search for: try_to_free_buffers

...optimize away the _lock from set_page_dirty if it's anon/hugetlbfs/tmpfs, would be nice if there was a clean way to do that. Now assuming we don't nak the use on ext4 VM_SHARED and we stick to set_page_dirty_lock for such case: could you recap how that __writepage ext4 crash was solved if try_to_free_buffers() run on a pinned GUP page (in our vhost case try_to_unmap would have gotten rid of the pins through the mmu notifier and the page would have been freed just fine). The first two things that come to mind is that we can easily forbid the try_to_free_buffers() if the page might be pinned by GUP, it...

On Thu, Mar 07, 2019 at 10:34:39AM -0500, Michael S. Tsirkin wrote: > On Thu, Mar 07, 2019 at 10:45:57AM +0800, Jason Wang wrote: > > > > On 2019/3/7 ??12:31, Michael S. Tsirkin wrote: > > > > +static void vhost_set_vmap_dirty(struct vhost_vmap *used) > > > > +{ > > > > + int i; > > > > + > > > > + for (i = 0; i <

On 2019/3/14 ??6:42, Michael S. Tsirkin wrote: >>>>> Which means after we fix vhost to add the flush_dcache_page after >>>>> kunmap, Parisc will get a double hit (but it also means Parisc >>>>> was the only one of those archs needed explicit cache flushes, >>>>> where vhost worked correctly so far.. so it kinds of proofs your

On 2019/3/14 ??6:42, Michael S. Tsirkin wrote: >>>>> Which means after we fix vhost to add the flush_dcache_page after >>>>> kunmap, Parisc will get a double hit (but it also means Parisc >>>>> was the only one of those archs needed explicit cache flushes, >>>>> where vhost worked correctly so far.. so it kinds of proofs your

On Thu, Mar 07, 2019 at 10:45:57AM +0800, Jason Wang wrote: > > On 2019/3/7 ??12:31, Michael S. Tsirkin wrote: > > > +static void vhost_set_vmap_dirty(struct vhost_vmap *used) > > > +{ > > > + int i; > > > + > > > + for (i = 0; i < used->npages; i++) > > > + set_page_dirty_lock(used->pages[i]); > > This seems to rely on

On Thu, Mar 07, 2019 at 10:45:57AM +0800, Jason Wang wrote: > > On 2019/3/7 ??12:31, Michael S. Tsirkin wrote: > > > +static void vhost_set_vmap_dirty(struct vhost_vmap *used) > > > +{ > > > + int i; > > > + > > > + for (i = 0; i < used->npages; i++) > > > + set_page_dirty_lock(used->pages[i]); > > This seems to rely on

...implementation that will work equally optimally on 32bit with <= 700M of RAM. Talking to Jerome about the set_page_dirty issue, he raised the point of what happens if two thread calls a mmu notifier invalidate simultaneously. The first mmu notifier could call set_page_dirty and then proceed in try_to_free_buffers or page_mkclean and then the concurrent mmu notifier that arrives second, then must not call set_page_dirty a second time. With KVM sptes mappings and vhost mappings you would call set_page_dirty (if you invoked gup with FOLL_WRITE) only when effectively tearing down any secondary mapping (you'...

...> set_page_dirty if it's anon/hugetlbfs/tmpfs, would be nice if there > was a clean way to do that. > > Now assuming we don't nak the use on ext4 VM_SHARED and we stick to > set_page_dirty_lock for such case: could you recap how that > __writepage ext4 crash was solved if try_to_free_buffers() run on a > pinned GUP page (in our vhost case try_to_unmap would have gotten rid > of the pins through the mmu notifier and the page would have been > freed just fine). So for the above the easiest thing is to call set_page_dirty() from the mmu notifier callback. It is always safe to us...

...> set_page_dirty if it's anon/hugetlbfs/tmpfs, would be nice if there > was a clean way to do that. > > Now assuming we don't nak the use on ext4 VM_SHARED and we stick to > set_page_dirty_lock for such case: could you recap how that > __writepage ext4 crash was solved if try_to_free_buffers() run on a > pinned GUP page (in our vhost case try_to_unmap would have gotten rid > of the pins through the mmu notifier and the page would have been > freed just fine). So for the above the easiest thing is to call set_page_dirty() from the mmu notifier callback. It is always safe to us...

