linux-next: manual merge of the tip tree with the bpf-next tree

linux-next: manual merge of the tip tree with the bpf-next tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 814 bytes --]

Hi all,

Today's linux-next merge of the tip tree got a conflict in:

  arch/x86/net/bpf_jit_comp.c

between commit:

  e782bdcf58c5 ("bpf, x64: remove ld_abs/ld_ind")

from the bpf-next tree and commit:

  5f26c50143f5 ("x86/bpf: Clean up non-standard comments, to make the code more readable")

from the tip tree.

I fixed it up (the former commit removed some code modified by the latter,
so I just removed it) and can carry the fix as necessary. This is now
fixed as far as linux-next is concerned, but any non trivial conflicts
should be mentioned to your upstream maintainer when your tree is
submitted for merging.  You may also want to consider cooperating with
the maintainer of the conflicting tree to minimise any particularly
complex conflicts.

-- 
Cheers,
Stephen Rothwell

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

Re: linux-next: manual merge of the tip tree with the bpf-next tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 1086 bytes --]

Hi all,

On Mon, 7 May 2018 12:09:09 +1000 Stephen Rothwell <sfr@canb.auug.org.au> wrote:
>
> Today's linux-next merge of the tip tree got a conflict in:
> 
>   arch/x86/net/bpf_jit_comp.c
> 
> between commit:
> 
>   e782bdcf58c5 ("bpf, x64: remove ld_abs/ld_ind")
> 
> from the bpf-next tree and commit:
> 
>   5f26c50143f5 ("x86/bpf: Clean up non-standard comments, to make the code more readable")
> 
> from the tip tree.
> 
> I fixed it up (the former commit removed some code modified by the latter,
> so I just removed it) and can carry the fix as necessary. This is now
> fixed as far as linux-next is concerned, but any non trivial conflicts
> should be mentioned to your upstream maintainer when your tree is
> submitted for merging.  You may also want to consider cooperating with
> the maintainer of the conflicting tree to minimise any particularly
> complex conflicts.

Actually the tip tree commit has been added to the bpf-next tree as a
different commit, so dropping it from the tip tree will clean this up.

-- 
Cheers,
Stephen Rothwell

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

Re: linux-next: manual merge of the tip tree with the bpf-next tree

[Daniel Borkmann]

On 05/07/2018 06:10 AM, Stephen Rothwell wrote:
> On Mon, 7 May 2018 12:09:09 +1000 Stephen Rothwell <sfr@canb.auug.org.au> wrote:
>>
>> Today's linux-next merge of the tip tree got a conflict in:
>>
>>   arch/x86/net/bpf_jit_comp.c
>>
>> between commit:
>>
>>   e782bdcf58c5 ("bpf, x64: remove ld_abs/ld_ind")
>>
>> from the bpf-next tree and commit:
>>
>>   5f26c50143f5 ("x86/bpf: Clean up non-standard comments, to make the code more readable")
>>
>> from the tip tree.
>>
>> I fixed it up (the former commit removed some code modified by the latter,
>> so I just removed it) and can carry the fix as necessary. This is now
>> fixed as far as linux-next is concerned, but any non trivial conflicts
>> should be mentioned to your upstream maintainer when your tree is
>> submitted for merging.  You may also want to consider cooperating with
>> the maintainer of the conflicting tree to minimise any particularly
>> complex conflicts.
> 
> Actually the tip tree commit has been added to the bpf-next tree as a
> different commit, so dropping it from the tip tree will clean this up.

Yep, it's been cherry-picked into bpf-next to avoid merge conflicts with
ongoing work.

Re: linux-next: manual merge of the tip tree with the bpf-next tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 1468 bytes --]

Hi all,

On Mon, 7 May 2018 10:15:45 +0200 Daniel Borkmann <daniel@iogearbox.net> wrote:
>
> On 05/07/2018 06:10 AM, Stephen Rothwell wrote:
> > On Mon, 7 May 2018 12:09:09 +1000 Stephen Rothwell <sfr@canb.auug.org.au> wrote:  
> >>
> >> Today's linux-next merge of the tip tree got a conflict in:
> >>
> >>   arch/x86/net/bpf_jit_comp.c
> >>
> >> between commit:
> >>
> >>   e782bdcf58c5 ("bpf, x64: remove ld_abs/ld_ind")
> >>
> >> from the bpf-next tree and commit:
> >>
> >>   5f26c50143f5 ("x86/bpf: Clean up non-standard comments, to make the code more readable")
> >>
> >> from the tip tree.
> >>
> >> I fixed it up (the former commit removed some code modified by the latter,
> >> so I just removed it) and can carry the fix as necessary. This is now
> >> fixed as far as linux-next is concerned, but any non trivial conflicts
> >> should be mentioned to your upstream maintainer when your tree is
> >> submitted for merging.  You may also want to consider cooperating with
> >> the maintainer of the conflicting tree to minimise any particularly
> >> complex conflicts.  
> > 
> > Actually the tip tree commit has been added to the bpf-next tree as a
> > different commit, so dropping it from the tip tree will clean this up.  
> 
> Yep, it's been cherry-picked into bpf-next to avoid merge conflicts with
> ongoing work.

This is now a conflict between the net-next tree and Linus' tree.

-- 
Cheers,
Stephen Rothwell

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

linux-next: manual merge of the tip tree with the bpf-next tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 2033 bytes --]

Hi all,

Today's linux-next merge of the tip tree got a conflict in:

  arch/x86/net/bpf_jit_comp.c

between commit:

  271de525e1d7 ("bpf: Remove prog->active check for bpf_lsm and bpf_iter")

from the bpf-next tree and commit:

  b2e9dfe54be4 ("x86/bpf: Emit call depth accounting if required")

from the tip tree.

I fixed it up (see below) and can carry the fix as necessary. This
is now fixed as far as linux-next is concerned, but any non trivial
conflicts should be mentioned to your upstream maintainer when your tree
is submitted for merging.  You may also want to consider cooperating
with the maintainer of the conflicting tree to minimise any particularly
complex conflicts.

-- 
Cheers,
Stephen Rothwell

diff --cc arch/x86/net/bpf_jit_comp.c
index cec5195602bc,f46b62029d91..000000000000
--- a/arch/x86/net/bpf_jit_comp.c
+++ b/arch/x86/net/bpf_jit_comp.c
@@@ -11,8 -11,8 +11,9 @@@
  #include <linux/bpf.h>
  #include <linux/memory.h>
  #include <linux/sort.h>
 +#include <linux/init.h>
  #include <asm/extable.h>
+ #include <asm/ftrace.h>
  #include <asm/set_memory.h>
  #include <asm/nospec-branch.h>
  #include <asm/text-patching.h>
@@@ -1930,7 -1869,7 +1948,7 @@@ static int invoke_bpf_prog(const struc
  	/* arg2: lea rsi, [rbp - ctx_cookie_off] */
  	EMIT4(0x48, 0x8D, 0x75, -run_ctx_off);
  
- 	if (emit_call(&prog, bpf_trampoline_enter(p), prog))
 -	if (emit_rsb_call(&prog, enter, prog))
++	if (emit_rsb_call(&prog, bpf_trampoline_enter(p), prog))
  		return -EINVAL;
  	/* remember prog start time returned by __bpf_prog_enter */
  	emit_mov_reg(&prog, true, BPF_REG_6, BPF_REG_0);
@@@ -1975,7 -1914,7 +1993,7 @@@
  	emit_mov_reg(&prog, true, BPF_REG_2, BPF_REG_6);
  	/* arg3: lea rdx, [rbp - run_ctx_off] */
  	EMIT4(0x48, 0x8D, 0x55, -run_ctx_off);
- 	if (emit_call(&prog, bpf_trampoline_exit(p), prog))
 -	if (emit_rsb_call(&prog, exit, prog))
++	if (emit_rsb_call(&prog, bpf_trampoline_exit(p), prog))
  		return -EINVAL;
  
