Fixes for bad PTR_TO_BTF_ID offset
Subsystems (7)
Kumar Kartikeya Dwivedi
maintainerv4 cover Fixes for bad PTR_TO_BTF_ID offset
This set fixes a bug related to bad var_off being permitted for kfunc call in case of PTR_TO_BTF_ID, consolidates offset checks for all register types allowed as helper or kfunc arguments into a common shared helper, and introduces a couple of other checks to harden the kfunc release logic and prevent future bugs. Some selftests are also included that fail in absence of these fixes, serving as demonstration of the issues being fixed.
Changelog:
----------
v3 -> v4:
v3: https://lore.kernel.org/bpf/[email protected]
* Update commit message for __diag patch to say clang instead of LLVM (Nathan)
* Address nits for check_func_arg_reg_off (Martin)
* Add comment for fixed_off_ok case, remove is_kfunc check (Martin)
v2 -> v3:
v2: https://lore.kernel.org/bpf/[email protected]
* Add my SoB to __diag for clang patch (Nathan)
v1 -> v2:
v1: https://lore.kernel.org/bpf/[email protected]
* Put reg->off check for release kfunc inside check_func_arg_reg_off, make the check a bit more readable
* Squash verifier selftests errstr update into patch 3 for bisect (Alexei)
* Include fix from Nathan for clang warning about missing prototypes
* Add unified __diag_ingore_all that works for both GCC/LLVM (Alexei)
Older discussion:
Kumar Kartikeya Dwivedi (7):
bpf: Add check_func_arg_reg_off function
bpf: Fix PTR_TO_BTF_ID var_off check
bpf: Disallow negative offset in check_ptr_off_reg
bpf: Harden register offset checks for release helpers and kfuncs compiler_types.h: Add unified __diag_ignore_all for GCC/LLVM
bpf: Replace __diag_ignore with unified __diag_ignore_all selftests/bpf: Add tests for kfunc register offset checks
Nathan Chancellor (1):
compiler-clang.h: Add __diag infrastructure for clang
include/linux/bpf_verifier.h | 4 + include/linux/compiler-clang.h | 25 +++++ include/linux/compiler-gcc.h | 3 + include/linux/compiler_types.h | 4 + kernel/bpf/btf.c | 40 ++++---- kernel/bpf/verifier.c | 94 +++++++++++++------ net/bpf/test_run.c | 15 ++- net/netfilter/nf_conntrack_bpf.c | 5 +- .../selftests/bpf/verifier/bounds_deduction.c | 2 +- tools/testing/selftests/bpf/verifier/calls.c | 83 ++++++++++++++++ tools/testing/selftests/bpf/verifier/ctx.c | 8 +- 11 files changed, 230 insertions(+), 53 deletions(-)
Signature · 1 line
2.35.1
Kumar Kartikeya Dwivedi
maintainerPATCH v4 1/8 bpf: Add check_func_arg_reg_off function
Lift the list of register types allowed for having fixed and variable offsets when passed as helper function arguments into a common helper, so that they can be reused for kfunc checks in later commits. Keeping a common helper aids maintainability and allows us to follow the same consistent rules across helpers and kfuncs. Also, convert check_func_arg to use this function.
--- include/linux/bpf_verifier.h | 3 ++ kernel/bpf/verifier.c | 69 +++++++++++++++++++++--------------- 2 files changed, 44 insertions(+), 28 deletions(-)
include/linux/bpf_verifier.h +3 -0
@@ -521,6 +521,9 @@ bpf_prog_offload_remove_insns(struct bpf_verifier_env *env, u32 off, u32 cnt);int check_ptr_off_reg (struct bpf_verifier_env *env,const struct bpf_reg_state *reg,int regno); +int check_func_arg_reg_off (struct bpf_verifier_env *env, +const struct bpf_reg_state *reg,int regno, +enum bpf_arg_type arg_type);int check_kfunc_mem_size_reg (struct bpf_verifier_env *env,struct bpf_reg_state *reg,u32 regno); int check_mem_reg(struct bpf_verifier_env *env, struct bpf_reg_state *reg,
kernel/bpf/verifier.c +41 -28
81 lines not on this page · raw message
Signature · 1 line
2.35.1
Kumar Kartikeya Dwivedi
maintainerPATCH v4 2/8 bpf: Fix PTR_TO_BTF_ID var_off check
In reply to Kumar Kartikeya DwivediWhen kfunc support was added, check_ctx_reg was called for PTR_TO_CTX register, but no offset checks were made for PTR_TO_BTF_ID. Only reg->off was taken into account by btf_struct_ids_match, which protected against type mismatch due to non-zero reg->off, but when reg->off was zero, a user could set the variable offset of the register and allow it to be passed to kfunc, leading to bad pointer being passed into the kernel.
Fix this by reusing the extracted helper check_func_arg_reg_off from previous commit, and make one call before checking all supported register types. Since the list is maintained, any future changes will be taken into account by updating check_func_arg_reg_off. This function prevents non-zero var_off to be set for PTR_TO_BTF_ID, but still allows a fixed non-zero reg->off, which is needed for type matching to work correctly when using pointer arithmetic.