...deactivate_page(page); + /*deactivate_page(page); + */ page_cache_release(page); fail: return -1; and --- fs/buffer.c.orig Tue Oct 30 11:15:43 2001 +++ fs/buffer.c Tue Oct 30 11:03:31 2001 @@ -2529,8 +2529,8 @@ busy_buffer_page: return 0; } EXPORT_SYMBOL(try_to_free_buffers); -EXPORT_SYMBOL(buffermem_pages); - +/* EXPORT_SYMBOL(buffermem_pages); +*/ /* ================== Debugging =================== */ void show_buffers(void) This make all compile and work basically.. Since I am not a real kernel hacker, I dont really know what is the impact of modifications (e...

...ing request at virtual address 00090020 Feb 6 14:26:47 bauhaus kernel: current->tss.cr3 = 01a61000, %%cr3 = 01a61000 Feb 6 14:26:47 bauhaus kernel: *pde = 00000000 Feb 6 14:26:47 bauhaus kernel: Oops: 0000 Feb 6 14:26:47 bauhaus kernel: CPU: 0 Feb 6 14:26:47 bauhaus kernel: EIP: 0010:[try_to_free_buffers+16/180] Feb 6 14:26:47 bauhaus kernel: EFLAGS: 00010296 Feb 6 14:26:47 bauhaus kernel: eax: 00009470 ebx: c037fc18 ecx: 0000cbc0 edx: 00000000 Feb 6 14:26:47 bauhaus kernel: esi: 00090000 edi: 00090000 ebp: c037fc18 esp: c1c67e1c Feb 6 14:26:47 bauhaus kernel: ds: 0018 es: 0018...

SYSTEM: rh6x based system, 2.2.19-6.2.7 rh errata kernel + 0.0.7a patch, I rebuilt rpm for i686; celeron466, 64MB, PIIX4. root fs is on software raid1 ext2, 6 additional fs's on software raid1 ext2. There's a 3rd HD, not mirrored, which is mounted ext3. EXT3-fs: mounted filesystem with ordered data mode. I enabled journal with tune2fs -j with unmounted fs. The 3 HDs are tuned with

...doesn't even > exist for transient pins like O_DIRECT (does it work by luck?), but > with mmu notifiers there are no long term pins anyway, so this works > normally and it's like the memory isn't pinned. In any case I think > that's a kernel bug in either __writepage or try_to_free_buffers, so I > would ignore it considering qemu will only use anon memory or tmpfs or > hugetlbfs as backing store for the virtio ring. It wouldn't make sense > for qemu to risk triggering I/O on a VM_SHARED ext4, so we shouldn't > be even exposed to what seems to be an orthogonal kern...

...VM_SHARED ext4 page doesn't even exist for transient pins like O_DIRECT (does it work by luck?), but with mmu notifiers there are no long term pins anyway, so this works normally and it's like the memory isn't pinned. In any case I think that's a kernel bug in either __writepage or try_to_free_buffers, so I would ignore it considering qemu will only use anon memory or tmpfs or hugetlbfs as backing store for the virtio ring. It wouldn't make sense for qemu to risk triggering I/O on a VM_SHARED ext4, so we shouldn't be even exposed to what seems to be an orthogonal kernel bug. I suppose w...

...VM_SHARED ext4 page doesn't even exist for transient pins like O_DIRECT (does it work by luck?), but with mmu notifiers there are no long term pins anyway, so this works normally and it's like the memory isn't pinned. In any case I think that's a kernel bug in either __writepage or try_to_free_buffers, so I would ignore it considering qemu will only use anon memory or tmpfs or hugetlbfs as backing store for the virtio ring. It wouldn't make sense for qemu to risk triggering I/O on a VM_SHARED ext4, so we shouldn't be even exposed to what seems to be an orthogonal kernel bug. I suppose w...

On Wed, Mar 06, 2019 at 02:18:12AM -0500, Jason Wang wrote: > +static const struct mmu_notifier_ops vhost_mmu_notifier_ops = { > + .invalidate_range = vhost_invalidate_range, > +}; > + > void vhost_dev_init(struct vhost_dev *dev, > struct vhost_virtqueue **vqs, int nvqs, int iov_limit) > { I also wonder here: when page is write protected then it does not look like

On Wed, Mar 06, 2019 at 02:18:12AM -0500, Jason Wang wrote: > +static const struct mmu_notifier_ops vhost_mmu_notifier_ops = { > + .invalidate_range = vhost_invalidate_range, > +}; > + > void vhost_dev_init(struct vhost_dev *dev, > struct vhost_virtqueue **vqs, int nvqs, int iov_limit) > { I also wonder here: when page is write protected then it does not look like