riscv: mm: still create swiotlb buffer for kmalloc() bouncing if required

PATCH RESEND Applied Jisheng Zhang 3 participants 3 messages Mar 25, 2024 – May 7, 2024 RISC-V

Share & tools

Jisheng Zhang

author

2024-03-25 11:13 UTC #0 lore raw

PATCH RESEND v3 1/1 riscv: mm: still create swiotlb buffer for kmalloc() bouncing if required

After commit f51f7a0fc2f4 ("riscv: enable DMA_BOUNCE_UNALIGNED_KMALLOC for !dma_coherent"), for non-coherent platforms with less than 4GB memory, we rely on users to pass "swiotlb=mmnn,force" kernel parameters to enable DMA bouncing for unaligned kmalloc() buffers. Now let's go
further: If no bouncing needed for ZONE_DMA, let kernel automatically allocate 1MB swiotlb buffer per 1GB of RAM for kmalloc() bouncing on non-coherent platforms, so that no need to pass "swiotlb=mmnn,force" any more.

The math of "1MB swiotlb buffer per 1GB of RAM for kmalloc() bouncing" is taken from arm64. Users can still force smaller swiotlb buffer by passing "swiotlb=mmnn".

Signed-off-by Jisheng Zhang <[email protected]>

Reviewed-by Alexandre Ghiti <[email protected]>

---
since v2:
- rebase on v6.8-rc1
- collect Reviewed-by tag

since v1
- fix build error if CONFIG_RISCV_DMA_NONCOHERENT=n

 arch/riscv/include/asm/cache.h |  2 +-
 arch/riscv/mm/init.c           | 16 +++++++++++++++-
 2 files changed, 16 insertions(+), 2 deletions(-)

arch/riscv/include/asm/cache.h +1 -1

index 2174fe7bac9a..570e9d8acad1 100644
@@ -26,8 +26,8 @@
 
 #ifndef __ASSEMBLY__
 
-#ifdef CONFIG_RISCV_DMA_NONCOHERENT
 extern int dma_cache_alignment;
+#ifdef CONFIG_RISCV_DMA_NONCOHERENT
 #define dma_get_cache_alignment dma_get_cache_alignment
 static inline int dma_get_cache_alignment(void)
 {

arch/riscv/mm/init.c +15 -1

index fe8e159394d8..ac5e7f64c05c 100644
@@ -161,11 +161,25 @@ static void print_vm_layout(void) { }
 
 void __init mem_init(void)
 {
+	bool swiotlb = max_pfn > PFN_DOWN(dma32_phys_limit);
 #ifdef CONFIG_FLATMEM
 	BUG_ON(!mem_map);
 #endif /* CONFIG_FLATMEM */
 
-	swiotlb_init(max_pfn > PFN_DOWN(dma32_phys_limit), SWIOTLB_VERBOSE);
+	if (IS_ENABLED(CONFIG_DMA_BOUNCE_UNALIGNED_KMALLOC) && !swiotlb &&
+	    dma_cache_alignment != 1) {
+		/*
+		 * If no bouncing needed for ZONE_DMA, allocate 1MB swiotlb
+		 * buffer per 1GB of RAM for kmalloc() bouncing on
+		 * non-coherent platforms.
+		 */
+		unsigned long size =
+			DIV_ROUND_UP(memblock_phys_mem_size(), 1024);
+		swiotlb_adjust_size(min(swiotlb_size_or_default(), size));
+		swiotlb = true;
+	}
+
+	swiotlb_init(swiotlb, SWIOTLB_VERBOSE);
 	memblock_free_all();
 
 	print_vm_layout();

Signature · 1 line

2.43.0

[email protected]

CI Applied

2024-04-28 22:00 UTC #1 lore raw

This patch was applied to riscv/linux.git (for-next) by Palmer Dabbelt <[email protected]>:

Hello:

This patch was applied to riscv/linux.git (for-next) by Palmer Dabbelt <[email protected]>:

Quoted 10 lines · On Mon, 25 Mar 2024 19:00:36 +0800 you wrote:

After commit f51f7a0fc2f4 ("riscv: enable DMA_BOUNCE_UNALIGNED_KMALLOC for !dma_coherent"), for non-coherent platforms with less than 4GB memory, we rely on users to pass "swiotlb=mmnn,force" kernel parameters to enable DMA bouncing for unaligned kmalloc() buffers. Now let's go
further: If no bouncing needed for ZONE_DMA, let kernel automatically allocate 1MB swiotlb buffer per 1GB of RAM for kmalloc() bouncing on non-coherent platforms, so that no need to pass "swiotlb=mmnn,force" any more.
[...]

