Re: [PATCH kernel v2] swiotlb: Half the size if allocation failed

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On 10/31/2022 9:13 AM, Alexey Kardashevskiy wrote:
At the moment the AMD encrypted platform reserves 6% of RAM for SWIOTLB
or 1GB, whichever is less. However it is possible that there is no block
big enough in the low memory which make SWIOTLB allocation fail and
the kernel continues without DMA. In such case a VM hangs on DMA.

This moves alloc+remap to a helper and calls it from a loop where
the size is halved on each iteration.

This updates default_nslabs on successful allocation which looks like
an oversight as not doing so should have broken callers of
swiotlb_size_or_default().

Signed-off-by: Alexey Kardashevskiy <aik@xxxxxxx>
--
Changes:
v2:
* moved alloc+remap to a helper as suggested
* removed "x86" and "amd" from subj

--
I hit the problem with
QEMU's "-m 16819M" where SWIOTLB was adjusted to
0x7e200 == 1,058,013,184 (slightly less than 1GB) while
0x7e180 still worked.

With guest errors enabled, there are many unassigned accesses from
virtio.
---
  kernel/dma/swiotlb.c | 66 +++++++++++++++++++++++++++-----------------
  1 file changed, 41 insertions(+), 25 deletions(-)

diff --git a/kernel/dma/swiotlb.c b/kernel/dma/swiotlb.c
index 339a990554e7..53fc6e7d3aa5 100644
--- a/kernel/dma/swiotlb.c
+++ b/kernel/dma/swiotlb.c
@@ -300,6 +300,36 @@ static void swiotlb_init_io_tlb_mem(struct io_tlb_mem *mem, phys_addr_t start,
  	return;
  }
+static void *swiotlb_memblock_alloc(unsigned long nslabs, unsigned int flags,
+				    int (*remap)(void *tlb, unsigned long nslabs))
+{
+	size_t bytes = PAGE_ALIGN(nslabs << IO_TLB_SHIFT);
+	void *tlb;
+
+	/*
+	 * By default allocate the bounce buffer memory from low memory, but
+	 * allow to pick a location everywhere for hypervisors with guest
+	 * memory encryption.
+	 */
+	if (flags & SWIOTLB_ANY)
+		tlb = memblock_alloc(bytes, PAGE_SIZE);
+	else
+		tlb = memblock_alloc_low(bytes, PAGE_SIZE);
+
+	if (!tlb) {
+		pr_warn("%s: Failed to allocate %zu bytes tlb structure\n", __func__, bytes);
+		return NULL;
+	}
+
+	if (remap && remap(tlb, nslabs) < 0) {
+		memblock_free(tlb, PAGE_ALIGN(bytes));
+		pr_warn("%s: Failed to remap %zu bytes\n", __func__, bytes);
+		return NULL;
+	}
+
+	return tlb;
+}
+
  /*
   * Statically reserve bounce buffer space and initialize bounce buffer data
   * structures for the software IO TLB used to implement the DMA API.
@@ -310,7 +340,6 @@ void __init swiotlb_init_remap(bool addressing_limit, unsigned int flags,
  	struct io_tlb_mem *mem = &io_tlb_default_mem;
  	unsigned long nslabs;
  	size_t alloc_size;
-	size_t bytes;
  	void *tlb;
if (!addressing_limit && !swiotlb_force_bounce)
@@ -325,32 +354,19 @@ void __init swiotlb_init_remap(bool addressing_limit, unsigned int flags,
  	if (!default_nareas)
  		swiotlb_adjust_nareas(num_possible_cpus());
- nslabs = default_nslabs;
-	/*
-	 * By default allocate the bounce buffer memory from low memory, but
-	 * allow to pick a location everywhere for hypervisors with guest
-	 * memory encryption.
-	 */
-retry:
-	bytes = PAGE_ALIGN(nslabs << IO_TLB_SHIFT);
-	if (flags & SWIOTLB_ANY)
-		tlb = memblock_alloc(bytes, PAGE_SIZE);
-	else
-		tlb = memblock_alloc_low(bytes, PAGE_SIZE);
-	if (!tlb) {
-		pr_warn("%s: failed to allocate tlb structure\n", __func__);
-		return;
-	}
-
-	if (remap && remap(tlb, nslabs) < 0) {
-		memblock_free(tlb, PAGE_ALIGN(bytes));
-
+	for (nslabs = default_nslabs;; ) {
+		tlb = swiotlb_memblock_alloc(nslabs, flags, remap);
+		if (tlb)
+			break;
+		if (nslabs <= IO_TLB_MIN_SLABS)
+			return;
  		nslabs = ALIGN(nslabs >> 1, IO_TLB_SEGSIZE);
-		if (nslabs >= IO_TLB_MIN_SLABS)
-			goto retry;
+	}
- pr_warn("%s: Failed to remap %zu bytes\n", __func__, bytes);
-		return;
+	if (default_nslabs != nslabs) {
+		pr_info("SWIOTLB bounce buffer size adjusted %lu -> %lu slabs",
+			default_nslabs, nslabs);
+		default_nslabs = nslabs;
  	}
alloc_size = PAGE_ALIGN(array_size(sizeof(*mem->slots), nslabs));

With memblock contiguous allocation under 4G memory fallback to lower order allocation, seems to fix the inconsistent state issue if the buffer is not at all allocated.

Feel free to add:
Reviewed-by: Pankaj Gupta <pankaj.gupta@xxxxxxx>



[Index of Archives]     [KVM ARM]     [KVM ia64]     [KVM ppc]     [Virtualization Tools]     [Spice Development]     [Libvirt]     [Libvirt Users]     [Linux USB Devel]     [Linux Audio Users]     [Yosemite Questions]     [Linux Kernel]     [Linux SCSI]     [XFree86]

  Powered by Linux