[PATCH v2 5/5] x86/mm: Drop usage of __flush_tlb_all() in kernel_physical_mapping_init()

[Dan Williams]

Commit f77084d96355 "x86/mm/pat: Disable preemption around
__flush_tlb_all()" addressed a case where __flush_tlb_all() is called
without preemption being disabled. It also left a warning to catch other
cases where preemption is not disabled. That warning triggers for the
memory hotplug path which is also used for persistent memory enabling:

 WARNING: CPU: 35 PID: 911 at ./arch/x86/include/asm/tlbflush.h:460
 RIP: 0010:__flush_tlb_all+0x1b/0x3a
 [..]
 Call Trace:
  phys_pud_init+0x29c/0x2bb
  kernel_physical_mapping_init+0xfc/0x219
  init_memory_mapping+0x1a5/0x3b0
  arch_add_memory+0x2c/0x50
  devm_memremap_pages+0x3aa/0x610
  pmem_attach_disk+0x585/0x700 [nd_pmem]

Andy wondered why a path that can sleep was using __flush_tlb_all() [1]
and Dave confirmed the expectation for TLB flush is for modifying /
invalidating existing pte entries, but not initial population [2]. Drop
the usage of __flush_tlb_all() in phys_{p4d,pud,pmd}_init() on the
expectation that this path is only ever populating empty entries for the
linear map. Note, at linear map teardown time there is a call to the
all-cpu flush_tlb_all() to invalidate the removed mappings.

[1]: https://lore.kernel.org/lkml/9DFD717D-857D-493D-A606-B635D72BAC21@xxxxxxxxxxxxxx
[2]: https://lore.kernel.org/lkml/749919a4-cdb1-48a3-adb4-adb81a5fa0b5@xxxxxxxxx

Fixes: f77084d96355 ("x86/mm/pat: Disable preemption around __flush_tlb_all()")
Cc: Kirill A. Shutemov <kirill.shutemov@xxxxxxxxxxxxxxx>
Cc: Sebastian Andrzej Siewior <bigeasy@xxxxxxxxxxxxx>
Cc: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
Cc: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
Cc: Borislav Petkov <bp@xxxxxxxxx>
Cc: <stable@xxxxxxxxxxxxxxx>
Reported-by: Andy Lutomirski <luto@xxxxxxxxxx>
Suggested-by: Dave Hansen <dave.hansen@xxxxxxxxxxxxxxx>
Signed-off-by: Dan Williams <dan.j.williams@xxxxxxxxx>
---
 arch/x86/mm/init_64.c |    6 ------
 1 file changed, 6 deletions(-)

diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c
index 3e25ac2793ef..484c1b92f078 100644
--- a/arch/x86/mm/init_64.c
+++ b/arch/x86/mm/init_64.c
@@ -584,7 +584,6 @@ phys_pud_init(pud_t *pud_page, unsigned long paddr, unsigned long paddr_end,
 							   paddr_end,
 							   page_size_mask,
 							   prot);
-				__flush_tlb_all();
 				continue;
 			}
 			/*
@@ -627,7 +626,6 @@ phys_pud_init(pud_t *pud_page, unsigned long paddr, unsigned long paddr_end,
 		pud_populate_safe(&init_mm, pud, pmd);
 		spin_unlock(&init_mm.page_table_lock);
 	}
-	__flush_tlb_all();
 
 	update_page_count(PG_LEVEL_1G, pages);
 
@@ -668,7 +666,6 @@ phys_p4d_init(p4d_t *p4d_page, unsigned long paddr, unsigned long paddr_end,
 			paddr_last = phys_pud_init(pud, paddr,
 					paddr_end,
 					page_size_mask);
-			__flush_tlb_all();
 			continue;
 		}
 
@@ -680,7 +677,6 @@ phys_p4d_init(p4d_t *p4d_page, unsigned long paddr, unsigned long paddr_end,
 		p4d_populate_safe(&init_mm, p4d, pud);
 		spin_unlock(&init_mm.page_table_lock);
 	}
-	__flush_tlb_all();
 
 	return paddr_last;
 }
@@ -733,8 +729,6 @@ kernel_physical_mapping_init(unsigned long paddr_start,
 	if (pgd_changed)
 		sync_global_pgds(vaddr_start, vaddr_end - 1);
 
-	__flush_tlb_all();
-
 	return paddr_last;
 }