kernel/bpf/core.c:900:5: warning: '__pte_index_size' is not defined, evaluates to 0
kernel test robot
build bottree: https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next.git master head: d0d131e3b655fd267d14bb1bed49e3f990a1465e commit: 643714ff18301f41bded2489e8f1f9d1b4782094 [5/7] bpf: cap BPF_PROG_PACK_SIZE to 2MB * num_possible_nodes() config: powerpc-allyesconfig (https://download.01.org/0day-ci/archive/20240309/[email protected]/config) compiler: clang version 19.0.0git (https://github.com/llvm/llvm-project 503c55e17037436dcd45ac69dea8967e67e3f5e8) reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20240309/[email protected]/reproduce)
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 <[email protected]>
| Closes: https://lore.kernel.org/oe-kbuild-all/[email protected]/
All warnings (new ones prefixed by >>):
In file included from kernel/bpf/core.c:21:
In file included from include/linux/filter.h:9:
In file included from include/linux/bpf.h:21:
In file included from include/linux/kallsyms.h:13:
In file included from include/linux/mm.h:2188:
include/linux/vmstat.h:508:43: warning: arithmetic between different enumeration types ('enum zone_stat_item' and 'enum numa_stat_item') [-Wenum-enum-conversion]
508 | return vmstat_text[NR_VM_ZONE_STAT_ITEMS +
| ~~~~~~~~~~~~~~~~~~~~~ ^
509 | item];
| ~~~~
include/linux/vmstat.h:515:43: warning: arithmetic between different enumeration types ('enum zone_stat_item' and 'enum numa_stat_item') [-Wenum-enum-conversion]
515 | return vmstat_text[NR_VM_ZONE_STAT_ITEMS +
| ~~~~~~~~~~~~~~~~~~~~~ ^
516 | NR_VM_NUMA_EVENT_ITEMS +
| ~~~~~~~~~~~~~~~~~~~~~~
include/linux/vmstat.h:522:36: warning: arithmetic between different enumeration types ('enum node_stat_item' and 'enum lru_list') [-Wenum-enum-conversion]
522 | return node_stat_name(NR_LRU_BASE + lru) + 3; // skip "nr_"
| ~~~~~~~~~~~ ^ ~~~
include/linux/vmstat.h:527:43: warning: arithmetic between different enumeration types ('enum zone_stat_item' and 'enum numa_stat_item') [-Wenum-enum-conversion]
527 | return vmstat_text[NR_VM_ZONE_STAT_ITEMS +
| ~~~~~~~~~~~~~~~~~~~~~ ^
528 | NR_VM_NUMA_EVENT_ITEMS +
| ~~~~~~~~~~~~~~~~~~~~~~
include/linux/vmstat.h:536:43: warning: arithmetic between different enumeration types ('enum zone_stat_item' and 'enum numa_stat_item') [-Wenum-enum-conversion]
536 | return vmstat_text[NR_VM_ZONE_STAT_ITEMS +
| ~~~~~~~~~~~~~~~~~~~~~ ^
537 | NR_VM_NUMA_EVENT_ITEMS +
| ~~~~~~~~~~~~~~~~~~~~~~
In file included from kernel/bpf/core.c:21:
In file included from include/linux/filter.h:9:
include/linux/bpf.h:739:48: warning: bitwise operation between different enumeration types ('enum bpf_type_flag' and 'enum bpf_arg_type') [-Wenum-enum-conversion]
739 | ARG_PTR_TO_MAP_VALUE_OR_NULL = PTR_MAYBE_NULL | ARG_PTR_TO_MAP_VALUE,
| ~~~~~~~~~~~~~~ ^ ~~~~~~~~~~~~~~~~~~~~
include/linux/bpf.h:740:43: warning: bitwise operation between different enumeration types ('enum bpf_type_flag' and 'enum bpf_arg_type') [-Wenum-enum-conversion]
740 | ARG_PTR_TO_MEM_OR_NULL = PTR_MAYBE_NULL | ARG_PTR_TO_MEM,
| ~~~~~~~~~~~~~~ ^ ~~~~~~~~~~~~~~
include/linux/bpf.h:741:43: warning: bitwise operation between different enumeration types ('enum bpf_type_flag' and 'enum bpf_arg_type') [-Wenum-enum-conversion]
741 | ARG_PTR_TO_CTX_OR_NULL = PTR_MAYBE_NULL | ARG_PTR_TO_CTX,
| ~~~~~~~~~~~~~~ ^ ~~~~~~~~~~~~~~
include/linux/bpf.h:742:45: warning: bitwise operation between different enumeration types ('enum bpf_type_flag' and 'enum bpf_arg_type') [-Wenum-enum-conversion]
742 | ARG_PTR_TO_SOCKET_OR_NULL = PTR_MAYBE_NULL | ARG_PTR_TO_SOCKET,
| ~~~~~~~~~~~~~~ ^ ~~~~~~~~~~~~~~~~~
include/linux/bpf.h:743:44: warning: bitwise operation between different enumeration types ('enum bpf_type_flag' and 'enum bpf_arg_type') [-Wenum-enum-conversion]
743 | ARG_PTR_TO_STACK_OR_NULL = PTR_MAYBE_NULL | ARG_PTR_TO_STACK,
| ~~~~~~~~~~~~~~ ^ ~~~~~~~~~~~~~~~~
include/linux/bpf.h:744:45: warning: bitwise operation between different enumeration types ('enum bpf_type_flag' and 'enum bpf_arg_type') [-Wenum-enum-conversion]
744 | ARG_PTR_TO_BTF_ID_OR_NULL = PTR_MAYBE_NULL | ARG_PTR_TO_BTF_ID,
| ~~~~~~~~~~~~~~ ^ ~~~~~~~~~~~~~~~~~
include/linux/bpf.h:748:38: warning: bitwise operation between different enumeration types ('enum bpf_type_flag' and 'enum bpf_arg_type') [-Wenum-enum-conversion]
748 | ARG_PTR_TO_UNINIT_MEM = MEM_UNINIT | ARG_PTR_TO_MEM,
| ~~~~~~~~~~ ^ ~~~~~~~~~~~~~~
include/linux/bpf.h:750:45: warning: bitwise operation between different enumeration types ('enum bpf_type_flag' and 'enum bpf_arg_type') [-Wenum-enum-conversion]
750 | ARG_PTR_TO_FIXED_SIZE_MEM = MEM_FIXED_SIZE | ARG_PTR_TO_MEM,
| ~~~~~~~~~~~~~~ ^ ~~~~~~~~~~~~~~
include/linux/bpf.h:773:48: warning: bitwise operation between different enumeration types ('enum bpf_type_flag' and 'enum bpf_return_type') [-Wenum-enum-conversion]
773 | RET_PTR_TO_MAP_VALUE_OR_NULL = PTR_MAYBE_NULL | RET_PTR_TO_MAP_VALUE,
| ~~~~~~~~~~~~~~ ^ ~~~~~~~~~~~~~~~~~~~~
include/linux/bpf.h:774:45: warning: bitwise operation between different enumeration types ('enum bpf_type_flag' and 'enum bpf_return_type') [-Wenum-enum-conversion]
774 | RET_PTR_TO_SOCKET_OR_NULL = PTR_MAYBE_NULL | RET_PTR_TO_SOCKET,
| ~~~~~~~~~~~~~~ ^ ~~~~~~~~~~~~~~~~~
include/linux/bpf.h:775:47: warning: bitwise operation between different enumeration types ('enum bpf_type_flag' and 'enum bpf_return_type') [-Wenum-enum-conversion]
775 | RET_PTR_TO_TCP_SOCK_OR_NULL = PTR_MAYBE_NULL | RET_PTR_TO_TCP_SOCK,
| ~~~~~~~~~~~~~~ ^ ~~~~~~~~~~~~~~~~~~~
include/linux/bpf.h:776:50: warning: bitwise operation between different enumeration types ('enum bpf_type_flag' and 'enum bpf_return_type') [-Wenum-enum-conversion]
776 | RET_PTR_TO_SOCK_COMMON_OR_NULL = PTR_MAYBE_NULL | RET_PTR_TO_SOCK_COMMON,
| ~~~~~~~~~~~~~~ ^ ~~~~~~~~~~~~~~~~~~~~~~
include/linux/bpf.h:778:49: warning: bitwise operation between different enumeration types ('enum bpf_type_flag' and 'enum bpf_return_type') [-Wenum-enum-conversion]
778 | RET_PTR_TO_DYNPTR_MEM_OR_NULL = PTR_MAYBE_NULL | RET_PTR_TO_MEM,
| ~~~~~~~~~~~~~~ ^ ~~~~~~~~~~~~~~
include/linux/bpf.h:779:45: warning: bitwise operation between different enumeration types ('enum bpf_type_flag' and 'enum bpf_return_type') [-Wenum-enum-conversion]
779 | RET_PTR_TO_BTF_ID_OR_NULL = PTR_MAYBE_NULL | RET_PTR_TO_BTF_ID,
| ~~~~~~~~~~~~~~ ^ ~~~~~~~~~~~~~~~~~
include/linux/bpf.h:780:43: warning: bitwise operation between different enumeration types ('enum bpf_type_flag' and 'enum bpf_return_type') [-Wenum-enum-conversion]
780 | RET_PTR_TO_BTF_ID_TRUSTED = PTR_TRUSTED | RET_PTR_TO_BTF_ID,
| ~~~~~~~~~~~ ^ ~~~~~~~~~~~~~~~~~
include/linux/bpf.h:891:44: warning: bitwise operation between different enumeration types ('enum bpf_type_flag' and 'enum bpf_reg_type') [-Wenum-enum-conversion]
891 | PTR_TO_MAP_VALUE_OR_NULL = PTR_MAYBE_NULL | PTR_TO_MAP_VALUE,
| ~~~~~~~~~~~~~~ ^ ~~~~~~~~~~~~~~~~
include/linux/bpf.h:892:42: warning: bitwise operation between different enumeration types ('enum bpf_type_flag' and 'enum bpf_reg_type') [-Wenum-enum-conversion]
892 | PTR_TO_SOCKET_OR_NULL = PTR_MAYBE_NULL | PTR_TO_SOCKET,
| ~~~~~~~~~~~~~~ ^ ~~~~~~~~~~~~~
include/linux/bpf.h:893:46: warning: bitwise operation between different enumeration types ('enum bpf_type_flag' and 'enum bpf_reg_type') [-Wenum-enum-conversion]
893 | PTR_TO_SOCK_COMMON_OR_NULL = PTR_MAYBE_NULL | PTR_TO_SOCK_COMMON,
| ~~~~~~~~~~~~~~ ^ ~~~~~~~~~~~~~~~~~~
include/linux/bpf.h:894:44: warning: bitwise operation between different enumeration types ('enum bpf_type_flag' and 'enum bpf_reg_type') [-Wenum-enum-conversion]
894 | PTR_TO_TCP_SOCK_OR_NULL = PTR_MAYBE_NULL | PTR_TO_TCP_SOCK,
| ~~~~~~~~~~~~~~ ^ ~~~~~~~~~~~~~~~
include/linux/bpf.h:895:42: warning: bitwise operation between different enumeration types ('enum bpf_type_flag' and 'enum bpf_reg_type') [-Wenum-enum-conversion]
895 | PTR_TO_BTF_ID_OR_NULL = PTR_MAYBE_NULL | PTR_TO_BTF_ID,
| ~~~~~~~~~~~~~~ ^ ~~~~~~~~~~~~~
>> kernel/bpf/core.c:900:5: warning: '__pte_index_size' is not defined, evaluates to 0 [-Wundef]
900 | #if PMD_SIZE <= (1 << 21)
| ^
arch/powerpc/include/asm/book3s/64/pgtable.h:208:26: note: expanded from macro 'PMD_SIZE'
208 | #define PMD_SIZE (1UL << PMD_SHIFT)
| ^
arch/powerpc/include/asm/book3s/64/pgtable.h:207:33: note: expanded from macro 'PMD_SHIFT'
207 | #define PMD_SHIFT (PAGE_SHIFT + PTE_INDEX_SIZE)
| ^
arch/powerpc/include/asm/book3s/64/pgtable.h:156:25: note: expanded from macro 'PTE_INDEX_SIZE'
156 | #define PTE_INDEX_SIZE __pte_index_size
| ^
26 warnings generated.
