dma: Fix max PFN arithmetic overflow on 32 bit systems
Subsystems (2)
PATCH v3 1/1 dma: Fix max PFN arithmetic overflow on 32 bit systems
The intermediate result of the old term (4UL * 1024 * 1024 * 1024) is 4 294 967 296 or 0x100000000 which is no problem on 64 bit systems. The patch does not change the later overall result of 0x100000 for MAX_DMA32_PFN. The new calculation yields the same result, but does not require 64 bit arithmetic.
On 32 bit systems the old calculation suffers from an arithmetic overflow in that intermediate term in braces: 4UL aka unsigned long int is 4 byte wide and an arithmetic overflow happens (the 0x100000000 does not fit in 4 bytes), the in braces result is truncated to zero, the following right shift does not alter that, so MAX_DMA32_PFN evaluates to 0 on 32 bit systems.
That wrong value is a problem in a comparision against MAX_DMA32_PFN in the init code for swiotlb in 'pci_swiotlb_detect_4gb()' to decide if swiotlb should be active. That comparison yields the opposite result, when compiling on 32 bit systems.
This was not possible before 1b7e03ef7570 ("x86, NUMA: Enable emulation on 32bit too") when that MAX_DMA32_PFN was first made visible to x86_32 (and which landed in v3.0).
In practice this wasn't a problem, unless you activated CONFIG_SWIOTLB on x86 (32 bit).
However for ARCH=x86 (32 bit) and if you have set CONFIG_IOMMU_INTEL, since c5a5dc4cbbf4 ("iommu/vt-d: Don't switch off swiotlb if bounce page is used") there's a dependency on CONFIG_SWIOTLB, which was not necessarily active before. That landed in v5.4, where we noticed it in the fli4l Linux distribution. We have CONFIG_IOMMU_INTEL active on both 32 and 64 bit kernel configs there (I could not find out why, so let's just say historical reasons).
The effect is at boot time 64 MiB (default size) were allocated for bounce buffers now, which is a noticeable amount of memory on small systems like pcengines ALIX 2D3 with 256 MiB memory, which are still frequently used as home routers.
We noticed this effect when migrating from kernel v4.19 (LTS) to v5.4 (LTS) in fli4l and got that kernel messages for example:
Linux version 5.4.22 (buildroot@buildroot) (gcc version 7.3.0 (Buildroot 2018.02.8)) #1 SMP Mon Nov 26 23:40:00 CET 2018
…
Memory: 183484K/261756K available (4594K kernel code, 393K rwdata, 1660K rodata, 536K init, 456K bss , 78272K reserved, 0K cma-reserved, 0K highmem) …
PCI-DMA: Using software bounce buffering for IO (SWIOTLB)
software IO TLB: mapped [mem 0x0bb78000-0x0fb78000] (64MB)
The initial analysis and the suggested fix was done by user 'sourcejedi' at stackoverflow and explicitly marked as GPLv2 for inclusion in the Linux kernel:
https://unix.stackexchange.com/a/520525/50007
The new calculation, which does not suffer from that overflow, is the same as for arch/mips now as suggested by Robin Murphy.
The fix was tested by fli4l users on round about two dozen different systems, including both 32 and 64 bit archs, bare metal and virtualized machines.
