| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <4854b569-5032-4b75-80a6-8c5822845dc7@baylibre.com>
Date: Mon, 24 Mar 2025 08:50:15 -0500
From: David Lechner <dlechner@...libre.com>
To: Dan Carpenter <dan.carpenter@...aro.org>, oe-kbuild@...ts.linux.dev,
Jonathan Cameron <jic23@...nel.org>, linux-iio@...r.kernel.org
Cc: lkp@...el.com, oe-kbuild-all@...ts.linux.dev,
Michael Hennerich <Michael.Hennerich@...log.com>,
Angelo Dureghello <adureghello@...libre.com>,
Alexandru Ardelean <aardelean@...libre.com>,
Beniamin Bia <beniamin.bia@...log.com>, Stefan Popa
<stefan.popa@...log.com>, linux-kernel@...r.kernel.org,
Nuno Sá <nuno.sa@...log.com>
Subject: Re: [PATCH v2 09/10] iio: adc: ad7606: dynamically allocate channel
info
On 3/22/25 12:25 PM, Dan Carpenter wrote:
> Hi David,
>
> kernel test robot noticed the following build warnings:
>
> url: https://github.com/intel-lab-lkp/linux/commits/David-Lechner/iio-adc-ad7606-check-for-NULL-before-calling-sw_mode_config/20250319-065737
> base: 9f36acefb2621d980734a5bb7d74e0e24e0af166
> patch link: https://lore.kernel.org/r/20250318-iio-adc-ad7606-improvements-v2-9-4b605427774c%40baylibre.com
> patch subject: [PATCH v2 09/10] iio: adc: ad7606: dynamically allocate channel info
> config: arm64-randconfig-r071-20250322 (https://download.01.org/0day-ci/archive/20250322/202503222246.RafigmhQ-lkp@intel.com/config)
> compiler: clang version 21.0.0git (https://github.com/llvm/llvm-project c2692afc0a92cd5da140dfcdfff7818a5b8ce997)
>
> If you fix the issue in a separate patch/commit (i.e. not just a new version of
> the same patch/commit), kindly add following tags
> | Reported-by: kernel test robot <lkp@...el.com>
> | Reported-by: Dan Carpenter <dan.carpenter@...aro.org>
> | Closes: https://lore.kernel.org/r/202503222246.RafigmhQ-lkp@intel.com/
>
> smatch warnings:
> drivers/iio/adc/ad7606.c:1270 ad7606_probe_channels() warn: potentially one past the end of array 'channels[i]'
>
> vim +1270 drivers/iio/adc/ad7606.c
>
> 87cf5705725eeb David Lechner 2025-03-18 1196 static int ad7606_probe_channels(struct iio_dev *indio_dev)
> e571c1902116a3 Alexandru Ardelean 2024-09-19 1197 {
> e571c1902116a3 Alexandru Ardelean 2024-09-19 1198 struct ad7606_state *st = iio_priv(indio_dev);
> 87cf5705725eeb David Lechner 2025-03-18 1199 struct device *dev = indio_dev->dev.parent;
> 87cf5705725eeb David Lechner 2025-03-18 1200 struct iio_chan_spec *channels;
> 87cf5705725eeb David Lechner 2025-03-18 1201 bool slow_bus;
> 87cf5705725eeb David Lechner 2025-03-18 1202 int ret, i;
> 87cf5705725eeb David Lechner 2025-03-18 1203
> 87cf5705725eeb David Lechner 2025-03-18 1204 slow_bus = !st->bops->iio_backend_config;
> 87cf5705725eeb David Lechner 2025-03-18 1205 indio_dev->num_channels = st->chip_info->num_adc_channels;
> 87cf5705725eeb David Lechner 2025-03-18 1206
> 87cf5705725eeb David Lechner 2025-03-18 1207 /* Slow buses also get 1 more channel for soft timestamp */
> 87cf5705725eeb David Lechner 2025-03-18 1208 if (slow_bus)
> 87cf5705725eeb David Lechner 2025-03-18 1209 indio_dev->num_channels++;
> 87cf5705725eeb David Lechner 2025-03-18 1210
> 87cf5705725eeb David Lechner 2025-03-18 1211 channels = devm_kcalloc(dev, indio_dev->num_channels, sizeof(*channels),
> 87cf5705725eeb David Lechner 2025-03-18 1212 GFP_KERNEL);
> 87cf5705725eeb David Lechner 2025-03-18 1213 if (!channels)
> f3838e934dfff2 Alexandru Ardelean 2024-09-19 1214 return -ENOMEM;
> f3838e934dfff2 Alexandru Ardelean 2024-09-19 1215
> 87cf5705725eeb David Lechner 2025-03-18 1216 for (i = 0; i < indio_dev->num_channels; i++) {
The fix is to change this line to:
for (i = 0; i < st->chip_info->num_adc_channels; i++) {
> 87cf5705725eeb David Lechner 2025-03-18 1217 struct iio_chan_spec *chan = &channels[i];
> 87cf5705725eeb David Lechner 2025-03-18 1218
> 87cf5705725eeb David Lechner 2025-03-18 1219 chan->type = IIO_VOLTAGE;
> 87cf5705725eeb David Lechner 2025-03-18 1220 chan->indexed = 1;
> 87cf5705725eeb David Lechner 2025-03-18 1221 chan->channel = i;