Here is the summary with links:
- [v3,RESEND] riscv: mm: still create swiotlb buffer for kmalloc() bouncing if required

    https://git.kernel.org/riscv/c/fc7a50eed986

You are awesome, thank you!

Signature · 2 lines

Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html

Geert Uytterhoeven

2024-05-07 15:08 UTC #2 lore raw

In reply to Jisheng Zhang

Hi Jisheng,

Quoted 9 lines · Jisheng Zhang

After commit f51f7a0fc2f4 ("riscv: enable DMA_BOUNCE_UNALIGNED_KMALLOC for !dma_coherent"), for non-coherent platforms with less than 4GB memory, we rely on users to pass "swiotlb=mmnn,force" kernel parameters to enable DMA bouncing for unaligned kmalloc() buffers. Now let's go
further: If no bouncing needed for ZONE_DMA, let kernel automatically allocate 1MB swiotlb buffer per 1GB of RAM for kmalloc() bouncing on non-coherent platforms, so that no need to pass "swiotlb=mmnn,force" any more.

Quoted 6 lines

The math of "1MB swiotlb buffer per 1GB of RAM for kmalloc() bouncing" is taken from arm64. Users can still force smaller swiotlb buffer by passing "swiotlb=mmnn".
Signed-off-by Jisheng Zhang <[email protected]>
Reviewed-by Alexandre Ghiti <[email protected]>

Thanks for your patch, which is now commit fc7a50eed9860d4b ("riscv:
mm: still create swiotlb buffer for kmalloc() bouncing if required") in riscv/for-next (next-20240429 and later).

On RZ/Five, which has 1 GiB of RAM, i.e. a bit less for free use:

Inode-cache hash table entries: 65536 (order: 7, 524288 bytes, linear) Built 1 zonelists, mobility grouping on. Total pages: 225792 mem auto-init: stack:off, heap alloc:off, heap free:off
+software IO TLB: SWIOTLB bounce buffer size adjusted to 0MB
^^^
+software IO TLB: area num 1.
+software IO TLB: mapped [mem 0x000000007ef56000-0x000000007f056000] (1MB)
^^^
Virtual kernel memory layout:

I was a bit intrigued by the "0MB". However, that seems to be correct:

     mem_init: memblock_phys_mem_size() = 939524096
     mem_init: size = 917504
     mem_init: swiotlb_size_or_default() = 67108864

and it must be rounded up to 1 MB later.

Apparently arm64 has the same discrepancy, which I never really noticed before (and I have no arm64 platforms with 1 GiB RAM or less):

     mem_init: memblock_phys_mem_size() = 2013265920
     mem_init: size = 1966080
     mem_init: swiotlb_size_or_default() = 67108864
     software IO TLB: SWIOTLB bounce buffer size adjusted to 1MB
                                                             ^^^
     software IO TLB: area num 2.
     software IO TLB: mapped [mem 0x00000000b9400000-0x00000000b9600000] (2MB)
                                                                         ^^^

Quoted 19 lines

arch/riscv/mm/init.c +8 -1

@@ -161,11 +161,25 @@ static void print_vm_layout(void) { }
 
 void __init mem_init(void)
 {
+       bool swiotlb = max_pfn > PFN_DOWN(dma32_phys_limit);
 #ifdef CONFIG_FLATMEM
        BUG_ON(!mem_map);
 #endif /* CONFIG_FLATMEM */
 
-       swiotlb_init(max_pfn > PFN_DOWN(dma32_phys_limit), SWIOTLB_VERBOSE);
+       if (IS_ENABLED(CONFIG_DMA_BOUNCE_UNALIGNED_KMALLOC) && !swiotlb &&
+           dma_cache_alignment != 1) {
+               /*
+                * If no bouncing needed for ZONE_DMA, allocate 1MB swiotlb
+                * buffer per 1GB of RAM for kmalloc() bouncing on
+                * non-coherent platforms.
+                */

arch/riscv/mm/init.c static void print_vm_layout(void) { } · line 185

Quoted 10 lines

arch/riscv/mm/init.c +7 -0

+               unsigned long size =
+                       DIV_ROUND_UP(memblock_phys_mem_size(), 1024);
+               swiotlb_adjust_size(min(swiotlb_size_or_default(), size));
+               swiotlb = true;
+       }
+
+       swiotlb_init(swiotlb, SWIOTLB_VERBOSE);
        memblock_free_all();
 
        print_vm_layout();

Gr{oetje,eeting}s,

Geert

Signature · 5 lines

Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds