* [PATCH v2] mm/codetag: move ref and tag null pointer check to alloc_tag_add
2024-10-18 15:29 [PATCH] mm/codetag: move ref and tag null pointer check to alloc_tag_ref_set Hao Ge
@ 2024-10-18 16:25 ` Hao Ge
2024-10-18 17:30 ` Suren Baghdasaryan
2024-10-20 7:04 ` [PATCH] mm/codetag: fix null pointer check logic for ref and tag Hao Ge
2024-10-20 7:08 ` [PATCH v3] " Hao Ge
2 siblings, 1 reply; 9+ messages in thread
From: Hao Ge @ 2024-10-18 16:25 UTC (permalink / raw)
To: surenb, kent.overstreet, akpm
Cc: yuzhao, linux-mm, linux-kernel, hao.ge, Hao Ge
From: Hao Ge <gehao@kylinos.cn>
When we compile and load lib/slub_kunit.c,it will cause a panic.
The root cause is that __kmalloc_cache_noprof was directly called
instead of kmem_cache_alloc,which resulted in no alloc_tag being
allocated.This caused current->alloc_tag to be null,leading to a
null pointer dereference in alloc_tag_ref_set.
Despite the fact that my colleague Pei Xiao will later fix the code
in slub_kunit.c,we still need to move the null pointer check for ref
and tag to alloc_tag_add here.
It is sufficient for us to issue a warning to the user;
It should not lead to a panic.
Here is the log for the panic:
[ 74.779373][ T2158] Unable to handle kernel NULL pointer dereference at virtual address 0000000000000020
[ 74.780130][ T2158] Mem abort info:
[ 74.780406][ T2158] ESR = 0x0000000096000004
[ 74.780756][ T2158] EC = 0x25: DABT (current EL), IL = 32 bits
[ 74.781225][ T2158] SET = 0, FnV = 0
[ 74.781529][ T2158] EA = 0, S1PTW = 0
[ 74.781836][ T2158] FSC = 0x04: level 0 translation fault
[ 74.782288][ T2158] Data abort info:
[ 74.782577][ T2158] ISV = 0, ISS = 0x00000004, ISS2 = 0x00000000
[ 74.783068][ T2158] CM = 0, WnR = 0, TnD = 0, TagAccess = 0
[ 74.783533][ T2158] GCS = 0, Overlay = 0, DirtyBit = 0, Xs = 0
[ 74.784010][ T2158] user pgtable: 4k pages, 48-bit VAs, pgdp=0000000105f34000
[ 74.784586][ T2158] [0000000000000020] pgd=0000000000000000, p4d=0000000000000000
[ 74.785293][ T2158] Internal error: Oops: 0000000096000004 [#1] SMP
[ 74.785805][ T2158] Modules linked in: slub_kunit kunit ip6t_rpfilter ip6t_REJECT nf_reject_ipv6 ipt_REJECT nf_reject_ipv4 xt_conntrack ebtable_nat ebtable_broute ip6table_nat ip6table_mangle 4
[ 74.790661][ T2158] CPU: 0 UID: 0 PID: 2158 Comm: kunit_try_catch Kdump: loaded Tainted: G W N 6.12.0-rc3+ #2
[ 74.791535][ T2158] Tainted: [W]=WARN, [N]=TEST
[ 74.791889][ T2158] Hardware name: QEMU KVM Virtual Machine, BIOS 0.0.0 02/06/2015
[ 74.792479][ T2158] pstate: 40400005 (nZcv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--)
[ 74.793101][ T2158] pc : alloc_tagging_slab_alloc_hook+0x120/0x270
[ 74.793607][ T2158] lr : alloc_tagging_slab_alloc_hook+0x120/0x270
[ 74.794095][ T2158] sp : ffff800084d33cd0
[ 74.794418][ T2158] x29: ffff800084d33cd0 x28: 0000000000000000 x27: 0000000000000000
[ 74.795095][ T2158] x26: 0000000000000000 x25: 0000000000000012 x24: ffff80007b30e314
[ 74.795822][ T2158] x23: ffff000390ff6f10 x22: 0000000000000000 x21: 0000000000000088
[ 74.796555][ T2158] x20: ffff000390285840 x19: fffffd7fc3ef7830 x18: ffffffffffffffff
[ 74.797283][ T2158] x17: ffff8000800e63b4 x16: ffff80007b33afc4 x15: ffff800081654c00
[ 74.798011][ T2158] x14: 0000000000000000 x13: 205d383531325420 x12: 5b5d383734363537
[ 74.798744][ T2158] x11: ffff800084d337e0 x10: 000000000000005d x9 : 00000000ffffffd0
[ 74.799476][ T2158] x8 : 7f7f7f7f7f7f7f7f x7 : ffff80008219d188 x6 : c0000000ffff7fff
[ 74.800206][ T2158] x5 : ffff0003fdbc9208 x4 : ffff800081edd188 x3 : 0000000000000001
[ 74.800932][ T2158] x2 : 0beaa6dee1ac5a00 x1 : 0beaa6dee1ac5a00 x0 : ffff80037c2cb000
[ 74.801656][ T2158] Call trace:
[ 74.801954][ T2158] alloc_tagging_slab_alloc_hook+0x120/0x270
[ 74.802494][ T2158] __kmalloc_cache_noprof+0x148/0x33c
[ 74.802976][ T2158] test_kmalloc_redzone_access+0x4c/0x104 [slub_kunit]
[ 74.803607][ T2158] kunit_try_run_case+0x70/0x17c [kunit]
[ 74.804124][ T2158] kunit_generic_run_threadfn_adapter+0x2c/0x4c [kunit]
[ 74.804768][ T2158] kthread+0x10c/0x118
[ 74.805141][ T2158] ret_from_fork+0x10/0x20
[ 74.805540][ T2158] Code: b9400a80 11000400 b9000a80 97ffd858 (f94012d3)
[ 74.806176][ T2158] SMP: stopping secondary CPUs
[ 74.808130][ T2158] Starting crashdump kernel...
Fixes: e0a955bf7f61 ("mm/codetag: add pgalloc_tag_copy()")
Signed-off-by: Hao Ge <gehao@kylinos.cn>
---
v2: Modify the errors in the title and commit message.
Remove the empty lines that were mistakenly added in version v1.
---
include/linux/alloc_tag.h | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/include/linux/alloc_tag.h b/include/linux/alloc_tag.h
index 1f0a9ff23a2c..8603e3a9df10 100644
--- a/include/linux/alloc_tag.h
+++ b/include/linux/alloc_tag.h
@@ -137,10 +137,6 @@ static inline void alloc_tag_sub_check(union codetag_ref *ref) {}
/* Caller should verify both ref and tag to be valid */
static inline void __alloc_tag_ref_set(union codetag_ref *ref, struct alloc_tag *tag)
{
- alloc_tag_add_check(ref, tag);
- if (!ref || !tag)
- return;
-
ref->ct = &tag->ct;
}
@@ -158,6 +154,10 @@ static inline void alloc_tag_ref_set(union codetag_ref *ref, struct alloc_tag *t
static inline void alloc_tag_add(union codetag_ref *ref, struct alloc_tag *tag, size_t bytes)
{
+ alloc_tag_add_check(ref, tag);
+ if (!ref || !tag)
+ return;
+
alloc_tag_ref_set(ref, tag);
this_cpu_add(tag->counters->bytes, bytes);
}
--
2.25.1
^ permalink raw reply [flat|nested] 9+ messages in thread* Re: [PATCH v2] mm/codetag: move ref and tag null pointer check to alloc_tag_add
2024-10-18 16:25 ` [PATCH v2] mm/codetag: move ref and tag null pointer check to alloc_tag_add Hao Ge
@ 2024-10-18 17:30 ` Suren Baghdasaryan
2024-10-19 0:46 ` Hao Ge
2024-10-19 0:56 ` Yu Zhao
0 siblings, 2 replies; 9+ messages in thread
From: Suren Baghdasaryan @ 2024-10-18 17:30 UTC (permalink / raw)
To: Hao Ge; +Cc: kent.overstreet, akpm, yuzhao, linux-mm, linux-kernel, Hao Ge
On Fri, Oct 18, 2024 at 9:26 AM Hao Ge <hao.ge@linux.dev> wrote:
>
> From: Hao Ge <gehao@kylinos.cn>
>
> When we compile and load lib/slub_kunit.c,it will cause a panic.
>
> The root cause is that __kmalloc_cache_noprof was directly called
> instead of kmem_cache_alloc,which resulted in no alloc_tag being
> allocated.This caused current->alloc_tag to be null,leading to a
> null pointer dereference in alloc_tag_ref_set.
>
> Despite the fact that my colleague Pei Xiao will later fix the code
> in slub_kunit.c,we still need to move the null pointer check for ref
> and tag to alloc_tag_add here.
> It is sufficient for us to issue a warning to the user;
> It should not lead to a panic.