> 87cf5705725eeb David Lechner 2025-03-18 1222 chan->scan_index = i;
> 87cf5705725eeb David Lechner 2025-03-18 1223 chan->scan_type.sign = 's';
> 87cf5705725eeb David Lechner 2025-03-18 1224 chan->scan_type.realbits = st->chip_info->bits;
> 87cf5705725eeb David Lechner 2025-03-18 1225 chan->scan_type.storagebits = st->chip_info->bits > 16 ? 32 : 16;
> 87cf5705725eeb David Lechner 2025-03-18 1226 chan->scan_type.endianness = IIO_CPU;
> f3838e934dfff2 Alexandru Ardelean 2024-09-19 1227
> 87cf5705725eeb David Lechner 2025-03-18 1228 if (indio_dev->modes & INDIO_DIRECT_MODE)
> 87cf5705725eeb David Lechner 2025-03-18 1229 chan->info_mask_separate |= BIT(IIO_CHAN_INFO_RAW);
> 87cf5705725eeb David Lechner 2025-03-18 1230
> 87cf5705725eeb David Lechner 2025-03-18 1231 if (st->sw_mode_en) {
> 87cf5705725eeb David Lechner 2025-03-18 1232 chan->info_mask_separate |= BIT(IIO_CHAN_INFO_SCALE);
> 87cf5705725eeb David Lechner 2025-03-18 1233 chan->info_mask_separate_available |=
> 87cf5705725eeb David Lechner 2025-03-18 1234 BIT(IIO_CHAN_INFO_SCALE);
> 87cf5705725eeb David Lechner 2025-03-18 1235
> 87cf5705725eeb David Lechner 2025-03-18 1236 /*
> 87cf5705725eeb David Lechner 2025-03-18 1237 * All chips with software mode support oversampling,
> 87cf5705725eeb David Lechner 2025-03-18 1238 * so we skip the oversampling_available check. And the
> 87cf5705725eeb David Lechner 2025-03-18 1239 * shared_by_type instead of shared_by_all on slow
> 87cf5705725eeb David Lechner 2025-03-18 1240 * buses is for backward compatibility.
> 87cf5705725eeb David Lechner 2025-03-18 1241 */
> 87cf5705725eeb David Lechner 2025-03-18 1242 if (slow_bus)
> 87cf5705725eeb David Lechner 2025-03-18 1243 chan->info_mask_shared_by_type |=
> 87cf5705725eeb David Lechner 2025-03-18 1244 BIT(IIO_CHAN_INFO_OVERSAMPLING_RATIO);
> 87cf5705725eeb David Lechner 2025-03-18 1245 else
> 87cf5705725eeb David Lechner 2025-03-18 1246 chan->info_mask_shared_by_all |=
> 87cf5705725eeb David Lechner 2025-03-18 1247 BIT(IIO_CHAN_INFO_OVERSAMPLING_RATIO);
> 87cf5705725eeb David Lechner 2025-03-18 1248
> 87cf5705725eeb David Lechner 2025-03-18 1249 chan->info_mask_shared_by_all_available |=
> 87cf5705725eeb David Lechner 2025-03-18 1250 BIT(IIO_CHAN_INFO_OVERSAMPLING_RATIO);
> 87cf5705725eeb David Lechner 2025-03-18 1251 } else {
> 87cf5705725eeb David Lechner 2025-03-18 1252 chan->info_mask_shared_by_type |=
> 87cf5705725eeb David Lechner 2025-03-18 1253 BIT(IIO_CHAN_INFO_SCALE);
> 87cf5705725eeb David Lechner 2025-03-18 1254
> 87cf5705725eeb David Lechner 2025-03-18 1255 if (st->chip_info->oversampling_avail)
> 87cf5705725eeb David Lechner 2025-03-18 1256 chan->info_mask_shared_by_all |=
> 87cf5705725eeb David Lechner 2025-03-18 1257 BIT(IIO_CHAN_INFO_OVERSAMPLING_RATIO);
> 87cf5705725eeb David Lechner 2025-03-18 1258 }
> 87cf5705725eeb David Lechner 2025-03-18 1259
> 87cf5705725eeb David Lechner 2025-03-18 1260 if (!slow_bus)
> 87cf5705725eeb David Lechner 2025-03-18 1261 chan->info_mask_shared_by_all |=
> 87cf5705725eeb David Lechner 2025-03-18 1262 BIT(IIO_CHAN_INFO_SAMP_FREQ);
> 87cf5705725eeb David Lechner 2025-03-18 1263
> 87cf5705725eeb David Lechner 2025-03-18 1264 ret = st->chip_info->scale_setup_cb(indio_dev, chan);
> e571c1902116a3 Alexandru Ardelean 2024-09-19 1265 if (ret)
> e571c1902116a3 Alexandru Ardelean 2024-09-19 1266 return ret;
> e571c1902116a3 Alexandru Ardelean 2024-09-19 1267 }
> e571c1902116a3 Alexandru Ardelean 2024-09-19 1268
> 87cf5705725eeb David Lechner 2025-03-18 1269 if (slow_bus)
> 87cf5705725eeb David Lechner 2025-03-18 @1270 channels[i] = (struct iio_chan_spec)IIO_CHAN_SOFT_TIMESTAMP(i);
> ^^^^^^^^^^^
> i is == indio_dev->num_channels so this is out of bounds by one element.
>
> 87cf5705725eeb David Lechner 2025-03-18 1271
> 87cf5705725eeb David Lechner 2025-03-18 1272 indio_dev->channels = channels;
> 87cf5705725eeb David Lechner 2025-03-18 1273
> e571c1902116a3 Alexandru Ardelean 2024-09-19 1274 return 0;
> e571c1902116a3 Alexandru Ardelean 2024-09-19 1275 }
>
Powered by blists - more mailing lists