* [LTP] [PATCH] genload: fix memory corruption in hogvm
@ 2024-06-10 8:56 Jiwei Sun
0 siblings, 0 replies; 4+ messages in thread
From: Jiwei Sun @ 2024-06-10 8:56 UTC (permalink / raw)
To: ltp, sunjw10; +Cc: ahuang12
From: Jiwei Sun <sunjw10@lenovo.com>
With the following command for doing memory stress test,
./genload -v --vm 10 --vm-chunks 4 --vm-bytes 1073741824
Some memory corruption issue was triggered,
malloc(): corrupted top size
The root cause of the issue is that allocated memory for ptr is less
than what is actually needed.
Signed-off-by: Jiwei Sun <sunjw10@lenovo.com>
---
tools/genload/genload.c | 2 +-
tools/genload/stress.c | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/tools/genload/genload.c b/tools/genload/genload.c
index 7f56d5272..9712e7828 100644
--- a/tools/genload/genload.c
+++ b/tools/genload/genload.c
@@ -642,7 +642,7 @@ int hogvm(long long forks, long long chunks, long long bytes)
usleep(backoff);
while (1) {
- ptr = (char **)malloc(chunks * 2);
+ ptr = (char **)malloc(chunks * 2 * sizeof(char *));
for (j = 0; chunks == 0 || j < chunks; j++) {
if ((ptr[j] =
(char *)malloc(bytes *
diff --git a/tools/genload/stress.c b/tools/genload/stress.c
index 7f56d5272..9712e7828 100644
--- a/tools/genload/stress.c
+++ b/tools/genload/stress.c
@@ -642,7 +642,7 @@ int hogvm(long long forks, long long chunks, long long bytes)
usleep(backoff);
while (1) {
- ptr = (char **)malloc(chunks * 2);
+ ptr = (char **)malloc(chunks * 2 * sizeof(char *));
for (j = 0; chunks == 0 || j < chunks; j++) {
if ((ptr[j] =
(char *)malloc(bytes *
--
2.27.0
--
Mailing list info: https://lists.linux.it/listinfo/ltp
^ permalink raw reply related [flat|nested] 4+ messages in thread* [LTP] [PATCH] genload: fix memory corruption in hogvm
@ 2024-06-11 2:58 Jiwei Sun
2024-06-11 12:11 ` Cyril Hrubis
0 siblings, 1 reply; 4+ messages in thread
From: Jiwei Sun @ 2024-06-11 2:58 UTC (permalink / raw)
To: ltp; +Cc: ahuang12
From: Jiwei Sun <sunjw10@lenovo.com>
With the following command for doing memory stress test,
./genload -v --vm 10 --vm-chunks 4 --vm-bytes 1073741824
Some memory corruption issue was triggered,
malloc(): corrupted top size
The root cause of the issue is that allocated memory for ptr is less
than what is actually needed.
Signed-off-by: Jiwei Sun <sunjw10@lenovo.com>
---
tools/genload/genload.c | 2 +-
tools/genload/stress.c | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/tools/genload/genload.c b/tools/genload/genload.c
index 7f56d5272..9712e7828 100644
--- a/tools/genload/genload.c
+++ b/tools/genload/genload.c
@@ -642,7 +642,7 @@ int hogvm(long long forks, long long chunks, long long bytes)
usleep(backoff);
while (1) {
- ptr = (char **)malloc(chunks * 2);
+ ptr = (char **)malloc(chunks * 2 * sizeof(char *));
for (j = 0; chunks == 0 || j < chunks; j++) {
if ((ptr[j] =
(char *)malloc(bytes *
diff --git a/tools/genload/stress.c b/tools/genload/stress.c
index 7f56d5272..9712e7828 100644
--- a/tools/genload/stress.c
+++ b/tools/genload/stress.c
@@ -642,7 +642,7 @@ int hogvm(long long forks, long long chunks, long long bytes)
usleep(backoff);
while (1) {
- ptr = (char **)malloc(chunks * 2);
+ ptr = (char **)malloc(chunks * 2 * sizeof(char *));
for (j = 0; chunks == 0 || j < chunks; j++) {
if ((ptr[j] =
(char *)malloc(bytes *
--
2.27.0
--
Mailing list info: https://lists.linux.it/listinfo/ltp
^ permalink raw reply related [flat|nested] 4+ messages in thread* Re: [LTP] [PATCH] genload: fix memory corruption in hogvm
2024-06-11 2:58 Jiwei Sun
@ 2024-06-11 12:11 ` Cyril Hrubis
2024-06-11 12:34 ` Jiwei Sun
0 siblings, 1 reply; 4+ messages in thread
From: Cyril Hrubis @ 2024-06-11 12:11 UTC (permalink / raw)
To: Jiwei Sun; +Cc: ahuang12, ltp
Hi!
