[PATCH AUTOSEL 5.4 1/6] ntfs: fix use-after-free in ntfs_ucsncmp()

[PATCH AUTOSEL 5.4 1/6] ntfs: fix use-after-free in ntfs_ucsncmp()

[Sasha Levin]

From: ChenXiaoSong <chenxiaosong2@huawei.com>

[ Upstream commit 38c9c22a85aeed28d0831f230136e9cf6fa2ed44 ]

Syzkaller reported use-after-free bug as follows:

==================================================================
BUG: KASAN: use-after-free in ntfs_ucsncmp+0x123/0x130
Read of size 2 at addr ffff8880751acee8 by task a.out/879

CPU: 7 PID: 879 Comm: a.out Not tainted 5.19.0-rc4-next-20220630-00001-gcc5218c8bd2c-dirty #7
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.16.0-0-gd239552ce722-prebuilt.qemu.org 04/01/2014
Call Trace:
 <TASK>
 dump_stack_lvl+0x1c0/0x2b0
 print_address_description.constprop.0.cold+0xd4/0x484
 print_report.cold+0x55/0x232
 kasan_report+0xbf/0xf0
 ntfs_ucsncmp+0x123/0x130
 ntfs_are_names_equal.cold+0x2b/0x41
 ntfs_attr_find+0x43b/0xb90
 ntfs_attr_lookup+0x16d/0x1e0
 ntfs_read_locked_attr_inode+0x4aa/0x2360
 ntfs_attr_iget+0x1af/0x220
 ntfs_read_locked_inode+0x246c/0x5120
 ntfs_iget+0x132/0x180
 load_system_files+0x1cc6/0x3480
 ntfs_fill_super+0xa66/0x1cf0
 mount_bdev+0x38d/0x460
 legacy_get_tree+0x10d/0x220
 vfs_get_tree+0x93/0x300
 do_new_mount+0x2da/0x6d0
 path_mount+0x496/0x19d0
 __x64_sys_mount+0x284/0x300
 do_syscall_64+0x3b/0xc0
 entry_SYSCALL_64_after_hwframe+0x46/0xb0
RIP: 0033:0x7f3f2118d9ea
Code: 48 8b 0d a9 f4 0b 00 f7 d8 64 89 01 48 83 c8 ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 49 89 ca b8 a5 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 76 f4 0b 00 f7 d8 64 89 01 48
RSP: 002b:00007ffc269deac8 EFLAGS: 00000202 ORIG_RAX: 00000000000000a5
RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007f3f2118d9ea
RDX: 0000000020000000 RSI: 0000000020000100 RDI: 00007ffc269dec00
RBP: 00007ffc269dec80 R08: 00007ffc269deb00 R09: 00007ffc269dec44
R10: 0000000000000000 R11: 0000000000000202 R12: 000055f81ab1d220
R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000000
 </TASK>

The buggy address belongs to the physical page:
page:0000000085430378 refcount:1 mapcount:1 mapping:0000000000000000 index:0x555c6a81d pfn:0x751ac
memcg:ffff888101f7e180
anon flags: 0xfffffc00a0014(uptodate|lru|mappedtodisk|swapbacked|node=0|zone=1|lastcpupid=0x1fffff)
raw: 000fffffc00a0014 ffffea0001bf2988 ffffea0001de2448 ffff88801712e201
raw: 0000000555c6a81d 0000000000000000 0000000100000000 ffff888101f7e180
page dumped because: kasan: bad access detected

Memory state around the buggy address:
 ffff8880751acd80: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
 ffff8880751ace00: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
>ffff8880751ace80: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
                                                          ^
 ffff8880751acf00: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
 ffff8880751acf80: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
==================================================================

The reason is that struct ATTR_RECORD->name_offset is 6485, end address of
name string is out of bounds.

Fix this by adding sanity check on end address of attribute name string.

