Skip to contentSkip to conversation

mfd: mt6370: Add the out-of-bound check to prevent the null pointer

PATCH In review cy_huang 2 participants 7 messages Oct 21, 2022 – Oct 26, 2022

Subsystems (2)
lore
Share & tools
lore

cy_huang

author
#0 lore raw

PATCH v2 1/1 mfd: mt6370: Add the out-of-bound check to prevent the null pointer

From: ChiYuan Huang <[email protected]>

This potential risk could happen at regmap_raw_read() or regmap_raw_write() when accessing the over-bound register address.

For testing, I try to reproduce it with a testing attribute file. Below's the issue trace log.

[41.314358] pc : i2c_smbus_xfer+0x58/0x120
[41.314371] lr : i2c_smbus_read_i2c_block_data+0x74/0xc0
[41.399677] Call trace:
[41.402153]  i2c_smbus_xfer+0x58/0x120
[41.405956]  i2c_smbus_read_i2c_block_data+0x74/0xc0
[41.410991]  mt6370_regmap_read+0x40/0x60 [mt6370]
[41.415855]  _regmap_raw_read+0xe4/0x278
[41.419834]  regmap_raw_read+0xec/0x240
[41.423721]  rg_bound_show+0xb0/0x120 [mt6370]
[41.428226]  dev_attr_show+0x3c/0x80
[41.431851]  sysfs_kf_seq_show+0xc4/0x150
[41.435916]  kernfs_seq_show+0x48/0x60
[41.439718]  seq_read_iter+0x11c/0x450
[41.443519]  kernfs_fop_read_iter+0x124/0x1c0
[41.447937]  vfs_read+0x1a8/0x288
[41.451296]  ksys_read+0x74/0x100
[41.454654]  __arm64_sys_read+0x24/0x30
[41.458541]  invoke_syscall+0x54/0x118
[41.462344]  el0_svc_common.constprop.4+0x94/0x128
[41.467202]  do_el0_svc+0x3c/0xd0
[41.470562]  el0_svc+0x20/0x60
[41.473658]  el0t_64_sync_handler+0x94/0xb8
[41.477899]  el0t_64_sync+0x15c/0x160
[41.481614] Code: 54000388 f9401262 aa1303e0 52800041 (f9400042)
[41.487793] ---[ end trace 0000000000000000 ]---
Fixes b2adf788e603 ("mfd: mt6370: Add MediaTek MT6370 support")

---
Dear reviewers:

In v2, I refined the patch title and commit message.

If there's still something improper, please kindly correct me.

---
 drivers/mfd/mt6370.c | 6 ++++++
 1 file changed, 6 insertions(+)
drivers/mfd/mt6370.c +6 -0
index cf19cce..acbf960 100644
@@ -190,6 +190,9 @@ static int mt6370_regmap_read(void *context, const void *reg_buf,
 	bank_idx = u8_buf[0];
 	bank_addr = u8_buf[1];
 
+	if (bank_idx >= MT6370_MAX_I2C)
+		return -EINVAL;
+
 	ret = i2c_smbus_read_i2c_block_data(info->i2c[bank_idx], bank_addr,
 					    val_size, val_buf);
 	if (ret < 0)
@@ -211,6 +214,9 @@ static int mt6370_regmap_write(void *context, const void *data, size_t count)
 	bank_idx = u8_buf[0];
 	bank_addr = u8_buf[1];
 
+	if (bank_idx >= MT6370_MAX_I2C)
+		return -EINVAL;
+
 	return i2c_smbus_write_i2c_block_data(info->i2c[bank_idx], bank_addr,
 					      len, data + MT6370_MAX_ADDRLEN);
 }
Signature · 1 line
2.7.4
#1 lore raw
Quoted 8 lines · cy_huang

From: ChiYuan Huang <[email protected]>

This potential risk could happen at regmap_raw_read() or regmap_raw_write() when accessing the over-bound register address.

For testing, I try to reproduce it with a testing attribute file. Below's the issue trace log.

It looks like you randomly cut the trace.
It's not what I meant and documentation suggests.

Quoted 25 lines
[41.314358] pc : i2c_smbus_xfer+0x58/0x120
[41.314371] lr : i2c_smbus_read_i2c_block_data+0x74/0xc0
[41.399677] Call trace:
[41.402153]  i2c_smbus_xfer+0x58/0x120
[41.405956]  i2c_smbus_read_i2c_block_data+0x74/0xc0
[41.410991]  mt6370_regmap_read+0x40/0x60 [mt6370]
[41.415855]  _regmap_raw_read+0xe4/0x278
[41.419834]  regmap_raw_read+0xec/0x240
[41.423721]  rg_bound_show+0xb0/0x120 [mt6370]
[41.428226]  dev_attr_show+0x3c/0x80
[41.431851]  sysfs_kf_seq_show+0xc4/0x150
[41.435916]  kernfs_seq_show+0x48/0x60
[41.439718]  seq_read_iter+0x11c/0x450
[41.443519]  kernfs_fop_read_iter+0x124/0x1c0
[41.447937]  vfs_read+0x1a8/0x288
[41.451296]  ksys_read+0x74/0x100
[41.454654]  __arm64_sys_read+0x24/0x30
[41.458541]  invoke_syscall+0x54/0x118
[41.462344]  el0_svc_common.constprop.4+0x94/0x128
[41.467202]  do_el0_svc+0x3c/0xd0
[41.470562]  el0_svc+0x20/0x60
[41.473658]  el0t_64_sync_handler+0x94/0xb8
[41.477899]  el0t_64_sync+0x15c/0x160
[41.481614] Code: 54000388 f9401262 aa1303e0 52800041 (f9400042)
[41.487793] ---[ end trace 0000000000000000 ]---

