- 06 Sep, 2022 5 commits
-
-
Michael Ellerman authored
Currently in pas_pci_init() a reference to the root node is leaked due to a missing of_node_put(). Instead just use of_root directly. Note that converting to of_find_compatible_node(NULL, ...) would not be entirely equivalent, because that would check the compatible property of the root node, whereas using of_root skips checking the root node and start the search at the first child of the root. Reported-by:
Liang He <windhl@126.com> Signed-off-by:
Michael Ellerman <mpe@ellerman.id.au> Link: https://lore.kernel.org/r/20220906010313.1296714-1-mpe@ellerman.id.au
-
Russell Currey authored
find_i2c_driver() contained the last usage of strlcpy() in arch/powerpc. The return value was used to check if strlen(src) >= n, for which strscpy() returns -E2BIG. Signed-off-by:
Russell Currey <ruscur@russell.cc> Signed-off-by:
-
Jilin Yuan authored
Delete the redundant word 'set'. Signed-off-by:
Jilin Yuan <yuanjilin@cdjrlc.com> Signed-off-by:
-
Jilin Yuan authored
Delete the redundant word 'the'. Signed-off-by:
-
Jilin Yuan authored
Delete the redundant word 'the'. Signed-off-by:
-
- 05 Sep, 2022 35 commits
-
-
Liang He authored
During the iteration of for_each_child_of_node(), we need to call of_node_put() for the old references stored in to 'ch_def' and 'ch_a' as their refcounters have been increased in last iteration. Signed-off-by:
-
Liang He authored
Call of_node_put() for the reference 'gparent' escaped out of the previous for_each_child_of_node() as it has increased the refcount. Signed-off-by:
-
Liang He authored
Call of_node_put() for the reference 'parent' returned by of_get_parent() which has increased the refcount. Signed-off-by:
-
Liang He authored
In probe_one_macio(), call of_node_put() for the refernece 'node' escaped out of the for_each_node_by_name() which has increased its refcount. While the 'node' will finally escaped into a global reference, we should still call of_node_put() in fail path which will stop global reference creation. Signed-off-by:
-
Liang He authored
In pcibios_alloc_controller(), 'phb' is allocated and escaped into global 'hose_list'. So we should call of_node_get() when a new reference created into 'phb->dn'. And when phb is freed, we should call of_node_put() on it. NOTE: This function is called in the iteration of for_each_xx in chrp_find_bridges() and pSeries_discover_phbs(). If there is no of_node_get(), the object may be prematurely freed. Signed-off-by:
-
Liang He authored
In these driver init functions, there are two kinds of errors: (1) missing of_put_node() for of_find_compatible_node()'s returned pointer (refcount incremented) in fail path or when it is not used anymore. (2) missing of_put_node() for 'for_each_xxx' loop's break Signed-off-by: -
Liang He authored
In fsl_pci_assign_primary(), we should remove the of_node_put() when breaking out of the for_each_matching_node() as the 'np' is escaped out by global 'fsl_pci_primary'. Signed-off-by:
-
Liang He authored
for_each_node_by_type() will automatically increase and decrease the refcount during the iteration. However, there is a reference escaped into global 'fsl_pci_primary' and we need to handle it. Signed-off-by:
-
Liang He authored
In ppc47x_init_irq(), we need to call of_node_put() when there is a break during the iteration of for_each_node_with_property() which will automatically increase and decrease the refcount. Signed-off-by:
-
Liang He authored
In pci_add_device_node_info(), use of_node_put() to drop the reference to 'parent' returned by of_get_parent() to keep refcount balance. Fixes: cca87d30 ("powerpc/pci: Refactor pci_dn") Co-authored-by:
Miaoqian Lin <linmq006@gmail.com> Signed-off-by:
Tyrel Datwyler <tyreld@linux.ibm.com> Link: https://lore.kernel.org/r/20220701131750.240170-1-windhl@126.com
-
Liang He authored
In pseries_cpuhp_cache_use_count() and pseries_cpuhp_detach_nodes(), we need carefully hold the reference returned by of_find_next_cache_node() and use it to call of_node_put() to keep refcount balance. Signed-off-by:
-
Liang He authored
We need a of_node_get() for of_find_compatible_node() to keep refcount balance. Signed-off-by:
-
Liang He authored
In mpc832x_spi_init(), hold the reference returned by of_find_compatible_node() and use it to call of_node_put() for refcount balance. Signed-off-by:
-
Liang He authored
In mpc5121_clk_provide_migration_support(), hold the reference returned by of_find_compatible_node() and use it to call of_node_put() for refcount balance. Signed-off-by:
-
Liang He authored
In smp_core99_setup(), add of_node_put() to drop the reference once it's no longer needed. Signed-off-by:
-
Liang He authored
In fsl_setup_msi_irqs(), use of_node_put() to drop the reference returned by of_parse_phandle(). Fixes: 895d603f ("powerpc/fsl_msi: add support for the fsl, msi property in PCI nodes") Co-authored-by:
-
Liang He authored
Add of_node_put() in various paths to drop references once they are no longer needed. Signed-off-by:
-
Liang He authored
In ls_uarts_init(), add an of_node_put() to keep refcount balance. Signed-off-by:
-
Liang He authored
In ibmebus_match_path(), use of_node_put() to drop the reference returned by of_find_node_by_path() before testing for equality of the pointers. Signed-off-by:
-
Liang He authored
In cell_iommu_init_disabled(), hold the reference returned by of_find_node_by_name() and use it to call of_node_put() for reference balance. Signed-off-by:
-
Liang He authored
There are several bugs as following: (1) In cbe_get_be_node(), hold the reference returned by of_find_xxx and of_get_xxx OF APIs and use it to call of_node_put(). (2) In cbe_fill_regs_map(), same as above. (3) In cbe_regs_init(), during the iteration of for_each_node_by_type(), the refcount of 'cpu' will be automatically increased and decreased. However, there is a reference escaped out into 'map->cpu_node' and it should be properly handled. Signed-off-by: -
Liang He authored
Use of_node_put() for of_find_node_by_path() and of_find_node_by_phandle() to keep refcount balance. Signed-off-by:
-
Liang He authored
In find_legacy_serial_ports(), of_find_node_by_path() will return a node pointer with refcount incremented. The reference should be dropped with of_node_put() when it is not used anymore. Signed-off-by:
-
Liang He authored
In update_events_in_group(), of_find_node_by_phandle() will return a node pointer with refcount incremented. The reference should be dropped with of_node_put() in the failure path or when it is not used anymore. Signed-off-by:
Athira Rajeev <atrajeev@linux.vnet.ibm.com> Link: https://lore.kernel.org/r/20220618071353.4059000-1-windhl@126.com
-
Liang He authored
Add missing of_node_put()s in various paths. Signed-off-by:
-
Liang He authored
In init_ioports(), of_find_node_by_name() will return a node pointer with refcount incremented. The reference should be dropped with of_node_put() when it is not used anymore. Signed-off-by:
-
Liang He authored
In maple_get_boot_time(), of_find_compatible_node() will return a node pointer with refcount incremented. The reference should be dropped with of_node_put() when it is not used anymore. Signed-off-by:
-
Liang He authored
In gpio_halt_probe(), of_find_matching_node() will return a node pointer with refcount incremented. The reference should be dropped with of_node_put() in the failure path. Signed-off-by:
-
Liang He authored
In media5200_init_irq(), of_find_compatible_node() will return a node pointer with refcount incremented. The reference should be dropped with of_node_put() in the failure path or when it is not used anymore. Don't worry about 'fpga_np == NULL' as of_node_put() can correctly handle that. Signed-off-by:
-
Liang He authored
In ksi8560_setup_arch(), of_find_compatible_node() will return a node pointer with refcount incremented. The reference should be dropped with of_node_put() when it is not used anymore. Signed-off-by:
-
Liang He authored
In mpc5121_clk_init(), of_find_compatible_node() will return a node pointer with refcount incremented. The reference should be dropped with of_node_put() when it is not used anymore. Signed-off-by:
-
Christophe Leroy authored
With recent binutils, the following warning appears: VDSO32L arch/powerpc/kernel/vdso/vdso32.so.dbg /opt/gcc-12.2.0-nolibc/powerpc64-linux/bin/../lib/gcc/powerpc64-linux/12.2.0/../../../../powerpc64-linux/bin/ld: warning: arch/powerpc/kernel/vdso/getcpu-32.o: missing .note.GNU-stack section implies executable stack /opt/gcc-12.2.0-nolibc/powerpc64-linux/bin/../lib/gcc/powerpc64-linux/12.2.0/../../../../powerpc64-linux/bin/ld: NOTE: This behaviour is deprecated and will be removed in a future version of the linker To avoid that, explicitly tell the linker we don't want executable stack. For more explanations, see commit ffcf9c57 ("x86: link vdso and boot with -z noexecstack --no-warn-rwx-segments") Signed-off-by:
Christophe Leroy <christophe.leroy@csgroup.eu> Signed-off-by:
-
Christophe Leroy authored
As reported by Nathan, the module_init() macro was not taken into account because the header was missing. That means spe_mathemu_init() was never called. This should have been detected by gcc at build time, but due to '-w' flag it went undetected. Removing that flag leads to many warnings hence errors. Fix those warnings then remove the -w flag. Reported-by:
Nathan Chancellor <nathan@kernel.org> Signed-off-by:
-
Nathan Chancellor authored
When building with a recent version of clang, there are a couple of errors around the call to module_init(): arch/powerpc/math-emu/math_efp.c:927:1: error: type specifier missing, defaults to 'int'; ISO C99 and later do not support implicit int [-Wimplicit-int] module_init(spe_mathemu_init); ^ int arch/powerpc/math-emu/math_efp.c:927:13: error: a parameter list without types is only allowed in a function definition module_init(spe_mathemu_init); ^ 2 errors generated. module_init() is a macro, which is not getting expanded because module.h is not included in this file. Add the include so that the macro can expand properly, clearing up the build failure. Fixes: ac6f1203 ("powerpc/85xx: Workaroudn e500 CPU erratum A005") [chleroy: added fixes tag] Reported-by:kernel test robot <lkp@intel.com> Signed-off-by:
-
Nicholas Piggin authored
dtl is the PAPR Dispatch Trace Log, which is entirely a pseries feature. The pseries platform alrady has a file dealing with the dtl, so move scanning for stolen time accounting there from kernel/time.c. Signed-off-by:
Nicholas Piggin <npiggin@gmail.com> Signed-off-by:
-