>
> Here is the log for the panic:
>
> [ 74.779373][ T2158] Unable to handle kernel NULL pointer dereference at virtual address 0000000000000020
> [ 74.780130][ T2158] Mem abort info:
> [ 74.780406][ T2158] ESR = 0x0000000096000004
> [ 74.780756][ T2158] EC = 0x25: DABT (current EL), IL = 32 bits
> [ 74.781225][ T2158] SET = 0, FnV = 0
> [ 74.781529][ T2158] EA = 0, S1PTW = 0
> [ 74.781836][ T2158] FSC = 0x04: level 0 translation fault
> [ 74.782288][ T2158] Data abort info:
> [ 74.782577][ T2158] ISV = 0, ISS = 0x00000004, ISS2 = 0x00000000
> [ 74.783068][ T2158] CM = 0, WnR = 0, TnD = 0, TagAccess = 0
> [ 74.783533][ T2158] GCS = 0, Overlay = 0, DirtyBit = 0, Xs = 0
> [ 74.784010][ T2158] user pgtable: 4k pages, 48-bit VAs, pgdp=0000000105f34000
> [ 74.784586][ T2158] [0000000000000020] pgd=0000000000000000, p4d=0000000000000000
> [ 74.785293][ T2158] Internal error: Oops: 0000000096000004 [#1] SMP
> [ 74.785805][ T2158] Modules linked in: slub_kunit kunit ip6t_rpfilter ip6t_REJECT nf_reject_ipv6 ipt_REJECT nf_reject_ipv4 xt_conntrack ebtable_nat ebtable_broute ip6table_nat ip6table_mangle 4
> [ 74.790661][ T2158] CPU: 0 UID: 0 PID: 2158 Comm: kunit_try_catch Kdump: loaded Tainted: G W N 6.12.0-rc3+ #2
> [ 74.791535][ T2158] Tainted: [W]=WARN, [N]=TEST
> [ 74.791889][ T2158] Hardware name: QEMU KVM Virtual Machine, BIOS 0.0.0 02/06/2015
> [ 74.792479][ T2158] pstate: 40400005 (nZcv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--)
> [ 74.793101][ T2158] pc : alloc_tagging_slab_alloc_hook+0x120/0x270
> [ 74.793607][ T2158] lr : alloc_tagging_slab_alloc_hook+0x120/0x270
> [ 74.794095][ T2158] sp : ffff800084d33cd0
> [ 74.794418][ T2158] x29: ffff800084d33cd0 x28: 0000000000000000 x27: 0000000000000000
> [ 74.795095][ T2158] x26: 0000000000000000 x25: 0000000000000012 x24: ffff80007b30e314
> [ 74.795822][ T2158] x23: ffff000390ff6f10 x22: 0000000000000000 x21: 0000000000000088
> [ 74.796555][ T2158] x20: ffff000390285840 x19: fffffd7fc3ef7830 x18: ffffffffffffffff
> [ 74.797283][ T2158] x17: ffff8000800e63b4 x16: ffff80007b33afc4 x15: ffff800081654c00
> [ 74.798011][ T2158] x14: 0000000000000000 x13: 205d383531325420 x12: 5b5d383734363537
> [ 74.798744][ T2158] x11: ffff800084d337e0 x10: 000000000000005d x9 : 00000000ffffffd0
> [ 74.799476][ T2158] x8 : 7f7f7f7f7f7f7f7f x7 : ffff80008219d188 x6 : c0000000ffff7fff
> [ 74.800206][ T2158] x5 : ffff0003fdbc9208 x4 : ffff800081edd188 x3 : 0000000000000001
> [ 74.800932][ T2158] x2 : 0beaa6dee1ac5a00 x1 : 0beaa6dee1ac5a00 x0 : ffff80037c2cb000
> [ 74.801656][ T2158] Call trace:
> [ 74.801954][ T2158] alloc_tagging_slab_alloc_hook+0x120/0x270
> [ 74.802494][ T2158] __kmalloc_cache_noprof+0x148/0x33c
> [ 74.802976][ T2158] test_kmalloc_redzone_access+0x4c/0x104 [slub_kunit]
> [ 74.803607][ T2158] kunit_try_run_case+0x70/0x17c [kunit]
> [ 74.804124][ T2158] kunit_generic_run_threadfn_adapter+0x2c/0x4c [kunit]
> [ 74.804768][ T2158] kthread+0x10c/0x118
> [ 74.805141][ T2158] ret_from_fork+0x10/0x20
> [ 74.805540][ T2158] Code: b9400a80 11000400 b9000a80 97ffd858 (f94012d3)
> [ 74.806176][ T2158] SMP: stopping secondary CPUs
> [ 74.808130][ T2158] Starting crashdump kernel...
>
> Fixes: e0a955bf7f61 ("mm/codetag: add pgalloc_tag_copy()")
> Signed-off-by: Hao Ge <gehao@kylinos.cn>
> ---
> v2: Modify the errors in the title and commit message.
> Remove the empty lines that were mistakenly added in version v1.
> ---
> include/linux/alloc_tag.h | 8 ++++----
> 1 file changed, 4 insertions(+), 4 deletions(-)
>
> diff --git a/include/linux/alloc_tag.h b/include/linux/alloc_tag.h
> index 1f0a9ff23a2c..8603e3a9df10 100644
> --- a/include/linux/alloc_tag.h
> +++ b/include/linux/alloc_tag.h
> @@ -137,10 +137,6 @@ static inline void alloc_tag_sub_check(union codetag_ref *ref) {}
> /* Caller should verify both ref and tag to be valid */
> static inline void __alloc_tag_ref_set(union codetag_ref *ref, struct alloc_tag *tag)
> {
> - alloc_tag_add_check(ref, tag);
> - if (!ref || !tag)
> - return;
> -
Unfortunately this change will result in __alloc_tag_ref_set() and
alloc_tag_ref_set() missing the following important check from
alloc_tag_sub_check():
WARN_ONCE(ref && ref->ct,
"alloc_tag was not cleared (got tag for %s:%u)\n",
ref->ct->filename, ref->ct->lineno);
I think the change below would fix this issue without the above
mentioned side-effect:
-static inline void __alloc_tag_ref_set(union codetag_ref *ref, struct
alloc_tag *tag)
+static inline bool __alloc_tag_ref_set(union codetag_ref *ref, struct
alloc_tag *tag)
{
alloc_tag_add_check(ref, tag);
if (!ref || !tag)
- return;
+ return false;
ref->ct = &tag->ct;
+ return true;
}
-static inline void alloc_tag_ref_set(union codetag_ref *ref, struct
alloc_tag *tag)
+static inline bool alloc_tag_ref_set(union codetag_ref *ref, struct
alloc_tag *tag)
{
- __alloc_tag_ref_set(ref, tag);
+ if (unlikely(!__alloc_tag_ref_set(ref, tag)))
+ return false;
+
/*
* We need in increment the call counter every time we have a new
* allocation or when we split a large allocation into smaller ones.
* Each new reference for every sub-allocation needs to increment call
* counter because when we free each part the counter will be
decremented.
*/
this_cpu_inc(tag->counters->calls);
+ return true;
}
static inline void alloc_tag_add(union codetag_ref *ref, struct
alloc_tag *tag, size_t bytes)
{
- alloc_tag_ref_set(ref, tag);
- this_cpu_add(tag->counters->bytes, bytes);
+ if (likely(alloc_tag_ref_set(ref, tag)))
+ this_cpu_add(tag->counters->bytes, bytes);
}
Could you please confirm this fix?
Thanks,
Suren.
> ref->ct = &tag->ct;
> }
>
> @@ -158,6 +154,10 @@ static inline void alloc_tag_ref_set(union codetag_ref *ref, struct alloc_tag *t
>
> static inline void alloc_tag_add(union codetag_ref *ref, struct alloc_tag *tag, size_t bytes)
> {
> + alloc_tag_add_check(ref, tag);
> + if (!ref || !tag)
> + return;
> +
> alloc_tag_ref_set(ref, tag);
> this_cpu_add(tag->counters->bytes, bytes);
> }
> --
> 2.25.1
>
^ permalink raw reply [flat|nested] 9+ messages in thread* Re: [PATCH v2] mm/codetag: move ref and tag null pointer check to alloc_tag_add
2024-10-18 17:30 ` Suren Baghdasaryan
@ 2024-10-19 0:46 ` Hao Ge
2024-10-19 21:17 ` Suren Baghdasaryan
2024-10-19 0:56 ` Yu Zhao
1 sibling, 1 reply; 9+ messages in thread
From: Hao Ge @ 2024-10-19 0:46 UTC (permalink / raw)
To: Suren Baghdasaryan
Cc: kent.overstreet, akpm, yuzhao, linux-mm, linux-kernel, Hao Ge
Hi Suren
On 10/19/24 01:30, Suren Baghdasaryan wrote:
> On Fri, Oct 18, 2024 at 9:26 AM Hao Ge <hao.ge@linux.dev> wrote:
>> From: Hao Ge <gehao@kylinos.cn>
>>
>> When we compile and load lib/slub_kunit.c,it will cause a panic.
>>
>> The root cause is that __kmalloc_cache_noprof was directly called
>> instead of kmem_cache_alloc,which resulted in no alloc_tag being
>> allocated.This caused current->alloc_tag to be null,leading to a
>> null pointer dereference in alloc_tag_ref_set.
>>
>> Despite the fact that my colleague Pei Xiao will later fix the code
>> in slub_kunit.c,we still need to move the null pointer check for ref
>> and tag to alloc_tag_add here.
>> It is sufficient for us to issue a warning to the user;
>> It should not lead to a panic.
>>
>> Here is the log for the panic:
>>
>> [ 74.779373][ T2158] Unable to handle kernel NULL pointer dereference at virtual address 0000000000000020
>> [ 74.780130][ T2158] Mem abort info:
>> [ 74.780406][ T2158] ESR = 0x0000000096000004
>> [ 74.780756][ T2158] EC = 0x25: DABT (current EL), IL = 32 bits
>> [ 74.781225][ T2158] SET = 0, FnV = 0
>> [ 74.781529][ T2158] EA = 0, S1PTW = 0
>> [ 74.781836][ T2158] FSC = 0x04: level 0 translation fault
>> [ 74.782288][ T2158] Data abort info:
>> [ 74.782577][ T2158] ISV = 0, ISS = 0x00000004, ISS2 = 0x00000000
>> [ 74.783068][ T2158] CM = 0, WnR = 0, TnD = 0, TagAccess = 0
>> [ 74.783533][ T2158] GCS = 0, Overlay = 0, DirtyBit = 0, Xs = 0
>> [ 74.784010][ T2158] user pgtable: 4k pages, 48-bit VAs, pgdp=0000000105f34000
>> [ 74.784586][ T2158] [0000000000000020] pgd=0000000000000000, p4d=0000000000000000
>> [ 74.785293][ T2158] Internal error: Oops: 0000000096000004 [#1] SMP
>> [ 74.785805][ T2158] Modules linked in: slub_kunit kunit ip6t_rpfilter ip6t_REJECT nf_reject_ipv6 ipt_REJECT nf_reject_ipv4 xt_conntrack ebtable_nat ebtable_broute ip6table_nat ip6table_mangle 4
>> [ 74.790661][ T2158] CPU: 0 UID: 0 PID: 2158 Comm: kunit_try_catch Kdump: loaded Tainted: G W N 6.12.0-rc3+ #2
>> [ 74.791535][ T2158] Tainted: [W]=WARN, [N]=TEST
>> [ 74.791889][ T2158] Hardware name: QEMU KVM Virtual Machine, BIOS 0.0.0 02/06/2015
>> [ 74.792479][ T2158] pstate: 40400005 (nZcv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--)
>> [ 74.793101][ T2158] pc : alloc_tagging_slab_alloc_hook+0x120/0x270
>> [ 74.793607][ T2158] lr : alloc_tagging_slab_alloc_hook+0x120/0x270
>> [ 74.794095][ T2158] sp : ffff800084d33cd0
>> [ 74.794418][ T2158] x29: ffff800084d33cd0 x28: 0000000000000000 x27: 0000000000000000
>> [ 74.795095][ T2158] x26: 0000000000000000 x25: 0000000000000012 x24: ffff80007b30e314
>> [ 74.795822][ T2158] x23: ffff000390ff6f10 x22: 0000000000000000 x21: 0000000000000088
>> [ 74.796555][ T2158] x20: ffff000390285840 x19: fffffd7fc3ef7830 x18: ffffffffffffffff
>> [ 74.797283][ T2158] x17: ffff8000800e63b4 x16: ffff80007b33afc4 x15: ffff800081654c00
>> [ 74.798011][ T2158] x14: 0000000000000000 x13: 205d383531325420 x12: 5b5d383734363537
>> [ 74.798744][ T2158] x11: ffff800084d337e0 x10: 000000000000005d x9 : 00000000ffffffd0
>> [ 74.799476][ T2158] x8 : 7f7f7f7f7f7f7f7f x7 : ffff80008219d188 x6 : c0000000ffff7fff
>> [ 74.800206][ T2158] x5 : ffff0003fdbc9208 x4 : ffff800081edd188 x3 : 0000000000000001
>> [ 74.800932][ T2158] x2 : 0beaa6dee1ac5a00 x1 : 0beaa6dee1ac5a00 x0 : ffff80037c2cb000
>> [ 74.801656][ T2158] Call trace:
>> [ 74.801954][ T2158] alloc_tagging_slab_alloc_hook+0x120/0x270
>> [ 74.802494][ T2158] __kmalloc_cache_noprof+0x148/0x33c
>> [ 74.802976][ T2158] test_kmalloc_redzone_access+0x4c/0x104 [slub_kunit]
>> [ 74.803607][ T2158] kunit_try_run_case+0x70/0x17c [kunit]
>> [ 74.804124][ T2158] kunit_generic_run_threadfn_adapter+0x2c/0x4c [kunit]
>> [ 74.804768][ T2158] kthread+0x10c/0x118
>> [ 74.805141][ T2158] ret_from_fork+0x10/0x20
>> [ 74.805540][ T2158] Code: b9400a80 11000400 b9000a80 97ffd858 (f94012d3)
>> [ 74.806176][ T2158] SMP: stopping secondary CPUs
>> [ 74.808130][ T2158] Starting crashdump kernel...
>>
>> Fixes: e0a955bf7f61 ("mm/codetag: add pgalloc_tag_copy()")
>> Signed-off-by: Hao Ge <gehao@kylinos.cn>
>> ---
>> v2: Modify the errors in the title and commit message.
>> Remove the empty lines that were mistakenly added in version v1.
>> ---
>> include/linux/alloc_tag.h | 8 ++++----
>> 1 file changed, 4 insertions(+), 4 deletions(-)
>>
>> diff --git a/include/linux/alloc_tag.h b/include/linux/alloc_tag.h
>> index 1f0a9ff23a2c..8603e3a9df10 100644
>> --- a/include/linux/alloc_tag.h
>> +++ b/include/linux/alloc_tag.h
>> @@ -137,10 +137,6 @@ static inline void alloc_tag_sub_check(union codetag_ref *ref) {}
>> /* Caller should verify both ref and tag to be valid */
>> static inline void __alloc_tag_ref_set(union codetag_ref *ref, struct alloc_tag *tag)
>> {
>> - alloc_tag_add_check(ref, tag);
>> - if (!ref || !tag)
>> - return;
>> -
> Unfortunately this change will result in __alloc_tag_ref_set() and
> alloc_tag_ref_set() missing the following important check from
> alloc_tag_sub_check():
>
Maybe I missed something here, I'm a bit confused. Can you give me an
example to explain it?
Thanks
Best regards
Hao
> WARN_ONCE(ref && ref->ct,
> "alloc_tag was not cleared (got tag for %s:%u)\n",
> ref->ct->filename, ref->ct->lineno);
>
> I think the change below would fix this issue without the above
> mentioned side-effect:
>
> -static inline void __alloc_tag_ref_set(union codetag_ref *ref, struct
> alloc_tag *tag)
> +static inline bool __alloc_tag_ref_set(union codetag_ref *ref, struct
> alloc_tag *tag)
> {
> alloc_tag_add_check(ref, tag);
> if (!ref || !tag)
> - return;
> + return false;
>
> ref->ct = &tag->ct;
> + return true;
> }
>
> -static inline void alloc_tag_ref_set(union codetag_ref *ref, struct
> alloc_tag *tag)
> +static inline bool alloc_tag_ref_set(union codetag_ref *ref, struct
> alloc_tag *tag)
> {
> - __alloc_tag_ref_set(ref, tag);
> + if (unlikely(!__alloc_tag_ref_set(ref, tag)))
> + return false;
> +
> /*
> * We need in increment the call counter every time we have a new
> * allocation or when we split a large allocation into smaller ones.
> * Each new reference for every sub-allocation needs to increment call
> * counter because when we free each part the counter will be
> decremented.
> */
> this_cpu_inc(tag->counters->calls);
> + return true;
> }
>
> static inline void alloc_tag_add(union codetag_ref *ref, struct
> alloc_tag *tag, size_t bytes)
> {
> - alloc_tag_ref_set(ref, tag);
> - this_cpu_add(tag->counters->bytes, bytes);
> + if (likely(alloc_tag_ref_set(ref, tag)))
> + this_cpu_add(tag->counters->bytes, bytes);
> }
>
> Could you please confirm this fix?
> Thanks,
> Suren.
>
>> ref->ct = &tag->ct;
>> }
>>
>> @@ -158,6 +154,10 @@ static inline void alloc_tag_ref_set(union codetag_ref *ref, struct alloc_tag *t
>>
>> static inline void alloc_tag_add(union codetag_ref *ref, struct alloc_tag *tag, size_t bytes)
>> {
>> + alloc_tag_add_check(ref, tag);
>> + if (!ref || !tag)
>> + return;
>> +
>> alloc_tag_ref_set(ref, tag);
>> this_cpu_add(tag->counters->bytes, bytes);
>> }
>> --
>> 2.25.1
>>
^ permalink raw reply [flat|nested] 9+ messages in thread* Re: [PATCH v2] mm/codetag: move ref and tag null pointer check to alloc_tag_add
2024-10-19 0:46 ` Hao Ge
@ 2024-10-19 21:17 ` Suren Baghdasaryan
0 siblings, 0 replies; 9+ messages in thread
From: Suren Baghdasaryan @ 2024-10-19 21:17 UTC (permalink / raw)
To: Hao Ge; +Cc: kent.overstreet, akpm, yuzhao, linux-mm, linux-kernel, Hao Ge
On Fri, Oct 18, 2024 at 5:47 PM Hao Ge <hao.ge@linux.dev> wrote:
>
> Hi Suren
>
>
> On 10/19/24 01:30, Suren Baghdasaryan wrote:
> > On Fri, Oct 18, 2024 at 9:26 AM Hao Ge <hao.ge@linux.dev> wrote:
> >> From: Hao Ge <gehao@kylinos.cn>
> >>
> >> When we compile and load lib/slub_kunit.c,it will cause a panic.
> >>
> >> The root cause is that __kmalloc_cache_noprof was directly called
> >> instead of kmem_cache_alloc,which resulted in no alloc_tag being
> >> allocated.This caused current->alloc_tag to be null,leading to a
> >> null pointer dereference in alloc_tag_ref_set.
> >>
> >> Despite the fact that my colleague Pei Xiao will later fix the code
> >> in slub_kunit.c,we still need to move the null pointer check for ref
> >> and tag to alloc_tag_add here.
> >> It is sufficient for us to issue a warning to the user;
> >> It should not lead to a panic.
> >>
> >> Here is the log for the panic:
> >>
> >> [ 74.779373][ T2158] Unable to handle kernel NULL pointer dereference at virtual address 0000000000000020
> >> [ 74.780130][ T2158] Mem abort info:
> >> [ 74.780406][ T2158] ESR = 0x0000000096000004
> >> [ 74.780756][ T2158] EC = 0x25: DABT (current EL), IL = 32 bits
> >> [ 74.781225][ T2158] SET = 0, FnV = 0
> >> [ 74.781529][ T2158] EA = 0, S1PTW = 0
> >> [ 74.781836][ T2158] FSC = 0x04: level 0 translation fault
> >> [ 74.782288][ T2158] Data abort info:
> >> [ 74.782577][ T2158] ISV = 0, ISS = 0x00000004, ISS2 = 0x00000000
> >> [ 74.783068][ T2158] CM = 0, WnR = 0, TnD = 0, TagAccess = 0
> >> [ 74.783533][ T2158] GCS = 0, Overlay = 0, DirtyBit = 0, Xs = 0
> >> [ 74.784010][ T2158] user pgtable: 4k pages, 48-bit VAs, pgdp=0000000105f34000
> >> [ 74.784586][ T2158] [0000000000000020] pgd=0000000000000000, p4d=0000000000000000
> >> [ 74.785293][ T2158] Internal error: Oops: 0000000096000004 [#1] SMP
> >> [ 74.785805][ T2158] Modules linked in: slub_kunit kunit ip6t_rpfilter ip6t_REJECT nf_reject_ipv6 ipt_REJECT nf_reject_ipv4 xt_conntrack ebtable_nat ebtable_broute ip6table_nat ip6table_mangle 4
> >> [ 74.790661][ T2158] CPU: 0 UID: 0 PID: 2158 Comm: kunit_try_catch Kdump: loaded Tainted: G W N 6.12.0-rc3+ #2
> >> [ 74.791535][ T2158] Tainted: [W]=WARN, [N]=TEST
> >> [ 74.791889][ T2158] Hardware name: QEMU KVM Virtual Machine, BIOS 0.0.0 02/06/2015
> >> [ 74.792479][ T2158] pstate: 40400005 (nZcv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--)
> >> [ 74.793101][ T2158] pc : alloc_tagging_slab_alloc_hook+0x120/0x270
> >> [ 74.793607][ T2158] lr : alloc_tagging_slab_alloc_hook+0x120/0x270
> >> [ 74.794095][ T2158] sp : ffff800084d33cd0
> >> [ 74.794418][ T2158] x29: ffff800084d33cd0 x28: 0000000000000000 x27: 0000000000000000
> >> [ 74.795095][ T2158] x26: 0000000000000000 x25: 0000000000000012 x24: ffff80007b30e314
> >> [ 74.795822][ T2158] x23: ffff000390ff6f10 x22: 0000000000000000 x21: 0000000000000088
> >> [ 74.796555][ T2158] x20: ffff000390285840 x19: fffffd7fc3ef7830 x18: ffffffffffffffff
> >> [ 74.797283][ T2158] x17: ffff8000800e63b4 x16: ffff80007b33afc4 x15: ffff800081654c00
> >> [ 74.798011][ T2158] x14: 0000000000000000 x13: 205d383531325420 x12: 5b5d383734363537
> >> [ 74.798744][ T2158] x11: ffff800084d337e0 x10: 000000000000005d x9 : 00000000ffffffd0
> >> [ 74.799476][ T2158] x8 : 7f7f7f7f7f7f7f7f x7 : ffff80008219d188 x6 : c0000000ffff7fff
> >> [ 74.800206][ T2158] x5 : ffff0003fdbc9208 x4 : ffff800081edd188 x3 : 0000000000000001
> >> [ 74.800932][ T2158] x2 : 0beaa6dee1ac5a00 x1 : 0beaa6dee1ac5a00 x0 : ffff80037c2cb000
> >> [ 74.801656][ T2158] Call trace:
> >> [ 74.801954][ T2158] alloc_tagging_slab_alloc_hook+0x120/0x270
> >> [ 74.802494][ T2158] __kmalloc_cache_noprof+0x148/0x33c
> >> [ 74.802976][ T2158] test_kmalloc_redzone_access+0x4c/0x104 [slub_kunit]
> >> [ 74.803607][ T2158] kunit_try_run_case+0x70/0x17c [kunit]
> >> [ 74.804124][ T2158] kunit_generic_run_threadfn_adapter+0x2c/0x4c [kunit]
> >> [ 74.804768][ T2158] kthread+0x10c/0x118
> >> [ 74.805141][ T2158] ret_from_fork+0x10/0x20
> >> [ 74.805540][ T2158] Code: b9400a80 11000400 b9000a80 97ffd858 (f94012d3)
> >> [ 74.806176][ T2158] SMP: stopping secondary CPUs
> >> [ 74.808130][ T2158] Starting crashdump kernel...
> >>
> >> Fixes: e0a955bf7f61 ("mm/codetag: add pgalloc_tag_copy()")
> >> Signed-off-by: Hao Ge <gehao@kylinos.cn>
> >> ---
> >> v2: Modify the errors in the title and commit message.
> >> Remove the empty lines that were mistakenly added in version v1.
> >> ---
> >> include/linux/alloc_tag.h | 8 ++++----
> >> 1 file changed, 4 insertions(+), 4 deletions(-)
> >>
> >> diff --git a/include/linux/alloc_tag.h b/include/linux/alloc_tag.h
> >> index 1f0a9ff23a2c..8603e3a9df10 100644
> >> --- a/include/linux/alloc_tag.h
> >> +++ b/include/linux/alloc_tag.h
> >> @@ -137,10 +137,6 @@ static inline void alloc_tag_sub_check(union codetag_ref *ref) {}
> >> /* Caller should verify both ref and tag to be valid */
> >> static inline void __alloc_tag_ref_set(union codetag_ref *ref, struct alloc_tag *tag)
> >> {
> >> - alloc_tag_add_check(ref, tag);
> >> - if (!ref || !tag)
> >> - return;
> >> -
> > Unfortunately this change will result in __alloc_tag_ref_set() and
> > alloc_tag_ref_set() missing the following important check from
> > alloc_tag_sub_check():
> >
>
> Maybe I missed something here, I'm a bit confused. Can you give me an
> example to explain it?
Sure. The warning I indicated would trigger when (ref != NULL &&
ref->ct != NULL), which means that we have a valid tag reference and
it's pointing to a valid tag. During allocation that is abnormal
because we expect that if the reference is valid (ref != NULL), it
should have been cleared (ref->ct == NULL). With your change we won't
get these warnings when calling __alloc_tag_ref_set() and
alloc_tag_ref_set(). Does that make sense?
>
> Thanks
>
>
> Best regards
> Hao
>
>
> > WARN_ONCE(ref && ref->ct,
> > "alloc_tag was not cleared (got tag for %s:%u)\n",
> > ref->ct->filename, ref->ct->lineno);
> >
> > I think the change below would fix this issue without the above
> > mentioned side-effect:
> >
> > -static inline void __alloc_tag_ref_set(union codetag_ref *ref, struct
> > alloc_tag *tag)
> > +static inline bool __alloc_tag_ref_set(union codetag_ref *ref, struct
> > alloc_tag *tag)
> > {
> > alloc_tag_add_check(ref, tag);
> > if (!ref || !tag)
> > - return;
> > + return false;
> >
> > ref->ct = &tag->ct;
> > + return true;
> > }
> >
> > -static inline void alloc_tag_ref_set(union codetag_ref *ref, struct
> > alloc_tag *tag)
> > +static inline bool alloc_tag_ref_set(union codetag_ref *ref, struct
> > alloc_tag *tag)
> > {
> > - __alloc_tag_ref_set(ref, tag);
> > + if (unlikely(!__alloc_tag_ref_set(ref, tag)))
> > + return false;
> > +
> > /*
> > * We need in increment the call counter every time we have a new
> > * allocation or when we split a large allocation into smaller ones.
> > * Each new reference for every sub-allocation needs to increment call
> > * counter because when we free each part the counter will be
> > decremented.
> > */
> > this_cpu_inc(tag->counters->calls);
> > + return true;
> > }
> >
> > static inline void alloc_tag_add(union codetag_ref *ref, struct
> > alloc_tag *tag, size_t bytes)
> > {
> > - alloc_tag_ref_set(ref, tag);
> > - this_cpu_add(tag->counters->bytes, bytes);
> > + if (likely(alloc_tag_ref_set(ref, tag)))
> > + this_cpu_add(tag->counters->bytes, bytes);
> > }
> >
> > Could you please confirm this fix?