[akpm@linux-foundation.org: coding-style cleanups]
[chenxiaosong2@huawei.com: cleanup suggested by Hawkins Jiawei]
  Link: https://lkml.kernel.org/r/20220709064511.3304299-1-chenxiaosong2@huawei.com
Link: https://lkml.kernel.org/r/20220707105329.4020708-1-chenxiaosong2@huawei.com
Signed-off-by: ChenXiaoSong <chenxiaosong2@huawei.com>
Signed-off-by: Hawkins Jiawei <yin31149@gmail.com>
Cc: Anton Altaparmakov <anton@tuxera.com>
Cc: ChenXiaoSong <chenxiaosong2@huawei.com>
Cc: Yongqiang Liu <liuyongqiang13@huawei.com>
Cc: Zhang Yi <yi.zhang@huawei.com>
Cc: Zhang Xiaoxu <zhangxiaoxu5@huawei.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 fs/ntfs/attrib.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/fs/ntfs/attrib.c b/fs/ntfs/attrib.c
index d563abc3e136..914e99173130 100644
--- a/fs/ntfs/attrib.c
+++ b/fs/ntfs/attrib.c
@@ -592,8 +592,12 @@ static int ntfs_attr_find(const ATTR_TYPE type, const ntfschar *name,
 		a = (ATTR_RECORD*)((u8*)ctx->attr +
 				le32_to_cpu(ctx->attr->length));
 	for (;;	a = (ATTR_RECORD*)((u8*)a + le32_to_cpu(a->length))) {
-		if ((u8*)a < (u8*)ctx->mrec || (u8*)a > (u8*)ctx->mrec +
-				le32_to_cpu(ctx->mrec->bytes_allocated))
+		u8 *mrec_end = (u8 *)ctx->mrec +
+		               le32_to_cpu(ctx->mrec->bytes_allocated);
+		u8 *name_end = (u8 *)a + le16_to_cpu(a->name_offset) +
+			       a->name_length * sizeof(ntfschar);
+		if ((u8*)a < (u8*)ctx->mrec || (u8*)a > mrec_end ||
+		    name_end > mrec_end)
 			break;
 		ctx->attr = a;
 		if (unlikely(le32_to_cpu(a->type) > le32_to_cpu(type) ||
-- 
2.35.1

[PATCH AUTOSEL 5.4 2/6] spi: spi-rspi: Fix PIO fallback on RZ platforms

[Sasha Levin]

From: Biju Das <biju.das.jz@bp.renesas.com>

[ Upstream commit b620aa3a7be346f04ae7789b165937615c6ee8d3 ]

RSPI IP on RZ/{A, G2L} SoC's has the same signal for both interrupt
and DMA transfer request. Setting DMARS register for DMA transfer
makes the signal to work as a DMA transfer request signal and
subsequent interrupt requests to the interrupt controller
are masked.

PIO fallback does not work as interrupt signal is disabled.

This patch fixes this issue by re-enabling the interrupts by
calling dmaengine_synchronize().

Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be>
Link: https://lore.kernel.org/r/20220721143449.879257-1-biju.das.jz@bp.renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/spi/spi-rspi.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/spi/spi-rspi.c b/drivers/spi/spi-rspi.c
index 0524741d73b9..8ae2ac40b4b2 100644
--- a/drivers/spi/spi-rspi.c
+++ b/drivers/spi/spi-rspi.c
@@ -595,6 +595,10 @@ static int rspi_dma_transfer(struct rspi_data *rspi, struct sg_table *tx,
 					       rspi->dma_callbacked, HZ);
 	if (ret > 0 && rspi->dma_callbacked) {
 		ret = 0;
+		if (tx)
+			dmaengine_synchronize(rspi->ctlr->dma_tx);
+		if (rx)
+			dmaengine_synchronize(rspi->ctlr->dma_rx);
 	} else {
 		if (!ret) {
 			dev_err(&rspi->ctlr->dev, "DMA timeout\n");
-- 
2.35.1

[PATCH AUTOSEL 5.4 3/6] mmu_gather: Let there be one tlb_{start,end}_vma() implementation

[Sasha Levin]

From: Peter Zijlstra <peterz@infradead.org>

[ Upstream commit 18ba064e42df3661e196ab58a23931fc732a420b ]

Now that architectures are no longer allowed to override
tlb_{start,end}_vma() re-arrange code so that there is only one
implementation for each of these functions.

This much simplifies trying to figure out what they actually do.

Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Will Deacon <will@kernel.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 include/asm-generic/tlb.h | 15 ++-------------
 1 file changed, 2 insertions(+), 13 deletions(-)

diff --git a/include/asm-generic/tlb.h b/include/asm-generic/tlb.h
index 268674c1d568..fe05a8562c52 100644
--- a/include/asm-generic/tlb.h
+++ b/include/asm-generic/tlb.h
@@ -321,8 +321,8 @@ static inline void __tlb_reset_range(struct mmu_gather *tlb)
 
 #ifdef CONFIG_MMU_GATHER_NO_RANGE
 
-#if defined(tlb_flush) || defined(tlb_start_vma) || defined(tlb_end_vma)
-#error MMU_GATHER_NO_RANGE relies on default tlb_flush(), tlb_start_vma() and tlb_end_vma()
+#if defined(tlb_flush)
+#error MMU_GATHER_NO_RANGE relies on default tlb_flush()
 #endif
 
