* [PATCH] RAS: Fix out-of-range section_length in ARM processor error handling
@ 2026-08-20 13:18 Abbott Liu
2026-08-22 9:49 ` Hanjun Guo
0 siblings, 1 reply; 3+ messages in thread
From: Abbott Liu @ 2026-08-20 13:18 UTC (permalink / raw)
To: tony.luck, bp, rafael.j.wysocki, mchehab+huawei, ardb, jic23,
guohanjun, linux-edac, linux-kernel
Cc: yangzhuohao1, douzhaolei, zouyipeng, wangbing6, nixiaoming,
liuwenliang
Commit 87880af2d24e ("APEI/GHES: ARM processor Error: don't go past
allocated memory") added bounds checks for malformed ARM processor
error records but contained several bugs:
- In log_arm_hw_error(), the ctx_info bounds check is inverted. The
condition `sz + (long)ctx_info - (long)err >= err->section_length`
adds ctx_info->size when the context header is already past the end
of the section instead of when it is within bounds. So change the
comparison to <=.
- The vsei_len < 0 error path did not verify the pei_len and ctx_len.
When vsei_len is negative, section_length is too small to hold the
full record, yet pei_len and ctx_len were derived from
err_info_num/context_info_num and may describe regions beyond the
(long)err .. err + section_length buffer. To prevent trace_arm_event
from reading past the allocated record, sanitize the parameters:
move the cpu lookup above this path so it is available for tracing,
recalculate ctx_len and pei_len based on section_length, limit them,
set the corresponding pointers to NULL and lengths to 0 when there
is no remaining space.
Fixes: 87880af2d24e ("APEI/GHES: ARM processor Error: don't go past allocated memory")
Signed-off-by: Abbott Liu <liuwenliang@huawei.com>
---
drivers/ras/ras.c | 34 ++++++++++++++++++++++------------
1 file changed, 22 insertions(+), 12 deletions(-)
diff --git a/drivers/ras/ras.c b/drivers/ras/ras.c
index 03df3db62334..e37bf11d0926 100644
--- a/drivers/ras/ras.c
+++ b/drivers/ras/ras.c
@@ -58,10 +58,10 @@ void log_arm_hw_error(struct cper_sec_proc_arm *err, const u8 sev)
struct cper_arm_err_info *err_info;
struct cper_arm_ctx_info *ctx_info;
u8 *ven_err_data;
- u32 ctx_len = 0;
+ s32 ctx_len = 0;
int n, sz, cpu;
s32 vsei_len;
- u32 pei_len;
+ s32 pei_len;
u8 *pei_err, *ctx_err;
pei_len = sizeof(struct cper_arm_err_info) * err->err_info_num;
@@ -74,27 +74,37 @@ void log_arm_hw_error(struct cper_sec_proc_arm *err, const u8 sev)
for (n = 0; n < err->context_info_num; n++) {
sz = sizeof(struct cper_arm_ctx_info);
- if (sz + (long)ctx_info - (long)err >= err->section_length)
+ if (sz + (long)ctx_info - (long)err <= err->section_length)
sz += ctx_info->size;
-
ctx_info = (struct cper_arm_ctx_info *)((long)ctx_info + sz);
ctx_len += sz;
}
+ cpu = GET_LOGICAL_INDEX(err->mpidr);
+ if (cpu < 0)
+ cpu = -1;
+
vsei_len = err->section_length - (sizeof(struct cper_sec_proc_arm) + pei_len + ctx_len);
if (vsei_len < 0) {
pr_warn(FW_BUG "section length: %d\n", err->section_length);
pr_warn(FW_BUG "section length is too small\n");
pr_warn(FW_BUG "firmware-generated error record is incorrect\n");
vsei_len = 0;
- }
- ven_err_data = (u8 *)ctx_info;
-
- cpu = GET_LOGICAL_INDEX(err->mpidr);
- if (cpu < 0)
- cpu = -1;
-
- trace_arm_event(err, pei_err, pei_len, ctx_err, ctx_len,
+ ven_err_data = NULL;
+ ctx_len = err->section_length - (sizeof(struct cper_sec_proc_arm) + pei_len);
