From: "Lad, Prabhakar" <[email protected]>
In the gen_pool_dma_alloc() the dma pointer can be NULL
and while assigning gen_pool_virt_to_phys(pool, vaddr) to
dma caused the following crash on da850 evm,
Unable to handle kernel NULL pointer dereference at virtual address 00000000
pgd = c0004000
[00000000] *pgd=00000000
Internal error: Oops: 805 [#1] PREEMPT ARM
Modules linked in:
CPU: 0 PID: 1 Comm: swapper Tainted: G W 3.13.0-rc1-00001-g0609e45-dirty #5
task: c4830000 ti: c4832000 task.ti: c4832000
PC is at gen_pool_dma_alloc+0x30/0x3c
LR is at gen_pool_virt_to_phys+0x74/0x80
pc : [<c01a73b8>] lr : [<c01a6ff8>] psr: 60000013
sp : c4833e38 ip : 00000000 fp : 00000000
r10: c0428928 r9 : c0430e1c r8 : c0444b70
r7 : c04436f0 r6 : c5840000 r5 : 00000000 r4 : c4802fe0
r3 : 80000000 r2 : c4833e18 r1 : c4832018 r0 : 80000000
Flags: nZCv IRQs on FIQs on Mode SVC_32 ISA ARM Segment kernel
Control: 0005317f Table: c0004000 DAC: 00000017
Process swapper (pid: 1, stack limit = 0xc48321c0)
Stack: (0xc4833e38 to 0xc4834000)
3e20: c0443700 c0461668
3e40: c0017df4 c0413f00 c0208b14 c0443700 c0444b5c c0485910 00000000 c0208b30
3e60: c0208b14 c0443700 00000001 c02072e0 c0443700 c0444b70 c0443734 00000000
3e80: 0000004f c0207500 c0444b70 c0207474 c4833e98 c0205cc4 c4822258 c48c9150
3ea0: 00000000 c0444b70 c4b2d780 c0459270 c4822200 c020644c c03a30d4 c019547c
3ec0: c0444b70 c0444b70 00000000 c040f01c c040c5c0 c0207a90 00000000 c0444b5c
3ee0: 00000001 c0208a90 c4832000 00000000 c040f01c c041298c c4832000 c040f038
3f00: 00000000 c0008848 c485ca80 c0468954 c4833f44 c4833f78 00000000 c0102b00
3f20: 00000000 c485ca20 00000000 c052c52c c03f1160 c00346c8 00000003 60000013
3f40: c039e7e8 00000000 00000007 00000007 c0445ca0 c0428924 c0428924 00000008
3f60: c04612e0 c040c5c0 0000004f c0428928 00000000 c040c4f8 00000007 00000007
3f80: c040c5c0 c003f3b0 00000000 c03021f8 00000000 00000000 00000000 00000000
3fa0: 00000000 c0302200 00000000 c0009710 00000000 00000000 00000000 00000000
3fc0: 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
3fe0: 00000000 00000000 00000000 00000000 00000013 00000000 00000001 00000000
[<c01a73b8>] (gen_pool_dma_alloc+0x30/0x3c) from [<c0413f00>] (davinci_pm_probe+0x40/0xa8)
[<c0413f00>] (davinci_pm_probe+0x40/0xa8) from [<c0208b30>] (platform_drv_probe+0x1c/0x4c)
[<c0208b30>] (platform_drv_probe+0x1c/0x4c) from [<c02072e0>] (driver_probe_device+0x98/0x22c)
[<c02072e0>] (driver_probe_device+0x98/0x22c) from [<c0207500>] (__driver_attach+0x8c/0x90)
[<c0207500>] (__driver_attach+0x8c/0x90) from [<c0205cc4>] (bus_for_each_dev+0x6c/0x8c)
[<c0205cc4>] (bus_for_each_dev+0x6c/0x8c) from [<c020644c>] (bus_add_driver+0x124/0x1d4)
[<c020644c>] (bus_add_driver+0x124/0x1d4) from [<c0207a90>] (driver_register+0x78/0xf8)
[<c0207a90>] (driver_register+0x78/0xf8) from [<c0208a90>] (platform_driver_probe+0x20/0xa4)
[<c0208a90>] (platform_driver_probe+0x20/0xa4) from [<c041298c>] (davinci_init_late+0xc/0x14)
[<c041298c>] (davinci_init_late+0xc/0x14) from [<c040f038>] (init_machine_late+0x1c/0x28)
[<c040f038>] (init_machine_late+0x1c/0x28) from [<c0008848>] (do_one_initcall+0x34/0x15c)
[<c0008848>] (do_one_initcall+0x34/0x15c) from [<c040c4f8>] (kernel_init_freeable+0xe4/0x1ac)
[<c040c4f8>] (kernel_init_freeable+0xe4/0x1ac) from [<c0302200>] (kernel_init+0x8/0xec)
[<c0302200>] (kernel_init+0x8/0xec) from [<c0009710>] (ret_from_fork+0x14/0x24)
Code: 0afffffa e1a00004 e1a01006 ebfffef2 (e5850000)
---[ end trace 24fd42a15d1cdd7d ]---
This patch fixes the above.