---
Notes:
v3:
- rewritten commit message to better explain that arithmetic overflow and added Fixes tag (Greg Kroah-Hartman)
- rebased on v5.7-rc7
v2:
- use the same calculation as with arch/mips (Robin Murphy)
arch/x86/include/asm/dma.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
arch/x86/include/asm/dma.h +1 -1
@@ -74,7 +74,7 @@#define MAX_DMA_PFN ((16UL * 1024 * 1024) >> PAGE_SHIFT)/* 4GB broken PCI/AGP hardware bus master zone */ -#define MAX_DMA32_PFN ((4UL * 1024 * 1024 * 1024) >> PAGE_SHIFT) +#define MAX_DMA32_PFN (1UL << (32 - PAGE_SHIFT))#ifdef CONFIG_X86_32 /* The maximum address that we can perform a DMA transfer to on this platform */
Signature · 1 line
2.20.1
Quoted 61 lines · Alexander Dahl
The intermediate result of the old term (4UL * 1024 * 1024 * 1024) is 4 294 967 296 or
0x100000000which is no problem on 64 bit systems. The patch does not change the later overall result of0x100000forMAX_DMA32_PFN. The new calculation yields the same result, but does not require 64 bit arithmetic.On 32 bit systems the old calculation suffers from an arithmetic overflow in that intermediate term in braces: 4UL aka unsigned long int is 4 byte wide and an arithmetic overflow happens (the
0x100000000does not fit in 4 bytes), the in braces result is truncated to zero, the following right shift does not alter that, soMAX_DMA32_PFNevaluates to 0 on 32 bit systems.That wrong value is a problem in a comparision against
MAX_DMA32_PFNin the init code for swiotlb in 'pci_swiotlb_detect_4gb()' to decide if swiotlb should be active. That comparison yields the opposite result, when compiling on 32 bit systems.This was not possible before
1b7e03ef7570("x86, NUMA: Enable emulation on 32bit too") when thatMAX_DMA32_PFNwas first made visible tox86_32(and which landed in v3.0).In practice this wasn't a problem, unless you activated
CONFIG_SWIOTLBon x86 (32 bit).However for ARCH=x86 (32 bit) and if you have set
CONFIG_IOMMU_INTEL, sincec5a5dc4cbbf4("iommu/vt-d: Don't switch off swiotlb if bounce page is used") there's a dependency onCONFIG_SWIOTLB, which was not necessarily active before. That landed in v5.4, where we noticed it in the fli4l Linux distribution. We haveCONFIG_IOMMU_INTELactive on both 32 and 64 bit kernel configs there (I could not find out why, so let's just say historical reasons).The effect is at boot time 64 MiB (default size) were allocated for bounce buffers now, which is a noticeable amount of memory on small systems like pcengines ALIX 2D3 with 256 MiB memory, which are still frequently used as home routers.
We noticed this effect when migrating from kernel v4.19 (LTS) to v5.4 (LTS) in fli4l and got that kernel messages for example:
Linux version 5.4.22 (buildroot@buildroot) (gcc version 7.3.0 (Buildroot 2018.02.8)) #1 SMP Mon Nov 26 23:40:00 CET 2018
…
Memory: 183484K/261756K available (4594K kernel code, 393K rwdata, 1660K rodata, 536K init, 456K bss , 78272K reserved, 0K cma-reserved, 0K highmem) …
PCI-DMA: Using software bounce buffering for IO (SWIOTLB)
software IO TLB: mapped [mem 0x0bb78000-0x0fb78000] (64MB)The initial analysis and the suggested fix was done by user 'sourcejedi' at stackoverflow and explicitly marked as GPLv2 for inclusion in the Linux kernel:
https://unix.stackexchange.com/a/520525/50007
The new calculation, which does not suffer from that overflow, is the same as for
arch/mipsnow as suggested by Robin Murphy.The fix was tested by fli4l users on round about two dozen different systems, including both 32 and 64 bit archs, bare metal and virtualized machines.
Quoted 7 lines
Fixes 1b7e03ef7570 ("x86, NUMA: Enable emulation on 32bit too")Reported-by Alan Jenkins <[email protected]>Suggested-by Robin Murphy <[email protected]>Signed-off-by Alexander Dahl <[email protected]>
[tip: x86/urgent] x86/dma: Fix max PFN arithmetic overflow on 32 bit systems
In reply to Alexander DahlThe following commit has been merged into the x86/urgent branch of tip:
The following commit has been merged into the x86/urgent branch of tip:
Commit-ID: 88743470668ef5eb6b7ba9e0f99888e5999bf172 Gitweb: https://git.kernel.org/tip/88743470668ef5eb6b7ba9e0f99888e5999bf172 Author: Alexander Dahl <[email protected]> AuthorDate: Tue, 26 May 2020 19:57:49 +02:00 Committer: Borislav Petkov <[email protected]> CommitterDate: Thu, 28 May 2020 20:21:32 +02:00
x86/dma: Fix max PFN arithmetic overflow on 32 bit systems
The intermediate result of the old term (4UL * 1024 * 1024 * 1024) is 4 294 967 296 or 0x100000000 which is no problem on 64 bit systems. The patch does not change the later overall result of 0x100000 for MAX_DMA32_PFN (after it has been shifted by PAGE_SHIFT). The new calculation yields the same result, but does not require 64 bit arithmetic.