 /*
@@ -342,17 +342,10 @@ static inline void tlb_flush(struct mmu_gather *tlb)
 static inline void
 tlb_update_vma_flags(struct mmu_gather *tlb, struct vm_area_struct *vma) { }
 
-#define tlb_end_vma tlb_end_vma
-static inline void tlb_end_vma(struct mmu_gather *tlb, struct vm_area_struct *vma) { }
-
 #else /* CONFIG_MMU_GATHER_NO_RANGE */
 
 #ifndef tlb_flush
 
-#if defined(tlb_start_vma) || defined(tlb_end_vma)
-#error Default tlb_flush() relies on default tlb_start_vma() and tlb_end_vma()
-#endif
-
 /*
  * When an architecture does not provide its own tlb_flush() implementation
  * but does have a reasonably efficient flush_vma_range() implementation
@@ -468,7 +461,6 @@ static inline unsigned long tlb_get_unmap_size(struct mmu_gather *tlb)
  * case where we're doing a full MM flush.  When we're doing a munmap,
  * the vmas are adjusted to only cover the region to be torn down.
  */
-#ifndef tlb_start_vma
 static inline void tlb_start_vma(struct mmu_gather *tlb, struct vm_area_struct *vma)
 {
 	if (tlb->fullmm)
@@ -477,9 +469,7 @@ static inline void tlb_start_vma(struct mmu_gather *tlb, struct vm_area_struct *
 	tlb_update_vma_flags(tlb, vma);
 	flush_cache_range(vma, vma->vm_start, vma->vm_end);
 }
-#endif
 
-#ifndef tlb_end_vma
 static inline void tlb_end_vma(struct mmu_gather *tlb, struct vm_area_struct *vma)
 {
 	if (tlb->fullmm)
@@ -493,7 +483,6 @@ static inline void tlb_end_vma(struct mmu_gather *tlb, struct vm_area_struct *vm
 	 */
 	tlb_flush_mmu_tlbonly(tlb);
 }
-#endif
 
 /*
  * tlb_flush_{pte|pmd|pud|p4d}_range() adjust the tlb->start and tlb->end,
-- 
2.35.1

[PATCH AUTOSEL 5.4 4/6] mmu_gather: Force tlb-flush VM_PFNMAP vmas

[Sasha Levin]

From: Peter Zijlstra <peterz@infradead.org>

[ Upstream commit b67fbebd4cf980aecbcc750e1462128bffe8ae15 ]

Jann reported a race between munmap() and unmap_mapping_range(), where
unmap_mapping_range() will no-op once unmap_vmas() has unlinked the
VMA; however munmap() will not yet have invalidated the TLBs.

Therefore unmap_mapping_range() will complete while there are still
(stale) TLB entries for the specified range.

Mitigate this by force flushing TLBs for VM_PFNMAP ranges.

Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Will Deacon <will@kernel.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 include/asm-generic/tlb.h | 31 ++++++++++++++++---------------
 1 file changed, 16 insertions(+), 15 deletions(-)

diff --git a/include/asm-generic/tlb.h b/include/asm-generic/tlb.h
index fe05a8562c52..c44f7ac97f19 100644
--- a/include/asm-generic/tlb.h
+++ b/include/asm-generic/tlb.h
@@ -271,6 +271,7 @@ struct mmu_gather {
 	 */
 	unsigned int		vma_exec : 1;
 	unsigned int		vma_huge : 1;
+	unsigned int		vma_pfn  : 1;
 
 	unsigned int		batch_count;
 
@@ -345,7 +346,6 @@ tlb_update_vma_flags(struct mmu_gather *tlb, struct vm_area_struct *vma) { }
 #else /* CONFIG_MMU_GATHER_NO_RANGE */
 
 #ifndef tlb_flush
-
 /*
  * When an architecture does not provide its own tlb_flush() implementation
  * but does have a reasonably efficient flush_vma_range() implementation
@@ -365,6 +365,9 @@ static inline void tlb_flush(struct mmu_gather *tlb)
 		flush_tlb_range(&vma, tlb->start, tlb->end);
 	}
 }
+#endif
+
+#endif /* CONFIG_MMU_GATHER_NO_RANGE */
 
 static inline void
 tlb_update_vma_flags(struct mmu_gather *tlb, struct vm_area_struct *vma)
@@ -382,17 +385,9 @@ tlb_update_vma_flags(struct mmu_gather *tlb, struct vm_area_struct *vma)
 	 */
 	tlb->vma_huge = !!(vma->vm_flags & VM_HUGETLB);
 	tlb->vma_exec = !!(vma->vm_flags & VM_EXEC);
+	tlb->vma_pfn  = !!(vma->vm_flags & (VM_PFNMAP|VM_MIXEDMAP));
 }
 