--
>> kernel/bpf/core.c:900:5: warning: '__pte_index_size' is not defined, evaluates to 0 [-Wundef]
900 | #if PMD_SIZE <= (1 << 21)
| ^
arch/powerpc/include/asm/book3s/64/pgtable.h:208:26: note: expanded from macro 'PMD_SIZE'
208 | #define PMD_SIZE (1UL << PMD_SHIFT)
| ^
arch/powerpc/include/asm/book3s/64/pgtable.h:207:33: note: expanded from macro 'PMD_SHIFT'
207 | #define PMD_SHIFT (PAGE_SHIFT + PTE_INDEX_SIZE)
| ^
arch/powerpc/include/asm/book3s/64/pgtable.h:156:25: note: expanded from macro 'PTE_INDEX_SIZE'
156 | #define PTE_INDEX_SIZE __pte_index_size
| ^1 warning generated.
vim +/__pte_index_size +900 kernel/bpf/core.c
891
892 /* PMD_SIZE is not available in some special config, e.g. ARCH=arm with
893 * CONFIG_MMU=n. Use PAGE_SIZE in these cases.
894 */
895 #ifdef PMD_SIZE
896 /* PMD_SIZE is really big for some archs. It doesn't make sense to
897 * reserve too much memory in one allocation. Cap BPF_PROG_PACK_SIZE to
898 * 2MiB * num_possible_nodes().
899 */
> 900 #if PMD_SIZE <= (1 << 21)
901 #define BPF_PROG_PACK_SIZE (PMD_SIZE * num_possible_nodes())
902 #else
903 #define BPF_PROG_PACK_SIZE ((1 << 21) * num_possible_nodes())
904 #endif
905 #else
906 #define BPF_PROG_PACK_SIZE PAGE_SIZE
907 #endif
908
Signature · 2 lines
0-DAY CI Kernel Test Service https://github.com/intel/lkp-tests/wiki
Puranjay,
Looks like I have to drop this patch for now,
since PMD_SIZE is not a constant on all archs
that can be evaluated by a preprocessor.
We need to find a different way.
Quoted 158 lines · kernel test robot
tree: https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next.git master head: d0d131e3b655fd267d14bb1bed49e3f990a1465e commit: 643714ff18301f41bded2489e8f1f9d1b4782094 [5/7] bpf: cap BPF_PROG_PACK_SIZE to 2MB * num_possible_nodes() config: powerpc-allyesconfig (https://download.01.org/0day-ci/archive/20240309/[email protected]/config) compiler: clang version 19.0.0git (https://github.com/llvm/llvm-project 503c55e17037436dcd45ac69dea8967e67e3f5e8) reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20240309/[email protected]/reproduce)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 <[email protected]>
| Closes: https://lore.kernel.org/oe-kbuild-all/[email protected]/All warnings (new ones prefixed by >>):
In file included from
kernel/bpf/core.c:21:
In file included frominclude/linux/filter.h:9:
In file included frominclude/linux/bpf.h:21:
In file included frominclude/linux/kallsyms.h:13:
In file included frominclude/linux/mm.h:2188:
include/linux/vmstat.h:508:43: warning: arithmetic between different enumeration types ('enumzone_stat_item' and 'enumnuma_stat_item') [-Wenum-enum-conversion]508 | return vmstat_text[NR_VM_ZONE_STAT_ITEMS + | ~~~~~~~~~~~~~~~~~~~~~ ^ 509 | item]; | ~~~~ include/linux/vmstat.h:515:43: warning: arithmetic between different enumeration types ('enum zone_stat_item' and 'enum numa_stat_item') [-Wenum-enum-conversion] 515 | return vmstat_text[NR_VM_ZONE_STAT_ITEMS + | ~~~~~~~~~~~~~~~~~~~~~ ^ 516 | NR_VM_NUMA_EVENT_ITEMS + | ~~~~~~~~~~~~~~~~~~~~~~ include/linux/vmstat.h:522:36: warning: arithmetic between different enumeration types ('enum node_stat_item' and 'enum lru_list') [-Wenum-enum-conversion] 522 | return node_stat_name(NR_LRU_BASE + lru) + 3; // skip "nr_" | ~~~~~~~~~~~ ^ ~~~ include/linux/vmstat.h:527:43: warning: arithmetic between different enumeration types ('enum zone_stat_item' and 'enum numa_stat_item') [-Wenum-enum-conversion] 527 | return vmstat_text[NR_VM_ZONE_STAT_ITEMS + | ~~~~~~~~~~~~~~~~~~~~~ ^ 528 | NR_VM_NUMA_EVENT_ITEMS + | ~~~~~~~~~~~~~~~~~~~~~~ include/linux/vmstat.h:536:43: warning: arithmetic between different enumeration types ('enum zone_stat_item' and 'enum numa_stat_item') [-Wenum-enum-conversion] 536 | return vmstat_text[NR_VM_ZONE_STAT_ITEMS + | ~~~~~~~~~~~~~~~~~~~~~ ^ 537 | NR_VM_NUMA_EVENT_ITEMS + | ~~~~~~~~~~~~~~~~~~~~~~ In file included from kernel/bpf/core.c:21: In file included from include/linux/filter.h:9: include/linux/bpf.h:739:48: warning: bitwise operation between different enumeration types ('enum bpf_type_flag' and 'enum bpf_arg_type') [-Wenum-enum-conversion] 739 | ARG_PTR_TO_MAP_VALUE_OR_NULL = PTR_MAYBE_NULL | ARG_PTR_TO_MAP_VALUE, | ~~~~~~~~~~~~~~ ^ ~~~~~~~~~~~~~~~~~~~~ include/linux/bpf.h:740:43: warning: bitwise operation between different enumeration types ('enum bpf_type_flag' and 'enum bpf_arg_type') [-Wenum-enum-conversion] 740 | ARG_PTR_TO_MEM_OR_NULL = PTR_MAYBE_NULL | ARG_PTR_TO_MEM, | ~~~~~~~~~~~~~~ ^ ~~~~~~~~~~~~~~ include/linux/bpf.h:741:43: warning: bitwise operation between different enumeration types ('enum bpf_type_flag' and 'enum bpf_arg_type') [-Wenum-enum-conversion] 741 | ARG_PTR_TO_CTX_OR_NULL = PTR_MAYBE_NULL | ARG_PTR_TO_CTX, | ~~~~~~~~~~~~~~ ^ ~~~~~~~~~~~~~~ include/linux/bpf.h:742:45: warning: bitwise operation between different enumeration types ('enum bpf_type_flag' and 'enum bpf_arg_type') [-Wenum-enum-conversion] 742 | ARG_PTR_TO_SOCKET_OR_NULL = PTR_MAYBE_NULL | ARG_PTR_TO_SOCKET, | ~~~~~~~~~~~~~~ ^ ~~~~~~~~~~~~~~~~~ include/linux/bpf.h:743:44: warning: bitwise operation between different enumeration types ('enum bpf_type_flag' and 'enum bpf_arg_type') [-Wenum-enum-conversion] 743 | ARG_PTR_TO_STACK_OR_NULL = PTR_MAYBE_NULL | ARG_PTR_TO_STACK, | ~~~~~~~~~~~~~~ ^ ~~~~~~~~~~~~~~~~ include/linux/bpf.h:744:45: warning: bitwise operation between different enumeration types ('enum bpf_type_flag' and 'enum bpf_arg_type') [-Wenum-enum-conversion] 744 | ARG_PTR_TO_BTF_ID_OR_NULL = PTR_MAYBE_NULL | ARG_PTR_TO_BTF_ID, | ~~~~~~~~~~~~~~ ^ ~~~~~~~~~~~~~~~~~ include/linux/bpf.h:748:38: warning: bitwise operation between different enumeration types ('enum bpf_type_flag' and 'enum bpf_arg_type') [-Wenum-enum-conversion] 