+ if (ctx_len < 0) {
+ ctx_len = 0;
+ ctx_err = NULL;
+ pei_len = err->section_length - sizeof(struct cper_sec_proc_arm);
+ if (pei_len < 0) {
+ pei_len = 0;
+ pei_err = NULL;
+ }
+ }
+ } else
+ ven_err_data = (u8 *)ctx_info;
+
+ trace_arm_event(err, pei_err, (u32)pei_len, ctx_err, (u32)ctx_len,
ven_err_data, (u32)vsei_len, sev, cpu);
}
--
2.43.0
^ permalink raw reply related [flat|nested] 3+ messages in thread
* Re: [PATCH] RAS: Fix out-of-range section_length in ARM processor error handling
2026-08-20 13:18 Abbott Liu
@ 2026-08-22 9:49 ` Hanjun Guo
0 siblings, 0 replies; 3+ messages in thread
From: Hanjun Guo @ 2026-08-22 9:49 UTC (permalink / raw)
To: Abbott Liu, tony.luck, bp, rafael.j.wysocki, mchehab+huawei, ardb,
jic23, linux-edac, linux-kernel
Cc: yangzhuohao1, douzhaolei, zouyipeng, wangbing6, nixiaoming
Hi Abbott,
On 2026/8/20 21:18, Abbott Liu wrote:
> Commit 87880af2d24e ("APEI/GHES: ARM processor Error: don't go past
> allocated memory") added bounds checks for malformed ARM processor
> error records but contained several bugs:
>
> - In log_arm_hw_error(), the ctx_info bounds check is inverted. The
> condition `sz + (long)ctx_info - (long)err >= err->section_length`
> adds ctx_info->size when the context header is already past the end
> of the section instead of when it is within bounds. So change the
> comparison to <=.
>
> - The vsei_len < 0 error path did not verify the pei_len and ctx_len.
> When vsei_len is negative, section_length is too small to hold the
> full record, yet pei_len and ctx_len were derived from
> err_info_num/context_info_num and may describe regions beyond the
> (long)err .. err + section_length buffer. To prevent trace_arm_event
> from reading past the allocated record, sanitize the parameters:
> move the cpu lookup above this path so it is available for tracing,
> recalculate ctx_len and pei_len based on section_length, limit them,
> set the corresponding pointers to NULL and lengths to 0 when there
> is no remaining space.
Please do this in two patches, one for a single issue fix.
>
> Fixes: 87880af2d24e ("APEI/GHES: ARM processor Error: don't go past allocated memory")
> Signed-off-by: Abbott Liu <liuwenliang@huawei.com>
> ---
> drivers/ras/ras.c | 34 ++++++++++++++++++++++------------
> 1 file changed, 22 insertions(+), 12 deletions(-)
>
> diff --git a/drivers/ras/ras.c b/drivers/ras/ras.c
> index 03df3db62334..e37bf11d0926 100644
> --- a/drivers/ras/ras.c
> +++ b/drivers/ras/ras.c
> @@ -58,10 +58,10 @@ void log_arm_hw_error(struct cper_sec_proc_arm *err, const u8 sev)
> struct cper_arm_err_info *err_info;
> struct cper_arm_ctx_info *ctx_info;
> u8 *ven_err_data;
> - u32 ctx_len = 0;
> + s32 ctx_len = 0;
> int n, sz, cpu;
> s32 vsei_len;
> - u32 pei_len;
> + s32 pei_len;
> u8 *pei_err, *ctx_err;
>
> pei_len = sizeof(struct cper_arm_err_info) * err->err_info_num;
> @@ -74,27 +74,37 @@ void log_arm_hw_error(struct cper_sec_proc_arm *err, const u8 sev)
> for (n = 0; n < err->context_info_num; n++) {
> sz = sizeof(struct cper_arm_ctx_info);
>
> - if (sz + (long)ctx_info - (long)err >= err->section_length)
> + if (sz + (long)ctx_info - (long)err <= err->section_length)
> sz += ctx_info->size;
> -
Please leave a empty line here as it is.