  	*pprog = prog;

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

linux-next: manual merge of the tip tree with the bpf-next tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 5871 bytes --]

Hi all,

Today's linux-next merge of the tip tree got a conflict in:

  tools/testing/selftests/bpf/prog_tests/usdt.c

between commit:

  69424097ee10 ("selftests/bpf: Enrich subtest_basic_usdt case in selftests to cover SIB handling logic")

from the bpf-next tree and commit:

  875e1705ad99 ("selftests/bpf: Add optimized usdt variant for basic usdt test")

from the tip tree.

I fixed it up (see below) and can carry the fix as necessary. This
is now fixed as far as linux-next is concerned, but any non trivial
conflicts should be mentioned to your upstream maintainer when your tree
is submitted for merging.  You may also want to consider cooperating
with the maintainer of the conflicting tree to minimise any particularly
complex conflicts.

-- 
Cheers,
Stephen Rothwell

diff --cc tools/testing/selftests/bpf/prog_tests/usdt.c
index 615e9c3e93bf,833eb87483a1..000000000000
--- a/tools/testing/selftests/bpf/prog_tests/usdt.c
+++ b/tools/testing/selftests/bpf/prog_tests/usdt.c
@@@ -40,73 -40,20 +40,80 @@@ static void __always_inline trigger_fun
  	}
  }
  
 +#if defined(__x86_64__) || defined(__i386__)
 +/*
 + * SIB (Scale-Index-Base) addressing format: "size@(base_reg, index_reg, scale)"
 + * - 'size' is the size in bytes of the array element, and its sign indicates
 + *   whether the type is signed (negative) or unsigned (positive).
 + * - 'base_reg' is the register holding the base address, normally rdx or edx
 + * - 'index_reg' is the register holding the index, normally rax or eax
 + * - 'scale' is the scaling factor (typically 1, 2, 4, or 8), which matches the
 + *    size of the element type.
 + *
 + * For example, for an array of 'short' (signed 2-byte elements), the SIB spec would be:
 + * - size: -2 (negative because 'short' is signed)
 + * - scale: 2 (since sizeof(short) == 2)
 + *
 + * The resulting SIB format: "-2@(%%rdx,%%rax,2)" for x86_64, "-2@(%%edx,%%eax,2)" for i386
 + */
 +static volatile short array[] = {-1, -2, -3, -4};
 +
 +#if defined(__x86_64__)
 +#define USDT_SIB_ARG_SPEC -2@(%%rdx,%%rax,2)
 +#else
 +#define USDT_SIB_ARG_SPEC -2@(%%edx,%%eax,2)
 +#endif
 +
 +unsigned short test_usdt_sib_semaphore SEC(".probes");
 +
 +static void trigger_sib_spec(void)
 +{
 +	/*
 +	 * Force SIB addressing with inline assembly.
 +	 *
 +	 * You must compile with -std=gnu99 or -std=c99 to use the
 +	 * STAP_PROBE_ASM macro.
 +	 *
 +	 * The STAP_PROBE_ASM macro generates a quoted string that gets
 +	 * inserted between the surrounding assembly instructions. In this
 +	 * case, USDT_SIB_ARG_SPEC is embedded directly into the instruction
 +	 * stream, creating a probe point between the asm statement boundaries.
 +	 * It works fine with gcc/clang.
 +	 *
 +	 * Register constraints:
 +	 * - "d"(array): Binds the 'array' variable to %rdx or %edx register
 +	 * - "a"(0): Binds the constant 0 to %rax or %eax register
 +	 * These ensure that when USDT_SIB_ARG_SPEC references %%rdx(%edx) and
 +	 * %%rax(%eax), they contain the expected values for SIB addressing.
 +	 *
 +	 * The "memory" clobber prevents the compiler from reordering memory
 +	 * accesses around the probe point, ensuring that the probe behavior
 +	 * is predictable and consistent.
 +	 */
 +	asm volatile(
 +		STAP_PROBE_ASM(test, usdt_sib, USDT_SIB_ARG_SPEC)
 +		:
 +		: "d"(array), "a"(0)
 +		: "memory"
 +	);
 +}
 +#endif
 +
- static void subtest_basic_usdt(void)
+ static void subtest_basic_usdt(bool optimized)
  {
  	LIBBPF_OPTS(bpf_usdt_opts, opts);
  	struct test_usdt *skel;
  	struct test_usdt__bss *bss;
- 	int err, i;
+ 	int err, i, called;
 +	const __u64 expected_cookie = 0xcafedeadbeeffeed;
  