> > Thanks,
> > Suren.
> >
> >> ref->ct = &tag->ct;
> >> }
> >>
> >> @@ -158,6 +154,10 @@ static inline void alloc_tag_ref_set(union codetag_ref *ref, struct alloc_tag *t
> >>
> >> static inline void alloc_tag_add(union codetag_ref *ref, struct alloc_tag *tag, size_t bytes)
> >> {
> >> + alloc_tag_add_check(ref, tag);
> >> + if (!ref || !tag)
> >> + return;
> >> +
> >> alloc_tag_ref_set(ref, tag);
> >> this_cpu_add(tag->counters->bytes, bytes);
> >> }
> >> --
> >> 2.25.1
> >>
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: [PATCH v2] mm/codetag: move ref and tag null pointer check to alloc_tag_add
2024-10-18 17:30 ` Suren Baghdasaryan
2024-10-19 0:46 ` Hao Ge
@ 2024-10-19 0:56 ` Yu Zhao
1 sibling, 0 replies; 9+ messages in thread
From: Yu Zhao @ 2024-10-19 0:56 UTC (permalink / raw)
To: Suren Baghdasaryan, Hao Ge
Cc: kent.overstreet, akpm, linux-mm, linux-kernel, Hao Ge
On Fri, Oct 18, 2024 at 11:30 AM Suren Baghdasaryan <surenb@google.com> wrote:
>
> On Fri, Oct 18, 2024 at 9:26 AM Hao Ge <hao.ge@linux.dev> wrote:
> >
> > From: Hao Ge <gehao@kylinos.cn>
> >
> > When we compile and load lib/slub_kunit.c,it will cause a panic.
> >
> > The root cause is that __kmalloc_cache_noprof was directly called
> > instead of kmem_cache_alloc,which resulted in no alloc_tag being
> > allocated.This caused current->alloc_tag to be null,leading to a
> > null pointer dereference in alloc_tag_ref_set.
> >
> > Despite the fact that my colleague Pei Xiao will later fix the code
> > in slub_kunit.c,we still need to move the null pointer check for ref
> > and tag to alloc_tag_add here.
> > It is sufficient for us to issue a warning to the user;
> > It should not lead to a panic.
> >
> > Here is the log for the panic:
> >
> > [ 74.779373][ T2158] Unable to handle kernel NULL pointer dereference at virtual address 0000000000000020
> > [ 74.780130][ T2158] Mem abort info:
> > [ 74.780406][ T2158] ESR = 0x0000000096000004
> > [ 74.780756][ T2158] EC = 0x25: DABT (current EL), IL = 32 bits
> > [ 74.781225][ T2158] SET = 0, FnV = 0
> > [ 74.781529][ T2158] EA = 0, S1PTW = 0
> > [ 74.781836][ T2158] FSC = 0x04: level 0 translation fault
> > [ 74.782288][ T2158] Data abort info:
> > [ 74.782577][ T2158] ISV = 0, ISS = 0x00000004, ISS2 = 0x00000000
> > [ 74.783068][ T2158] CM = 0, WnR = 0, TnD = 0, TagAccess = 0
> > [ 74.783533][ T2158] GCS = 0, Overlay = 0, DirtyBit = 0, Xs = 0
> > [ 74.784010][ T2158] user pgtable: 4k pages, 48-bit VAs, pgdp=0000000105f34000
> > [ 74.784586][ T2158] [0000000000000020] pgd=0000000000000000, p4d=0000000000000000
> > [ 74.785293][ T2158] Internal error: Oops: 0000000096000004 [#1] SMP
> > [ 74.785805][ T2158] Modules linked in: slub_kunit kunit ip6t_rpfilter ip6t_REJECT nf_reject_ipv6 ipt_REJECT nf_reject_ipv4 xt_conntrack ebtable_nat ebtable_broute ip6table_nat ip6table_mangle 4
> > [ 74.790661][ T2158] CPU: 0 UID: 0 PID: 2158 Comm: kunit_try_catch Kdump: loaded Tainted: G W N 6.12.0-rc3+ #2
> > [ 74.791535][ T2158] Tainted: [W]=WARN, [N]=TEST
> > [ 74.791889][ T2158] Hardware name: QEMU KVM Virtual Machine, BIOS 0.0.0 02/06/2015
> > [ 74.792479][ T2158] pstate: 40400005 (nZcv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--)
> > [ 74.793101][ T2158] pc : alloc_tagging_slab_alloc_hook+0x120/0x270
> > [ 74.793607][ T2158] lr : alloc_tagging_slab_alloc_hook+0x120/0x270
> > [ 74.794095][ T2158] sp : ffff800084d33cd0
> > [ 74.794418][ T2158] x29: ffff800084d33cd0 x28: 0000000000000000 x27: 0000000000000000
> > [ 74.795095][ T2158] x26: 0000000000000000 x25: 0000000000000012 x24: ffff80007b30e314
> > [ 74.795822][ T2158] x23: ffff000390ff6f10 x22: 0000000000000000 x21: 0000000000000088
> > [ 74.796555][ T2158] x20: ffff000390285840 x19: fffffd7fc3ef7830 x18: ffffffffffffffff
> > [ 74.797283][ T2158] x17: ffff8000800e63b4 x16: ffff80007b33afc4 x15: ffff800081654c00
> > [ 74.798011][ T2158] x14: 0000000000000000 x13: 205d383531325420 x12: 5b5d383734363537
> > [ 74.798744][ T2158] x11: ffff800084d337e0 x10: 000000000000005d x9 : 00000000ffffffd0
> > [ 74.799476][ T2158] x8 : 7f7f7f7f7f7f7f7f x7 : ffff80008219d188 x6 : c0000000ffff7fff
> > [ 74.800206][ T2158] x5 : ffff0003fdbc9208 x4 : ffff800081edd188 x3 : 0000000000000001
> > [ 74.800932][ T2158] x2 : 0beaa6dee1ac5a00 x1 : 0beaa6dee1ac5a00 x0 : ffff80037c2cb000
> > [ 74.801656][ T2158] Call trace:
> > [ 74.801954][ T2158] alloc_tagging_slab_alloc_hook+0x120/0x270
> > [ 74.802494][ T2158] __kmalloc_cache_noprof+0x148/0x33c
> > [ 74.802976][ T2158] test_kmalloc_redzone_access+0x4c/0x104 [slub_kunit]
> > [ 74.803607][ T2158] kunit_try_run_case+0x70/0x17c [kunit]
> > [ 74.804124][ T2158] kunit_generic_run_threadfn_adapter+0x2c/0x4c [kunit]
> > [ 74.804768][ T2158] kthread+0x10c/0x118
> > [ 74.805141][ T2158] ret_from_fork+0x10/0x20
> > [ 74.805540][ T2158] Code: b9400a80 11000400 b9000a80 97ffd858 (f94012d3)
> > [ 74.806176][ T2158] SMP: stopping secondary CPUs
> > [ 74.808130][ T2158] Starting crashdump kernel...
Thank for catching this!
> > Fixes: e0a955bf7f61 ("mm/codetag: add pgalloc_tag_copy()")
> > Signed-off-by: Hao Ge <gehao@kylinos.cn>
> > ---
> > v2: Modify the errors in the title and commit message.
> > Remove the empty lines that were mistakenly added in version v1.
> > ---
> > include/linux/alloc_tag.h | 8 ++++----
> > 1 file changed, 4 insertions(+), 4 deletions(-)
> >
> > diff --git a/include/linux/alloc_tag.h b/include/linux/alloc_tag.h
> > index 1f0a9ff23a2c..8603e3a9df10 100644
> > --- a/include/linux/alloc_tag.h
> > +++ b/include/linux/alloc_tag.h
> > @@ -137,10 +137,6 @@ static inline void alloc_tag_sub_check(union codetag_ref *ref) {}
> > /* Caller should verify both ref and tag to be valid */
> > static inline void __alloc_tag_ref_set(union codetag_ref *ref, struct alloc_tag *tag)
> > {
> > - alloc_tag_add_check(ref, tag);
> > - if (!ref || !tag)
> > - return;
> > -
>
> Unfortunately this change will result in __alloc_tag_ref_set() and
> alloc_tag_ref_set() missing the following important check
Correct, and therefore this patch is a good fix.
> from alloc_tag_sub_check():
>
> WARN_ONCE(ref && ref->ct,
> "alloc_tag was not cleared (got tag for %s:%u)\n",
> ref->ct->filename, ref->ct->lineno);
>
> I think the change below would fix this issue without the above
> mentioned side-effect:
I had this in my mind too when I first saw this patch earlier today.
> -static inline void __alloc_tag_ref_set(union codetag_ref *ref, struct
> alloc_tag *tag)
> +static inline bool __alloc_tag_ref_set(union codetag_ref *ref, struct
> alloc_tag *tag)
> {
> alloc_tag_add_check(ref, tag);
> if (!ref || !tag)
> - return;
> + return false;
>
> ref->ct = &tag->ct;
> + return true;
> }
>
> -static inline void alloc_tag_ref_set(union codetag_ref *ref, struct
> alloc_tag *tag)
> +static inline bool alloc_tag_ref_set(union codetag_ref *ref, struct
> alloc_tag *tag)
> {
> - __alloc_tag_ref_set(ref, tag);
> + if (unlikely(!__alloc_tag_ref_set(ref, tag)))
> + return false;
> +
> /*
> * We need in increment the call counter every time we have a new
> * allocation or when we split a large allocation into smaller ones.
> * Each new reference for every sub-allocation needs to increment call
> * counter because when we free each part the counter will be
> decremented.
> */
> this_cpu_inc(tag->counters->calls);
> + return true;
> }
>
> static inline void alloc_tag_add(union codetag_ref *ref, struct
> alloc_tag *tag, size_t bytes)
> {
> - alloc_tag_ref_set(ref, tag);
> - this_cpu_add(tag->counters->bytes, bytes);
> + if (likely(alloc_tag_ref_set(ref, tag)))
> + this_cpu_add(tag->counters->bytes, bytes);
> }
>
> Could you please confirm this fix?
After that, please feel free to add
Acked-by: Yu Zhao <yuzhao@google.com>
(Of course with Suggested-by: Suren).
> Thanks,
> Suren.
>
> > ref->ct = &tag->ct;
> > }
> >
> > @@ -158,6 +154,10 @@ static inline void alloc_tag_ref_set(union codetag_ref *ref, struct alloc_tag *t
> >
> > static inline void alloc_tag_add(union codetag_ref *ref, struct alloc_tag *tag, size_t bytes)
> > {
> > + alloc_tag_add_check(ref, tag);
> > + if (!ref || !tag)
> > + return;
> > +
> > alloc_tag_ref_set(ref, tag);
> > this_cpu_add(tag->counters->bytes, bytes);
> > }
> > --
> > 2.25.1
> >
^ permalink raw reply [flat|nested] 9+ messages in thread
* [PATCH] mm/codetag: fix null pointer check logic for ref and tag
2024-10-18 15:29 [PATCH] mm/codetag: move ref and tag null pointer check to alloc_tag_ref_set Hao Ge
2024-10-18 16:25 ` [PATCH v2] mm/codetag: move ref and tag null pointer check to alloc_tag_add Hao Ge
@ 2024-10-20 7:04 ` Hao Ge
2024-10-20 7:08 ` [PATCH v3] " Hao Ge
2 siblings, 0 replies; 9+ messages in thread
From: Hao Ge @ 2024-10-20 7:04 UTC (permalink / raw)
To: surenb, kent.overstreet, akpm
Cc: yuzhao, linux-mm, linux-kernel, hao.ge, Hao Ge
From: Hao Ge <gehao@kylinos.cn>
When we compile and load lib/slub_kunit.c,it will cause a panic.
The root cause is that __kmalloc_cache_noprof was directly called
instead of kmem_cache_alloc,which resulted in no alloc_tag being
allocated.This caused current->alloc_tag to be null,leading to a
null pointer dereference in alloc_tag_ref_set.
Despite the fact that my colleague Pei Xiao will later fix the code
in slub_kunit.c,we still need fix null pointer check logic for ref
and tag to avoid panic caused by a null pointer dereference.
Here is the log for the panic:
[ 74.779373][ T2158] Unable to handle kernel NULL pointer dereference at virtual address 0000000000000020
[ 74.780130][ T2158] Mem abort info:
[ 74.780406][ T2158] ESR = 0x0000000096000004
[ 74.780756][ T2158] EC = 0x25: DABT (current EL), IL = 32 bits
[ 74.781225][ T2158] SET = 0, FnV = 0
[ 74.781529][ T2158] EA = 0, S1PTW = 0
[ 74.781836][ T2158] FSC = 0x04: level 0 translation fault
[ 74.782288][ T2158] Data abort info:
[ 74.782577][ T2158] ISV = 0, ISS = 0x00000004, ISS2 = 0x00000000
[ 74.783068][ T2158] CM = 0, WnR = 0, TnD = 0, TagAccess = 0
[ 74.783533][ T2158] GCS = 0, Overlay = 0, DirtyBit = 0, Xs = 0
[ 74.784010][ T2158] user pgtable: 4k pages, 48-bit VAs, pgdp=0000000105f34000
[ 74.784586][ T2158] [0000000000000020] pgd=0000000000000000, p4d=0000000000000000
[ 74.785293][ T2158] Internal error: Oops: 0000000096000004 [#1] SMP
[ 74.785805][ T2158] Modules linked in: slub_kunit kunit ip6t_rpfilter ip6t_REJECT nf_reject_ipv6 ipt_REJECT nf_reject_ipv4 xt_conntrack ebtable_nat ebtable_broute ip6table_nat ip6table_mangle 4
[ 74.790661][ T2158] CPU: 0 UID: 0 PID: 2158 Comm: kunit_try_catch Kdump: loaded Tainted: G W N 6.12.0-rc3+ #2
[ 74.791535][ T2158] Tainted: [W]=WARN, [N]=TEST
[ 74.791889][ T2158] Hardware name: QEMU KVM Virtual Machine, BIOS 0.0.0 02/06/2015
[ 74.792479][ T2158] pstate: 40400005 (nZcv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--)
[ 74.793101][ T2158] pc : alloc_tagging_slab_alloc_hook+0x120/0x270
[ 74.793607][ T2158] lr : alloc_tagging_slab_alloc_hook+0x120/0x270
[ 74.794095][ T2158] sp : ffff800084d33cd0
[ 74.794418][ T2158] x29: ffff800084d33cd0 x28: 0000000000000000 x27: 0000000000000000
[ 74.795095][ T2158] x26: 0000000000000000 x25: 0000000000000012 x24: ffff80007b30e314
[ 74.795822][ T2158] x23: ffff000390ff6f10 x22: 0000000000000000 x21: 0000000000000088
[ 74.796555][ T2158] x20: ffff000390285840 x19: fffffd7fc3ef7830 x18: ffffffffffffffff
[ 74.797283][ T2158] x17: ffff8000800e63b4 x16: ffff80007b33afc4 x15: ffff800081654c00
[ 74.798011][ T2158] x14: 0000000000000000 x13: 205d383531325420 x12: 5b5d383734363537
[ 74.798744][ T2158] x11: ffff800084d337e0 x10: 000000000000005d x9 : 00000000ffffffd0
[ 74.799476][ T2158] x8 : 7f7f7f7f7f7f7f7f x7 : ffff80008219d188 x6 : c0000000ffff7fff
[ 74.800206][ T2158] x5 : ffff0003fdbc9208 x4 : ffff800081edd188 x3 : 0000000000000001
[ 74.800932][ T2158] x2 : 0beaa6dee1ac5a00 x1 : 0beaa6dee1ac5a00 x0 : ffff80037c2cb000
[ 74.801656][ T2158] Call trace:
[ 74.801954][ T2158] alloc_tagging_slab_alloc_hook+0x120/0x270
[ 74.802494][ T2158] __kmalloc_cache_noprof+0x148/0x33c
[ 74.802976][ T2158] test_kmalloc_redzone_access+0x4c/0x104 [slub_kunit]
[ 74.803607][ T2158] kunit_try_run_case+0x70/0x17c [kunit]
[ 74.804124][ T2158] kunit_generic_run_threadfn_adapter+0x2c/0x4c [kunit]
[ 74.804768][ T2158] kthread+0x10c/0x118
[ 74.805141][ T2158] ret_from_fork+0x10/0x20
[ 74.805540][ T2158] Code: b9400a80 11000400 b9000a80 97ffd858 (f94012d3)
[ 74.806176][ T2158] SMP: stopping secondary CPUs
[ 74.808130][ T2158] Starting crashdump kernel...
Fixes: e0a955bf7f61 ("mm/codetag: add pgalloc_tag_copy()")
Suggested-by: Suren Baghdasaryan <surenb@google.com>
Acked-by: Yu Zhao <yuzhao@google.com>
Signed-off-by: Hao Ge <gehao@kylinos.cn>
---
v3: Modify the code logic based on Suren's suggestions.
Update title and commit message due to code logic change.
Add Suggested-by: Suren Baghdasaryan <surenb@google.com>
Add Acked-by: Yu Zhao <yuzhao@google.com>
v2: Modify the errors in the title and commit message.
Remove the empty lines that were mistakenly added in version v1.
---
include/linux/alloc_tag.h | 16 ++++++++++------
1 file changed, 10 insertions(+), 6 deletions(-)
diff --git a/include/linux/alloc_tag.h b/include/linux/alloc_tag.h
index 1f0a9ff23a2c..941deffc590d 100644
--- a/include/linux/alloc_tag.h
+++ b/include/linux/alloc_tag.h
@@ -135,18 +135,21 @@ static inline void alloc_tag_sub_check(union codetag_ref *ref) {}
#endif
/* Caller should verify both ref and tag to be valid */
-static inline void __alloc_tag_ref_set(union codetag_ref *ref, struct alloc_tag *tag)
+static inline bool __alloc_tag_ref_set(union codetag_ref *ref, struct alloc_tag *tag)
{
alloc_tag_add_check(ref, tag);
if (!ref || !tag)
- return;
+ return false;
ref->ct = &tag->ct;
+ return true;
}
-static inline void alloc_tag_ref_set(union codetag_ref *ref, struct alloc_tag *tag)
+static inline bool alloc_tag_ref_set(union codetag_ref *ref, struct alloc_tag *tag)
{
- __alloc_tag_ref_set(ref, tag);
+ if (unlikely(!__alloc_tag_ref_set(ref, tag)))
+ return false;
+
/*
* We need in increment the call counter every time we have a new
* allocation or when we split a large allocation into smaller ones.