Thanks
Hanjun
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH] RAS: Fix out-of-range section_length in ARM processor error handling
@ 2026-08-25 12:21 Liuwenliang (Abbott Liu)
0 siblings, 0 replies; 3+ messages in thread
From: Liuwenliang (Abbott Liu) @ 2026-08-25 12:21 UTC (permalink / raw)
To: Guohanjun (Hanjun Guo), tony.luck@intel.com, bp@alien8.de,
rafael.j.wysocki@intel.com, mchehab+huawei@kernel.org,
ardb@kernel.org, jic23@kernel.org, linux-edac@vger.kernel.org,
linux-kernel@vger.kernel.org
Cc: yangzhuohao (A), douzhaolei, zouyipeng, Wangbing, Nixiaoming,
Liuwenliang (Abbott Liu)
Hi Hanjun, thanks for your review.
>Hi Abbott,
>
>On 2026/8/20 21:18, Abbott Liu wrote:
>> Commit 87880af2d24e ("APEI/GHES: ARM processor Error: don't go past
>> allocated memory") added bounds checks for malformed ARM processor
>> error records but contained several bugs:
>>
>> - In log_arm_hw_error(), the ctx_info bounds check is inverted. The
>> condition `sz + (long)ctx_info - (long)err >= err->section_length`
>> adds ctx_info->size when the context header is already past the end
>> of the section instead of when it is within bounds. So change the
>> comparison to <=.
>>
>> - The vsei_len < 0 error path did not verify the pei_len and ctx_len.
>> When vsei_len is negative, section_length is too small to hold the
>> full record, yet pei_len and ctx_len were derived from
>> err_info_num/context_info_num and may describe regions beyond the
>> (long)err .. err + section_length buffer. To prevent trace_arm_event
>> from reading past the allocated record, sanitize the parameters:
>> move the cpu lookup above this path so it is available for tracing,
>> recalculate ctx_len and pei_len based on section_length, limit them,
>> set the corresponding pointers to NULL and lengths to 0 when there
>> is no remaining space.
>
>Please do this in two patches, one for a single issue fix.
I will change it in next version.
>>
>> Fixes: 87880af2d24e ("APEI/GHES: ARM processor Error: don't go past allocated memory")
>> Signed-off-by: Abbott Liu <liuwenliang@huawei.com>
>> ---
>> drivers/ras/ras.c | 34 ++++++++++++++++++++++------------
>> 1 file changed, 22 insertions(+), 12 deletions(-)
>>
>> diff --git a/drivers/ras/ras.c b/drivers/ras/ras.c
>> index 03df3db62334..e37bf11d0926 100644
>> --- a/drivers/ras/ras.c
>> +++ b/drivers/ras/ras.c
>> @@ -58,10 +58,10 @@ void log_arm_hw_error(struct cper_sec_proc_arm *err, const u8 sev)
>> struct cper_arm_err_info *err_info;
>> struct cper_arm_ctx_info *ctx_info;
>> u8 *ven_err_data;
>> - u32 ctx_len = 0;
>> + s32 ctx_len = 0;
>> int n, sz, cpu;
>> s32 vsei_len;
>> - u32 pei_len;
>> + s32 pei_len;
>> u8 *pei_err, *ctx_err;
>>
>> pei_len = sizeof(struct cper_arm_err_info) * err->err_info_num;
>> @@ -74,27 +74,37 @@ void log_arm_hw_error(struct cper_sec_proc_arm *err, const u8 sev)
>> for (n = 0; n < err->context_info_num; n++) {
>> sz = sizeof(struct cper_arm_ctx_info);
>>
>> - if (sz + (long)ctx_info - (long)err >= err->section_length)
>> + if (sz + (long)ctx_info - (long)err <= err->section_length)
>> sz += ctx_info->size;
>> -
>
>Please leave a empty line here as it is.
I will change it in next version.
>Thanks
>Hanjun
>
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2026-08-25 12:21 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-08-25 12:21 [PATCH] RAS: Fix out-of-range section_length in ARM processor error handling Liuwenliang (Abbott Liu)
-- strict thread matches above, loose matches on Subject: below --
2026-08-20 13:18 Abbott Liu
2026-08-22 9:49 ` Hanjun Guo
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox