mm/alloc_tag: clear codetag for pages allocated before page_ext initialization

Due to initialization ordering, page_ext is allocated and initialized relatively late during boot. Some pages have already been allocated and freed before page_ext becomes available, leaving their codetag uninitialized.

A clear example is in init_section_page_ext(): alloc_page_ext() calls kmemleak_alloc(). If the slab cache has no free objects, it falls back to the buddy allocator to allocate memory. However, at this point page_ext is not yet fully initialized, so these newly allocated pages have no codetag set. These pages may later be reclaimed by KASAN, which causes the warning to trigger when they are freed because their codetag ref is still empty.

Use a global array to track pages allocated before page_ext is fully initialized. The array size is fixed at 8192 entries, and will emit a warning if this limit is exceeded. When page_ext initialization completes, set their codetag to empty to avoid warnings when they are freed later.

Fixes 93d5440ece3c ("alloc_tag: uninline code gated by mem_alloc_profiling_key in page allocator")

+#ifdef CONFIG_MEM_ALLOC_PROFILING_DEBUG

Quoted 13 lines · On Thu, 26 Mar 2026 22:05:54 +0800 Hao Ge <[email protected]> wrote:

Due to initialization ordering, page_ext is allocated and initialized relatively late during boot. Some pages have already been allocated and freed before page_ext becomes available, leaving their codetag uninitialized.

A clear example is in init_section_page_ext(): alloc_page_ext() calls kmemleak_alloc(). If the slab cache has no free objects, it falls back to the buddy allocator to allocate memory. However, at this point page_ext is not yet fully initialized, so these newly allocated pages have no codetag set. These pages may later be reclaimed by KASAN, which causes the warning to trigger when they are freed because their codetag ref is still empty.

Quoted 6 lines

Use a global array to track pages allocated before page_ext is fully initialized. The array size is fixed at 8192 entries, and will emit a warning if this limit is exceeded. When page_ext initialization completes, set their codetag to empty to avoid warnings when they are freed later.

Thanks. I'll queue this for review and test.

But where will I queue it?

Quoted 3 lines

Fixes 93d5440ece3c ("alloc_tag: uninline code gated by mem_alloc_profiling_key in page allocator")

A year ago, so a cc:stable might be needed.

Quoted 1 line

+#ifdef CONFIG_MEM_ALLOC_PROFILING_DEBUG

otoh, it appears that the bug only hits with CONFIG_MEM_ALLOC_PROFILING_DEBUG=y? If so, I'll add that (important) info to the changelog.

Do people use CONFIG_MEM_ALLOC_PROFILING_DEBUG much? Is a backport really needed?

Either way, it seems that this isn't a very urgent issue so I'm inclined to add it to the 7.1-rc1 pile, perhaps with a cc:stable.

Please all share your thoughts with me, thanks.

Quoted 3 lines

Do people use CONFIG_MEM_ALLOC_PROFILING_DEBUG much? Is a backport really needed?

IMO backport would be good.

Quoted 5 lines

Either way, it seems that this isn't a very urgent issue so I'm inclined to add it to the 7.1-rc1 pile, perhaps with a cc:stable.

Please all share your thoughts with me, thanks.

I'm reviewing and testing the patch and there is a race and a couple of smaller issues. I'll post a reply later today.

Quoted 5 lines · On Thu, 26 Mar 2026 18:19:56 -0700 Suren Baghdasaryan <[email protected]> wrote:

Quoted 3 lines

Do people use CONFIG_MEM_ALLOC_PROFILING_DEBUG much? Is a backport really needed?

IMO backport would be good.

OK, thanks, I'll slap a cc:stable on it and keep it in the 7.1.rc1 queue.

I added

"This bug occurs when CONFIG_MEM_ALLOC_PROFILING_DEBUG=y."

somewhere in the changelog but that's pretty lame and could be expanded upon in a respin.

Quoted 6 lines

Quoted 5 lines

Either way, it seems that this isn't a very urgent issue so I'm inclined to add it to the 7.1-rc1 pile, perhaps with a cc:stable.

Please all share your thoughts with me, thanks.

Quoted 2 lines

I'm reviewing and testing the patch and there is a race and a couple of smaller issues. I'll post a reply later today.

Cool. I was going to keep it in there for mm-new testing (which is light). But I guess there isn't much value in this so I'll drop v2.

Sashiko had a bunch of nags, but I think you're checking that routinely?
https://sashiko.dev/#/patchset/20260326140554.191996-1-hao.ge%40linux.dev

Quoted 23 lines · Andrew Morton

Quoted 13 lines · On Thu, 26 Mar 2026 22:05:54 +0800 Hao Ge <[email protected]> wrote:

Due to initialization ordering, page_ext is allocated and initialized relatively late during boot. Some pages have already been allocated and freed before page_ext becomes available, leaving their codetag uninitialized.

A clear example is in init_section_page_ext(): alloc_page_ext() calls kmemleak_alloc(). If the slab cache has no free objects, it falls back to the buddy allocator to allocate memory. However, at this point page_ext is not yet fully initialized, so these newly allocated pages have no codetag set. These pages may later be reclaimed by KASAN, which causes the warning to trigger when they are freed because their codetag ref is still empty.

Quoted 6 lines

Use a global array to track pages allocated before page_ext is fully initialized. The array size is fixed at 8192 entries, and will emit a warning if this limit is exceeded. When page_ext initialization completes, set their codetag to empty to avoid warnings when they are freed later.

Thanks. I'll queue this for review and test.

Quoted 1 line

But where will I queue it?

I don't think it's extra urgent. It is visible only when debugging with CONFIG_MEM_ALLOC_PROFILING_DEBUG.

Quoted 1 line

Quoted 1 line

Fixes 93d5440ece3c ("alloc_tag: uninline code gated by mem_alloc_profiling_key in page allocator")

Hmm. I'm not sure that's the right patch. Technically the problem exists once we introduced CONFIG_MEM_ALLOC_PROFILING_DEBUG. I'll double-check.

Quoted 6 lines

A year ago, so a cc:stable might be needed.

Quoted 1 line

+#ifdef CONFIG_MEM_ALLOC_PROFILING_DEBUG

otoh, it appears that the bug only hits with CONFIG_MEM_ALLOC_PROFILING_DEBUG=y? If so, I'll add that (important) info to the changelog.

Correct, it affects only CONFIG_MEM_ALLOC_PROFILING_DEBUG=y and only if !mem_profiling_compressed.

Quoted 2 lines

Do people use CONFIG_MEM_ALLOC_PROFILING_DEBUG much? Is a backport really needed?

IMO backport would be good.

Quoted 4 lines

Either way, it seems that this isn't a very urgent issue so I'm inclined to add it to the 7.1-rc1 pile, perhaps with a cc:stable.

Please all share your thoughts with me, thanks.

I'm reviewing and testing the patch and there is a race and a couple of smaller issues. I'll post a reply later today.

Due to initialization ordering, page_ext is allocated and initialized relatively late during boot. Some pages have already been allocated and freed before page_ext becomes available, leaving their codetag uninitialized.

A clear example is in init_section_page_ext(): alloc_page_ext() calls kmemleak_alloc(). If the slab cache has no free objects, it falls back to the buddy allocator to allocate memory. However, at this point page_ext is not yet fully initialized, so these newly allocated pages have no codetag set. These pages may later be reclaimed by KASAN, which causes the warning to trigger when they are freed because their codetag ref is still empty.

Use a global array to track pages allocated before page_ext is fully initialized. The array size is fixed at 8192 entries, and will emit a warning if this limit is exceeded. When page_ext initialization completes, set their codetag to empty to avoid warnings when they are freed later.

The following warning is observed when this issue occurs:

[    9.582133] ------------[ cut here ]------------
[    9.582137] alloc_tag was not set
[    9.582139] WARNING: ./include/linux/alloc_tag.h:164 at __pgalloc_tag_sub+0x40f/0x550, CPU#5: systemd/1
[    9.582190] CPU: 5 UID: 0 PID: 1 Comm: systemd Not tainted 7.0.0-rc4 #1 PREEMPT(lazy)
[    9.582192] Hardware name: Red Hat KVM, BIOS rel-1.16.3-0-ga6ed6b701f0a-prebuilt.qemu.org 04/01/2014
[    9.582194] RIP: 0010:__pgalloc_tag_sub+0x40f/0x550
[    9.582196] Code: 00 00 4c 29 e5 48 8b 05 1f 88 56 05 48 8d 4c ad 00 48 8d 2c c8 e9 87 fd ff ff 0f 0b 0f 0b e9 f3 fe ff ff 48 8d 3d 61 2f ed 03 <67> 48 0f b9 3a e9 b3 fd ff ff 0f 0b eb e4 e8 5e cd 14 02 4c 89 c7
[    9.582197] RSP: 0018:ffffc9000001f940 EFLAGS: 00010246
[    9.582200] RAX: dffffc0000000000 RBX: 1ffff92000003f2b RCX: 1ffff110200d806c
[    9.582201] RDX: ffff8881006c0360 RSI: 0000000000000004 RDI: ffffffff9bc7b460
[    9.582202] RBP: 0000000000000000 R08: 0000000000000000 R09: fffffbfff3a62324
[    9.582203] R10: ffffffff9d311923 R11: 0000000000000000 R12: ffffea0004001b00
[    9.582204] R13: 0000000000002000 R14: ffffea0000000000 R15: ffff8881006c0360
[    9.582206] FS:  00007ffbbcf2d940(0000) GS:ffff888450479000(0000) knlGS:0000000000000000
[    9.582208] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[    9.582210] CR2: 000055ee3aa260d0 CR3: 0000000148b67005 CR4: 0000000000770ef0
[    9.582211] PKRU: 55555554
[    9.582212] Call Trace:
[    9.582213]  <TASK>
[    9.582214]  ? __pfx___pgalloc_tag_sub+0x10/0x10
[    9.582216]  ? check_bytes_and_report+0x68/0x140
[    9.582219]  __free_frozen_pages+0x2e4/0x1150
[    9.582221]  ? __free_slab+0xc2/0x2b0
[    9.582224]  qlist_free_all+0x4c/0xf0
[    9.582227]  kasan_quarantine_reduce+0x15d/0x180
[    9.582229]  __kasan_slab_alloc+0x69/0x90
[    9.582232]  kmem_cache_alloc_noprof+0x14a/0x500
[    9.582234]  do_getname+0x96/0x310
[    9.582237]  do_readlinkat+0x91/0x2f0
[    9.582239]  ? __pfx_do_readlinkat+0x10/0x10
[    9.582240]  ? get_random_bytes_user+0x1df/0x2c0
[    9.582244]  __x64_sys_readlinkat+0x96/0x100
[    9.582246]  do_syscall_64+0xce/0x650
[    9.582250]  ? __x64_sys_getrandom+0x13a/0x1e0
[    9.582252]  ? __pfx___x64_sys_getrandom+0x10/0x10
[    9.582254]  ? do_syscall_64+0x114/0x650
[    9.582255]  ? ksys_read+0xfc/0x1d0
[    9.582258]  ? __pfx_ksys_read+0x10/0x10
[    9.582260]  ? do_syscall_64+0x114/0x650
[    9.582262]  ? do_syscall_64+0x114/0x650
[    9.582264]  ? __pfx_fput_close_sync+0x10/0x10
[    9.582266]  ? file_close_fd_locked+0x178/0x2a0
[    9.582268]  ? __x64_sys_faccessat2+0x96/0x100
[    9.582269]  ? __x64_sys_close+0x7d/0xd0
[    9.582271]  ? do_syscall_64+0x114/0x650
[    9.582273]  ? do_syscall_64+0x114/0x650
[    9.582275]  ? clear_bhb_loop+0x50/0xa0
[    9.582277]  ? clear_bhb_loop+0x50/0xa0
[    9.582279]  entry_SYSCALL_64_after_hwframe+0x76/0x7e
[    9.582280] RIP: 0033:0x7ffbbda345ee
[    9.582282] Code: 0f 1f 40 00 48 8b 15 29 38 0d 00 f7 d8 64 89 02 48 c7 c0 ff ff ff ff c3 0f 1f 40 00 f3 0f 1e fa 49 89 ca b8 0b 01 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d fa 37 0d 00 f7 d8 64 89 01 48
[    9.582284] RSP: 002b:00007ffe2ad8de58 EFLAGS: 00000202 ORIG_RAX: 000000000000010b
[    9.582286] RAX: ffffffffffffffda RBX: 000055ee3aa25570 RCX: 00007ffbbda345ee
[    9.582287] RDX: 000055ee3aa25570 RSI: 00007ffe2ad8dee0 RDI: 00000000ffffff9c
[    9.582288] RBP: 0000000000001000 R08: 0000000000000003 R09: 0000000000001001
[    9.582289] R10: 0000000000001000 R11: 0000000000000202 R12: 0000000000000033
[    9.582290] R13: 00007ffe2ad8dee0 R14: 00000000ffffff9c R15: 00007ffe2ad8deb0
[    9.582292]  </TASK>
[    9.582293] ---[ end trace 0000000000000000 ]---

Fixes 93d5440ece3c ("alloc_tag: uninline code gated by mem_alloc_profiling_key in page allocator")

Suggested-by Suren Baghdasaryan <[email protected]>

Signed-off-by Hao Ge <[email protected]>

---
v2:
- Replace spin_lock_irqsave() with atomic_try_cmpxchg() to avoid potential deadlock in NMI context
- Change EARLY_ALLOC_PFN_MAX from 256 to 8192
- Add pr_warn_once() when the limit is exceeded
- Check ref.ct before clearing to avoid overwriting valid tags
- Use function pointer (alloc_tag_add_early_pfn_ptr) instead of state

---
 include/linux/alloc_tag.h   |  2 +
 include/linux/pgalloc_tag.h |  2 +-
 lib/alloc_tag.c             | 92 +++++++++++++++++++++++++++++++++++++
 mm/page_alloc.c             |  7 +++
 4 files changed, 102 insertions(+), 1 deletion(-)

include/linux/alloc_tag.h +0 -0

index d40ac39bfbe8..bf226c2be2ad 100644

include/linux/alloc_tag.h +1 -0

+
 #define ALLOC_TAG_SECTION_NAME "alloc_tags"
 
 struct codetag_bytes {

include/linux/pgalloc_tag.h +1 -1

index 38a82d65e58e..951d33362268 100644
@@ -181,7 +181,7 @@ static inline struct alloc_tag *__pgalloc_tag_get(struct page *page)
 
        if (get_page_tag_ref(page, &ref, &handle)) {
                alloc_tag_sub_check(&ref);
-               if (ref.ct)
+               if (ref.ct && !is_codetag_empty(&ref))
                        tag = ct_to_alloc_tag(ref.ct);
                put_page_tag_ref(handle);
        }

lib/alloc_tag.c +31 -0

index 58991ab09d84..7b1812768af9 100644
@@ -6,6 +6,7 @@
 #include <linux/kallsyms.h>
 #include <linux/module.h>
 #include <linux/page_ext.h>
+#include <linux/pgalloc_tag.h>
 #include <linux/proc_fs.h>
 #include <linux/seq_buf.h>
 #include <linux/seq_file.h>

@@ -26,6 +27,96 @@ static bool mem_profiling_support;
 
 static struct codetag_type *alloc_tag_cttype;
 
+#ifdef CONFIG_MEM_ALLOC_PROFILING_DEBUG
+
+/*
+ * Track page allocations before page_ext is initialized.
+ * Some pages are allocated before page_ext becomes available, leaving
+ * their codetag uninitialized. Track these early PFNs so we can clear
+ * their codetag refs later to avoid warnings when they are freed.
+ *
+ * Early allocations include:
+ *   - Base allocations independent of CPU count
+ *   - Per-CPU allocations (e.g., CPU hotplug callbacks during smp_init,
+ *     such as trace ring buffers, scheduler per-cpu data)
+ *
+ * For simplicity, we fix the size to 8192.
+ * If insufficient, a warning will be triggered to alert the user.
+ */
+#define EARLY_ALLOC_PFN_MAX            8192
+
+static unsigned long early_pfns[EARLY_ALLOC_PFN_MAX] __initdata;
+static atomic_t early_pfn_count __initdata = ATOMIC_INIT(0);
+
+static void __init __alloc_tag_add_early_pfn(unsigned long pfn)
+{
+       int old_idx, new_idx;
+
+       do {
+               old_idx = atomic_read(&early_pfn_count);
+               if (old_idx >= EARLY_ALLOC_PFN_MAX) {
+                       pr_warn_once("Early page allocations before page_ext init exceeded EARLY_ALLOC_PFN_MAX (%d)\n",
+                                     EARLY_ALLOC_PFN_MAX);

lib/alloc_tag.c +4 -0

+
+void alloc_tag_add_early_pfn(unsigned long pfn)
+{
+       if (static_key_enabled(&mem_profiling_compressed))

lib/alloc_tag.c +5 -0

+       for (i = 0; i < atomic_read(&early_pfn_count); i++) {
+               unsigned long pfn = early_pfns[i];
+
+               if (pfn_valid(pfn)) {
+                       struct page *page = pfn_to_page(pfn);

mm/page_alloc.c +0 -1

        }
 }
 
--

2.25.1

Quoted 104 lines · Hao Ge

Due to initialization ordering, page_ext is allocated and initialized relatively late during boot. Some pages have already been allocated and freed before page_ext becomes available, leaving their codetag uninitialized.

A clear example is in init_section_page_ext(): alloc_page_ext() calls kmemleak_alloc(). If the slab cache has no free objects, it falls back to the buddy allocator to allocate memory. However, at this point page_ext is not yet fully initialized, so these newly allocated pages have no codetag set. These pages may later be reclaimed by KASAN, which causes the warning to trigger when they are freed because their codetag ref is still empty.

Use a global array to track pages allocated before page_ext is fully initialized. The array size is fixed at 8192 entries, and will emit a warning if this limit is exceeded. When page_ext initialization completes, set their codetag to empty to avoid warnings when they are freed later.

The following warning is observed when this issue occurs:

[    9.582133] ------------[ cut here ]------------
[    9.582137] alloc_tag was not set
[    9.582139] WARNING: ./include/linux/alloc_tag.h:164 at __pgalloc_tag_sub+0x40f/0x550, CPU#5: systemd/1
[    9.582190] CPU: 5 UID: 0 PID: 1 Comm: systemd Not tainted 7.0.0-rc4 #1 PREEMPT(lazy)
[    9.582192] Hardware name: Red Hat KVM, BIOS rel-1.16.3-0-ga6ed6b701f0a-prebuilt.qemu.org 04/01/2014
[    9.582194] RIP: 0010:__pgalloc_tag_sub+0x40f/0x550
[    9.582196] Code: 00 00 4c 29 e5 48 8b 05 1f 88 56 05 48 8d 4c ad 00 48 8d 2c c8 e9 87 fd ff ff 0f 0b 0f 0b e9 f3 fe ff ff 48 8d 3d 61 2f ed 03 <67> 48 0f b9 3a e9 b3 fd ff ff 0f 0b eb e4 e8 5e cd 14 02 4c 89 c7
[    9.582197] RSP: 0018:ffffc9000001f940 EFLAGS: 00010246
[    9.582200] RAX: dffffc0000000000 RBX: 1ffff92000003f2b RCX: 1ffff110200d806c
[    9.582201] RDX: ffff8881006c0360 RSI: 0000000000000004 RDI: ffffffff9bc7b460
[    9.582202] RBP: 0000000000000000 R08: 0000000000000000 R09: fffffbfff3a62324
[    9.582203] R10: ffffffff9d311923 R11: 0000000000000000 R12: ffffea0004001b00
[    9.582204] R13: 0000000000002000 R14: ffffea0000000000 R15: ffff8881006c0360
[    9.582206] FS:  00007ffbbcf2d940(0000) GS:ffff888450479000(0000) knlGS:0000000000000000
[    9.582208] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[    9.582210] CR2: 000055ee3aa260d0 CR3: 0000000148b67005 CR4: 0000000000770ef0
[    9.582211] PKRU: 55555554
[    9.582212] Call Trace:
[    9.582213]  <TASK>
[    9.582214]  ? __pfx___pgalloc_tag_sub+0x10/0x10
[    9.582216]  ? check_bytes_and_report+0x68/0x140
[    9.582219]  __free_frozen_pages+0x2e4/0x1150
[    9.582221]  ? __free_slab+0xc2/0x2b0
[    9.582224]  qlist_free_all+0x4c/0xf0
[    9.582227]  kasan_quarantine_reduce+0x15d/0x180
[    9.582229]  __kasan_slab_alloc+0x69/0x90
[    9.582232]  kmem_cache_alloc_noprof+0x14a/0x500
[    9.582234]  do_getname+0x96/0x310
[    9.582237]  do_readlinkat+0x91/0x2f0
[    9.582239]  ? __pfx_do_readlinkat+0x10/0x10
[    9.582240]  ? get_random_bytes_user+0x1df/0x2c0
[    9.582244]  __x64_sys_readlinkat+0x96/0x100
[    9.582246]  do_syscall_64+0xce/0x650
[    9.582250]  ? __x64_sys_getrandom+0x13a/0x1e0
[    9.582252]  ? __pfx___x64_sys_getrandom+0x10/0x10
[    9.582254]  ? do_syscall_64+0x114/0x650
[    9.582255]  ? ksys_read+0xfc/0x1d0
[    9.582258]  ? __pfx_ksys_read+0x10/0x10
[    9.582260]  ? do_syscall_64+0x114/0x650
[    9.582262]  ? do_syscall_64+0x114/0x650
[    9.582264]  ? __pfx_fput_close_sync+0x10/0x10
[    9.582266]  ? file_close_fd_locked+0x178/0x2a0
[    9.582268]  ? __x64_sys_faccessat2+0x96/0x100
[    9.582269]  ? __x64_sys_close+0x7d/0xd0
[    9.582271]  ? do_syscall_64+0x114/0x650
[    9.582273]  ? do_syscall_64+0x114/0x650
[    9.582275]  ? clear_bhb_loop+0x50/0xa0
[    9.582277]  ? clear_bhb_loop+0x50/0xa0
[    9.582279]  entry_SYSCALL_64_after_hwframe+0x76/0x7e
[    9.582280] RIP: 0033:0x7ffbbda345ee
[    9.582282] Code: 0f 1f 40 00 48 8b 15 29 38 0d 00 f7 d8 64 89 02 48 c7 c0 ff ff ff ff c3 0f 1f 40 00 f3 0f 1e fa 49 89 ca b8 0b 01 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d fa 37 0d 00 f7 d8 64 89 01 48
[    9.582284] RSP: 002b:00007ffe2ad8de58 EFLAGS: 00000202 ORIG_RAX: 000000000000010b
[    9.582286] RAX: ffffffffffffffda RBX: 000055ee3aa25570 RCX: 00007ffbbda345ee
[    9.582287] RDX: 000055ee3aa25570 RSI: 00007ffe2ad8dee0 RDI: 00000000ffffff9c
[    9.582288] RBP: 0000000000001000 R08: 0000000000000003 R09: 0000000000001001
[    9.582289] R10: 0000000000001000 R11: 0000000000000202 R12: 0000000000000033
[    9.582290] R13: 00007ffe2ad8dee0 R14: 00000000ffffff9c R15: 00007ffe2ad8deb0
[    9.582292]  </TASK>
[    9.582293] ---[ end trace 0000000000000000 ]---

Fixes 93d5440ece3c ("alloc_tag: uninline code gated by mem_alloc_profiling_key in page allocator")

Suggested-by Suren Baghdasaryan <[email protected]>

Signed-off-by Hao Ge <[email protected]>

---
v2:
- Replace spin_lock_irqsave() with atomic_try_cmpxchg() to avoid potential deadlock in NMI context
- Change EARLY_ALLOC_PFN_MAX from 256 to 8192
- Add pr_warn_once() when the limit is exceeded
- Check ref.ct before clearing to avoid overwriting valid tags
- Use function pointer (alloc_tag_add_early_pfn_ptr) instead of state

---
 include/linux/alloc_tag.h   |  2 +
 include/linux/pgalloc_tag.h |  2 +-
 lib/alloc_tag.c             | 92 +++++++++++++++++++++++++++++++++++++
 mm/page_alloc.c             |  7 +++
 4 files changed, 102 insertions(+), 1 deletion(-)

include/linux/alloc_tag.h +0 -0

index d40ac39bfbe8..bf226c2be2ad 100644

include/linux/alloc_tag.h static inline void set_codetag_empty(union codetag_ref *ref) · line 77

Quoted 4 lines

include/linux/alloc_tag.h +1 -0

@@ -74,6 +74,8 @@ static inline void set_codetag_empty(union codetag_ref *ref)
 
 #ifdef CONFIG_MEM_ALLOC_PROFILING
 
+void alloc_tag_add_early_pfn(unsigned long pfn);

Although this works, the usual approach is have it defined this way in the header file:

#if CONFIG_MEM_ALLOC_PROFILING_DEBUG
void alloc_tag_add_early_pfn(unsigned long pfn);
#else
static inline void alloc_tag_add_early_pfn(unsigned long pfn) {} #endif

Quoted 42 lines

include/linux/alloc_tag.h +1 -0

+
 #define ALLOC_TAG_SECTION_NAME "alloc_tags"
 
 struct codetag_bytes {

include/linux/pgalloc_tag.h +1 -1

index 38a82d65e58e..951d33362268 100644
@@ -181,7 +181,7 @@ static inline struct alloc_tag *__pgalloc_tag_get(struct page *page)
 
        if (get_page_tag_ref(page, &ref, &handle)) {
                alloc_tag_sub_check(&ref);
-               if (ref.ct)
+               if (ref.ct && !is_codetag_empty(&ref))
                        tag = ct_to_alloc_tag(ref.ct);
                put_page_tag_ref(handle);
        }

lib/alloc_tag.c +10 -0

index 58991ab09d84..7b1812768af9 100644
@@ -6,6 +6,7 @@
 #include <linux/kallsyms.h>
 #include <linux/module.h>
 #include <linux/page_ext.h>
+#include <linux/pgalloc_tag.h>
 #include <linux/proc_fs.h>
 #include <linux/seq_buf.h>
 #include <linux/seq_file.h>

@@ -26,6 +27,96 @@ static bool mem_profiling_support;
 
 static struct codetag_type *alloc_tag_cttype;
 
+#ifdef CONFIG_MEM_ALLOC_PROFILING_DEBUG
+
+/*
+ * Track page allocations before page_ext is initialized.
+ * Some pages are allocated before page_ext becomes available, leaving
+ * their codetag uninitialized. Track these early PFNs so we can clear
+ * their codetag refs later to avoid warnings when they are freed.
+ *
+ * Early allocations include:

lib/alloc_tag.c static bool mem_profiling_support; · line 46

Quoted 8 lines

lib/alloc_tag.c +8 -0

+ *   - Base allocations independent of CPU count
+ *   - Per-CPU allocations (e.g., CPU hotplug callbacks during smp_init,
+ *     such as trace ring buffers, scheduler per-cpu data)
+ *
+ * For simplicity, we fix the size to 8192.
+ * If insufficient, a warning will be triggered to alert the user.
+ */
+#define EARLY_ALLOC_PFN_MAX            8192

Quoted 13 lines

lib/alloc_tag.c +13 -0

+
+static unsigned long early_pfns[EARLY_ALLOC_PFN_MAX] __initdata;
+static atomic_t early_pfn_count __initdata = ATOMIC_INIT(0);
+
+static void __init __alloc_tag_add_early_pfn(unsigned long pfn)
+{
+       int old_idx, new_idx;
+
+       do {
+               old_idx = atomic_read(&early_pfn_count);
+               if (old_idx >= EARLY_ALLOC_PFN_MAX) {
+                       pr_warn_once("Early page allocations before page_ext init exceeded EARLY_ALLOC_PFN_MAX (%d)\n",
+                                     EARLY_ALLOC_PFN_MAX);

lib/alloc_tag.c static bool mem_profiling_support; · line 69

Quoted 10 lines

lib/alloc_tag.c +10 -0

+                       return;
+               }
+               new_idx = old_idx + 1;
+       } while (!atomic_try_cmpxchg(&early_pfn_count, &old_idx, new_idx));
+
+       early_pfns[old_idx] = pfn;
+}
+
+static void (*alloc_tag_add_early_pfn_ptr)(unsigned long pfn) __refdata =
+               __alloc_tag_add_early_pfn;

So, there is a possible race between clear_early_alloc_pfn_tag_refs() and __alloc_tag_add_early_pfn(). I think the easiest way to resolve this is using RCU. It's easier to show that with the code:

typedef void (*alloc_tag_add_func)(unsigned long pfn);

static alloc_tag_add_func __rcu alloc_tag_add_early_pfn_ptr __refdata =

                __alloc_tag_add_early_pfn;

void alloc_tag_add_early_pfn(unsigned long pfn)

{
        alloc_tag_add_func alloc_tag_add;

        if (static_key_enabled(&mem_profiling_compressed))
                return;

        rcu_read_lock();
        alloc_tag_add = rcu_dereference(alloc_tag_add_early_pfn_ptr);
        if (alloc_tag_add)
                alloc_tag_add(pfn);
        rcu_read_unlock();
}

static void __init clear_early_alloc_pfn_tag_refs(void)

{
        unsigned int i;

        if (static_key_enabled(&mem_profiling_compressed))
                return;

       rcu_assign_pointer(alloc_tag_add_early_pfn_ptr, NULL);
        /* Make sure we are not racing with __alloc_tag_add_early_pfn() */
        synchronize_rcu();
        ...
}

So, clear_early_alloc_pfn_tag_refs() resets alloc_tag_add_early_pfn_ptr to NULL before starting its loop and alloc_tag_add_early_pfn() calls __alloc_tag_add_early_pfn() in RCU read section. This way you know that after synchronize_rcu() nobody is or will be executing __alloc_tag_add_early_pfn() anymore. synchronize_rcu() can increase boot time but this happens only with CONFIG_MEM_ALLOC_PROFILING_DEBUG, so should be acceptable.

Quoted 4 lines

lib/alloc_tag.c +4 -0

+
+void alloc_tag_add_early_pfn(unsigned long pfn)
+{
+       if (static_key_enabled(&mem_profiling_compressed))

lib/alloc_tag.c static bool mem_profiling_support; · line 83

Quoted 10 lines

lib/alloc_tag.c +10 -0

+               return;
+
+       if (alloc_tag_add_early_pfn_ptr)
+               alloc_tag_add_early_pfn_ptr(pfn);
+}
+
+static void __init clear_early_alloc_pfn_tag_refs(void)
+{
+       unsigned int i;
+

I included this in the code I suggested above but just as a reminder, here we also need:

      if (static_key_enabled(&mem_profiling_compressed))
               return;

Quoted 5 lines

lib/alloc_tag.c +5 -0

+       for (i = 0; i < atomic_read(&early_pfn_count); i++) {
+               unsigned long pfn = early_pfns[i];
+
+               if (pfn_valid(pfn)) {
+                       struct page *page = pfn_to_page(pfn);

lib/alloc_tag.c static bool mem_profiling_support; · line 98

Quoted 10 lines

lib/alloc_tag.c +10 -0

+                       union pgtag_ref_handle handle;
+                       union codetag_ref ref;
+
+                       if (get_page_tag_ref(page, &ref, &handle)) {
+                               /*
+                                * An early-allocated page could be freed and reallocated
+                                * after its page_ext is initialized but before we clear it.
+                                * In that case, it already has a valid tag set.
+                                * We should not overwrite that valid tag with CODETAG_EMPTY.
+                                */

You don't really solve this race here. See explanation below.

lib/alloc_tag.c static bool mem_profiling_support; · line 103

Quoted 5 lines

lib/alloc_tag.c +5 -0

+                               if (ref.ct) {
+                                       put_page_tag_ref(handle);
+                                       continue;
+                               }
+

Between the above "if (ref.ct)" check and below set_codetag_empty() an allocation can change the ref.ct value to a valid reference (because page_ext already exists) and you will override it with CODETAG_EMPTY. I think we have two options:
1. Just let that override happen and lose accounting for that racing allocation. I think that's preferred option since the race is not likely and extra complexity is not worth it IMO.
2. Do clear_page_tag_ref() here but atomically. Something like clear_page_tag_ref_if_null() calling update_page_tag_ref_if_null() which calls cmpxchg(&ref->ct, NULL, CODETAG_EMPTY).

If you agree with option #1 then please update the comment above highlighting this smaller race and that we are ok with it.

lib/alloc_tag.c static bool mem_profiling_support; · line 113

Quoted 10 lines

lib/alloc_tag.c +10 -0

+                               set_codetag_empty(&ref);
+                               update_page_tag_ref(handle, &ref);
+                               put_page_tag_ref(handle);
+                       }
+               }
+
+       }
+
+       atomic_set(&early_pfn_count, 0);
+       alloc_tag_add_early_pfn_ptr = NULL;

Once we did that RCU synchronization we don't need the above resets. early_pfn_count won't be used anymore and alloc_tag_add_early_pfn_ptr is already NULL.

mm/page_alloc.c void __pgalloc_tag_add(struct page *page, struct task_struct *task,

Quoted 22 lines

lib/alloc_tag.c +7 -0

+}
+#else /* !CONFIG_MEM_ALLOC_PROFILING_DEBUG */
+inline void alloc_tag_add_early_pfn(unsigned long pfn) {}
+static inline void __init clear_early_alloc_pfn_tag_refs(void) {}
+#endif
+
 #ifdef CONFIG_ARCH_MODULE_NEEDS_WEAK_PER_CPU
 DEFINE_PER_CPU(struct alloc_tag_counters, _shared_alloc_tag);
 EXPORT_SYMBOL(_shared_alloc_tag);

