Re: [PATCH V5 4/7] arm64: mm: Offset TTBR1 to allow 52-bit PTRS_PER_PGD
From: Suzuki K Poulose <suzuki.poulose@arm.com>
Date: 2018-12-07 12:02:51
Also in:
linux-mm
On 12/06/2018 10:50 PM, Steve Capper wrote:
quoted hunk ↗ jump to hunk
Enabling 52-bit VAs on arm64 requires that the PGD table expands from 64 entries (for the 48-bit case) to 1024 entries. This quantity, PTRS_PER_PGD is used as follows to compute which PGD entry corresponds to a given virtual address, addr: pgd_index(addr) -> (addr >> PGDIR_SHIFT) & (PTRS_PER_PGD - 1) Userspace addresses are prefixed by 0's, so for a 48-bit userspace address, uva, the following is true: (uva >> PGDIR_SHIFT) & (1024 - 1) == (uva >> PGDIR_SHIFT) & (64 - 1) In other words, a 48-bit userspace address will have the same pgd_index when using PTRS_PER_PGD = 64 and 1024. Kernel addresses are prefixed by 1's so, given a 48-bit kernel address, kva, we have the following inequality: (kva >> PGDIR_SHIFT) & (1024 - 1) != (kva >> PGDIR_SHIFT) & (64 - 1) In other words a 48-bit kernel virtual address will have a different pgd_index when using PTRS_PER_PGD = 64 and 1024. If, however, we note that: kva = 0xFFFF << 48 + lower (where lower[63:48] == 0b) and, PGDIR_SHIFT = 42 (as we are dealing with 64KB PAGE_SIZE) We can consider: (kva >> PGDIR_SHIFT) & (1024 - 1) - (kva >> PGDIR_SHIFT) & (64 - 1) = (0xFFFF << 6) & 0x3FF - (0xFFFF << 6) & 0x3F // "lower" cancels out = 0x3C0 In other words, one can switch PTRS_PER_PGD to the 52-bit value globally provided that they increment ttbr1_el1 by 0x3C0 * 8 = 0x1E00 bytes when running with 48-bit kernel VAs (TCR_EL1.T1SZ = 16). For kernel configuration where 52-bit userspace VAs are possible, this patch offsets ttbr1_el1 and sets PTRS_PER_PGD corresponding to the 52-bit value. Suggested-by: Catalin Marinas <catalin.marinas@arm.com> Signed-off-by: Steve Capper <redacted> --- Changed in V5, removed ttbr1 save/restore logic for software PAN as hardware PAN is a mandatory ARMv8.1 feature anyway. The logic to enable 52-bit VAs has also been changed to depend on ARM64_PAN || !ARM64_SW_TTBR0_PAN (in a later patch) This patch is new in V4 of the series --- arch/arm64/include/asm/assembler.h | 23 +++++++++++++++++++++++ arch/arm64/include/asm/pgtable-hwdef.h | 9 +++++++++ arch/arm64/kernel/head.S | 1 + arch/arm64/kernel/hibernate-asm.S | 1 + arch/arm64/mm/proc.S | 4 ++++ 5 files changed, 38 insertions(+)diff --git a/arch/arm64/include/asm/assembler.h b/arch/arm64/include/asm/assembler.h index 6142402c2eb4..e2fe378d2a63 100644 --- a/arch/arm64/include/asm/assembler.h +++ b/arch/arm64/include/asm/assembler.h@@ -515,6 +515,29 @@ USER(\label, ic ivau, \tmp2) // invalidate I line PoU mrs \rd, sp_el0 .endm +/* + * Offset ttbr1 to allow for 48-bit kernel VAs set with 52-bit PTRS_PER_PGD. + * orr is used as it can cover the immediate value (and is idempotent). + * In future this may be nop'ed out when dealing with 52-bit kernel VAs. + * ttbr: Value of ttbr to set, modified. + */ + .macro offset_ttbr1, ttbr +#ifdef CONFIG_ARM64_52BIT_VA + orr \ttbr, \ttbr, #TTBR1_BADDR_4852_OFFSET +#endif + .endm + +/* + * Perform the reverse of offset_ttbr1. + * bic is used as it can cover the immediate value and, in future, won't need + * to be nop'ed out when dealing with 52-bit kernel VAs. + */ + .macro restore_ttbr1, ttbr +#ifdef CONFIG_ARM64_52BIT_VA + bic \ttbr, \ttbr, #TTBR1_BADDR_4852_OFFSET +#endif + .endm +
The above operation is safe as long as the TTBR1_BADDR_4852_OFFSET is aligned to 2^6 or more. Otherwise we could corrupt the Bits[51:48] of the BADDR stored in TTBR1[5:2] and thus the TTBR1:BADDR must be aligned to 64bytes minimum as per v8.2LVA restrictions. Since we have restricted the VA_BITS to 48, we should be safe here. Do we need a BUILD_BUG_ON() or something to check if this is still valid? Eitherway, Reviewed-by: Suzuki K Poulose <suzuki.poulose@arm.com> _______________________________________________ linux-arm-kernel mailing list linux-arm-kernel@lists.infradead.org http://lists.infradead.org/mailman/listinfo/linux-arm-kernel