@@ -154,12 +157,13 @@ static inline void alloc_tag_ref_set(union codetag_ref *ref, struct alloc_tag *t
* counter because when we free each part the counter will be decremented.
*/
this_cpu_inc(tag->counters->calls);
+ return true;
}
static inline void alloc_tag_add(union codetag_ref *ref, struct alloc_tag *tag, size_t bytes)
{
- alloc_tag_ref_set(ref, tag);
- this_cpu_add(tag->counters->bytes, bytes);
+ if (likely(alloc_tag_ref_set(ref, tag)))
+ this_cpu_add(tag->counters->bytes, bytes);
}
static inline void alloc_tag_sub(union codetag_ref *ref, size_t bytes)
--
2.25.1
^ permalink raw reply [flat|nested] 9+ messages in thread* [PATCH v3] mm/codetag: fix null pointer check logic for ref and tag
2024-10-18 15:29 [PATCH] mm/codetag: move ref and tag null pointer check to alloc_tag_ref_set Hao Ge
2024-10-18 16:25 ` [PATCH v2] mm/codetag: move ref and tag null pointer check to alloc_tag_add Hao Ge
2024-10-20 7:04 ` [PATCH] mm/codetag: fix null pointer check logic for ref and tag Hao Ge
@ 2024-10-20 7:08 ` Hao Ge
2024-10-21 18:11 ` Suren Baghdasaryan
2 siblings, 1 reply; 9+ messages in thread
From: Hao Ge @ 2024-10-20 7:08 UTC (permalink / raw)
To: surenb, kent.overstreet, akpm
Cc: yuzhao, linux-mm, linux-kernel, hao.ge, Hao Ge
From: Hao Ge <gehao@kylinos.cn>
When we compile and load lib/slub_kunit.c,it will cause a panic.
The root cause is that __kmalloc_cache_noprof was directly called
instead of kmem_cache_alloc,which resulted in no alloc_tag being
allocated.This caused current->alloc_tag to be null,leading to a
null pointer dereference in alloc_tag_ref_set.
Despite the fact that my colleague Pei Xiao will later fix the code
in slub_kunit.c,we still need fix null pointer check logic for ref
and tag to avoid panic caused by a null pointer dereference.
Here is the log for the panic:
[ 74.779373][ T2158] Unable to handle kernel NULL pointer dereference at virtual address 0000000000000020
[ 74.780130][ T2158] Mem abort info:
[ 74.780406][ T2158] ESR = 0x0000000096000004
[ 74.780756][ T2158] EC = 0x25: DABT (current EL), IL = 32 bits
[ 74.781225][ T2158] SET = 0, FnV = 0
[ 74.781529][ T2158] EA = 0, S1PTW = 0
[ 74.781836][ T2158] FSC = 0x04: level 0 translation fault
[ 74.782288][ T2158] Data abort info:
[ 74.782577][ T2158] ISV = 0, ISS = 0x00000004, ISS2 = 0x00000000
[ 74.783068][ T2158] CM = 0, WnR = 0, TnD = 0, TagAccess = 0
[ 74.783533][ T2158] GCS = 0, Overlay = 0, DirtyBit = 0, Xs = 0
[ 74.784010][ T2158] user pgtable: 4k pages, 48-bit VAs, pgdp=0000000105f34000
[ 74.784586][ T2158] [0000000000000020] pgd=0000000000000000, p4d=0000000000000000
[ 74.785293][ T2158] Internal error: Oops: 0000000096000004 [#1] SMP
[ 74.785805][ T2158] Modules linked in: slub_kunit kunit ip6t_rpfilter ip6t_REJECT nf_reject_ipv6 ipt_REJECT nf_reject_ipv4 xt_conntrack ebtable_nat ebtable_broute ip6table_nat ip6table_mangle 4
[ 74.790661][ T2158] CPU: 0 UID: 0 PID: 2158 Comm: kunit_try_catch Kdump: loaded Tainted: G W N 6.12.0-rc3+ #2
[ 74.791535][ T2158] Tainted: [W]=WARN, [N]=TEST
[ 74.791889][ T2158] Hardware name: QEMU KVM Virtual Machine, BIOS 0.0.0 02/06/2015
[ 74.792479][ T2158] pstate: 40400005 (nZcv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--)
[ 74.793101][ T2158] pc : alloc_tagging_slab_alloc_hook+0x120/0x270
[ 74.793607][ T2158] lr : alloc_tagging_slab_alloc_hook+0x120/0x270
[ 74.794095][ T2158] sp : ffff800084d33cd0
[ 74.794418][ T2158] x29: ffff800084d33cd0 x28: 0000000000000000 x27: 0000000000000000
[ 74.795095][ T2158] x26: 0000000000000000 x25: 0000000000000012 x24: ffff80007b30e314
[ 74.795822][ T2158] x23: ffff000390ff6f10 x22: 0000000000000000 x21: 0000000000000088
[ 74.796555][ T2158] x20: ffff000390285840 x19: fffffd7fc3ef7830 x18: ffffffffffffffff
[ 74.797283][ T2158] x17: ffff8000800e63b4 x16: ffff80007b33afc4 x15: ffff800081654c00
[ 74.798011][ T2158] x14: 0000000000000000 x13: 205d383531325420 x12: 5b5d383734363537
[ 74.798744][ T2158] x11: ffff800084d337e0 x10: 000000000000005d x9 : 00000000ffffffd0
[ 74.799476][ T2158] x8 : 7f7f7f7f7f7f7f7f x7 : ffff80008219d188 x6 : c0000000ffff7fff
[ 74.800206][ T2158] x5 : ffff0003fdbc9208 x4 : ffff800081edd188 x3 : 0000000000000001
[ 74.800932][ T2158] x2 : 0beaa6dee1ac5a00 x1 : 0beaa6dee1ac5a00 x0 : ffff80037c2cb000
[ 74.801656][ T2158] Call trace:
[ 74.801954][ T2158] alloc_tagging_slab_alloc_hook+0x120/0x270
[ 74.802494][ T2158] __kmalloc_cache_noprof+0x148/0x33c
[ 74.802976][ T2158] test_kmalloc_redzone_access+0x4c/0x104 [slub_kunit]
[ 74.803607][ T2158] kunit_try_run_case+0x70/0x17c [kunit]
[ 74.804124][ T2158] kunit_generic_run_threadfn_adapter+0x2c/0x4c [kunit]
[ 74.804768][ T2158] kthread+0x10c/0x118
[ 74.805141][ T2158] ret_from_fork+0x10/0x20
[ 74.805540][ T2158] Code: b9400a80 11000400 b9000a80 97ffd858 (f94012d3)
[ 74.806176][ T2158] SMP: stopping secondary CPUs
[ 74.808130][ T2158] Starting crashdump kernel...
Fixes: e0a955bf7f61 ("mm/codetag: add pgalloc_tag_copy()")
Suggested-by: Suren Baghdasaryan <surenb@google.com>
Acked-by: Yu Zhao <yuzhao@google.com>
Signed-off-by: Hao Ge <gehao@kylinos.cn>
---
v3: Modify the code logic based on Suren's suggestions.
Update title and commit message due to code logic change.
Add Suggested-by: Suren Baghdasaryan <surenb@google.com>
Add Acked-by: Yu Zhao <yuzhao@google.com>
v2: Modify the errors in the title and commit message.
Remove the empty lines that were mistakenly added in version v1.
---
include/linux/alloc_tag.h | 16 ++++++++++------
1 file changed, 10 insertions(+), 6 deletions(-)
diff --git a/include/linux/alloc_tag.h b/include/linux/alloc_tag.h
index 1f0a9ff23a2c..941deffc590d 100644
--- a/include/linux/alloc_tag.h
+++ b/include/linux/alloc_tag.h
@@ -135,18 +135,21 @@ static inline void alloc_tag_sub_check(union codetag_ref *ref) {}
#endif
/* Caller should verify both ref and tag to be valid */
-static inline void __alloc_tag_ref_set(union codetag_ref *ref, struct alloc_tag *tag)
+static inline bool __alloc_tag_ref_set(union codetag_ref *ref, struct alloc_tag *tag)
{
alloc_tag_add_check(ref, tag);
if (!ref || !tag)
- return;
+ return false;
ref->ct = &tag->ct;
+ return true;
}
-static inline void alloc_tag_ref_set(union codetag_ref *ref, struct alloc_tag *tag)
+static inline bool alloc_tag_ref_set(union codetag_ref *ref, struct alloc_tag *tag)
{
- __alloc_tag_ref_set(ref, tag);
+ if (unlikely(!__alloc_tag_ref_set(ref, tag)))
+ return false;
+
/*
* We need in increment the call counter every time we have a new
* allocation or when we split a large allocation into smaller ones.
@@ -154,12 +157,13 @@ static inline void alloc_tag_ref_set(union codetag_ref *ref, struct alloc_tag *t
* counter because when we free each part the counter will be decremented.