748 | ARG_PTR_TO_UNINIT_MEM = MEM_UNINIT | ARG_PTR_TO_MEM, | ~~~~~~~~~~ ^ ~~~~~~~~~~~~~~ include/linux/bpf.h:750:45: warning: bitwise operation between different enumeration types ('enum bpf_type_flag' and 'enum bpf_arg_type') [-Wenum-enum-conversion] 750 | ARG_PTR_TO_FIXED_SIZE_MEM = MEM_FIXED_SIZE | ARG_PTR_TO_MEM, | ~~~~~~~~~~~~~~ ^ ~~~~~~~~~~~~~~ include/linux/bpf.h:773:48: warning: bitwise operation between different enumeration types ('enum bpf_type_flag' and 'enum bpf_return_type') [-Wenum-enum-conversion] 773 | RET_PTR_TO_MAP_VALUE_OR_NULL = PTR_MAYBE_NULL | RET_PTR_TO_MAP_VALUE, | ~~~~~~~~~~~~~~ ^ ~~~~~~~~~~~~~~~~~~~~ include/linux/bpf.h:774:45: warning: bitwise operation between different enumeration types ('enum bpf_type_flag' and 'enum bpf_return_type') [-Wenum-enum-conversion] 774 | RET_PTR_TO_SOCKET_OR_NULL = PTR_MAYBE_NULL | RET_PTR_TO_SOCKET, | ~~~~~~~~~~~~~~ ^ ~~~~~~~~~~~~~~~~~ include/linux/bpf.h:775:47: warning: bitwise operation between different enumeration types ('enum bpf_type_flag' and 'enum bpf_return_type') [-Wenum-enum-conversion] 775 | RET_PTR_TO_TCP_SOCK_OR_NULL = PTR_MAYBE_NULL | RET_PTR_TO_TCP_SOCK, | ~~~~~~~~~~~~~~ ^ ~~~~~~~~~~~~~~~~~~~ include/linux/bpf.h:776:50: warning: bitwise operation between different enumeration types ('enum bpf_type_flag' and 'enum bpf_return_type') [-Wenum-enum-conversion] 776 | RET_PTR_TO_SOCK_COMMON_OR_NULL = PTR_MAYBE_NULL | RET_PTR_TO_SOCK_COMMON, | ~~~~~~~~~~~~~~ ^ ~~~~~~~~~~~~~~~~~~~~~~ include/linux/bpf.h:778:49: warning: bitwise operation between different enumeration types ('enum bpf_type_flag' and 'enum bpf_return_type') [-Wenum-enum-conversion] 778 | RET_PTR_TO_DYNPTR_MEM_OR_NULL = PTR_MAYBE_NULL | RET_PTR_TO_MEM, | ~~~~~~~~~~~~~~ ^ ~~~~~~~~~~~~~~ include/linux/bpf.h:779:45: warning: bitwise operation between different enumeration types ('enum bpf_type_flag' and 'enum bpf_return_type') [-Wenum-enum-conversion] 779 | RET_PTR_TO_BTF_ID_OR_NULL = PTR_MAYBE_NULL | RET_PTR_TO_BTF_ID, | ~~~~~~~~~~~~~~ ^ ~~~~~~~~~~~~~~~~~ include/linux/bpf.h:780:43: warning: bitwise operation between different enumeration types ('enum bpf_type_flag' and 'enum bpf_return_type') [-Wenum-enum-conversion] 780 | RET_PTR_TO_BTF_ID_TRUSTED = PTR_TRUSTED | RET_PTR_TO_BTF_ID, | ~~~~~~~~~~~ ^ ~~~~~~~~~~~~~~~~~
include/linux/bpf.h:891:44: warning: bitwise operation between different enumeration types ('enumbpf_type_flag' and 'enumbpf_reg_type') [-Wenum-enum-conversion]
891 |PTR_TO_MAP_VALUE_OR_NULL=PTR_MAYBE_NULL|PTR_TO_MAP_VALUE,
| ~~~~~~~~~~~~~~ ^ ~~~~~~~~~~~~~~~~
include/linux/bpf.h:892:42: warning: bitwise operation between different enumeration types ('enumbpf_type_flag' and 'enumbpf_reg_type') [-Wenum-enum-conversion]
892 |PTR_TO_SOCKET_OR_NULL=PTR_MAYBE_NULL|PTR_TO_SOCKET,
| ~~~~~~~~~~~~~~ ^ ~~~~~~~~~~~~~