...

Quoted 1 line

If there's still something improper, please kindly correct me.

https://www.kernel.org/doc/html/latest/process/submitting-patches.html#backtraces-in-commit-messages

Signature · 2 lines
With Best Regards,
Andy Shevchenko
#2 lore raw

Andy Shevchenko <[email protected]> 於 2022年10月21日 週五 下午4:34寫道:

Quoted 14 lines · cy_huang
Quoted 8 lines

From: ChiYuan Huang <[email protected]>

This potential risk could happen at regmap_raw_read() or regmap_raw_write() when accessing the over-bound register address.

For testing, I try to reproduce it with a testing attribute file. Below's the issue trace log.

It looks like you randomly cut the trace.
It's not what I meant and documentation suggests.

I checked the submitting-patch.rst.
To satisfy the requirement for 70-75 chars per line, I only keep the important log.

May I ask what you mean for the 'trim'?
Is it 'Still keep 70-75 per line and cut the characters that's over the limit to the next line"?

Quoted 35 lines
Quoted 25 lines
[41.314358] pc : i2c_smbus_xfer+0x58/0x120
[41.314371] lr : i2c_smbus_read_i2c_block_data+0x74/0xc0
[41.399677] Call trace:
[41.402153]  i2c_smbus_xfer+0x58/0x120
[41.405956]  i2c_smbus_read_i2c_block_data+0x74/0xc0
[41.410991]  mt6370_regmap_read+0x40/0x60 [mt6370]
[41.415855]  _regmap_raw_read+0xe4/0x278
[41.419834]  regmap_raw_read+0xec/0x240
[41.423721]  rg_bound_show+0xb0/0x120 [mt6370]
[41.428226]  dev_attr_show+0x3c/0x80
[41.431851]  sysfs_kf_seq_show+0xc4/0x150
[41.435916]  kernfs_seq_show+0x48/0x60
[41.439718]  seq_read_iter+0x11c/0x450
[41.443519]  kernfs_fop_read_iter+0x124/0x1c0
[41.447937]  vfs_read+0x1a8/0x288
[41.451296]  ksys_read+0x74/0x100
[41.454654]  __arm64_sys_read+0x24/0x30
[41.458541]  invoke_syscall+0x54/0x118
[41.462344]  el0_svc_common.constprop.4+0x94/0x128
[41.467202]  do_el0_svc+0x3c/0xd0
[41.470562]  el0_svc+0x20/0x60
[41.473658]  el0t_64_sync_handler+0x94/0xb8
[41.477899]  el0t_64_sync+0x15c/0x160
[41.481614] Code: 54000388 f9401262 aa1303e0 52800041 (f9400042)
[41.487793] ---[ end trace 0000000000000000 ]---

...

Quoted 1 line

If there's still something improper, please kindly correct me.

https://www.kernel.org/doc/html/latest/process/submitting-patches.html#backtraces-in-commit-messages

--
With Best Regards,
Andy Shevchenko

#3 lore raw
Quoted 6 lines · ChiYuan Huang

Andy Shevchenko <[email protected]> 於 2022年10月21日 週五 下午4:34寫道:

Quoted 5 lines · cy_huang

It looks like you randomly cut the trace.
It's not what I meant and documentation suggests.

Quoted 9 lines

I checked the submitting-patch.rst.
To satisfy the requirement for 70-75 chars per line, I only keep the important log.

May I ask what you mean for the 'trim'?
Is it 'Still keep 70-75 per line and cut the characters that's over the limit to the next line"?

Quoted 1 line

https://www.kernel.org/doc/html/latest/process/submitting-patches.html#backtraces-in-commit-messages

Have you had a chance to read this section of the document?

Signature · 2 lines
With Best Regards,
Andy Shevchenko
#4 lore raw

Andy Shevchenko <[email protected]> 於 2022年10月21日 週五 下午5:14寫道:

Quoted 12 lines · ChiYuan Huang
Quoted 10 lines

Andy Shevchenko <[email protected]> 於 2022年10月21日 週五 下午4:34寫道:

Quoted 5 lines · cy_huang

It looks like you randomly cut the trace.
It's not what I meant and documentation suggests.

