user/sven/linux.git/scripts/coccinelle, branch v5.15.27

scripts: coccinelle: allow list_entry_is_head() to use pos

2021-08-08T20:00:49Z

Currently use_after_iter.cocci generates false positives for code of the following form: ~~~ list_for_each_entry(d, &ddata->irq_list, node) { if (irq == d->irq) break; } if (list_entry_is_head(d, &ddata->irq_list, node)) return IRQ_NONE; ~~~ [This specific example comes from drivers/power/supply/cpcap-battery.c] Most list macros use list_entry_is_head() as loop exit condition meaning it is not unsafe to reuse pos (a.k.a. d) in the code above. Let's avoid reporting these cases. Signed-off-by: Daniel Thompson Signed-off-by: Julia Lawall

coccinelle: api: rename kzfree to kfree_sensitive

2021-08-05T11:01:09Z

Commit 453431a54934 ("mm, treewide: rename kzfree() to kfree_sensitive()") renamed kzfree() to kfree_sensitive(), it should be applied to coccinelle. Signed-off-by: Weizhao Ouyang Signed-off-by: Julia Lawall Acked-by: Denis Efremov

coccinelle: api: remove kobj_to_dev.cocci script

2021-06-21T18:45:49Z

Using kobj_to_dev() instead of container_of() is not universally accepted among maintainers as an improvement. The warning leads to repeated patch submissions that won't be accepted. Remove the script. Cc: Christoph Hellwig Cc: Jens Axboe Cc: Denis Efremov Cc: Julia Lawall Signed-off-by: Keith Busch Signed-off-by: Julia Lawall Acked-by: Jens Axboe Acked-by: Denis Efremov

coccinelle: misc: minmax: suppress patch generation for err returns

2021-05-01T19:27:10Z

There is a standard idiom for "if 'ret' holds an error, return it": return ret < 0 ? ret : 0; Developers prefer to keep the things as they are because stylistic change to "return min(ret, 0);" breaks readability. Let's suppress automatic generation for this type of patches. Signed-off-by: Denis Efremov

drop unneeded *s

2021-05-01T19:23:57Z

Kfree.cocci only supports org and report mode, so the *s (used for context mode) are not useful. Signed-off-by: Julia Lawall

coccinelle: irqf_oneshot: reduce the severity due to false positives

2021-05-01T19:23:57Z

The IRQF_ONESHOT should be present for threaded IRQ using default primary handler. However intetrupt of many child devices, e.g. children of MFD, is nested thus the IRQF_ONESHOT is not needed. The coccinelle message about error misleads submitters and reviewers about the severity of the issue, so make it a warning and mention possible false positive. Signed-off-by: Krzysztof Kozlowski Signed-off-by: Julia Lawall

coccinelle: misc: add swap script

2021-05-01T19:23:56Z

Check for opencoded swap() implementation. Signed-off-by: Denis Efremov Signed-off-by: Julia Lawall

coccinelle: misc: update uninitialized_var.cocci documentation

2021-05-01T19:23:56Z

Remove the documentation link from the warning message because commit 3942ea7a10c9 ("deprecated.rst: Remove now removed uninitialized_var") removed the section from documentation. Update the rule documentation accordingly. Signed-off-by: Denis Efremov Signed-off-by: Julia Lawall

coccinelle: misc: restrict patch mode in flexible_array.cocci

2021-05-01T19:23:56Z

Skip patches generation for structs with a single field. Changing a zero-length array to a flexible array member in a struct with no named members breaks the compilation. However, reporting such cases is still valuable, e.g. commit 637464c59e0b ("ACPI: NFIT: Fix flexible_array.cocci warnings"). Signed-off-by: Denis Efremov Signed-off-by: Julia Lawall

coccinelle: misc: add minmax script

2021-05-01T19:23:56Z

Check for opencoded min(), max() implementations. Signed-off-by: Denis Efremov Signed-off-by: Julia Lawall