[PATCH] [PATCH 4.9] staging: android: ion: fix page is NULL

[Cheng Chao]

I notice that v4.9.283 has released, but this patch is not merged.
It's exactly a bug.

On Sat, Sep 11, 2021 at 10:40 PM Cheng Chao <cs.os.kernel at gmail.com> wrote:
>
> the bug has been introduced by the patch:
>    https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/drivers/staging/android/ion/ion_system_heap.c?h=v4.9.282&id=e7f63771b60e7802c5a9b437c5ab1a8e33a0bb35
>
> On Sat, Sep 11, 2021 at 10:20 PM Cheng Chao <cs.os.kernel at gmail.com> wrote:
> >
> > for longterm v4.4.283, the code has checked if (!page)
> >    https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/tree/drivers/staging/android/ion/ion_system_heap.c?h=v4.4.283
> >
> >
> > static struct page *alloc_buffer_page(struct ion_system_heap *heap,
> >      struct ion_buffer *buffer,
> >      unsigned long order)
> > {
> > ...
> > if (!cached) {
> > page = ion_page_pool_alloc(pool);
> > } else {
> > ...
> > page = alloc_pages(gfp_flags | __GFP_COMP, order);
> > if (!page)    <---
> >    return NULL; <---
> >
> > ion_pages_sync_for_device(NULL, page, PAGE_SIZE << order,
> > ...
> > }
> >
> > ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
> > for longterm v4.14.246, has no ion_pages_sync_for_device after
> > ion_page_pool_alloc
> > https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/tree/drivers/staging/android/ion/ion_system_heap.c?h=v4.14.246
> >
> >
> > static struct page *alloc_buffer_page(struct ion_system_heap *heap,
> >      struct ion_buffer *buffer,
> >      unsigned long order)
> > {
> > ...
> > if (!cached)
> > pool = heap->uncached_pools[order_to_index(order)];
> > else
> > pool = heap->cached_pools[order_to_index(order)];
> >
> > page = ion_page_pool_alloc(pool);
> >
> > return page;
> > }
> >
> > ----------------------------------------------------------------------------------------------------------------------------------------------------------
> > after longterm v4.19.206(include),
> >
> > static struct page *alloc_buffer_page(struct ion_system_heap *heap,
> >      struct ion_buffer *buffer,
> >      unsigned long order)
> > {
> > struct ion_page_pool *pool = heap->pools[order_to_index(order)];
> >
> > return ion_page_pool_alloc(pool);
> > }
> >
> > On Sat, Sep 11, 2021 at 7:54 PM Greg KH <gregkh at linuxfoundation.org> wrote:
> > >
> > > On Sat, Sep 11, 2021 at 07:21:15PM +0800, Cheng Chao wrote:
> > > > kernel panic is here:
> > > >
> > > > Unable to handle kernel paging request at virtual address b0380000
> > > > pgd = d9d94000
> > > > [b0380000] *pgd=00000000
> > > > Internal error: Oops: 2805 [#1] PREEMPT SMP ARM
> > > > ...
> > > > task: daa2dd00 task.stack: da194000
> > > > PC is at v7_dma_clean_range+0x1c/0x34
> > > > LR is at arm_dma_sync_single_for_device+0x44/0x58
> > > > pc : [<c011aa0c>]    lr : [<c011645c>]    psr: 200f0013
> > > > sp : da195da0  ip : dc1f9000  fp : c1043dc4
> > > > r10: 00000000  r9 : c16f1f58  r8 : 00000001
> > > > r7 : c1621f94  r6 : c0116418  r5 : 00000000  r4 : c011aa58
> > > > r3 : 0000003f  r2 : 00000040  r1 : b0480000  r0 : b0380000
> > > > Flags: nzCv  IRQs on  FIQs on  Mode SVC_32  ISA ARM  Segment none
> > > > Control: 10c5383d  Table: 19d9406a  DAC: 00000051
> > > > ...
> > > > [<c011aa0c>] (v7_dma_clean_range) from [<c011645c>] (arm_dma_sync_single_for_device+0x44/0x58)
> > > > [<c011645c>] (arm_dma_sync_single_for_device) from [<c0117088>] (arm_dma_sync_sg_for_device+0x50/0x7c)
> > > > [<c0117088>] (arm_dma_sync_sg_for_device) from [<c0c033c4>] (ion_pages_sync_for_device+0xb0/0xec)
> > > > [<c0c033c4>] (ion_pages_sync_for_device) from [<c0c054ac>] (ion_system_heap_allocate+0x2a0/0x2e0)
> > > > [<c0c054ac>] (ion_system_heap_allocate) from [<c0c02c78>] (ion_alloc+0x12c/0x494)
> > > > [<c0c02c78>] (ion_alloc) from [<c0c03eac>] (ion_ioctl+0x510/0x63c)
> > > > [<c0c03eac>] (ion_ioctl) from [<c027c4b0>] (do_vfs_ioctl+0xa8/0x9b4)
> > > > [<c027c4b0>] (do_vfs_ioctl) from [<c027ce28>] (SyS_ioctl+0x6c/0x7c)
> > > > [<c027ce28>] (SyS_ioctl) from [<c0108a40>] (ret_fast_syscall+0x0/0x48)
> > > > Code: e3a02004 e1a02312 e2423001 e1c00003 (ee070f3a)
> > > > ---[ end trace 89278304932c0e87 ]---
> > > > Kernel panic - not syncing: Fatal exception
> > > >
> > > > Signed-off-by: Cheng Chao <cs.os.kernel at gmail.com>
> > > > ---
> > > >  drivers/staging/android/ion/ion_system_heap.c | 2 +-
> > > >  1 file changed, 1 insertion(+), 1 deletion(-)
> > > >
> > > > diff --git a/drivers/staging/android/ion/ion_system_heap.c b/drivers/staging/android/ion/ion_system_heap.c
> > > > index 22c481f2ae4f..2a35b99cf628 100644
> > > > --- a/drivers/staging/android/ion/ion_system_heap.c
> > > > +++ b/drivers/staging/android/ion/ion_system_heap.c
> > > > @@ -75,7 +75,7 @@ static struct page *alloc_buffer_page(struct ion_system_heap *heap,
> > > >
> > > >       page = ion_page_pool_alloc(pool);
> > > >
> > > > -     if (cached)
> > > > +     if (page && cached)
> > > >               ion_pages_sync_for_device(NULL, page, PAGE_SIZE << order,
> > > >                                         DMA_BIDIRECTIONAL);
> > > >       return page;
> > > > --
> > > > 2.26.3
> > > >
> > >
> > > Why is this just a 4.9 patch?  Ion didn't get removed until 5.11, so
> > > wouldn't this be an issue for anything 5.10 and older?
> > >
> > > thanks,
> > >
> > > greg k-h