ARG_DONTCARE is passed as arg_type, since kfunc doesn't support accepting a ARG_PTR_TO_ALLOC_MEM without relying on size of parameter type from BTF (in case of pointer), or using a mem, len pair. The forcing of offset check for ARG_PTR_TO_ALLOC_MEM is done because ringbuf helpers obtain the size from the header located at the beginning of the memory region, hence any changes to the original pointer shouldn't be allowed. In case of kfunc, size is always known, either at verification time, or using the length parameter, hence this forcing is not required.
Since this check will happen once already for PTR_TO_CTX, remove the check_ptr_off_reg call inside its block.
--- kernel/bpf/btf.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-)
kernel/bpf/btf.c +6 -3
@@ -5726,7 +5726,7 @@ static int btf_check_func_arg_match(struct bpf_verifier_env *env,const char *func_name, *ref_tname;const struct btf_type *t, *ref_t;const struct btf_param *args; -int ref_regno =0 ; +int ref_regno =0 , ret;bool rel = false; t =btf_type_by_id (btf, func_id);
@@ -5776,6 +5776,11 @@ static int btf_check_func_arg_match(struct bpf_verifier_env *env, ref_t =btf_type_skip_modifiers (btf, t->type, &ref_id); ref_tname =btf_name_by_offset (btf, ref_t->name_off); + + ret =check_func_arg_reg_off (env, reg, regno,ARG_DONTCARE ); +if (ret <0 ) +return ret; +if (btf_get_prog_ctx_type (log, btf, t, env->prog->type, i)) {/* If function expects ctx type in BTF check that caller
@@ -5787,8 +5792,6 @@ static int btf_check_func_arg_match(struct bpf_verifier_env *env, i, btf_type_str(t));return -EINVAL ; } -if (check_ptr_off_reg (env, reg, regno)) -return -EINVAL ; }else if (is_kfunc && (reg->type ==PTR_TO_BTF_ID || (reg2btf_ids[base_type (reg->type)] && !type_flag (reg->type)))) {const struct btf_type *reg_ref_t;
Signature · 1 line
2.35.1
Kumar Kartikeya Dwivedi
maintainerPATCH v4 3/8 bpf: Disallow negative offset in check_ptr_off_reg
In reply to Kumar Kartikeya Dwivedicheck_ptr_off_reg only allows fixed offset to be set for PTR_TO_BTF_ID, where reg->off < 0 doesn't make sense. This would shift the pointer backwards, and fails later in btf_struct_ids_match or btf_struct_walk due to out of bounds access (since offset is interpreted as unsigned).
Improve the verifier by rejecting this case by using a better error message for BPF helpers and kfunc, by putting a check inside the check_func_arg_reg_off function.
Also, update existing verifier selftests to work with new error string.
--- kernel/bpf/verifier.c | 6 ++++++ tools/testing/selftests/bpf/verifier/bounds_deduction.c | 2 +- tools/testing/selftests/bpf/verifier/ctx.c | 8 ++++---- 3 files changed, 11 insertions(+), 5 deletions(-)
kernel/bpf/verifier.c +6 -0
@@ -3990,6 +3990,12 @@ static int __check_ptr_off_reg(struct bpf_verifier_env *env,* is only allowed in its original, unmodified form. */ +if (reg->off <0 ) { +verbose (env,"negative offset %s ptr R%d off=%d disallowed\n" , +reg_type_str (env, reg->type), regno, reg->off); +return -EACCES ; + } +if (!fixed_off_ok && reg->off) {verbose (env,"dereference of modified %s ptr R%d off=%d disallowed\n" ,reg_type_str (env, reg->type), regno, reg->off);
tools/testing/selftests/bpf/verifier/bounds_deduction.c +1 -1
@@ -105,7 +105,7 @@ BPF_EXIT_INSN(), }, .errstr_unpriv = "R1 has pointer with unsupported alu operation", - .errstr = "dereference of modified ctx ptr", + .errstr = "negative offset ctx ptr R1 off=-1 disallowed", .result = REJECT, .flags = F_NEEDS_EFFICIENT_UNALIGNED_ACCESS, },
tools/testing/selftests/bpf/verifier/ctx.c +4 -4
@@ -58,7 +58,7 @@ }, .prog_type = BPF_PROG_TYPE_SCHED_CLS, .result = REJECT, - .errstr = "dereference of modified ctx ptr", + .errstr = "negative offset ctx ptr R1 off=-612 disallowed", }, { "pass modified ctx pointer to helper, 2",
@@ -71,8 +71,8 @@ }, .result_unpriv = REJECT, .result = REJECT, - .errstr_unpriv = "dereference of modified ctx ptr", - .errstr = "dereference of modified ctx ptr", + .errstr_unpriv = "negative offset ctx ptr R1 off=-612 disallowed", + .errstr = "negative offset ctx ptr R1 off=-612 disallowed", }, { "pass modified ctx pointer to helper, 3",
@@ -141,7 +141,7 @@ .prog_type = BPF_PROG_TYPE_CGROUP_SOCK_ADDR, .expected_attach_type = BPF_CGROUP_UDP6_SENDMSG, .result = REJECT, - .errstr = "dereference of modified ctx ptr", + .errstr = "negative offset ctx ptr R1 off=-612 disallowed", }, { "pass ctx or null check, 5: null (connect)",
Signature · 1 line
2.35.1
Kumar Kartikeya Dwivedi
maintainerPATCH v4 4/8 bpf: Harden register offset checks for release helpers and kfuncs
In reply to Kumar Kartikeya DwivediLet's ensure that the PTR_TO_BTF_ID reg being passed in to release BPF helpers and kfuncs always has its offset set to 0. While not a real problem now, there's a very real possibility this will become a problem when more and more kfuncs are exposed, and more BPF helpers are added which can release PTR_TO_BTF_ID.