-#else
-
-static inline void
-tlb_update_vma_flags(struct mmu_gather *tlb, struct vm_area_struct *vma) { }
-
-#endif
-
-#endif /* CONFIG_MMU_GATHER_NO_RANGE */
-
 static inline void tlb_flush_mmu_tlbonly(struct mmu_gather *tlb)
 {
 	if (!tlb->end)
@@ -476,12 +471,18 @@ static inline void tlb_end_vma(struct mmu_gather *tlb, struct vm_area_struct *vm
 		return;
 
 	/*
-	 * Do a TLB flush and reset the range at VMA boundaries; this avoids
-	 * the ranges growing with the unused space between consecutive VMAs,
-	 * but also the mmu_gather::vma_* flags from tlb_start_vma() rely on
-	 * this.
+	 * VM_PFNMAP is more fragile because the core mm will not track the
+	 * page mapcount -- there might not be page-frames for these PFNs after
+	 * all. Force flush TLBs for such ranges to avoid munmap() vs
+	 * unmap_mapping_range() races.
 	 */
-	tlb_flush_mmu_tlbonly(tlb);
+	if (tlb->vma_pfn || !IS_ENABLED(CONFIG_MMU_GATHER_MERGE_VMAS)) {
+		/*
+		 * Do a TLB flush and reset the range at VMA boundaries; this avoids
+		 * the ranges growing with the unused space between consecutive VMAs.
+		 */
+		tlb_flush_mmu_tlbonly(tlb);
+	}
 }
 
 /*
-- 
2.35.1

[PATCH AUTOSEL 5.4 5/6] netfilter: nf_tables: add rescheduling points during loop detection walks

[Sasha Levin]

From: Florian Westphal <fw@strlen.de>

[ Upstream commit 81ea010667417ef3f218dfd99b69769fe66c2b67 ]

Add explicit rescheduling points during ruleset walk.

Switching to a faster algorithm is possible but this is a much
smaller change, suitable for nf tree.

Link: https://bugzilla.netfilter.org/show_bug.cgi?id=1460
Signed-off-by: Florian Westphal <fw@strlen.de>
Acked-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 net/netfilter/nf_tables_api.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c
index 58a7d89719b1..d60067c5221e 100644
--- a/net/netfilter/nf_tables_api.c
+++ b/net/netfilter/nf_tables_api.c
@@ -2684,6 +2684,8 @@ int nft_chain_validate(const struct nft_ctx *ctx, const struct nft_chain *chain)
 			if (err < 0)
 				return err;
 		}
+
+		cond_resched();
 	}
 
 	return 0;
@@ -7310,9 +7312,13 @@ static int nf_tables_check_loops(const struct nft_ctx *ctx,
 				break;
 			}
 		}
+
+		cond_resched();
 	}
 
 	list_for_each_entry(set, &ctx->table->sets, list) {
+		cond_resched();
+
 		if (!nft_is_active_next(ctx->net, set))
 			continue;
 		if (!(set->flags & NFT_SET_MAP) ||
-- 
2.35.1

[PATCH AUTOSEL 5.4 6/6] ARM: findbit: fix overflowing offset

[Sasha Levin]

From: "Russell King (Oracle)" <rmk+kernel@armlinux.org.uk>

[ Upstream commit ec85bd369fd2bfaed6f45dd678706429d4f75b48 ]

When offset is larger than the size of the bit array, we should not
attempt to access the array as we can perform an access beyond the
end of the array. Fix this by changing the pre-condition.

Using "cmp r2, r1; bhs ..." covers us for the size == 0 case, since
this will always take the branch when r1 is zero, irrespective of
the value of r2. This means we can fix this bug without adding any
additional code!