include/linux/bpf.h:893:46: warning: bitwise operation between different enumeration types ('enumbpf_type_flag' and 'enumbpf_reg_type') [-Wenum-enum-conversion]
893 |PTR_TO_SOCK_COMMON_OR_NULL=PTR_MAYBE_NULL|PTR_TO_SOCK_COMMON,
| ~~~~~~~~~~~~~~ ^ ~~~~~~~~~~~~~~~~~~
include/linux/bpf.h:894:44: warning: bitwise operation between different enumeration types ('enumbpf_type_flag' and 'enumbpf_reg_type') [-Wenum-enum-conversion]
894 |PTR_TO_TCP_SOCK_OR_NULL=PTR_MAYBE_NULL|PTR_TO_TCP_SOCK,
| ~~~~~~~~~~~~~~ ^ ~~~~~~~~~~~~~~~
include/linux/bpf.h:895:42: warning: bitwise operation between different enumeration types ('enumbpf_type_flag' and 'enumbpf_reg_type') [-Wenum-enum-conversion]
895 |PTR_TO_BTF_ID_OR_NULL=PTR_MAYBE_NULL|PTR_TO_BTF_ID,
| ~~~~~~~~~~~~~~ ^ ~~~~~~~~~~~~~Quoted 1 line
Quoted 1 line
kernel/bpf/core.c:900:5: warning: '__pte_index_size' is not defined, evaluates to 0 [-Wundef]900 | #if PMD_SIZE <= (1 << 21) | ^ arch/powerpc/include/asm/book3s/64/pgtable.h:208:26: note: expanded from macro 'PMD_SIZE' 208 | #define PMD_SIZE (1UL << PMD_SHIFT) | ^ arch/powerpc/include/asm/book3s/64/pgtable.h:207:33: note: expanded from macro 'PMD_SHIFT' 207 | #define PMD_SHIFT (PAGE_SHIFT + PTE_INDEX_SIZE) | ^ arch/powerpc/include/asm/book3s/64/pgtable.h:156:25: note: expanded from macro 'PTE_INDEX_SIZE' 156 | #define PTE_INDEX_SIZE __pte_index_size | ^26 warnings generated.
--Quoted 1 line
Quoted 1 line
kernel/bpf/core.c:900:5: warning: '__pte_index_size' is not defined, evaluates to 0 [-Wundef]900 | #if PMD_SIZE <= (1 << 21) | ^ arch/powerpc/include/asm/book3s/64/pgtable.h:208:26: note: expanded from macro 'PMD_SIZE' 208 | #define PMD_SIZE (1UL << PMD_SHIFT) | ^ arch/powerpc/include/asm/book3s/64/pgtable.h:207:33: note: expanded from macro 'PMD_SHIFT' 207 | #define PMD_SHIFT (PAGE_SHIFT + PTE_INDEX_SIZE) | ^ arch/powerpc/include/asm/book3s/64/pgtable.h:156:25: note: expanded from macro 'PTE_INDEX_SIZE' 156 | #define PTE_INDEX_SIZE __pte_index_size | ^1 warning generated.
vim +/__pte_index_size +900
kernel/bpf/core.c891
892 /*PMD_SIZEis not available in some special config, e.g. ARCH=arm with
893 *CONFIG_MMU=n. UsePAGE_SIZEin these cases.
894 */
895 #ifdefPMD_SIZE
896 /*PMD_SIZEis really big for some archs. It doesn't make sense to897 * reserve too much memory in one allocation. Cap BPF_PROG_PACK_SIZE to 898 * 2MiB * num_possible_nodes(). 899 */Quoted 1 line
900 #if
PMD_SIZE<= (1 << 21)901 #define BPF_PROG_PACK_SIZE (PMD_SIZE * num_possible_nodes()) 902 #else 903 #define BPF_PROG_PACK_SIZE ((1 << 21) * num_possible_nodes()) 904 #endif 905 #else 906 #define BPF_PROG_PACK_SIZE PAGE_SIZE 907 #endif908
--
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki
Hi Alexei,
Quoted 9 lines · Alexei Starovoitov
Puranjay,
Looks like I have to drop this patch for now,
sincePMD_SIZEis not a constant on all archs
that can be evaluated by a preprocessor.We need to find a different way.
How about we define it like:
-#define BPF_PROG_PACK_SIZE (PMD_SIZE * num_possible_nodes()) +/*
+ * PMD_SIZE is really big for some archs. It doesn't make sense to
+ * reserve too much memory in one allocation. Cap BPF_PROG_PACK_SIZE to
+ * 2MiB * num_possible_nodes().
+ */
+#define BPF_PROG_PACK_SIZE ((PMD_SIZE <= (1 << 21)) ? (PMD_SIZE * num_possible_nodes()) \
+
: ((1 << 21) * num_possible_nodes()))
So, it will be computed at runtime. This adds more performance overhead but this is called very infrequently
so shouldn't matter.
Or we can hardcode it as 2MB * num_possible_nodes() because PMD size in most architectures will be 2MB or larger but never smaller.
Thanks,
Puranjay
Quoted 25 lines · Puranjay Mohan
Hi Alexei,
Quoted 9 lines · Alexei Starovoitov
Puranjay,
Looks like I have to drop this patch for now,
sincePMD_SIZEis not a constant on all archs
that can be evaluated by a preprocessor.We need to find a different way.
How about we define it like:
(quoted diff) +6
-1-#define BPF_PROG_PACK_SIZE (PMD_SIZE * num_possible_nodes()) +/* + * PMD_SIZE is really big for some archs. It doesn't make sense to + * reserve too much memory in one allocation. Cap BPF_PROG_PACK_SIZE to + * 2MiB * num_possible_nodes(). + */ +#define BPF_PROG_PACK_SIZE ((PMD_SIZE <= (1 << 21)) ? (PMD_SIZE *
num_possible_nodes()) \
+
Quoted 8 lines
: ((1 << 21) * num_possible_nodes()))So, it will be computed at runtime. This adds more performance overhead but this is called very infrequently
so shouldn't matter.Or we can hardcode it as
2MB * num_possible_nodes()because PMD size in most architectures will be 2MB or larger but never smaller.
I think hard coding is cleaner and less surprising. I'd like to hear what Song thinks about it.
Song Liu
reviewerQuoted 33 lines · Alexei Starovoitov
Quoted 31 lines · Puranjay Mohan
Hi Alexei,
Quoted 9 lines · Alexei Starovoitov
Puranjay,
Looks like I have to drop this patch for now,
sincePMD_SIZEis not a constant on all archs
that can be evaluated by a preprocessor.We need to find a different way.
How about we define it like:
(quoted diff) +6
-1-#define BPF_PROG_PACK_SIZE (PMD_SIZE * num_possible_nodes()) +/* + * PMD_SIZE is really big for some archs. It doesn't make sense to + * reserve too much memory in one allocation. Cap BPF_PROG_PACK_SIZE to + * 2MiB * num_possible_nodes(). + */ +#define BPF_PROG_PACK_SIZE ((PMD_SIZE <= (1 << 21)) ? (PMD_SIZE *
num_possible_nodes()) \
+: ((1 << 21) * num_possible_nodes()))So, it will be computed at runtime. This adds more performance overhead but this is called very infrequently
so shouldn't matter.
Quoted 5 lines
Quoted 2 lines
Or we can hardcode it as
2MB * num_possible_nodes()because PMD size in most architectures will be 2MB or larger but never smaller.I think hard coding is cleaner and less surprising. I'd like to hear what Song thinks about it.
Agreed. Let's just hard code it to 2MiB * num_possible_nodes().
Thanks,
Song
Quoted 7 lines
Quoted 2 lines
I think hard coding is cleaner and less surprising. I'd like to hear what Song thinks about it.
Agreed. Let's just hard code it to 2MiB *
num_possible_nodes().Thanks,
Song
Here it is: https://lore.kernel.org/all/[email protected]/
Thanks,
Puranjay