I checked the submitting-patch.rst.
To satisfy the requirement for 70-75 chars per line, I only keep the important log.

Quoted 8 lines
Quoted 5 lines

May I ask what you mean for the 'trim'?
Is it 'Still keep 70-75 per line and cut the characters that's over the limit to the next line"?

Quoted 1 line

https://www.kernel.org/doc/html/latest/process/submitting-patches.html#backtraces-in-commit-messages

Have you had a chance to read this section of the document?

OK, get it.
Is the below text enough to express this problem?

Ex. Testing as below (mt6370 register range from 0 to 0x1ff)
rg_bound_show()

        regmap_raw_read(regmap, 0x200, &val, sizeof(val));

pstate: 60000005 (nZCv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--)

pc : i2c_smbus_xfer+0x58/0x120
lr : i2c_smbus_read_i2c_block_data+0x74/0xc0
Call trace:
 i2c_smbus_xfer+0x58/0x120
 i2c_smbus_read_i2c_block_data+0x74/0xc0
 mt6370_regmap_read+0x40/0x60
 _regmap_raw_read+0xe4/0x278
 regmap_raw_read+0xec/0x240
 rg_bound_show+0xb0/0x120
Quoted 3 lines

--
With Best Regards,
Andy Shevchenko

#5 lore raw
Quoted 4 lines · ChiYuan Huang

Andy Shevchenko <[email protected]> 於 2022年10月21日 週五 下午5:14寫道:

Quoted 3 lines · ChiYuan Huang
Quoted 2 lines

Andy Shevchenko <[email protected]> 於 2022年10月21日 週五 下午4:34寫道:

Quoted 1 line · cy_huang

...

Quoted 6 lines
Quoted 4 lines
Quoted 1 line
Quoted 1 line

https://www.kernel.org/doc/html/latest/process/submitting-patches.html#backtraces-in-commit-messages

Have you had a chance to read this section of the document?

OK, get it.
Is the below text enough to express this problem?

It is you who decides, because I don't know what exact problem this will represent.

Quoted 5 lines

Ex. Testing as below (mt6370 register range from 0 to 0x1ff)
rg_bound_show()

        regmap_raw_read(regmap, 0x200, &val, sizeof(val));

pstate: 60000005 (nZCv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--)

For example, why is this line here?

Quoted 9 lines
pc : i2c_smbus_xfer+0x58/0x120
lr : i2c_smbus_read_i2c_block_data+0x74/0xc0
Call trace:
 i2c_smbus_xfer+0x58/0x120
 i2c_smbus_read_i2c_block_data+0x74/0xc0
 mt6370_regmap_read+0x40/0x60
 _regmap_raw_read+0xe4/0x278
 regmap_raw_read+0xec/0x240
 rg_bound_show+0xb0/0x120

The rule of thumb is that it's okay to shrink to ~4-5 lines (in most cases).

Signature · 2 lines
With Best Regards,
Andy Shevchenko
#6 lore raw

Andy Shevchenko <[email protected]> 於 2022年10月21日 週五 下午6:35寫道:

Quoted 37 lines · ChiYuan Huang
Quoted 4 lines

Andy Shevchenko <[email protected]> 於 2022年10月21日 週五 下午5:14寫道:

Quoted 3 lines · ChiYuan Huang
Quoted 2 lines

Andy Shevchenko <[email protected]> 於 2022年10月21日 週五 下午4:34寫道:

Quoted 1 line · cy_huang

...

Quoted 6 lines
Quoted 4 lines
Quoted 1 line
Quoted 1 line

https://www.kernel.org/doc/html/latest/process/submitting-patches.html#backtraces-in-commit-messages

Have you had a chance to read this section of the document?

OK, get it.
Is the below text enough to express this problem?

It is you who decides, because I don't know what exact problem this will represent.

Quoted 5 lines

Ex. Testing as below (mt6370 register range from 0 to 0x1ff)
rg_bound_show()

        regmap_raw_read(regmap, 0x200, &val, sizeof(val));

pstate: 60000005 (nZCv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--)

For example, why is this line here?

Quoted 9 lines
pc : i2c_smbus_xfer+0x58/0x120
lr : i2c_smbus_read_i2c_block_data+0x74/0xc0
Call trace:
 i2c_smbus_xfer+0x58/0x120
 i2c_smbus_read_i2c_block_data+0x74/0xc0
 mt6370_regmap_read+0x40/0x60
 _regmap_raw_read+0xe4/0x278
 regmap_raw_read+0xec/0x240
 rg_bound_show+0xb0/0x120
Quoted 2 lines

The rule of thumb is that it's okay to shrink to ~4-5 lines (in most cases).

At first, I start this thread due to the original thread seems stopped Since it already restarted, please forget this one.
Just follow the restaed thread for the discussion.
https://lore.kernel.org/all/Y1aCiReTZDbPp%2FrS@kadam/

Quoted 3 lines

--
With Best Regards,
Andy Shevchenko