s390/ism: Properly fix receive message buffer allocation
Subsystems (4)
PATCH 1/1 s390/ism: Properly fix receive message buffer allocation
Since [1], dma_alloc_coherent() does not accept requests for GFP_COMP anymore, even on archs that may be able to fulfill this. Functionality that relied on the receive buffer being a compound page broke at that point:
The SMC-D protocol, that utilizes the ism device driver, passes receive buffers to the splice processor in a struct splice_pipe_desc with a single entry list of struct pages. As the buffer is no longer a compound page, the splice processor now rejects requests to handle more than a page worth of data.
Replace dma_alloc_coherent() and allocate a buffer with folio_alloc and create a DMA map for it with dma_map_page(). Since only receive buffers on ISM devices use DMA, qualify the mapping as FROM_DEVICE. Since ISM devices are available on arch s390, only, and on that arch all DMA is coherent, there is no need to introduce and export some kind of dma_sync_to_cpu() method to be called by the SMC-D protocol layer.
Analogously, replace dma_free_coherent by a two step dma_unmap_page, then folio_put to free the receive buffer.
[1] https://lore.kernel.org/all/[email protected]/
---
Hi all,
this is the next iteration (v3 so to speak) of the fix [2] whose v2 [3] was picked - then reverted - while there was still one comment outstanding. Here the last review comment is addressed and proper error checking is done after allocating the folio.
Please consider this complete patch for inclusion into the next 6.9 rc.
Thank you,
Gerd
[2] https://lore.kernel.org/all/[email protected]/
[3] https://lore.kernel.org/all/[email protected]/
https://lore.kernel.org/all/[email protected]/
--- drivers/s390/net/ism_drv.c | 37 ++++++++++++++++++++++++++++--------- 1 file changed, 28 insertions(+), 9 deletions(-)
drivers/s390/net/ism_drv.c +28 -9
@@ -292,13 +292,16 @@ static int ism_read_local_gid(struct ism_dev *ism)static void ism_free_dmb (struct ism_dev *ism,struct ism_dmb *dmb) {clear_bit (dmb->sba_idx, ism->sba_bitmap); -dma_free_coherent (&ism->pdev->dev, dmb->dmb_len, - dmb->cpu_addr, dmb->dma_addr); +dma_unmap_page (&ism->pdev->dev, dmb->dma_addr, dmb->dmb_len, +DMA_FROM_DEVICE ); +folio_put (virt_to_folio (dmb->cpu_addr)); }static int ism_alloc_dmb (struct ism_dev *ism,struct ism_dmb *dmb) { +struct folio *folio;unsigned long bit; +int rc;if (PAGE_ALIGN (dmb->dmb_len) >dma_get_max_seg_size (&ism->pdev->dev))return -EINVAL ;
@@ -315,14 +318,30 @@ static int ism_alloc_dmb(struct ism_dev *ism, struct ism_dmb *dmb) test_and_set_bit(dmb->sba_idx, ism->sba_bitmap)) return -EINVAL; - dmb->cpu_addr =dma_alloc_coherent (&ism->pdev->dev, dmb->dmb_len, - &dmb->dma_addr, -GFP_KERNEL | __GFP_NOWARN | - __GFP_NOMEMALLOC | __GFP_NORETRY); -if (!dmb->cpu_addr) -clear_bit (dmb->sba_idx, ism->sba_bitmap); + folio =folio_alloc (GFP_KERNEL | __GFP_NOWARN | __GFP_NOMEMALLOC | + __GFP_NORETRY,get_order (dmb->dmb_len)); -return dmb->cpu_addr ?0 : -ENOMEM ; +if (!folio) { + rc = -ENOMEM ; + goto out_bit; + } + + dmb->cpu_addr =folio_address (folio); + dmb->dma_addr =dma_map_page (&ism->pdev->dev, +virt_to_page (dmb->cpu_addr),0 , + dmb->dmb_len,DMA_FROM_DEVICE ); +if (dma_mapping_error (&ism->pdev->dev, dmb->dma_addr)) { + rc = -ENOMEM ; + goto out_free; + } + +return 0 ; + +out_free: +kfree (dmb->cpu_addr); +out_bit: +clear_bit (dmb->sba_idx, ism->sba_bitmap); +return rc; } int ism_register_dmb(struct ism_dev *ism, struct ism_dmb *dmb,
Signature · 1 line
2.44.0
This patch was applied to netdev/net.git (main) by David S. Miller <[email protected]>:
Hello:
This patch was applied to netdev/net.git (main)
by David S. Miller <[email protected]>:
Quoted 10 lines · On Mon, 15 Apr 2024 15:15:07 +0200 you wrote:
Since [1],
dma_alloc_coherent()does not accept requests forGFP_COMPanymore, even on archs that may be able to fulfill this. Functionality that relied on the receive buffer being a compound page broke at that point:
The SMC-D protocol, that utilizes the ism device driver, passes receive buffers to the splice processor in a structsplice_pipe_descwith a single entry list of struct pages. As the buffer is no longer a compound page, the splice processor now rejects requests to handle more than a page worth of data.[...]
Here is the summary with links:
- [net] s390/ism: Properly fix receive message buffer allocation
https://git.kernel.org/netdev/net/c/83781384a96b
You are awesome, thank you!
Signature · 2 lines
Deet-doot-dot, I am a bot. https://korg.docs.kernel.org/patchwork/pwbot.html