Currently we use bottom-up allocation after sparse_init(), the reason is
sparse_init() need a lot of memory, and bottom-up allocation may exhaust
precious low memory (below 4GB). On the other hand, SWIOTLB and CMA need
low memories for DMA32, so swiotlb_init() and dma_contiguous_reserve()
need bottom-up allocation.

Since swiotlb_init() and dma_contiguous_reserve() are both called in
arch_mem_init(), we no longer need bottom-up allocation after that. So
we set the allocation policy to top-down at the end of arch_mem_init(),
in order to avoid later memory allocations (such as KASAN) exhaust low
memory.

This solve at least two problems:
1. Some buggy BIOSes use 0xfd000000~0xfe000000 for secondary CPUs, but
   didn't reserve this range, which causes smpboot failures.
2. Some DMA32 devices, such as Loongson-DRM and OHCI, cannot work with
   KASAN enabled.

Cc: stable@vger.kernel.org
Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
---
 arch/loongarch/kernel/setup.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/arch/loongarch/kernel/setup.c b/arch/loongarch/kernel/setup.c
index 20cb6f306456..2b260d15b2e2 100644
--- a/arch/loongarch/kernel/setup.c
+++ b/arch/loongarch/kernel/setup.c
@@ -421,6 +421,7 @@ static void __init arch_mem_init(char **cmdline_p)
 				   PFN_UP(__pa_symbol(&__nosave_end)));
 
 	memblock_dump_all();
+	memblock_set_bottom_up(false);
 
 	early_memtest(PFN_PHYS(ARCH_PFN_OFFSET), PFN_PHYS(max_low_pfn));
 }
-- 
2.47.3

From: Tiezhu Yang <yangtiezhu@loongson.cn>

According to Documentation/dev-tools/kasan.rst, software KASAN modes use
compiler instrumentation to insert validity checks. Such instrumentation
might be incompatible with some parts of the kernel, and therefore needs
to be disabled, just use the attribute __no_sanitize_address to disable
instrumentation for the low level function setup_ptwalker().

Otherwise bringing up the secondary CPUs failed when CONFIG_KASAN is set
(especially when PTW is enabled), here are the call chains:

    smpboot_entry()
      start_secondary()
        cpu_probe()
          per_cpu_trap_init()
            tlb_init()
              setup_tlb_handler()
                setup_ptwalker()

The reason is the PGD registers are configured in setup_ptwalker(), but
KASAN instrumentation may cause TLB exceptions before that.

Cc: stable@vger.kernel.org
Signed-off-by: Tiezhu Yang <yangtiezhu@loongson.cn>
Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
---
 arch/loongarch/mm/tlb.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/loongarch/mm/tlb.c b/arch/loongarch/mm/tlb.c
index 4014c4469587..aaf7d685cc2a 100644
--- a/arch/loongarch/mm/tlb.c
+++ b/arch/loongarch/mm/tlb.c
@@ -202,7 +202,7 @@ void __update_tlb(struct vm_area_struct *vma, unsigned long address, pte_t *ptep
 	local_irq_restore(flags);
 }
 
-static void setup_ptwalker(void)
+static void __no_sanitize_address setup_ptwalker(void)
 {
 	unsigned long pwctl0, pwctl1;
 	unsigned long pgd_i = 0, pgd_w = 0;
-- 
2.47.3

From: Tiezhu Yang <yangtiezhu@loongson.cn>

kasan_init_generic() indicates that kasan is fully initialized, so it
should be put at end of kasan_init().

Otherwise bringing up the primary CPU failed when CONFIG_KASAN is set
on PTW-enabled systems, here are the call chains:

    kernel_entry()
      start_kernel()
        setup_arch()
          kasan_init()
            kasan_init_generic()

The reason is PTW-enabled systems have speculative accesses which means
memory accesses to the shadow memory after kasan_init() may be executed
by hardware before. However, accessing shadow memory is safe only after
kasan fully initialized because kasan_init() uses a temporary PGD table
until we have populated all levels of shadow page tables and writen the
PGD register. Moving kasan_init_generic() later can defer the occasion
of kasan_enabled(), so as to avoid speculative accesses on shadow pages.

After moving kasan_init_generic() to the end, kasan_init() can no longer
call kasan_mem_to_shadow() for shadow address conversion because it will
always return kasan_early_shadow_page. On the other hand, we should keep
the current logic of kasan_mem_to_shadow() for both the early and final
stage because there may be instrumentation before kasan_init().

To solve this, we factor out a new mem_to_shadow() function from current
kasan_mem_to_shadow() for the shadow address conversion in kasan_init().