Signed-off-by: Lad, Prabhakar <[email protected]>
---
lib/genalloc.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/lib/genalloc.c b/lib/genalloc.c
index dda3116..f48163f 100644
--- a/lib/genalloc.c
+++ b/lib/genalloc.c
@@ -334,7 +334,8 @@ void *gen_pool_dma_alloc(struct gen_pool *pool, size_t size, dma_addr_t *dma)
if (!vaddr)
return NULL;
- *dma = gen_pool_virt_to_phys(pool, vaddr);
+ if (dma)
+ *dma = gen_pool_virt_to_phys(pool, vaddr);
return (void *)vaddr;
}
--
1.7.9.5
Hi Prabhakar,
On 23 January 2014 15:26, Prabhakar Lad <[email protected]> wrote:
> From: "Lad, Prabhakar" <[email protected]>
>
> In the gen_pool_dma_alloc() the dma pointer can be NULL
> and while assigning gen_pool_virt_to_phys(pool, vaddr) to
> dma caused the following crash on da850 evm,
>
[snip]
>
> Signed-off-by: Lad, Prabhakar <[email protected]>
> ---
> lib/genalloc.c | 3 ++-
> 1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/lib/genalloc.c b/lib/genalloc.c
> index dda3116..f48163f 100644
> --- a/lib/genalloc.c
> +++ b/lib/genalloc.c
> @@ -334,7 +334,8 @@ void *gen_pool_dma_alloc(struct gen_pool *pool, size_t size, dma_addr_t *dma)
> if (!vaddr)
> return NULL;
>
> - *dma = gen_pool_virt_to_phys(pool, vaddr);
> + if (dma)
> + *dma = gen_pool_virt_to_phys(pool, vaddr);
Wouldn't it be better to return (with error/message) if dma is NULL
rather than silently ignore it?
>
> return (void *)vaddr;
--
With warm regards,
Sachin
Hi Sachin,
On Thu, Jan 23, 2014 at 3:38 PM, Sachin Kamat <[email protected]> wrote:
> Hi Prabhakar,
>
> On 23 January 2014 15:26, Prabhakar Lad <[email protected]> wrote:
>> From: "Lad, Prabhakar" <[email protected]>
>>
>> In the gen_pool_dma_alloc() the dma pointer can be NULL
>> and while assigning gen_pool_virt_to_phys(pool, vaddr) to
>> dma caused the following crash on da850 evm,
>>
> [snip]
>>
>> Signed-off-by: Lad, Prabhakar <[email protected]>
>> ---
>> lib/genalloc.c | 3 ++-
>> 1 file changed, 2 insertions(+), 1 deletion(-)
>>
>> diff --git a/lib/genalloc.c b/lib/genalloc.c
>> index dda3116..f48163f 100644
>> --- a/lib/genalloc.c
>> +++ b/lib/genalloc.c
>> @@ -334,7 +334,8 @@ void *gen_pool_dma_alloc(struct gen_pool *pool, size_t size, dma_addr_t *dma)
>> if (!vaddr)
>> return NULL;
>>
>> - *dma = gen_pool_virt_to_phys(pool, vaddr);
>> + if (dma)
>> + *dma = gen_pool_virt_to_phys(pool, vaddr);
>
> Wouldn't it be better to return (with error/message) if dma is NULL
> rather than silently ignore it?
>
I am not sure if returning here with error is OK,
may be just adding a warning message could be OK ?
Regards,
--Prabhakar Lad
On Thu, 23 Jan 2014 15:51:31 +0530 Prabhakar Lad <[email protected]> wrote:
> Hi Sachin,
>
> On Thu, Jan 23, 2014 at 3:38 PM, Sachin Kamat <[email protected]> wrote:
> > Hi Prabhakar,
> >
> > On 23 January 2014 15:26, Prabhakar Lad <[email protected]> wrote:
> >> From: "Lad, Prabhakar" <[email protected]>
> >>
> >> In the gen_pool_dma_alloc() the dma pointer can be NULL
> >> and while assigning gen_pool_virt_to_phys(pool, vaddr) to
> >> dma caused the following crash on da850 evm,
> >>
> > [snip]
> >>
> >> Signed-off-by: Lad, Prabhakar <[email protected]>
> >> ---
> >> lib/genalloc.c | 3 ++-
> >> 1 file changed, 2 insertions(+), 1 deletion(-)
> >>
> >> diff --git a/lib/genalloc.c b/lib/genalloc.c
> >> index dda3116..f48163f 100644
> >> --- a/lib/genalloc.c
> >> +++ b/lib/genalloc.c
> >> @@ -334,7 +334,8 @@ void *gen_pool_dma_alloc(struct gen_pool *pool, size_t size, dma_addr_t *dma)
> >> if (!vaddr)
> >> return NULL;
> >>
> >> - *dma = gen_pool_virt_to_phys(pool, vaddr);
> >> + if (dma)
> >> + *dma = gen_pool_virt_to_phys(pool, vaddr);
> >
> > Wouldn't it be better to return (with error/message) if dma is NULL
> > rather than silently ignore it?