Previous commits already protected against non-zero var_off. One of the case we are concerned about now is when we have a type that can be returned by e.g. an acquire kfunc:
struct foo {
int a;
int b;
struct bar b;
};... and struct bar is also a type that can be returned by another acquire kfunc.
Then, doing the following sequence:
struct foo *f = bpf_get_foo(); // acquire kfunc if (!f) return 0; bpf_put_bar(&f->b); // release kfunc
... would work with the current code, since the btf_struct_ids_match takes reg->off into account for matching pointer type with release kfunc argument type, but would obviously be incorrect, and most likely lead to a kernel crash. A test has been included later to prevent regressions in this area.
--- include/linux/bpf_verifier.h | 3 ++- kernel/bpf/btf.c | 33 +++++++++++++++++++-------------- kernel/bpf/verifier.c | 25 ++++++++++++++++++++++--- 3 files changed, 43 insertions(+), 18 deletions(-)
include/linux/bpf_verifier.h +2 -1
@@ -523,7 +523,8 @@ int check_ptr_off_reg(struct bpf_verifier_env *env, const struct bpf_reg_state *reg, int regno);int check_func_arg_reg_off (struct bpf_verifier_env *env,const struct bpf_reg_state *reg,int regno, -enum bpf_arg_type arg_type); +enum bpf_arg_type arg_type, +bool is_release_func);int check_kfunc_mem_size_reg (struct bpf_verifier_env *env,struct bpf_reg_state *reg,u32 regno); int check_mem_reg(struct bpf_verifier_env *env, struct bpf_reg_state *reg,
kernel/bpf/btf.c +19 -14
@@ -5753,6 +5753,10 @@ static int btf_check_func_arg_match(struct bpf_verifier_env *env,return -EINVAL ; } +/* Only kfunc can be release func */ +if (is_kfunc) + rel =btf_kfunc_id_set_contains (btf,resolve_prog_type (env->prog), +BTF_KFUNC_TYPE_RELEASE , func_id);/* check that BTF function arguments match actual types that the * verifier sees. */
@@ -5777,7 +5781,7 @@ static int btf_check_func_arg_match(struct bpf_verifier_env *env, ref_t =btf_type_skip_modifiers (btf, t->type, &ref_id); ref_tname =btf_name_by_offset (btf, ref_t->name_off); - ret =check_func_arg_reg_off (env, reg, regno,ARG_DONTCARE ); + ret =check_func_arg_reg_off (env, reg, regno,ARG_DONTCARE , rel);if (ret <0 )return ret;
@@ -5809,7 +5813,11 @@ static int btf_check_func_arg_match(struct bpf_verifier_env *env,if (reg->type ==PTR_TO_BTF_ID ) { reg_btf = reg->btf; reg_ref_id = reg->btf_id; -/* Ensure only one argument is referenced PTR_TO_BTF_ID */ +/* Ensure only one argument is referenced +* PTR_TO_BTF_ID, check_func_arg_reg_off relies +* on only one referenced register being allowed +* for kfuncs. +*/ if (reg->ref_obj_id) {if (ref_obj_id) { bpf_log(log, "verifier internal error: more than one arg with ref_obj_id R%d %u %u\n",
@@ -5891,18 +5899,15 @@ static int btf_check_func_arg_match(struct bpf_verifier_env *env,/* Either both are set, or neither */ WARN_ON_ONCE ((ref_obj_id && !ref_regno) || (!ref_obj_id && ref_regno)); -if (is_kfunc) { - rel =btf_kfunc_id_set_contains (btf,resolve_prog_type (env->prog), -BTF_KFUNC_TYPE_RELEASE , func_id); -/* We already made sure ref_obj_id is set only for one argument */ -if (rel && !ref_obj_id) { -bpf_log (log,"release kernel function %s expects refcounted PTR_TO_BTF_ID\n" , - func_name); -return -EINVAL ; - } -/* Allow (!rel && ref_obj_id), so that passing such referenced PTR_TO_BTF_ID to -* other kfuncs works -*/ +/* We already made sure ref_obj_id is set only for one argument. We do +* allow (!rel && ref_obj_id), so that passing such referenced +* PTR_TO_BTF_ID to other kfuncs works. Note that rel is only true when +* is_kfunc is true. +*/ +if (rel && !ref_obj_id) { +bpf_log (log,"release kernel function %s expects refcounted PTR_TO_BTF_ID\n" , + func_name); +return -EINVAL ; }/* returns argument register number > 0 in case of reference release kfunc */ return rel ? ref_regno :0 ;
kernel/bpf/verifier.c +22 -3
@@ -5367,10 +5367,11 @@ static int check_reg_type(struct bpf_verifier_env *env, u32 regno,int check_func_arg_reg_off (struct bpf_verifier_env *env,const struct bpf_reg_state *reg,int regno, -enum bpf_arg_type arg_type) +enum bpf_arg_type arg_type, +bool is_release_func) { +bool fixed_off_ok = false, release_reg;enum bpf_reg_type type = reg->type; -bool fixed_off_ok = false;switch ((u32 )type) {case SCALAR_VALUE :
