The patch below does not apply to the 5.15-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to stable@vger.kernel.org.
To reproduce the conflict and resubmit, you may use the following commands:
git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-5.15.y git checkout FETCH_HEAD git cherry-pick -x 3bf983e4e93ce8e6d69e9d63f52a66ec0856672e # <resolve conflicts, build, test, etc.> git commit -s git send-email --to 'stable@vger.kernel.org' --in-reply-to '2023112321-tightrope-reimburse-f004@gregkh' --subject-prefix 'PATCH 5.15.y' HEAD^..
Possible dependencies:
3bf983e4e93c ("powerpc/pseries/iommu: enable_ddw incorrectly returns direct mapping for SR-IOV device") fb4ee2b30cd0 ("powerpc/pseries/ddw: simplify enable_ddw()")
thanks,
greg k-h
------------------ original commit in Linus's tree ------------------
From 3bf983e4e93ce8e6d69e9d63f52a66ec0856672e Mon Sep 17 00:00:00 2001 From: Gaurav Batra gbatra@linux.vnet.ibm.com Date: Mon, 2 Oct 2023 22:08:02 -0500 Subject: [PATCH] powerpc/pseries/iommu: enable_ddw incorrectly returns direct mapping for SR-IOV device
When a device is initialized, the driver invokes dma_supported() twice - first for streaming mappings followed by coherent mappings. For an SR-IOV device, default window is deleted and DDW created. With vPMEM enabled, TCE mappings are dynamically created for both vPMEM and SR-IOV device. There are no direct mappings.
First time when dma_supported() is called with 64 bit mask, DDW is created and marked as dynamic window. The second time dma_supported() is called, enable_ddw() finds existing window for the device and incorrectly returns it as "direct mapping".
This only happens when size of DDW is big enough to map max LPAR memory.
This results in streaming TCEs to not get dynamically mapped, since code incorrently assumes these are already pre-mapped. The adapter initially comes up but goes down due to EEH.
Fixes: 381ceda88c4c ("powerpc/pseries/iommu: Make use of DDW for indirect mapping") Cc: stable@vger.kernel.org # v5.15+ Signed-off-by: Gaurav Batra gbatra@linux.vnet.ibm.com Signed-off-by: Michael Ellerman mpe@ellerman.id.au Link: https://msgid.link/20231003030802.47914-1-gbatra@linux.vnet.ibm.com
diff --git a/arch/powerpc/platforms/pseries/iommu.c b/arch/powerpc/platforms/pseries/iommu.c index 16d93b580f61..496e16c588aa 100644 --- a/arch/powerpc/platforms/pseries/iommu.c +++ b/arch/powerpc/platforms/pseries/iommu.c @@ -914,7 +914,8 @@ static int remove_ddw(struct device_node *np, bool remove_prop, const char *win_ return 0; }
-static bool find_existing_ddw(struct device_node *pdn, u64 *dma_addr, int *window_shift) +static bool find_existing_ddw(struct device_node *pdn, u64 *dma_addr, int *window_shift, + bool *direct_mapping) { struct dma_win *window; const struct dynamic_dma_window_prop *dma64; @@ -927,6 +928,7 @@ static bool find_existing_ddw(struct device_node *pdn, u64 *dma_addr, int *windo dma64 = window->prop; *dma_addr = be64_to_cpu(dma64->dma_base); *window_shift = be32_to_cpu(dma64->window_shift); + *direct_mapping = window->direct; found = true; break; } @@ -1270,10 +1272,8 @@ static bool enable_ddw(struct pci_dev *dev, struct device_node *pdn)
mutex_lock(&dma_win_init_mutex);
- if (find_existing_ddw(pdn, &dev->dev.archdata.dma_offset, &len)) { - direct_mapping = (len >= max_ram_len); + if (find_existing_ddw(pdn, &dev->dev.archdata.dma_offset, &len, &direct_mapping)) goto out_unlock; - }
/* * If we already went through this for a previous function of