On 32 bit systems the old calculation suffers from an arithmetic overflow in that intermediate term in braces: 4UL aka unsigned long int is 4 byte wide and an arithmetic overflow happens (the 0x100000000 does not fit in 4 bytes), the in braces result is truncated to zero, the following right shift does not alter that, so MAX_DMA32_PFN evaluates to 0 on 32 bit systems.
That wrong value is a problem in a comparision against MAX_DMA32_PFN in the init code for swiotlb in pci_swiotlb_detect_4gb() to decide if swiotlb should be active. That comparison yields the opposite result, when compiling on 32 bit systems.
This was not possible before
1b7e03ef7570 ("x86, NUMA: Enable emulation on 32bit too")
when that MAX_DMA32_PFN was first made visible to x86_32 (and which landed in v3.0).
In practice this wasn't a problem, unless CONFIG_SWIOTLB is active on x86-32.
However if one has set CONFIG_IOMMU_INTEL, since
c5a5dc4cbbf4 ("iommu/vt-d: Don't switch off swiotlb if bounce page is used")
there's a dependency on CONFIG_SWIOTLB, which was not necessarily active before. That landed in v5.4, where we noticed it in the fli4l Linux distribution. We have CONFIG_IOMMU_INTEL active on both 32 and 64 bit kernel configs there (I could not find out why, so let's just say historical reasons).
The effect is at boot time 64 MiB (default size) were allocated for bounce buffers now, which is a noticeable amount of memory on small systems like pcengines ALIX 2D3 with 256 MiB memory, which are still frequently used as home routers.
We noticed this effect when migrating from kernel v4.19 (LTS) to v5.4 (LTS) in fli4l and got that kernel messages for example:
Linux version 5.4.22 (buildroot@buildroot) (gcc version 7.3.0 (Buildroot 2018.02.8)) #1 SMP Mon Nov 26 23:40:00 CET 2018
…
Memory: 183484K/261756K available (4594K kernel code, 393K rwdata, 1660K rodata, 536K init, 456K bss , 78272K reserved, 0K cma-reserved, 0K highmem) …
PCI-DMA: Using software bounce buffering for IO (SWIOTLB)
software IO TLB: mapped [mem 0x0bb78000-0x0fb78000] (64MB)
The initial analysis and the suggested fix was done by user 'sourcejedi' at stackoverflow and explicitly marked as GPLv2 for inclusion in the Linux kernel:
https://unix.stackexchange.com/a/520525/50007
The new calculation, which does not suffer from that overflow, is the same as for arch/mips now as suggested by Robin Murphy.
The fix was tested by fli4l users on round about two dozen different systems, including both 32 and 64 bit archs, bare metal and virtualized machines.
[ bp: Massage commit message. ]
--- arch/x86/include/asm/dma.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
arch/x86/include/asm/dma.h +1 -1
@@ -74,7 +74,7 @@#define MAX_DMA_PFN ((16UL * 1024 * 1024) >> PAGE_SHIFT)/* 4GB broken PCI/AGP hardware bus master zone */ -#define MAX_DMA32_PFN ((4UL * 1024 * 1024 * 1024) >> PAGE_SHIFT) +#define MAX_DMA32_PFN (1UL << (32 - PAGE_SHIFT))#ifdef CONFIG_X86_32 /* The maximum address that we can perform a DMA transfer to on this platform */