@@ -5395,6 +5396,21 @@ int check_func_arg_reg_off(struct bpf_verifier_env *env,* fixed offset. */ case PTR_TO_BTF_ID : +/* When referenced PTR_TO_BTF_ID is passed to release function, +* it's fixed offset must be 0. We rely on the property that +* only one referenced register can be passed to BPF helpers and +* kfuncs. In the other cases, fixed offset can be non-zero. +*/ + release_reg = is_release_func && reg->ref_obj_id; +if (release_reg && reg->off) { +verbose (env,"R%d must have zero offset when passed to release func\n" , + regno); +return -EINVAL ; + } +/* For release_reg == true, fixed_off_ok must be false, but we +* already checked and rejected reg->off != 0 above, so set to +* true to allow fixed offset for all other cases. +*/ fixed_off_ok = true;break ;default :
@@ -5452,11 +5468,14 @@ static int check_func_arg(struct bpf_verifier_env *env, u32 arg,if (err)return err; - err =check_func_arg_reg_off (env, reg, regno, arg_type); + err =check_func_arg_reg_off (env, reg, regno, arg_type,is_release_function (meta->func_id));if (err)return err; skip_type_check: +/* check_func_arg_reg_off relies on only one referenced register being +* allowed for BPF helpers. +*/ if (reg->ref_obj_id) {if (meta->ref_obj_id) { verbose(env, "verifier internal error: more than one arg with ref_obj_id R%d %u %u\n",
Signature · 1 line
2.35.1
Kumar Kartikeya Dwivedi
maintainerPATCH v4 5/8 compiler-clang.h: Add __diag infrastructure for clang
In reply to Kumar Kartikeya DwivediFrom: Nathan Chancellor <[email protected]>
Add __diag macros similar to those in compiler-gcc.h, so that warnings that need to be adjusted for specific cases but not globally can be ignored when building with clang.
--- include/linux/compiler-clang.h | 22 ++++++++++++++++++++++ 1 file changed, 22 insertions(+)
include/linux/compiler-clang.h +22 -0
@@ -68,3 +68,25 @@#define __nocfi __attribute__((__no_sanitize__("cfi")))#define __cficanonical __attribute__((__cfi_canonical_jump_table__)) + +/* +* Turn individual warnings and errors on and off locally, depending +* on version. +*/ +#define __diag_clang (version, severity, s) \ + __diag_clang_ ## version(__diag_clang_ ## severity s) + +/* Severity used in pragma directives */ +#define __diag_clang_ignore ignored +#define __diag_clang_warn warning +#define __diag_clang_error error + +#define __diag_str1 (s) #s +#define __diag_str (s) __diag_str1(s) +#define __diag (s) _Pragma(__diag_str(clang diagnostic s)) + +#if CONFIG_CLANG_VERSION >=110000 +#define __diag_clang_11 (s) __diag(s) +#else +#define __diag_clang_11 (s) +#endif
Signature · 1 line
2.35.1
Nick Desaulniers
reviewerRe: [PATCH bpf-next v4 5/8] compiler-clang.h: Add __diag infrastructure for clang
include/linux/compiler-clang.h
Quoted 21 lines · Kumar Kartikeya Dwivedi
From: Nathan Chancellor <[email protected]>
Add
__diagmacros similar to those incompiler-gcc.h, so that warnings that need to be adjusted for specific cases but not globally can be ignored when building with clang.Cc Nathan Chancellor <[email protected]>Cc Nick Desaulniers <[email protected]>Signed-off-by Nathan Chancellor <[email protected]>[ Kartikeya: wrote commit message ]Signed-off-by Kumar Kartikeya Dwivedi <[email protected]>--- include/linux/compiler-clang.h | 22 ++++++++++++++++++++++ 1 file changed, 22 insertions(+)include/linux/compiler-clang.h +0
-0
The equivalent functionality for GCC has
357 #ifndef __diag_GCC
358 #define __diag_GCC(version, severity, string)
359 #endif
in include/linux/compiler_types.h. Should this patch as well? (at least #define __diag_clang`)?
Quoted 6 lines
include/linux/compiler-clang.h +2
-0@@ -68,3 +68,25 @@#define __nocfi __attribute__((__no_sanitize__("cfi")))#define __cficanonical __attribute__((__cfi_canonical_jump_table__)) + +/*
include/linux/compiler-clang.h line 82
Quoted 10 lines
include/linux/compiler-clang.h +10
-0+* Turn individual warnings and errors on and off locally, depending +* on version. +*/ +#define __diag_clang (version, severity, s) \ + __diag_clang_ ## version(__diag_clang_ ## severity s) + +/* Severity used in pragma directives */ +#define __diag_clang_ignore ignored +#define __diag_clang_warn warning +#define __diag_clang_error error
These severities match GCC. I wonder if rather than copy+pasting these over, we could rework __diag_ignore, __diag_warn, and __diag_error to not invoke a compiler-suffixed macro and rather pass the compiler along (or make it implicit since we know CONFIG_CC_IS_CLANG vs CONFIG_CC_IS_GCC)? We can probably land this than follow up on better code-reuse between compilers for diagnostics.