Cc: stable@vger.kernel.org
Signed-off-by: Tiezhu Yang <yangtiezhu@loongson.cn>
Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
---
 arch/loongarch/mm/kasan_init.c | 78 +++++++++++++++++-----------------
 1 file changed, 40 insertions(+), 38 deletions(-)

diff --git a/arch/loongarch/mm/kasan_init.c b/arch/loongarch/mm/kasan_init.c
index 170da98ad4f5..0fc02ca06457 100644
--- a/arch/loongarch/mm/kasan_init.c
+++ b/arch/loongarch/mm/kasan_init.c
@@ -40,39 +40,43 @@ static pgd_t kasan_pg_dir[PTRS_PER_PGD] __initdata __aligned(PAGE_SIZE);
 #define __pte_none(early, pte) (early ? pte_none(pte) : \
 ((pte_val(pte) & _PFN_MASK) == (unsigned long)__pa(kasan_early_shadow_page)))
 
-void *kasan_mem_to_shadow(const void *addr)
+static void *mem_to_shadow(const void *addr)
 {
-	if (!kasan_enabled()) {
+	unsigned long offset = 0;
+	unsigned long maddr = (unsigned long)addr;
+	unsigned long xrange = (maddr >> XRANGE_SHIFT) & 0xffff;
+
+	if (maddr >= FIXADDR_START)
 		return (void *)(kasan_early_shadow_page);
-	} else {
-		unsigned long maddr = (unsigned long)addr;
-		unsigned long xrange = (maddr >> XRANGE_SHIFT) & 0xffff;
-		unsigned long offset = 0;
-
-		if (maddr >= FIXADDR_START)
-			return (void *)(kasan_early_shadow_page);
-
-		maddr &= XRANGE_SHADOW_MASK;
-		switch (xrange) {
-		case XKPRANGE_CC_SEG:
-			offset = XKPRANGE_CC_SHADOW_OFFSET;
-			break;
-		case XKPRANGE_UC_SEG:
-			offset = XKPRANGE_UC_SHADOW_OFFSET;
-			break;
-		case XKPRANGE_WC_SEG:
-			offset = XKPRANGE_WC_SHADOW_OFFSET;
-			break;
-		case XKVRANGE_VC_SEG:
-			offset = XKVRANGE_VC_SHADOW_OFFSET;
-			break;
-		default:
-			WARN_ON(1);
-			return NULL;
-		}
 
-		return (void *)((maddr >> KASAN_SHADOW_SCALE_SHIFT) + offset);
+	maddr &= XRANGE_SHADOW_MASK;
+	switch (xrange) {
+	case XKPRANGE_CC_SEG:
+		offset = XKPRANGE_CC_SHADOW_OFFSET;
+		break;
+	case XKPRANGE_UC_SEG:
+		offset = XKPRANGE_UC_SHADOW_OFFSET;
+		break;
+	case XKPRANGE_WC_SEG:
+		offset = XKPRANGE_WC_SHADOW_OFFSET;
+		break;
+	case XKVRANGE_VC_SEG:
+		offset = XKVRANGE_VC_SHADOW_OFFSET;
+		break;
+	default:
+		WARN_ON(1);
+		return NULL;
 	}
+
+	return (void *)((maddr >> KASAN_SHADOW_SCALE_SHIFT) + offset);
+}
+
+void *kasan_mem_to_shadow(const void *addr)
+{
+	if (kasan_enabled())
+		return mem_to_shadow(addr);
+	else
+		return (void *)(kasan_early_shadow_page);
 }
 
 const void *kasan_shadow_to_mem(const void *shadow_addr)
@@ -293,11 +297,8 @@ void __init kasan_init(void)
 	/* Maps everything to a single page of zeroes */
 	kasan_pgd_populate(KASAN_SHADOW_START, KASAN_SHADOW_END, NUMA_NO_NODE, true);
 
-	kasan_populate_early_shadow(kasan_mem_to_shadow((void *)VMALLOC_START),
-					kasan_mem_to_shadow((void *)KFENCE_AREA_END));
-
-	/* Enable KASAN here before kasan_mem_to_shadow(). */
-	kasan_init_generic();
+	kasan_populate_early_shadow(mem_to_shadow((void *)VMALLOC_START),
+					mem_to_shadow((void *)KFENCE_AREA_END));
 
 	/* Populate the linear mapping */
 	for_each_mem_range(i, &pa_start, &pa_end) {
@@ -307,13 +308,13 @@ void __init kasan_init(void)
 		if (start >= end)
 			break;
 
-		kasan_map_populate((unsigned long)kasan_mem_to_shadow(start),
-			(unsigned long)kasan_mem_to_shadow(end), NUMA_NO_NODE);
+		kasan_map_populate((unsigned long)mem_to_shadow(start),
+			(unsigned long)mem_to_shadow(end), NUMA_NO_NODE);
 	}
 
 	/* Populate modules mapping */
-	kasan_map_populate((unsigned long)kasan_mem_to_shadow((void *)MODULES_VADDR),
-		(unsigned long)kasan_mem_to_shadow((void *)MODULES_END), NUMA_NO_NODE);
+	kasan_map_populate((unsigned long)mem_to_shadow((void *)MODULES_VADDR),
+		(unsigned long)mem_to_shadow((void *)MODULES_END), NUMA_NO_NODE);
 	/*
 	 * KAsan may reuse the contents of kasan_early_shadow_pte directly, so we
 	 * should make sure that it maps the zero page read-only.
@@ -328,4 +329,5 @@ void __init kasan_init(void)
 
 	/* At this point kasan is fully initialized. Enable error messages */
 	init_task.kasan_depth = 0;
+	kasan_init_generic();
 }
-- 
2.47.3