> >
> I am not sure if returning here with error is OK,
> may be just adding a warning message could be OK ?
The patch look OK as-is to me. `dma' is a second return value from
gen_pool_dma_alloc() and this patch extends the gen_pool_dma_alloc()
interface by making that return value optional. That's good for
callers who don't want the physical address, and they can call
gen_pool_virt_to_phys() at a later time to get the physical address
anyway.
>From my reading, 3.13.x kernels will need this patch.
I suppose we should document the API change:
--- a/lib/genalloc.c~lib-genallocc-add-check-gen_pool_dma_alloc-if-dma-pointer-is-not-null-fix
+++ a/lib/genalloc.c
@@ -316,7 +316,7 @@ EXPORT_SYMBOL(gen_pool_alloc);
* gen_pool_dma_alloc - allocate special memory from the pool for DMA usage
* @pool: pool to allocate from
* @size: number of bytes to allocate from the pool
- * @dma: dma-view physical address
+ * @dma: dma-view physical address return value. Use NULL if unneeded.
*
* Allocate the requested number of bytes from the specified pool.
* Uses the pool allocation function (with first-fit algorithm by default).
_
On 24 January 2014 02:53, Andrew Morton <[email protected]> wrote:
> On Thu, 23 Jan 2014 15:51:31 +0530 Prabhakar Lad <[email protected]> wrote:
>
>> Hi Sachin,
>>
>> On Thu, Jan 23, 2014 at 3:38 PM, Sachin Kamat <[email protected]> wrote:
>> > Hi Prabhakar,
>> >
>> > On 23 January 2014 15:26, Prabhakar Lad <[email protected]> wrote:
>> >> From: "Lad, Prabhakar" <[email protected]>
>> >>
>> >> In the gen_pool_dma_alloc() the dma pointer can be NULL
>> >> and while assigning gen_pool_virt_to_phys(pool, vaddr) to
>> >> dma caused the following crash on da850 evm,
>> >>
>> > [snip]
>> >>
>> >> Signed-off-by: Lad, Prabhakar <[email protected]>
>> >> ---
>> >> lib/genalloc.c | 3 ++-
>> >> 1 file changed, 2 insertions(+), 1 deletion(-)
>> >>
>> >> diff --git a/lib/genalloc.c b/lib/genalloc.c
>> >> index dda3116..f48163f 100644
>> >> --- a/lib/genalloc.c
>> >> +++ b/lib/genalloc.c
>> >> @@ -334,7 +334,8 @@ void *gen_pool_dma_alloc(struct gen_pool *pool, size_t size, dma_addr_t *dma)
>> >> if (!vaddr)
>> >> return NULL;
>> >>
>> >> - *dma = gen_pool_virt_to_phys(pool, vaddr);
>> >> + if (dma)
>> >> + *dma = gen_pool_virt_to_phys(pool, vaddr);
>> >
>> > Wouldn't it be better to return (with error/message) if dma is NULL
>> > rather than silently ignore it?
>> >
>> I am not sure if returning here with error is OK,
>> may be just adding a warning message could be OK ?
>
> The patch look OK as-is to me. `dma' is a second return value from
> gen_pool_dma_alloc() and this patch extends the gen_pool_dma_alloc()
> interface by making that return value optional. That's good for
> callers who don't want the physical address, and they can call
> gen_pool_virt_to_phys() at a later time to get the physical address
> anyway.
>
> From my reading, 3.13.x kernels will need this patch.
>
> I suppose we should document the API change:
>
> --- a/lib/genalloc.c~lib-genallocc-add-check-gen_pool_dma_alloc-if-dma-pointer-is-not-null-fix
> +++ a/lib/genalloc.c
> @@ -316,7 +316,7 @@ EXPORT_SYMBOL(gen_pool_alloc);
> * gen_pool_dma_alloc - allocate special memory from the pool for DMA usage
> * @pool: pool to allocate from
> * @size: number of bytes to allocate from the pool
> - * @dma: dma-view physical address
> + * @dma: dma-view physical address return value. Use NULL if unneeded.
Makes sense. Thanks Andrew.
--
With warm regards,
Sachin