> Signed-off-by: Jiwei Sun <sunjw10@lenovo.com>
> ---
> tools/genload/genload.c | 2 +-
> tools/genload/stress.c | 2 +-
> 2 files changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/tools/genload/genload.c b/tools/genload/genload.c
> index 7f56d5272..9712e7828 100644
> --- a/tools/genload/genload.c
> +++ b/tools/genload/genload.c
> @@ -642,7 +642,7 @@ int hogvm(long long forks, long long chunks, long long bytes)
> usleep(backoff);
>
> while (1) {
> - ptr = (char **)malloc(chunks * 2);
> + ptr = (char **)malloc(chunks * 2 * sizeof(char *));
Good catch, however shouldn't this be just chunks * sizeof(char*) ?
> for (j = 0; chunks == 0 || j < chunks; j++) {
> if ((ptr[j] =
> (char *)malloc(bytes *
> diff --git a/tools/genload/stress.c b/tools/genload/stress.c
> index 7f56d5272..9712e7828 100644
> --- a/tools/genload/stress.c
> +++ b/tools/genload/stress.c
> @@ -642,7 +642,7 @@ int hogvm(long long forks, long long chunks, long long bytes)
> usleep(backoff);
>
> while (1) {
> - ptr = (char **)malloc(chunks * 2);
> + ptr = (char **)malloc(chunks * 2 * sizeof(char *));
Here as well.
> for (j = 0; chunks == 0 || j < chunks; j++) {
> if ((ptr[j] =
> (char *)malloc(bytes *
> --
> 2.27.0
>
>
> --
> Mailing list info: https://lists.linux.it/listinfo/ltp
--
Cyril Hrubis
chrubis@suse.cz
--
Mailing list info: https://lists.linux.it/listinfo/ltp
^ permalink raw reply [flat|nested] 4+ messages in thread* Re: [LTP] [PATCH] genload: fix memory corruption in hogvm
2024-06-11 12:11 ` Cyril Hrubis
@ 2024-06-11 12:34 ` Jiwei Sun
0 siblings, 0 replies; 4+ messages in thread
From: Jiwei Sun @ 2024-06-11 12:34 UTC (permalink / raw)
To: Cyril Hrubis; +Cc: ahuang12, ltp
Hi Cyril,
On 6/11/24 20:11, Cyril Hrubis wrote:
> Hi!
>> Signed-off-by: Jiwei Sun <sunjw10@lenovo.com>
>> ---
>> tools/genload/genload.c | 2 +-
>> tools/genload/stress.c | 2 +-
>> 2 files changed, 2 insertions(+), 2 deletions(-)
>>
>> diff --git a/tools/genload/genload.c b/tools/genload/genload.c
>> index 7f56d5272..9712e7828 100644
>> --- a/tools/genload/genload.c
>> +++ b/tools/genload/genload.c
>> @@ -642,7 +642,7 @@ int hogvm(long long forks, long long chunks, long long bytes)
>> usleep(backoff);
>>
>> while (1) {
>> - ptr = (char **)malloc(chunks * 2);
>> + ptr = (char **)malloc(chunks * 2 * sizeof(char *));
>
> Good catch, however shouldn't this be just chunks * sizeof(char*) ?
Yes indeed, totally agree with you, "* 2" is redundant,
thanks for your review and reply. And also, if the chunks is 0,
the memory will be corrupted too. I will modify it in the v2 patch.
Thanks,
Regards,
Jiwei
>
>> for (j = 0; chunks == 0 || j < chunks; j++) {
>> if ((ptr[j] =
>> (char *)malloc(bytes *
>> diff --git a/tools/genload/stress.c b/tools/genload/stress.c
>> index 7f56d5272..9712e7828 100644
>> --- a/tools/genload/stress.c
>> +++ b/tools/genload/stress.c
>> @@ -642,7 +642,7 @@ int hogvm(long long forks, long long chunks, long long bytes)
>> usleep(backoff);
>>
>> while (1) {
>> - ptr = (char **)malloc(chunks * 2);
>> + ptr = (char **)malloc(chunks * 2 * sizeof(char *));
>
> Here as well.
>
>> for (j = 0; chunks == 0 || j < chunks; j++) {
>> if ((ptr[j] =
>> (char *)malloc(bytes *
>> --
>> 2.27.0
>>
>>
>> --
>> Mailing list info: https://lists.linux.it/listinfo/ltp
>
--
Mailing list info: https://lists.linux.it/listinfo/ltp
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2024-06-11 12:34 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2024-06-10 8:56 [LTP] [PATCH] genload: fix memory corruption in hogvm Jiwei Sun
-- strict thread matches above, loose matches on Subject: below --
2024-06-11 2:58 Jiwei Sun
2024-06-11 12:11 ` Cyril Hrubis
2024-06-11 12:34 ` Jiwei Sun
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.