@@ -760,6 +851,7 @@ static __init bool need_page_alloc_tagging(void)
 
 static __init void init_page_alloc_tagging(void)
 {
+       clear_early_alloc_pfn_tag_refs();
 }
 
 struct page_ext_operations page_alloc_tagging_ops = {

mm/page_alloc.c +0 -0

index 2d4b6f1a554e..8f9bda04403b 100644
@@ -1293,6 +1293,13 @@ void __pgalloc_tag_add(struct page *page, struct task_struct *task,

In here let's mark the normal branch as "likely":

(quoted diff) +1 -1

-        if (get_page_tag_ref(page, &ref, &handle)) {
+        if (likely(get_page_tag_ref(page, &ref, &handle))) {

mm/page_alloc.c void __pgalloc_tag_add(struct page *page, struct task_struct *task, · line 1302

Quoted 11 lines

mm/page_alloc.c +7 -0

                alloc_tag_add(&ref, task->alloc_tag, PAGE_SIZE * nr);
                update_page_tag_ref(handle, &ref);
                put_page_tag_ref(handle);
+       } else {
+               /*
+                * page_ext is not available yet, record the pfn so we can
+                * clear the tag ref later when page_ext is initialized.
+                */
+               alloc_tag_add_early_pfn(page_to_pfn(page));
+               alloc_tag_set_inaccurate(current->alloc_tag);

Here we should be using task->alloc_tag instead of current->alloc_tag but we also need to check that task->alloc_tag != NULL.

Quoted 6 lines

mm/page_alloc.c +0 -1

        }
 }
 
--

2.25.1

Quoted 159 lines · Suren Baghdasaryan

Quoted 103 lines · Hao Ge

Due to initialization ordering, page_ext is allocated and initialized relatively late during boot. Some pages have already been allocated and freed before page_ext becomes available, leaving their codetag uninitialized.

A clear example is in init_section_page_ext(): alloc_page_ext() calls kmemleak_alloc(). If the slab cache has no free objects, it falls back to the buddy allocator to allocate memory. However, at this point page_ext is not yet fully initialized, so these newly allocated pages have no codetag set. These pages may later be reclaimed by KASAN, which causes the warning to trigger when they are freed because their codetag ref is still empty.

Use a global array to track pages allocated before page_ext is fully initialized. The array size is fixed at 8192 entries, and will emit a warning if this limit is exceeded. When page_ext initialization completes, set their codetag to empty to avoid warnings when they are freed later.

The following warning is observed when this issue occurs:

[    9.582133] ------------[ cut here ]------------
[    9.582137] alloc_tag was not set
[    9.582139] WARNING: ./include/linux/alloc_tag.h:164 at __pgalloc_tag_sub+0x40f/0x550, CPU#5: systemd/1
[    9.582190] CPU: 5 UID: 0 PID: 1 Comm: systemd Not tainted 7.0.0-rc4 #1 PREEMPT(lazy)
[    9.582192] Hardware name: Red Hat KVM, BIOS rel-1.16.3-0-ga6ed6b701f0a-prebuilt.qemu.org 04/01/2014
[    9.582194] RIP: 0010:__pgalloc_tag_sub+0x40f/0x550
[    9.582196] Code: 00 00 4c 29 e5 48 8b 05 1f 88 56 05 48 8d 4c ad 00 48 8d 2c c8 e9 87 fd ff ff 0f 0b 0f 0b e9 f3 fe ff ff 48 8d 3d 61 2f ed 03 <67> 48 0f b9 3a e9 b3 fd ff ff 0f 0b eb e4 e8 5e cd 14 02 4c 89 c7
[    9.582197] RSP: 0018:ffffc9000001f940 EFLAGS: 00010246
[    9.582200] RAX: dffffc0000000000 RBX: 1ffff92000003f2b RCX: 1ffff110200d806c
[    9.582201] RDX: ffff8881006c0360 RSI: 0000000000000004 RDI: ffffffff9bc7b460
[    9.582202] RBP: 0000000000000000 R08: 0000000000000000 R09: fffffbfff3a62324
[    9.582203] R10: ffffffff9d311923 R11: 0000000000000000 R12: ffffea0004001b00
[    9.582204] R13: 0000000000002000 R14: ffffea0000000000 R15: ffff8881006c0360
[    9.582206] FS:  00007ffbbcf2d940(0000) GS:ffff888450479000(0000) knlGS:0000000000000000
[    9.582208] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[    9.582210] CR2: 000055ee3aa260d0 CR3: 0000000148b67005 CR4: 0000000000770ef0
[    9.582211] PKRU: 55555554
[    9.582212] Call Trace:
[    9.582213]  <TASK>
[    9.582214]  ? __pfx___pgalloc_tag_sub+0x10/0x10
[    9.582216]  ? check_bytes_and_report+0x68/0x140
[    9.582219]  __free_frozen_pages+0x2e4/0x1150
[    9.582221]  ? __free_slab+0xc2/0x2b0
[    9.582224]  qlist_free_all+0x4c/0xf0
[    9.582227]  kasan_quarantine_reduce+0x15d/0x180
[    9.582229]  __kasan_slab_alloc+0x69/0x90
[    9.582232]  kmem_cache_alloc_noprof+0x14a/0x500
[    9.582234]  do_getname+0x96/0x310
[    9.582237]  do_readlinkat+0x91/0x2f0
[    9.582239]  ? __pfx_do_readlinkat+0x10/0x10
[    9.582240]  ? get_random_bytes_user+0x1df/0x2c0
[    9.582244]  __x64_sys_readlinkat+0x96/0x100
[    9.582246]  do_syscall_64+0xce/0x650
[    9.582250]  ? __x64_sys_getrandom+0x13a/0x1e0
[    9.582252]  ? __pfx___x64_sys_getrandom+0x10/0x10
[    9.582254]  ? do_syscall_64+0x114/0x650
[    9.582255]  ? ksys_read+0xfc/0x1d0
[    9.582258]  ? __pfx_ksys_read+0x10/0x10
[    9.582260]  ? do_syscall_64+0x114/0x650
[    9.582262]  ? do_syscall_64+0x114/0x650
[    9.582264]  ? __pfx_fput_close_sync+0x10/0x10
[    9.582266]  ? file_close_fd_locked+0x178/0x2a0
[    9.582268]  ? __x64_sys_faccessat2+0x96/0x100
[    9.582269]  ? __x64_sys_close+0x7d/0xd0
[    9.582271]  ? do_syscall_64+0x114/0x650
[    9.582273]  ? do_syscall_64+0x114/0x650
[    9.582275]  ? clear_bhb_loop+0x50/0xa0
[    9.582277]  ? clear_bhb_loop+0x50/0xa0
[    9.582279]  entry_SYSCALL_64_after_hwframe+0x76/0x7e
[    9.582280] RIP: 0033:0x7ffbbda345ee
[    9.582282] Code: 0f 1f 40 00 48 8b 15 29 38 0d 00 f7 d8 64 89 02 48 c7 c0 ff ff ff ff c3 0f 1f 40 00 f3 0f 1e fa 49 89 ca b8 0b 01 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d fa 37 0d 00 f7 d8 64 89 01 48
[    9.582284] RSP: 002b:00007ffe2ad8de58 EFLAGS: 00000202 ORIG_RAX: 000000000000010b
[    9.582286] RAX: ffffffffffffffda RBX: 000055ee3aa25570 RCX: 00007ffbbda345ee
[    9.582287] RDX: 000055ee3aa25570 RSI: 00007ffe2ad8dee0 RDI: 00000000ffffff9c
[    9.582288] RBP: 0000000000001000 R08: 0000000000000003 R09: 0000000000001001
[    9.582289] R10: 0000000000001000 R11: 0000000000000202 R12: 0000000000000033
[    9.582290] R13: 00007ffe2ad8dee0 R14: 00000000ffffff9c R15: 00007ffe2ad8deb0
[    9.582292]  </TASK>
[    9.582293] ---[ end trace 0000000000000000 ]---

Fixes 93d5440ece3c ("alloc_tag: uninline code gated by mem_alloc_profiling_key in page allocator")

Suggested-by Suren Baghdasaryan <[email protected]>

Signed-off-by Hao Ge <[email protected]>

---
v2:
- Replace spin_lock_irqsave() with atomic_try_cmpxchg() to avoid potential deadlock in NMI context
- Change EARLY_ALLOC_PFN_MAX from 256 to 8192
- Add pr_warn_once() when the limit is exceeded
- Check ref.ct before clearing to avoid overwriting valid tags
- Use function pointer (alloc_tag_add_early_pfn_ptr) instead of state

---
  include/linux/alloc_tag.h   |  2 +
  include/linux/pgalloc_tag.h |  2 +-
  lib/alloc_tag.c             | 92 +++++++++++++++++++++++++++++++++++++
  mm/page_alloc.c             |  7 +++
  4 files changed, 102 insertions(+), 1 deletion(-)

include/linux/alloc_tag.h +0 -0

index d40ac39bfbe8..bf226c2be2ad 100644

include/linux/alloc_tag.h static inline void set_codetag_empty(union codetag_ref *ref) · line 77

Quoted 4 lines

include/linux/alloc_tag.h +1 -0

@@ -74,6 +74,8 @@ static inline void set_codetag_empty(union codetag_ref *ref)
 
  #ifdef CONFIG_MEM_ALLOC_PROFILING
 
+void alloc_tag_add_early_pfn(unsigned long pfn);

Although this works, the usual approach is have it defined this way in the header file:

#if CONFIG_MEM_ALLOC_PROFILING_DEBUG
void alloc_tag_add_early_pfn(unsigned long pfn);
#else
static inline void alloc_tag_add_early_pfn(unsigned long pfn) {} #endif

Quoted 42 lines

include/linux/alloc_tag.h +1 -0

+
  #define ALLOC_TAG_SECTION_NAME "alloc_tags"
 
  struct codetag_bytes {

include/linux/pgalloc_tag.h +1 -1

index 38a82d65e58e..951d33362268 100644
@@ -181,7 +181,7 @@ static inline struct alloc_tag *__pgalloc_tag_get(struct page *page)
 
         if (get_page_tag_ref(page, &ref, &handle)) {
                 alloc_tag_sub_check(&ref);
-               if (ref.ct)
+               if (ref.ct && !is_codetag_empty(&ref))
                         tag = ct_to_alloc_tag(ref.ct);
                 put_page_tag_ref(handle);
         }

lib/alloc_tag.c +10 -0

index 58991ab09d84..7b1812768af9 100644
@@ -6,6 +6,7 @@
  #include <linux/kallsyms.h>
  #include <linux/module.h>
  #include <linux/page_ext.h>
+#include <linux/pgalloc_tag.h>
  #include <linux/proc_fs.h>
  #include <linux/seq_buf.h>
  #include <linux/seq_file.h>

@@ -26,6 +27,96 @@ static bool mem_profiling_support;
 
  static struct codetag_type *alloc_tag_cttype;
 
+#ifdef CONFIG_MEM_ALLOC_PROFILING_DEBUG
+
+/*
+ * Track page allocations before page_ext is initialized.
+ * Some pages are allocated before page_ext becomes available, leaving
+ * their codetag uninitialized. Track these early PFNs so we can clear
+ * their codetag refs later to avoid warnings when they are freed.
+ *
+ * Early allocations include:

Quoted 8 lines

lib/alloc_tag.c static bool mem_profiling_support; · line 46

Quoted 8 lines

lib/alloc_tag.c +8 -0

+ *   - Base allocations independent of CPU count
+ *   - Per-CPU allocations (e.g., CPU hotplug callbacks during smp_init,
+ *     such as trace ring buffers, scheduler per-cpu data)
+ *
+ * For simplicity, we fix the size to 8192.
+ * If insufficient, a warning will be triggered to alert the user.
+ */
+#define EARLY_ALLOC_PFN_MAX            8192

Forgot to mention that we will need to do something about this limit using dynamic allocation. I was thinking we could allocate pages dynamically (with a GFP flag similar to ___GFP_NO_OBJ_EXT to avoid recursion), linking them via page->lru and then freeing them at the end of clear_early_alloc_pfn_tag_refs(). That adds more complexity but solves this limit problem. However all this can be done as a followup patch.

Quoted 182 lines

Quoted 13 lines

lib/alloc_tag.c +13 -0

+
+static unsigned long early_pfns[EARLY_ALLOC_PFN_MAX] __initdata;
+static atomic_t early_pfn_count __initdata = ATOMIC_INIT(0);
+
+static void __init __alloc_tag_add_early_pfn(unsigned long pfn)
+{
+       int old_idx, new_idx;
+
+       do {
+               old_idx = atomic_read(&early_pfn_count);
+               if (old_idx >= EARLY_ALLOC_PFN_MAX) {
+                       pr_warn_once("Early page allocations before page_ext init exceeded EARLY_ALLOC_PFN_MAX (%d)\n",
+                                     EARLY_ALLOC_PFN_MAX);

lib/alloc_tag.c static bool mem_profiling_support; · line 69

Quoted 10 lines

lib/alloc_tag.c +10 -0

+                       return;
+               }
+               new_idx = old_idx + 1;
+       } while (!atomic_try_cmpxchg(&early_pfn_count, &old_idx, new_idx));
+
+       early_pfns[old_idx] = pfn;
+}
+
+static void (*alloc_tag_add_early_pfn_ptr)(unsigned long pfn) __refdata =
+               __alloc_tag_add_early_pfn;

So, there is a possible race between clear_early_alloc_pfn_tag_refs() and __alloc_tag_add_early_pfn(). I think the easiest way to resolve this is using RCU. It's easier to show that with the code:

typedef void (*alloc_tag_add_func)(unsigned long pfn);

static alloc_tag_add_func __rcu alloc_tag_add_early_pfn_ptr __refdata =

                 __alloc_tag_add_early_pfn;

void alloc_tag_add_early_pfn(unsigned long pfn)

{
         alloc_tag_add_func alloc_tag_add;

         if (static_key_enabled(&mem_profiling_compressed))
                 return;

         rcu_read_lock();
         alloc_tag_add = rcu_dereference(alloc_tag_add_early_pfn_ptr);
         if (alloc_tag_add)
                 alloc_tag_add(pfn);
         rcu_read_unlock();
}

static void __init clear_early_alloc_pfn_tag_refs(void)

{
         unsigned int i;

         if (static_key_enabled(&mem_profiling_compressed))
                 return;

        rcu_assign_pointer(alloc_tag_add_early_pfn_ptr, NULL);
         /* Make sure we are not racing with __alloc_tag_add_early_pfn() */
         synchronize_rcu();
         ...
}

So, clear_early_alloc_pfn_tag_refs() resets alloc_tag_add_early_pfn_ptr to NULL before starting its loop and alloc_tag_add_early_pfn() calls __alloc_tag_add_early_pfn() in RCU read section. This way you know that after synchronize_rcu() nobody is or will be executing __alloc_tag_add_early_pfn() anymore. synchronize_rcu() can increase boot time but this happens only with CONFIG_MEM_ALLOC_PROFILING_DEBUG, so should be acceptable.

Quoted 4 lines

lib/alloc_tag.c +4 -0

+
+void alloc_tag_add_early_pfn(unsigned long pfn)
+{
+       if (static_key_enabled(&mem_profiling_compressed))

lib/alloc_tag.c static bool mem_profiling_support; · line 83

Quoted 10 lines

lib/alloc_tag.c +10 -0

+               return;
+
+       if (alloc_tag_add_early_pfn_ptr)
+               alloc_tag_add_early_pfn_ptr(pfn);
+}
+
+static void __init clear_early_alloc_pfn_tag_refs(void)
+{
+       unsigned int i;
+

I included this in the code I suggested above but just as a reminder, here we also need:

       if (static_key_enabled(&mem_profiling_compressed))
                return;

Quoted 5 lines

lib/alloc_tag.c +5 -0

+       for (i = 0; i < atomic_read(&early_pfn_count); i++) {
+               unsigned long pfn = early_pfns[i];
+
+               if (pfn_valid(pfn)) {
+                       struct page *page = pfn_to_page(pfn);

lib/alloc_tag.c static bool mem_profiling_support; · line 98

Quoted 10 lines

lib/alloc_tag.c +10 -0

+                       union pgtag_ref_handle handle;
+                       union codetag_ref ref;
+
+                       if (get_page_tag_ref(page, &ref, &handle)) {
+                               /*
+                                * An early-allocated page could be freed and reallocated
+                                * after its page_ext is initialized but before we clear it.
+                                * In that case, it already has a valid tag set.
+                                * We should not overwrite that valid tag with CODETAG_EMPTY.
+                                */

You don't really solve this race here. See explanation below.

lib/alloc_tag.c static bool mem_profiling_support; · line 103

Quoted 5 lines

lib/alloc_tag.c +5 -0

+                               if (ref.ct) {
+                                       put_page_tag_ref(handle);
+                                       continue;
+                               }
+

Between the above "if (ref.ct)" check and below set_codetag_empty() an allocation can change the ref.ct value to a valid reference (because page_ext already exists) and you will override it with CODETAG_EMPTY. I think we have two options:
1. Just let that override happen and lose accounting for that racing allocation. I think that's preferred option since the race is not likely and extra complexity is not worth it IMO.
2. Do clear_page_tag_ref() here but atomically. Something like clear_page_tag_ref_if_null() calling update_page_tag_ref_if_null() which calls cmpxchg(&ref->ct, NULL, CODETAG_EMPTY).

If you agree with option #1 then please update the comment above highlighting this smaller race and that we are ok with it.

lib/alloc_tag.c static bool mem_profiling_support; · line 113

Quoted 10 lines

lib/alloc_tag.c +10 -0

+                               set_codetag_empty(&ref);
+                               update_page_tag_ref(handle, &ref);
+                               put_page_tag_ref(handle);
+                       }
+               }
+
+       }
+
+       atomic_set(&early_pfn_count, 0);
+       alloc_tag_add_early_pfn_ptr = NULL;

Once we did that RCU synchronization we don't need the above resets. early_pfn_count won't be used anymore and alloc_tag_add_early_pfn_ptr is already NULL.

mm/page_alloc.c void __pgalloc_tag_add(struct page *page, struct task_struct *task,

Quoted 22 lines

lib/alloc_tag.c +7 -0

+}
+#else /* !CONFIG_MEM_ALLOC_PROFILING_DEBUG */
+inline void alloc_tag_add_early_pfn(unsigned long pfn) {}
+static inline void __init clear_early_alloc_pfn_tag_refs(void) {}
+#endif
+
  #ifdef CONFIG_ARCH_MODULE_NEEDS_WEAK_PER_CPU
  DEFINE_PER_CPU(struct alloc_tag_counters, _shared_alloc_tag);
  EXPORT_SYMBOL(_shared_alloc_tag);

@@ -760,6 +851,7 @@ static __init bool need_page_alloc_tagging(void)
 
  static __init void init_page_alloc_tagging(void)
  {
+       clear_early_alloc_pfn_tag_refs();
  }
 
  struct page_ext_operations page_alloc_tagging_ops = {

mm/page_alloc.c +0 -0

index 2d4b6f1a554e..8f9bda04403b 100644
@@ -1293,6 +1293,13 @@ void __pgalloc_tag_add(struct page *page, struct task_struct *task,

In here let's mark the normal branch as "likely":

(quoted diff) +1 -1

-        if (get_page_tag_ref(page, &ref, &handle)) {
+        if (likely(get_page_tag_ref(page, &ref, &handle))) {

mm/page_alloc.c void __pgalloc_tag_add(struct page *page, struct task_struct *task, · line 1302

Quoted 11 lines

mm/page_alloc.c +7 -0

                 alloc_tag_add(&ref, task->alloc_tag, PAGE_SIZE * nr);
                 update_page_tag_ref(handle, &ref);
                 put_page_tag_ref(handle);
+       } else {
+               /*
+                * page_ext is not available yet, record the pfn so we can
+                * clear the tag ref later when page_ext is initialized.
+                */
+               alloc_tag_add_early_pfn(page_to_pfn(page));
+               alloc_tag_set_inaccurate(current->alloc_tag);

Here we should be using task->alloc_tag instead of current->alloc_tag but we also need to check that task->alloc_tag != NULL.

Quoted 6 lines

mm/page_alloc.c +0 -1

         }
  }
 
--

2.25.1