+ #define TRIGGER(x) ({			\
+ 	trigger_func(x);		\
+ 	if (optimized)			\
+ 		trigger_func(x);	\
+ 	optimized ? 2 : 1;		\
+ 	})
+ 
  	skel = test_usdt__open_and_load();
  	if (!ASSERT_OK_PTR(skel, "skel_open"))
  		return;
@@@ -126,22 -73,13 +133,22 @@@
  	if (!ASSERT_OK_PTR(skel->links.usdt0, "usdt0_link"))
  		goto cleanup;
  
 +#if defined(__x86_64__) || defined(__i386__)
 +	opts.usdt_cookie = expected_cookie;
 +	skel->links.usdt_sib = bpf_program__attach_usdt(skel->progs.usdt_sib,
 +							 0 /*self*/, "/proc/self/exe",
 +							 "test", "usdt_sib", &opts);
 +	if (!ASSERT_OK_PTR(skel->links.usdt_sib, "usdt_sib_link"))
 +		goto cleanup;
 +#endif
 +
- 	trigger_func(1);
+ 	called = TRIGGER(1);
  
- 	ASSERT_EQ(bss->usdt0_called, 1, "usdt0_called");
- 	ASSERT_EQ(bss->usdt3_called, 1, "usdt3_called");
- 	ASSERT_EQ(bss->usdt12_called, 1, "usdt12_called");
+ 	ASSERT_EQ(bss->usdt0_called, called, "usdt0_called");
+ 	ASSERT_EQ(bss->usdt3_called, called, "usdt3_called");
+ 	ASSERT_EQ(bss->usdt12_called, called, "usdt12_called");
  
 -	ASSERT_EQ(bss->usdt0_cookie, 0xcafedeadbeeffeed, "usdt0_cookie");
 +	ASSERT_EQ(bss->usdt0_cookie, expected_cookie, "usdt0_cookie");
  	ASSERT_EQ(bss->usdt0_arg_cnt, 0, "usdt0_arg_cnt");
  	ASSERT_EQ(bss->usdt0_arg_ret, -ENOENT, "usdt0_arg_ret");
  	ASSERT_EQ(bss->usdt0_arg_size, -ENOENT, "usdt0_arg_size");
@@@ -225,18 -163,9 +232,19 @@@
  	ASSERT_EQ(bss->usdt3_args[1], 42, "usdt3_arg2");
  	ASSERT_EQ(bss->usdt3_args[2], (uintptr_t)&bla, "usdt3_arg3");
  
 +#if defined(__x86_64__) || defined(__i386__)
 +	trigger_sib_spec();
 +	ASSERT_EQ(bss->usdt_sib_called, 1, "usdt_sib_called");
 +	ASSERT_EQ(bss->usdt_sib_cookie, expected_cookie, "usdt_sib_cookie");
 +	ASSERT_EQ(bss->usdt_sib_arg_cnt, 1, "usdt_sib_arg_cnt");
 +	ASSERT_EQ(bss->usdt_sib_arg, nums[0], "usdt_sib_arg");
 +	ASSERT_EQ(bss->usdt_sib_arg_ret, 0, "usdt_sib_arg_ret");
 +	ASSERT_EQ(bss->usdt_sib_arg_size, sizeof(nums[0]), "usdt_sib_arg_size");
 +#endif
 +
  cleanup:
  	test_usdt__destroy(skel);
+ #undef TRIGGER
  }
  
  unsigned short test_usdt_100_semaphore SEC(".probes");

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

Re: linux-next: manual merge of the tip tree with the bpf-next tree

[Alexei Starovoitov]

On Thu, Sep 11, 2025 at 7:41 PM Stephen Rothwell <sfr@canb.auug.org.au> wrote:
>
> Hi all,
>
> Today's linux-next merge of the tip tree got a conflict in:
>
>   tools/testing/selftests/bpf/prog_tests/usdt.c
>
> between commit:
>
>   69424097ee10 ("selftests/bpf: Enrich subtest_basic_usdt case in selftests to cover SIB handling logic")
>
> from the bpf-next tree and commit:
>
>   875e1705ad99 ("selftests/bpf: Add optimized usdt variant for basic usdt test")
>
> from the tip tree.
>
> I fixed it up (see below) and can carry the fix as necessary. This
> is now fixed as far as linux-next is concerned, but any non trivial
> conflicts should be mentioned to your upstream maintainer when your tree
> is submitted for merging.  You may also want to consider cooperating
> with the maintainer of the conflicting tree to minimise any particularly
> complex conflicts.

Thanks for headsup. Looks good.

Re: linux-next: manual merge of the tip tree with the bpf-next tree

[Jiri Olsa]

On Fri, Sep 12, 2025 at 12:40:59PM +1000, Stephen Rothwell wrote:
> Hi all,
> 
> Today's linux-next merge of the tip tree got a conflict in:
> 
>   tools/testing/selftests/bpf/prog_tests/usdt.c
> 
> between commit:
> 
>   69424097ee10 ("selftests/bpf: Enrich subtest_basic_usdt case in selftests to cover SIB handling logic")
> 
> from the bpf-next tree and commit:
> 
>   875e1705ad99 ("selftests/bpf: Add optimized usdt variant for basic usdt test")
> 
> from the tip tree.
> 
> I fixed it up (see below) and can carry the fix as necessary. This
> is now fixed as far as linux-next is concerned, but any non trivial
> conflicts should be mentioned to your upstream maintainer when your tree
> is submitted for merging.  You may also want to consider cooperating
> with the maintainer of the conflicting tree to minimise any particularly
> complex conflicts.

hi,
fwiw the conflict was mentioned in here:
  https://lore.kernel.org/bpf/aMAiMrLlfmG9FbQ3@krava/

the fix looks good, thanks

jirka

> 
> -- 
> Cheers,
> Stephen Rothwell
> 
> diff --cc tools/testing/selftests/bpf/prog_tests/usdt.c
> index 615e9c3e93bf,833eb87483a1..000000000000
> --- a/tools/testing/selftests/bpf/prog_tests/usdt.c
> +++ b/tools/testing/selftests/bpf/prog_tests/usdt.c
> @@@ -40,73 -40,20 +40,80 @@@ static void __always_inline trigger_fun
>   	}
>   }
>   
>  +#if defined(__x86_64__) || defined(__i386__)
>  +/*
>  + * SIB (Scale-Index-Base) addressing format: "size@(base_reg, index_reg, scale)"
>  + * - 'size' is the size in bytes of the array element, and its sign indicates
>  + *   whether the type is signed (negative) or unsigned (positive).
>  + * - 'base_reg' is the register holding the base address, normally rdx or edx
>  + * - 'index_reg' is the register holding the index, normally rax or eax
>  + * - 'scale' is the scaling factor (typically 1, 2, 4, or 8), which matches the
>  + *    size of the element type.
>  + *
>  + * For example, for an array of 'short' (signed 2-byte elements), the SIB spec would be:
>  + * - size: -2 (negative because 'short' is signed)
>  + * - scale: 2 (since sizeof(short) == 2)
>  + *
>  + * The resulting SIB format: "-2@(%%rdx,%%rax,2)" for x86_64, "-2@(%%edx,%%eax,2)" for i386
>  + */
>  +static volatile short array[] = {-1, -2, -3, -4};
>  +
>  +#if defined(__x86_64__)
>  +#define USDT_SIB_ARG_SPEC -2@(%%rdx,%%rax,2)
>  +#else
>  +#define USDT_SIB_ARG_SPEC -2@(%%edx,%%eax,2)
>  +#endif
>  +
>  +unsigned short test_usdt_sib_semaphore SEC(".probes");
>  +
>  +static void trigger_sib_spec(void)
>  +{
>  +	/*
>  +	 * Force SIB addressing with inline assembly.
>  +	 *
>  +	 * You must compile with -std=gnu99 or -std=c99 to use the
>  +	 * STAP_PROBE_ASM macro.
>  +	 *
>  +	 * The STAP_PROBE_ASM macro generates a quoted string that gets
>  +	 * inserted between the surrounding assembly instructions. In this
>  +	 * case, USDT_SIB_ARG_SPEC is embedded directly into the instruction
>  +	 * stream, creating a probe point between the asm statement boundaries.
>  +	 * It works fine with gcc/clang.
>  +	 *
>  +	 * Register constraints:
>  +	 * - "d"(array): Binds the 'array' variable to %rdx or %edx register
>  +	 * - "a"(0): Binds the constant 0 to %rax or %eax register
>  +	 * These ensure that when USDT_SIB_ARG_SPEC references %%rdx(%edx) and
>  +	 * %%rax(%eax), they contain the expected values for SIB addressing.
>  +	 *
>  +	 * The "memory" clobber prevents the compiler from reordering memory
>  +	 * accesses around the probe point, ensuring that the probe behavior
>  +	 * is predictable and consistent.
>  +	 */
>  +	asm volatile(
>  +		STAP_PROBE_ASM(test, usdt_sib, USDT_SIB_ARG_SPEC)
>  +		:
>  +		: "d"(array), "a"(0)
>  +		: "memory"
>  +	);
>  +}
>  +#endif
>  +
> - static void subtest_basic_usdt(void)
> + static void subtest_basic_usdt(bool optimized)
>   {
>   	LIBBPF_OPTS(bpf_usdt_opts, opts);
>   	struct test_usdt *skel;
>   	struct test_usdt__bss *bss;
> - 	int err, i;
> + 	int err, i, called;
>  +	const __u64 expected_cookie = 0xcafedeadbeeffeed;
>   
> + #define TRIGGER(x) ({			\
> + 	trigger_func(x);		\
> + 	if (optimized)			\
> + 		trigger_func(x);	\
> + 	optimized ? 2 : 1;		\
> + 	})
> + 
>   	skel = test_usdt__open_and_load();
>   	if (!ASSERT_OK_PTR(skel, "skel_open"))
>   		return;
> @@@ -126,22 -73,13 +133,22 @@@
>   	if (!ASSERT_OK_PTR(skel->links.usdt0, "usdt0_link"))
>   		goto cleanup;
>   
>  +#if defined(__x86_64__) || defined(__i386__)
>  +	opts.usdt_cookie = expected_cookie;
>  +	skel->links.usdt_sib = bpf_program__attach_usdt(skel->progs.usdt_sib,
>  +							 0 /*self*/, "/proc/self/exe",
>  +							 "test", "usdt_sib", &opts);
>  +	if (!ASSERT_OK_PTR(skel->links.usdt_sib, "usdt_sib_link"))
>  +		goto cleanup;
>  +#endif
>  +
> - 	trigger_func(1);
> + 	called = TRIGGER(1);
>   
> - 	ASSERT_EQ(bss->usdt0_called, 1, "usdt0_called");
> - 	ASSERT_EQ(bss->usdt3_called, 1, "usdt3_called");
> - 	ASSERT_EQ(bss->usdt12_called, 1, "usdt12_called");
> + 	ASSERT_EQ(bss->usdt0_called, called, "usdt0_called");
> + 	ASSERT_EQ(bss->usdt3_called, called, "usdt3_called");
> + 	ASSERT_EQ(bss->usdt12_called, called, "usdt12_called");
>   
>  -	ASSERT_EQ(bss->usdt0_cookie, 0xcafedeadbeeffeed, "usdt0_cookie");
>  +	ASSERT_EQ(bss->usdt0_cookie, expected_cookie, "usdt0_cookie");
>   	ASSERT_EQ(bss->usdt0_arg_cnt, 0, "usdt0_arg_cnt");
>   	ASSERT_EQ(bss->usdt0_arg_ret, -ENOENT, "usdt0_arg_ret");
>   	ASSERT_EQ(bss->usdt0_arg_size, -ENOENT, "usdt0_arg_size");
> @@@ -225,18 -163,9 +232,19 @@@
>   	ASSERT_EQ(bss->usdt3_args[1], 42, "usdt3_arg2");
>   	ASSERT_EQ(bss->usdt3_args[2], (uintptr_t)&bla, "usdt3_arg3");
>   
>  +#if defined(__x86_64__) || defined(__i386__)
>  +	trigger_sib_spec();
>  +	ASSERT_EQ(bss->usdt_sib_called, 1, "usdt_sib_called");
>  +	ASSERT_EQ(bss->usdt_sib_cookie, expected_cookie, "usdt_sib_cookie");
>  +	ASSERT_EQ(bss->usdt_sib_arg_cnt, 1, "usdt_sib_arg_cnt");
>  +	ASSERT_EQ(bss->usdt_sib_arg, nums[0], "usdt_sib_arg");
>  +	ASSERT_EQ(bss->usdt_sib_arg_ret, 0, "usdt_sib_arg_ret");
>  +	ASSERT_EQ(bss->usdt_sib_arg_size, sizeof(nums[0]), "usdt_sib_arg_size");
>  +#endif
>  +
>   cleanup:
>   	test_usdt__destroy(skel);
> + #undef TRIGGER
>   }
>   
>   unsigned short test_usdt_100_semaphore SEC(".probes");

linux-next: manual merge of the tip tree with the bpf-next tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 2043 bytes --]

Hi all,

Today's linux-next merge of the tip tree got a conflict in:

  kernel/bpf/stackmap.c

between commit:

  e17d62fedd10 ("bpf: Refactor stack map trace depth calculation into helper function")

from the bpf-next tree and commit:

  c69993ecdd4d ("perf: Support deferred user unwind")

from the tip tree.

I fixed it up (see below) and can carry the fix as necessary. This
is now fixed as far as linux-next is concerned, but any non trivial
conflicts should be mentioned to your upstream maintainer when your tree
is submitted for merging.  You may also want to consider cooperating
with the maintainer of the conflicting tree to minimise any particularly
complex conflicts.

-- 
Cheers,
Stephen Rothwell

diff --cc kernel/bpf/stackmap.c
index 2365541c81dd,8f1dacaf01fe..000000000000
--- a/kernel/bpf/stackmap.c
+++ b/kernel/bpf/stackmap.c
@@@ -333,9 -310,12 +333,9 @@@ BPF_CALL_3(bpf_get_stackid, struct pt_r
  			       BPF_F_FAST_STACK_CMP | BPF_F_REUSE_STACKID)))
  		return -EINVAL;
  
 -	max_depth += skip;
 -	if (max_depth > sysctl_perf_event_max_stack)
 -		max_depth = sysctl_perf_event_max_stack;
 -
 +	max_depth = stack_map_calculate_max_depth(map->value_size, elem_size, flags);
  	trace = get_perf_callchain(regs, kernel, user, max_depth,
- 				   false, false);
+ 				   false, false, 0);
  
  	if (unlikely(!trace))
  		/* couldn't fetch the stack trace */
@@@ -463,15 -446,13 +463,15 @@@ static long __bpf_get_stack(struct pt_r
  	if (may_fault)
  		rcu_read_lock(); /* need RCU for perf's callchain below */
  
 -	if (trace_in)
 +	if (trace_in) {
  		trace = trace_in;
 -	else if (kernel && task)
 +		trace->nr = min_t(u32, trace->nr, max_depth);
 +	} else if (kernel && task) {
  		trace = get_callchain_entry_for_task(task, max_depth);
 -	else
 +	} else {
  		trace = get_perf_callchain(regs, kernel, user, max_depth,
- 					   crosstask, false);
+ 					   crosstask, false, 0);
 +	}
  
  	if (unlikely(!trace) || trace->nr < skip) {
  		if (may_fault)

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

Re: linux-next: manual merge of the tip tree with the bpf-next tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 2359 bytes --]

Hi all,

On Wed, 5 Nov 2025 13:31:59 +1100 Stephen Rothwell <sfr@canb.auug.org.au> wrote:
>
> Today's linux-next merge of the tip tree got a conflict in:
> 
>   kernel/bpf/stackmap.c
> 
> between commit:
> 
>   e17d62fedd10 ("bpf: Refactor stack map trace depth calculation into helper function")
> 
> from the bpf-next tree and commit:
> 
>   c69993ecdd4d ("perf: Support deferred user unwind")
> 
> from the tip tree.
> 
> I fixed it up (see below) and can carry the fix as necessary. This
> is now fixed as far as linux-next is concerned, but any non trivial
> conflicts should be mentioned to your upstream maintainer when your tree
> is submitted for merging.  You may also want to consider cooperating
> with the maintainer of the conflicting tree to minimise any particularly
> complex conflicts.
> 
> -- 
> Cheers,
> Stephen Rothwell
> 
> diff --cc kernel/bpf/stackmap.c
> index 2365541c81dd,8f1dacaf01fe..000000000000
> --- a/kernel/bpf/stackmap.c
> +++ b/kernel/bpf/stackmap.c
> @@@ -333,9 -310,12 +333,9 @@@ BPF_CALL_3(bpf_get_stackid, struct pt_r
>   			       BPF_F_FAST_STACK_CMP | BPF_F_REUSE_STACKID)))
>   		return -EINVAL;
>   
>  -	max_depth += skip;
>  -	if (max_depth > sysctl_perf_event_max_stack)
>  -		max_depth = sysctl_perf_event_max_stack;
>  -
>  +	max_depth = stack_map_calculate_max_depth(map->value_size, elem_size, flags);
>   	trace = get_perf_callchain(regs, kernel, user, max_depth,
> - 				   false, false);
> + 				   false, false, 0);
>   
>   	if (unlikely(!trace))
>   		/* couldn't fetch the stack trace */
> @@@ -463,15 -446,13 +463,15 @@@ static long __bpf_get_stack(struct pt_r
>   	if (may_fault)
>   		rcu_read_lock(); /* need RCU for perf's callchain below */
>   
>  -	if (trace_in)
>  +	if (trace_in) {
>   		trace = trace_in;
>  -	else if (kernel && task)
>  +		trace->nr = min_t(u32, trace->nr, max_depth);
>  +	} else if (kernel && task) {
>   		trace = get_callchain_entry_for_task(task, max_depth);
>  -	else
>  +	} else {
>   		trace = get_perf_callchain(regs, kernel, user, max_depth,
> - 					   crosstask, false);
> + 					   crosstask, false, 0);
>  +	}
>   
>   	if (unlikely(!trace) || trace->nr < skip) {
>   		if (may_fault)

This is now a conflict between the bpf-next tree and Linus' tree.

-- 
Cheers,
Stephen Rothwell

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]