <jglisse@...hat.com>,
        Jason Gunthorpe <jgg@...lanox.com>,
        Ben Skeggs <bskeggs@...hat.com>,
        "Yang, Philip" <Philip.Yang@....com>
CC:     Ira Weiny <ira.weiny@...el.com>,
        "linux-mm@...ck.org" <linux-mm@...ck.org>,
        "nouveau@...ts.freedesktop.org" <nouveau@...ts.freedesktop.org>,
        "dri-devel@...ts.freedesktop.org" <dri-devel@...ts.freedesktop.org>,
        "linux-nvdimm@...ts.01.org" <linux-nvdimm@...ts.01.org>,
        "linux-pci@...r.kernel.org" <linux-pci@...r.kernel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH 19/22] mm: always return EBUSY for invalid ranges in
 hmm_range_{fault,snapshot}

On 2019-07-01 2:20 a.m., Christoph Hellwig wrote:
> We should not have two different error codes for the same condition.  In
> addition this really complicates the code due to the special handling of
> EAGAIN that drops the mmap_sem due to the FAULT_FLAG_ALLOW_RETRY logic
> in the core vm.

I think the comment above hmm_range_snapshot needs an update. Also 
Documentation/vm/hmm.rst shows some example code using 
hmm_range_snapshot that retries on -EAGAIN. That would need to be 
updated to use -EBUSY or remove the retry logic altogether.

Other than that, this patch is Reviewed-by: Felix Kuehling 
<Felix.Kuehling@....com>

Philip, this means we should remove our retry logic again in 
amdgpu_ttm_tt_get_user_pages. According to the comment above 
hmm_range_fault, it can only return -EAGAIN if the block parameter is 
false. I think this statement is now actually true. We set block=true, 
so we can't get -EAGAIN. On -EBUSY we can let 
amdgpu_amdkfd_restore_userptr_worker schedule the retry (which it does 
already anyway).

Regards,
   Felix


>
> Signed-off-by: Christoph Hellwig <hch@....de>
> ---
>   mm/hmm.c | 8 +++-----
>   1 file changed, 3 insertions(+), 5 deletions(-)
>
> diff --git a/mm/hmm.c b/mm/hmm.c
> index c85ed7d4e2ce..d125df698e2b 100644
> --- a/mm/hmm.c
> +++ b/mm/hmm.c
> @@ -974,7 +974,7 @@ long hmm_range_snapshot(struct hmm_range *range)
>   	do {
>   		/* If range is no longer valid force retry. */
>   		if (!range->valid)
> -			return -EAGAIN;
> +			return -EBUSY;
>   
>   		vma = find_vma(hmm->mm, start);
>   		if (vma == NULL || (vma->vm_flags & device_vma))
> @@ -1069,10 +1069,8 @@ long hmm_range_fault(struct hmm_range *range, bool block)
>   
>   	do {
>   		/* If range is no longer valid force retry. */
> -		if (!range->valid) {
> -			up_read(&hmm->mm->mmap_sem);
> -			return -EAGAIN;
> -		}
> +		if (!range->valid)
> +			return -EBUSY;
>   
>   		vma = find_vma(hmm->mm, start);
>   		if (vma == NULL || (vma->vm_flags & device_vma))