Quoted 13 lines
include/linux/compiler-clang.h +10
-1+ +#define __diag_str1 (s) #s +#define __diag_str (s) __diag_str1(s) +#define __diag (s) _Pragma(__diag_str(clang diagnostic s)) + +#if CONFIG_CLANG_VERSION >=110000 +#define __diag_clang_11 (s) __diag(s) +#else +#define __diag_clang_11 (s) +#endif --2.35.1
Signature · 2 lines
Thanks, ~Nick Desaulniers
Nick Desaulniers
reviewerRe: [PATCH bpf-next v4 5/8] compiler-clang.h: Add __diag infrastructure for clang
Also, please cc me and the llvm mailing list on all changes to include/linux/compiler-clang.h. I see now on lore there's further patches here than just this single patch I was cc'ed on.
Quoted 73 lines · Nick Desaulniers
include/linux/compiler-clang.h
Quoted 21 lines · Kumar Kartikeya Dwivedi
From: Nathan Chancellor <[email protected]>
Add
__diagmacros similar to those incompiler-gcc.h, so that warnings that need to be adjusted for specific cases but not globally can be ignored when building with clang.Cc Nathan Chancellor <[email protected]>Cc Nick Desaulniers <[email protected]>Signed-off-by Nathan Chancellor <[email protected]>[ Kartikeya: wrote commit message ]Signed-off-by Kumar Kartikeya Dwivedi <[email protected]>--- include/linux/compiler-clang.h | 22 ++++++++++++++++++++++ 1 file changed, 22 insertions(+)include/linux/compiler-clang.h +0
-0The equivalent functionality for GCC has
357 #ifndef__diag_GCC
358 #define__diag_GCC(version, severity, string)
359 #endif
ininclude/linux/compiler_types.h. Should this patch as well? (at least #define __diag_clang`)?Quoted 6 lines
include/linux/compiler-clang.h +2
-0@@ -68,3 +68,25 @@#define __nocfi __attribute__((__no_sanitize__("cfi")))#define __cficanonical __attribute__((__cfi_canonical_jump_table__)) + +/* include/linux/compiler-clang.h line 82
Quoted 10 lines
include/linux/compiler-clang.h +10
-0+* Turn individual warnings and errors on and off locally, depending +* on version. +*/ +#define __diag_clang (version, severity, s) \ + __diag_clang_ ## version(__diag_clang_ ## severity s) + +/* Severity used in pragma directives */ +#define __diag_clang_ignore ignored +#define __diag_clang_warn warning +#define __diag_clang_error errorThese severities match GCC. I wonder if rather than copy+pasting these over, we could rework
__diag_ignore,__diag_warn, and__diag_errorto not invoke a compiler-suffixed macro and rather pass the compiler along (or make it implicit since we knowCONFIG_CC_IS_CLANGvsCONFIG_CC_IS_GCC)? We can probably land this than follow up on better code-reuse between compilers for diagnostics.Quoted 11 lines
include/linux/compiler-clang.h +10
-1+ +#define __diag_str1 (s) #s +#define __diag_str (s) __diag_str1(s) +#define __diag (s) _Pragma(__diag_str(clang diagnostic s)) + +#if CONFIG_CLANG_VERSION >=110000 +#define __diag_clang_11 (s) __diag(s) +#else +#define __diag_clang_11 (s) +#endif --Quoted 2 lines
2.35.1
--
Thanks,
~Nick Desaulniers
Signature · 2 lines
Thanks, ~Nick Desaulniers
Kumar Kartikeya Dwivedi
maintainerRe: [PATCH bpf-next v4 5/8] compiler-clang.h: Add __diag infrastructure for clang
Quoted 4 lines · Nick Desaulniers
Also, please cc me and the llvm mailing list on all changes to
include/linux/compiler-clang.h. I see now on lore there's further patches here than just this single patch I was cc'ed on.
Mea culpa, sorry about that.
Quoted 21 lines
Quoted 19 lines · Nick Desaulniers
Quoted 17 lines · Kumar Kartikeya Dwivedi
From: Nathan Chancellor <[email protected]>
Add
__diagmacros similar to those incompiler-gcc.h, so that warnings that need to be adjusted for specific cases but not globally can be ignored when building with clang.Cc Nathan Chancellor <[email protected]>Cc Nick Desaulniers <[email protected]>Signed-off-by Nathan Chancellor <[email protected]>[ Kartikeya: wrote commit message ]Signed-off-by Kumar Kartikeya Dwivedi <[email protected]>--- include/linux/compiler-clang.h | 22 ++++++++++++++++++++++ 1 file changed, 22 insertions(+)
Quoted 12 lines
Quoted 12 lines
include/linux/compiler-clang.h
Quoted 4 lines
include/linux/compiler-clang.h +0
-0The equivalent functionality for GCC has
357 #ifndef__diag_GCC
358 #define__diag_GCC(version, severity, string)
359 #endif
ininclude/linux/compiler_types.h. Should this patch as well? (at least #define __diag_clang`)?
I left it out because there are no users for it, no code is doing __diag_ignore(clang, ...), all of the current ones are hardcoding e.g. __diag_ignore(GCC, 8, ...).
Quoted 16 lines
Quoted 16 lines
Quoted 16 lines
include/linux/compiler-clang.h +12
-0@@ -68,3 +68,25 @@#define __nocfi __attribute__((__no_sanitize__("cfi")))#define __cficanonical __attribute__((__cfi_canonical_jump_table__)) + +/* +* Turn individual warnings and errors on and off locally, depending +* on version. +*/ +#define __diag_clang (version, severity, s) \ + __diag_clang_ ## version(__diag_clang_ ## severity s) + +/* Severity used in pragma directives */ +#define __diag_clang_ignore ignored +#define __diag_clang_warn warning +#define __diag_clang_error error
Quoted 8 lines
Quoted 8 lines
These severities match GCC. I wonder if rather than copy+pasting these over, we could rework
__diag_ignore,__diag_warn, and__diag_errorto not invoke a compiler-suffixed macro and rather pass the compiler along (or make it implicit since we knowCONFIG_CC_IS_CLANGvsCONFIG_CC_IS_GCC)? We can probably land this than follow up on better code-reuse between compilers for diagnostics.
That was the idea with __diag_ignore_all in the next patch. FYI this series has already been applied.
Quoted 21 lines
Quoted 10 lines
Quoted 10 lines
include/linux/compiler-clang.h +10
-0+ +#define __diag_str1 (s) #s +#define __diag_str (s) __diag_str1(s) +#define __diag (s) _Pragma(__diag_str(clang diagnostic s)) + +#if CONFIG_CLANG_VERSION >=110000 +#define __diag_clang_11 (s) __diag(s) +#else +#define __diag_clang_11 (s) +#endif Quoted 8 lines
Quoted 3 lines
--
2.35.1--
Thanks,
~Nick Desaulniers
Quoted 3 lines
--
Thanks,
~Nick Desaulniers
--
Kartikeya
Kumar Kartikeya Dwivedi
maintainerPATCH v4 6/8 compiler_types.h: Add unified __diag_ignore_all for GCC/LLVM
In reply to Kumar Kartikeya DwivediAdd a __diag_ignore_all macro, to ignore warnings for both GCC and LLVM, without having to specify the compiler type and version. By default, GCC 8 and clang 11 are used. This will be used by bpf subsystem to ignore -Wmissing-prototypes warning for functions that are meant to be global functions so that they are in vmlinux BTF, but don't have a prototype.
--- include/linux/compiler-clang.h | 3 +++ include/linux/compiler-gcc.h | 3 +++ include/linux/compiler_types.h | 4 ++++ 3 files changed, 10 insertions(+)
include/linux/compiler-clang.h +3 -0
@@ -90,3 +90,6 @@#else #define __diag_clang_11 (s)#endif + +#define __diag_ignore_all (option, comment) \ + __diag_clang(11, ignore, option)
include/linux/compiler-gcc.h +3 -0
@@ -151,6 +151,9 @@#define __diag_GCC_8 (s)#endif +#define __diag_ignore_all (option, comment) \ + __diag_GCC(8, ignore, option) +/* * Prior to 9.1, -Wno-alloc-size-larger-than (and therefore the "alloc_size" * attribute) do not work, and must be disabled.
include/linux/compiler_types.h +4 -0
@@ -371,4 +371,8 @@ struct ftrace_likely_data {#define __diag_error (compiler, version, option, comment) \ __diag_ ## compiler(version, error, option) +#ifndef __diag_ignore_all +#define __diag_ignore_all (option, comment) +#endif +#endif /* __LINUX_COMPILER_TYPES_H */
Signature · 1 line
2.35.1
Nick Desaulniers
reviewerRe: [PATCH bpf-next v4 6/8] compiler_types.h: Add unified __diag_ignore_all for GCC/LLVM
Quoted 30 lines · Kumar Kartikeya Dwivedi
Add a
__diag_ignore_allmacro, to ignore warnings for both GCC and LLVM, without having to specify the compiler type and version. By default, GCC 8 and clang 11 are used. This will be used by bpf subsystem to ignore -Wmissing-prototypes warning for functions that are meant to be global functions so that they are in vmlinux BTF, but don't have a prototype.Signed-off-by Kumar Kartikeya Dwivedi <[email protected]>--- include/linux/compiler-clang.h | 3 +++ include/linux/compiler-gcc.h | 3 +++ include/linux/compiler_types.h | 4 ++++ 3 files changed, 10 insertions(+)include/linux/compiler-clang.h +3
-0@@ -90,3 +90,6 @@#else #define __diag_clang_11 (s)#endif + +#define __diag_ignore_all (option, comment) \ + __diag_clang(11, ignore, option)include/linux/compiler-gcc.h +0
-0
include/linux/compiler-gcc.h line 155
Quoted 5 lines
include/linux/compiler-gcc.h +2
-0@@ -151,6 +151,9 @@#define __diag_GCC_8 (s)#endif +#define __diag_ignore_all (option, comment) \ + __diag_GCC(8, ignore, option)
While this approach will work for clang, it doesn't seem scalable for GCC. Documentation/process/changes.rst documents that we support gcc 5.1+. This approach will only disable diagnostics for gcc 8+.
Quoted 20 lines
include/linux/compiler-gcc.h +1
-0+/* * Prior to 9.1, -Wno-alloc-size-larger-than (and therefore the "alloc_size" * attribute) do not work, and must be disabled. include/linux/compiler_types.h +4
-0@@ -371,4 +371,8 @@ struct ftrace_likely_data {#define __diag_error (compiler, version, option, comment) \ __diag_ ## compiler(version, error, option) +#ifndef __diag_ignore_all +#define __diag_ignore_all (option, comment) +#endif +#endif /* __LINUX_COMPILER_TYPES_H */ Signature · 1 line
2.35.1
Kumar Kartikeya Dwivedi
maintainerRe: [PATCH bpf-next v4 6/8] compiler_types.h: Add unified __diag_ignore_all for GCC/LLVM
Quoted 34 lines · Nick Desaulniers
Quoted 33 lines · Kumar Kartikeya Dwivedi
Add a
__diag_ignore_allmacro, to ignore warnings for both GCC and LLVM, without having to specify the compiler type and version. By default, GCC 8 and clang 11 are used. This will be used by bpf subsystem to ignore -Wmissing-prototypes warning for functions that are meant to be global functions so that they are in vmlinux BTF, but don't have a prototype.Signed-off-by Kumar Kartikeya Dwivedi <[email protected]>--- include/linux/compiler-clang.h | 3 +++ include/linux/compiler-gcc.h | 3 +++ include/linux/compiler_types.h | 4 ++++ 3 files changed, 10 insertions(+)include/linux/compiler-clang.h +3
-0@@ -90,3 +90,6 @@#else #define __diag_clang_11 (s)#endif + +#define __diag_ignore_all (option, comment) \ + __diag_clang(11, ignore, option)include/linux/compiler-gcc.h +0
-0@@ -151,6 +151,9 @@#define __diag_GCC_8 (s)#endif
Quoted 7 lines
include/linux/compiler-gcc.h line 155
Quoted 2 lines
include/linux/compiler-gcc.h +2
-0+#define __diag_ignore_all (option, comment) \ + __diag_GCC(8, ignore, option)While this approach will work for clang, it doesn't seem scalable for GCC.
Documentation/process/changes.rstdocuments that we support gcc 5.1+. This approach will only disable diagnostics for gcc 8+.
ISTM the original commit adding these macros only defined them for GCC 8+, so coverage for previous versions is already not there, unrelated to this change. I am not sure what the reason for that was, though.
Quoted 12 lines
Quoted 12 lines
include/linux/compiler-gcc.h +1
-0+/* * Prior to 9.1, -Wno-alloc-size-larger-than (and therefore the "alloc_size" * attribute) do not work, and must be disabled. include/linux/compiler_types.h +0
-0@@ -371,4 +371,8 @@ struct ftrace_likely_data {#define __diag_error (compiler, version, option, comment) \ __diag_ ## compiler(version, error, option)
Quoted 8 lines
Quoted 8 lines
include/linux/compiler_types.h +4
-1+#ifndef __diag_ignore_all +#define __diag_ignore_all (option, comment) +#endif +#endif /* __LINUX_COMPILER_TYPES_H */ --2.35.1
--
Kartikeya
Kumar Kartikeya Dwivedi
maintainerPATCH v4 7/8 bpf: Replace __diag_ignore with unified __diag_ignore_all
In reply to Kumar Kartikeya DwivediCurrently, -Wmissing-prototypes warning is ignored for GCC, but not clang. This leads to clang build warning in W=1 mode. Since the flag used by both compilers is same, we can use the unified __diag_ignore_all macro that works for all supported versions and compilers which have __diag macro support (currently GCC >= 8.0, and Clang >= 11.0).
Also add nf_conntrack_bpf.h include to prevent missing prototype warning for register_nf_conntrack_bpf.
--- net/bpf/test_run.c | 4 ++-- net/netfilter/nf_conntrack_bpf.c | 5 +++-- 2 files changed, 5 insertions(+), 4 deletions(-)
net/bpf/test_run.c +2 -2
@@ -201,8 +201,8 @@ static int bpf_test_finish(const union bpf_attr *kattr,* future. */ __diag_push (); -__diag_ignore (GCC ,8 ,"-Wmissing-prototypes" , -"Global functions as their definitions will be in vmlinux BTF" ); +__diag_ignore_all ("-Wmissing-prototypes" , +"Global functions as their definitions will be in vmlinux BTF" ); int noinline bpf_fentry_test1(int a) {return a +1 ;
net/netfilter/nf_conntrack_bpf.c +3 -2
@@ -12,6 +12,7 @@#include <linux/btf_ids.h> #include <linux/net_namespace.h> #include <net/netfilter/nf_conntrack.h> +#include <net/netfilter/nf_conntrack_bpf.h> #include <net/netfilter/nf_conntrack_core.h> /* bpf_ct_opts - Options for CT lookup helpers
@@ -102,8 +103,8 @@ static struct nf_conn *__bpf_nf_ct_lookup(struct net *net, }__diag_push (); -__diag_ignore (GCC ,8 ,"-Wmissing-prototypes" , -"Global functions as their definitions will be in nf_conntrack BTF" ); +__diag_ignore_all ("-Wmissing-prototypes" , +"Global functions as their definitions will be in nf_conntrack BTF" );/* bpf_xdp_ct_lookup - Lookup CT entry for the given tuple, and acquire a * reference to it
Signature · 1 line
2.35.1
Kumar Kartikeya Dwivedi
maintainerPATCH v4 8/8 selftests/bpf: Add tests for kfunc register offset checks
In reply to Kumar Kartikeya DwivediInclude a few verifier selftests that test against the problems being fixed by previous commits, i.e. release kfunc always require PTR_TO_BTF_ID fixed and var_off to be 0, and negative offset is not permitted and returns a helpful error message.