Tested-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 arch/arm/lib/findbit.S | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

diff --git a/arch/arm/lib/findbit.S b/arch/arm/lib/findbit.S
index b5e8b9ae4c7d..7fd3600db8ef 100644
--- a/arch/arm/lib/findbit.S
+++ b/arch/arm/lib/findbit.S
@@ -40,8 +40,8 @@ ENDPROC(_find_first_zero_bit_le)
  * Prototype: int find_next_zero_bit(void *addr, unsigned int maxbit, int offset)
  */
 ENTRY(_find_next_zero_bit_le)
-		teq	r1, #0
-		beq	3b
+		cmp	r2, r1
+		bhs	3b
 		ands	ip, r2, #7
 		beq	1b			@ If new byte, goto old routine
  ARM(		ldrb	r3, [r0, r2, lsr #3]	)
@@ -81,8 +81,8 @@ ENDPROC(_find_first_bit_le)
  * Prototype: int find_next_zero_bit(void *addr, unsigned int maxbit, int offset)
  */
 ENTRY(_find_next_bit_le)
-		teq	r1, #0
-		beq	3b
+		cmp	r2, r1
+		bhs	3b
 		ands	ip, r2, #7
 		beq	1b			@ If new byte, goto old routine
  ARM(		ldrb	r3, [r0, r2, lsr #3]	)
@@ -115,8 +115,8 @@ ENTRY(_find_first_zero_bit_be)
 ENDPROC(_find_first_zero_bit_be)
 
 ENTRY(_find_next_zero_bit_be)
-		teq	r1, #0
-		beq	3b
+		cmp	r2, r1
+		bhs	3b
 		ands	ip, r2, #7
 		beq	1b			@ If new byte, goto old routine
 		eor	r3, r2, #0x18		@ big endian byte ordering
@@ -149,8 +149,8 @@ ENTRY(_find_first_bit_be)
 ENDPROC(_find_first_bit_be)
 
 ENTRY(_find_next_bit_be)
-		teq	r1, #0
-		beq	3b
+		cmp	r2, r1
+		bhs	3b
 		ands	ip, r2, #7
 		beq	1b			@ If new byte, goto old routine
 		eor	r3, r2, #0x18		@ big endian byte ordering
-- 
2.35.1


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

[PATCH AUTOSEL 5.4 6/6] ARM: findbit: fix overflowing offset

[Sasha Levin]

From: "Russell King (Oracle)" <rmk+kernel@armlinux.org.uk>

[ Upstream commit ec85bd369fd2bfaed6f45dd678706429d4f75b48 ]

When offset is larger than the size of the bit array, we should not
attempt to access the array as we can perform an access beyond the
end of the array. Fix this by changing the pre-condition.

Using "cmp r2, r1; bhs ..." covers us for the size == 0 case, since
this will always take the branch when r1 is zero, irrespective of
the value of r2. This means we can fix this bug without adding any
additional code!

Tested-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 arch/arm/lib/findbit.S | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

diff --git a/arch/arm/lib/findbit.S b/arch/arm/lib/findbit.S
index b5e8b9ae4c7d..7fd3600db8ef 100644
--- a/arch/arm/lib/findbit.S
+++ b/arch/arm/lib/findbit.S
@@ -40,8 +40,8 @@ ENDPROC(_find_first_zero_bit_le)
  * Prototype: int find_next_zero_bit(void *addr, unsigned int maxbit, int offset)
  */
 ENTRY(_find_next_zero_bit_le)
-		teq	r1, #0
-		beq	3b
+		cmp	r2, r1
+		bhs	3b
 		ands	ip, r2, #7
 		beq	1b			@ If new byte, goto old routine
  ARM(		ldrb	r3, [r0, r2, lsr #3]	)
@@ -81,8 +81,8 @@ ENDPROC(_find_first_bit_le)
  * Prototype: int find_next_zero_bit(void *addr, unsigned int maxbit, int offset)
  */
 ENTRY(_find_next_bit_le)
-		teq	r1, #0
-		beq	3b
+		cmp	r2, r1
+		bhs	3b
 		ands	ip, r2, #7
 		beq	1b			@ If new byte, goto old routine
  ARM(		ldrb	r3, [r0, r2, lsr #3]	)
@@ -115,8 +115,8 @@ ENTRY(_find_first_zero_bit_be)
 ENDPROC(_find_first_zero_bit_be)
 
 ENTRY(_find_next_zero_bit_be)
-		teq	r1, #0
-		beq	3b
+		cmp	r2, r1
+		bhs	3b
 		ands	ip, r2, #7
 		beq	1b			@ If new byte, goto old routine
 		eor	r3, r2, #0x18		@ big endian byte ordering
@@ -149,8 +149,8 @@ ENTRY(_find_first_bit_be)
 ENDPROC(_find_first_bit_be)
 
 ENTRY(_find_next_bit_be)
-		teq	r1, #0
-		beq	3b
+		cmp	r2, r1
+		bhs	3b
 		ands	ip, r2, #7
 		beq	1b			@ If new byte, goto old routine
 		eor	r3, r2, #0x18		@ big endian byte ordering
-- 
2.35.1