*/
this_cpu_inc(tag->counters->calls);
+ return true;
}
static inline void alloc_tag_add(union codetag_ref *ref, struct alloc_tag *tag, size_t bytes)
{
- alloc_tag_ref_set(ref, tag);
- this_cpu_add(tag->counters->bytes, bytes);
+ if (likely(alloc_tag_ref_set(ref, tag)))
+ this_cpu_add(tag->counters->bytes, bytes);
}
static inline void alloc_tag_sub(union codetag_ref *ref, size_t bytes)
--
2.25.1
^ permalink raw reply [flat|nested] 9+ messages in thread* Re: [PATCH v3] mm/codetag: fix null pointer check logic for ref and tag
2024-10-20 7:08 ` [PATCH v3] " Hao Ge
@ 2024-10-21 18:11 ` Suren Baghdasaryan
0 siblings, 0 replies; 9+ messages in thread
From: Suren Baghdasaryan @ 2024-10-21 18:11 UTC (permalink / raw)
To: Hao Ge; +Cc: kent.overstreet, akpm, yuzhao, linux-mm, linux-kernel, Hao Ge
On Sun, Oct 20, 2024 at 12:08 AM Hao Ge <hao.ge@linux.dev> wrote:
>
> From: Hao Ge <gehao@kylinos.cn>
>
> When we compile and load lib/slub_kunit.c,it will cause a panic.
>
> The root cause is that __kmalloc_cache_noprof was directly called
> instead of kmem_cache_alloc,which resulted in no alloc_tag being
> allocated.This caused current->alloc_tag to be null,leading to a
> null pointer dereference in alloc_tag_ref_set.
>
> Despite the fact that my colleague Pei Xiao will later fix the code
> in slub_kunit.c,we still need fix null pointer check logic for ref
> and tag to avoid panic caused by a null pointer dereference.
>
> Here is the log for the panic:
>
> [ 74.779373][ T2158] Unable to handle kernel NULL pointer dereference at virtual address 0000000000000020
> [ 74.780130][ T2158] Mem abort info:
> [ 74.780406][ T2158] ESR = 0x0000000096000004
> [ 74.780756][ T2158] EC = 0x25: DABT (current EL), IL = 32 bits
> [ 74.781225][ T2158] SET = 0, FnV = 0
> [ 74.781529][ T2158] EA = 0, S1PTW = 0
> [ 74.781836][ T2158] FSC = 0x04: level 0 translation fault
> [ 74.782288][ T2158] Data abort info:
> [ 74.782577][ T2158] ISV = 0, ISS = 0x00000004, ISS2 = 0x00000000
> [ 74.783068][ T2158] CM = 0, WnR = 0, TnD = 0, TagAccess = 0
> [ 74.783533][ T2158] GCS = 0, Overlay = 0, DirtyBit = 0, Xs = 0
> [ 74.784010][ T2158] user pgtable: 4k pages, 48-bit VAs, pgdp=0000000105f34000
> [ 74.784586][ T2158] [0000000000000020] pgd=0000000000000000, p4d=0000000000000000
> [ 74.785293][ T2158] Internal error: Oops: 0000000096000004 [#1] SMP
> [ 74.785805][ T2158] Modules linked in: slub_kunit kunit ip6t_rpfilter ip6t_REJECT nf_reject_ipv6 ipt_REJECT nf_reject_ipv4 xt_conntrack ebtable_nat ebtable_broute ip6table_nat ip6table_mangle 4
> [ 74.790661][ T2158] CPU: 0 UID: 0 PID: 2158 Comm: kunit_try_catch Kdump: loaded Tainted: G W N 6.12.0-rc3+ #2
> [ 74.791535][ T2158] Tainted: [W]=WARN, [N]=TEST
> [ 74.791889][ T2158] Hardware name: QEMU KVM Virtual Machine, BIOS 0.0.0 02/06/2015
> [ 74.792479][ T2158] pstate: 40400005 (nZcv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--)
> [ 74.793101][ T2158] pc : alloc_tagging_slab_alloc_hook+0x120/0x270
> [ 74.793607][ T2158] lr : alloc_tagging_slab_alloc_hook+0x120/0x270
> [ 74.794095][ T2158] sp : ffff800084d33cd0
> [ 74.794418][ T2158] x29: ffff800084d33cd0 x28: 0000000000000000 x27: 0000000000000000
> [ 74.795095][ T2158] x26: 0000000000000000 x25: 0000000000000012 x24: ffff80007b30e314
> [ 74.795822][ T2158] x23: ffff000390ff6f10 x22: 0000000000000000 x21: 0000000000000088
> [ 74.796555][ T2158] x20: ffff000390285840 x19: fffffd7fc3ef7830 x18: ffffffffffffffff
> [ 74.797283][ T2158] x17: ffff8000800e63b4 x16: ffff80007b33afc4 x15: ffff800081654c00
> [ 74.798011][ T2158] x14: 0000000000000000 x13: 205d383531325420 x12: 5b5d383734363537
> [ 74.798744][ T2158] x11: ffff800084d337e0 x10: 000000000000005d x9 : 00000000ffffffd0
> [ 74.799476][ T2158] x8 : 7f7f7f7f7f7f7f7f x7 : ffff80008219d188 x6 : c0000000ffff7fff
> [ 74.800206][ T2158] x5 : ffff0003fdbc9208 x4 : ffff800081edd188 x3 : 0000000000000001
> [ 74.800932][ T2158] x2 : 0beaa6dee1ac5a00 x1 : 0beaa6dee1ac5a00 x0 : ffff80037c2cb000
> [ 74.801656][ T2158] Call trace:
> [ 74.801954][ T2158] alloc_tagging_slab_alloc_hook+0x120/0x270
> [ 74.802494][ T2158] __kmalloc_cache_noprof+0x148/0x33c
> [ 74.802976][ T2158] test_kmalloc_redzone_access+0x4c/0x104 [slub_kunit]
> [ 74.803607][ T2158] kunit_try_run_case+0x70/0x17c [kunit]
> [ 74.804124][ T2158] kunit_generic_run_threadfn_adapter+0x2c/0x4c [kunit]
> [ 74.804768][ T2158] kthread+0x10c/0x118
> [ 74.805141][ T2158] ret_from_fork+0x10/0x20
> [ 74.805540][ T2158] Code: b9400a80 11000400 b9000a80 97ffd858 (f94012d3)
> [ 74.806176][ T2158] SMP: stopping secondary CPUs
> [ 74.808130][ T2158] Starting crashdump kernel...
>
> Fixes: e0a955bf7f61 ("mm/codetag: add pgalloc_tag_copy()")
> Suggested-by: Suren Baghdasaryan <surenb@google.com>
> Acked-by: Yu Zhao <yuzhao@google.com>
> Signed-off-by: Hao Ge <gehao@kylinos.cn>
Thanks for the fix!
Acked-by: Suren Baghdasaryan <surenb@google.com>
> ---
> v3: Modify the code logic based on Suren's suggestions.
> Update title and commit message due to code logic change.
> Add Suggested-by: Suren Baghdasaryan <surenb@google.com>
> Add Acked-by: Yu Zhao <yuzhao@google.com>
>
> v2: Modify the errors in the title and commit message.
> Remove the empty lines that were mistakenly added in version v1.
> ---
> include/linux/alloc_tag.h | 16 ++++++++++------
> 1 file changed, 10 insertions(+), 6 deletions(-)
>
> diff --git a/include/linux/alloc_tag.h b/include/linux/alloc_tag.h
> index 1f0a9ff23a2c..941deffc590d 100644
> --- a/include/linux/alloc_tag.h
> +++ b/include/linux/alloc_tag.h
> @@ -135,18 +135,21 @@ static inline void alloc_tag_sub_check(union codetag_ref *ref) {}
> #endif
>
> /* Caller should verify both ref and tag to be valid */
> -static inline void __alloc_tag_ref_set(union codetag_ref *ref, struct alloc_tag *tag)
> +static inline bool __alloc_tag_ref_set(union codetag_ref *ref, struct alloc_tag *tag)
> {
> alloc_tag_add_check(ref, tag);
> if (!ref || !tag)
> - return;
> + return false;
>
> ref->ct = &tag->ct;
> + return true;
> }
>
> -static inline void alloc_tag_ref_set(union codetag_ref *ref, struct alloc_tag *tag)
> +static inline bool alloc_tag_ref_set(union codetag_ref *ref, struct alloc_tag *tag)
> {
> - __alloc_tag_ref_set(ref, tag);
> + if (unlikely(!__alloc_tag_ref_set(ref, tag)))
> + return false;
> +
> /*
> * We need in increment the call counter every time we have a new
> * allocation or when we split a large allocation into smaller ones.
> @@ -154,12 +157,13 @@ static inline void alloc_tag_ref_set(union codetag_ref *ref, struct alloc_tag *t
> * counter because when we free each part the counter will be decremented.
> */
> this_cpu_inc(tag->counters->calls);
> + return true;
> }
>
> static inline void alloc_tag_add(union codetag_ref *ref, struct alloc_tag *tag, size_t bytes)
> {
> - alloc_tag_ref_set(ref, tag);
> - this_cpu_add(tag->counters->bytes, bytes);
> + if (likely(alloc_tag_ref_set(ref, tag)))
> + this_cpu_add(tag->counters->bytes, bytes);
> }
>
> static inline void alloc_tag_sub(union codetag_ref *ref, size_t bytes)
> --
> 2.25.1
>
^ permalink raw reply [flat|nested] 9+ messages in thread