--- net/bpf/test_run.c | 11 +++ tools/testing/selftests/bpf/verifier/calls.c | 83 ++++++++++++++++++++ 2 files changed, 94 insertions(+)
net/bpf/test_run.c +11 -0
@@ -270,9 +270,14 @@ struct sock * noinline bpf_kfunc_call_test3(struct sock *sk)return sk; } +struct prog_test_member { +u64 c; +}; +struct prog_test_ref_kfunc {int a;int b; +struct prog_test_member memb;struct prog_test_ref_kfunc *next; };
@@ -295,6 +300,10 @@ noinline void bpf_kfunc_call_test_release(struct prog_test_ref_kfunc *p) { } +noinline void bpf_kfunc_call_memb_release(struct prog_test_member *p) +{ +} +struct prog_test_pass1 {int x0;struct {
@@ -379,6 +388,7 @@ BTF_ID(func, bpf_kfunc_call_test2) BTF_ID(func, bpf_kfunc_call_test3) BTF_ID(func, bpf_kfunc_call_test_acquire) BTF_ID(func, bpf_kfunc_call_test_release) +BTF_ID(func, bpf_kfunc_call_memb_release) BTF_ID(func, bpf_kfunc_call_test_pass_ctx) BTF_ID(func, bpf_kfunc_call_test_pass1)BTF_ID (func, bpf_kfunc_call_test_pass2)
@@ -396,6 +406,7 @@ BTF_SET_END(test_sk_acquire_kfunc_ids) BTF_SET_START(test_sk_release_kfunc_ids) BTF_ID(func, bpf_kfunc_call_test_release) +BTF_ID(func, bpf_kfunc_call_memb_release) BTF_SET_END(test_sk_release_kfunc_ids) BTF_SET_START(test_sk_ret_null_kfunc_ids)
tools/testing/selftests/bpf/verifier/calls.c +83 -0
89 lines not on this page · raw message
Signature · 1 line
2.35.1
Quoted 6 lines · Kumar Kartikeya Dwivedi
This set fixes a bug related to bad
var_offbeing permitted for kfunc call in case ofPTR_TO_BTF_ID, consolidates offset checks for all register types allowed as helper or kfunc arguments into a common shared helper, and introduces a couple of other checks to harden the kfunc release logic and prevent future bugs. Some selftests are also included that fail in absence of these fixes, serving as demonstration of the issues being fixed.
This series was applied to bpf/bpf-next.git (master) by Alexei Starovoitov <[email protected]>:
Hello:
This series was applied to bpf/bpf-next.git (master) by Alexei Starovoitov <[email protected]>:
Quoted 8 lines · On Sat, 5 Mar 2022 04:16:37 +0530 you wrote:
This set fixes a bug related to bad
var_offbeing permitted for kfunc call in case ofPTR_TO_BTF_ID, consolidates offset checks for all register types allowed as helper or kfunc arguments into a common shared helper, and introduces a couple of other checks to harden the kfunc release logic and prevent future bugs. Some selftests are also included that fail in absence of these fixes, serving as demonstration of the issues being fixed.[...]
Here is the summary with links:
- [bpf-next,v4,1/8] bpf: Add check_func_arg_reg_off function
https://git.kernel.org/bpf/bpf-next/c/25b35dd28138 - [bpf-next,v4,2/8] bpf: Fix PTR_TO_BTF_ID var_off check https://git.kernel.org/bpf/bpf-next/c/655efe5089f0 - [bpf-next,v4,3/8] bpf: Disallow negative offset in check_ptr_off_reg https://git.kernel.org/bpf/bpf-next/c/e1fad0ff46b3 - [bpf-next,v4,4/8] bpf: Harden register offset checks for release helpers and kfuncs https://git.kernel.org/bpf/bpf-next/c/24d5bb806c7e - [bpf-next,v4,5/8] compiler-clang.h: Add __diag infrastructure for clang https://git.kernel.org/bpf/bpf-next/c/f014a00bbeb0 - [bpf-next,v4,6/8] compiler_types.h: Add unified __diag_ignore_all for GCC/LLVM https://git.kernel.org/bpf/bpf-next/c/4d1ea705d797 - [bpf-next,v4,7/8] bpf: Replace __diag_ignore with unified __diag_ignore_all https://git.kernel.org/bpf/bpf-next/c/0b206c6d1066 - [bpf-next,v4,8/8] selftests/bpf: Add tests for kfunc register offset checks https://git.kernel.org/bpf/bpf-next/c/8218ccb5bd68
You are awesome, thank you!
Signature · 2 lines
Deet-doot-dot, I am a bot. https://korg.docs.kernel.org/patchwork/pwbot.html