December 2025 - Linux-stable-mirror

[PATCH v2 2/5] usb: phy: isp1301: fix non-OF device reference imbalance

by Johan Hovold

A recent change fixing a device reference leak in a UDC driver introduced a potential use-after-free in the non-OF case as the isp1301_get_client() helper only increases the reference count for the returned I2C device in the OF case. Increment the reference count also for non-OF so that the caller can decrement it unconditionally. Note that this is inherently racy just as using the returned I2C device is since nothing is preventing the PHY driver from being unbound while in use. Fixes: c84117912bdd ("USB: lpc32xx_udc: Fix error handling in probe") Cc: stable(a)vger.kernel.org Cc: Ma Ke <make24(a)iscas.ac.cn> Signed-off-by: Johan Hovold <johan(a)kernel.org> --- drivers/usb/phy/phy-isp1301.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/usb/phy/phy-isp1301.c b/drivers/usb/phy/phy-isp1301.c index f9b5c411aee4..2940f0c84e1b 100644 --- a/drivers/usb/phy/phy-isp1301.c +++ b/drivers/usb/phy/phy-isp1301.c @@ -149,7 +149,12 @@ struct i2c_client *isp1301_get_client(struct device_node *node) return client; /* non-DT: only one ISP1301 chip supported */ - return isp1301_i2c_client; + if (isp1301_i2c_client) { + get_device(&isp1301_i2c_client->dev); + return isp1301_i2c_client; + } + + return NULL; } EXPORT_SYMBOL_GPL(isp1301_get_client); -- 2.51.2

1 week, 5 days

2
2
0 0

[PATCH] ocfs2: handle OCFS2_SUPER_BLOCK_FL flag in system dinode

by Prithvi Tambewagh

When ocfs2_populate_inode() is called during mount process, if the flag OCFS2_SUPER_BLOCK_FL is set in on-disk system dinode, then BUG() is triggered, causing kernel to panic. This is indicative of metadata corruption. This is fixed by calling ocfs2_error() to print the error log and the corresponding inode is marked as 'bad', so that it is not used further during the mount process. It is ensured that the fact of that inode being bad is propagated to caller ocfs2_populate_inode() i.e. ocfs2_read_locked_inode() using is_bad_inode() and further behind along the call trace as well. Reported-by: syzbot+779d072a1067a8b1a917(a)syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=779d072a1067a8b1a917 Tested-by: syzbot+779d072a1067a8b1a917(a)syzkaller.appspotmail.com Cc: stable(a)vger.kernel.org Signed-off-by: Prithvi Tambewagh <activprithvi(a)gmail.com> --- fs/ocfs2/inode.c | 14 ++++++++++++-- 1 file changed, 12 insertions(+), 2 deletions(-) diff --git a/fs/ocfs2/inode.c b/fs/ocfs2/inode.c index 12e5d1f73325..f439dc801845 100644 --- a/fs/ocfs2/inode.c +++ b/fs/ocfs2/inode.c @@ -347,7 +347,12 @@ void ocfs2_populate_inode(struct inode *inode, struct ocfs2_dinode *fe, } else if (fe->i_flags & cpu_to_le32(OCFS2_SUPER_BLOCK_FL)) { /* we can't actually hit this as read_inode can't * handle superblocks today ;-) */ - BUG(); + ocfs2_error(sb, + "System Inode %llu has " + "OCFS2_SUPER_BLOCK_FL set", + (unsigned long long)le64_to_cpu(fe->i_blkno)); + make_bad_inode(inode); + return; } switch (inode->i_mode & S_IFMT) { @@ -555,6 +560,11 @@ static int ocfs2_read_locked_inode(struct inode *inode, ocfs2_populate_inode(inode, fe, 0); + if (is_bad_inode(inode)) { + status = -EIO; + goto bail; + } + BUG_ON(args->fi_blkno != le64_to_cpu(fe->i_blkno)); if (buffer_dirty(bh) && !buffer_jbd(bh)) { @@ -576,7 +586,7 @@ static int ocfs2_read_locked_inode(struct inode *inode, if (can_lock) ocfs2_inode_unlock(inode, lock_level); - if (status < 0) + if (status < 0 && !is_bad_inode(inode)) make_bad_inode(inode); brelse(bh); base-commit: d76bb1ebb5587f66b0f8b8099bfbb44722bc08b3 -- 2.43.0

1 week, 5 days

3
2
0 0

FAILED: patch "[PATCH] hsr: hold rcu and dev lock for hsr_get_port_ndev" failed to apply to 6.12-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 6.12-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(a)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-6.12.y git checkout FETCH_HEAD git cherry-pick -x 847748fc66d08a89135a74e29362a66ba4e3ab15 # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2025121829-aloof-cresting-f057@gregkh' --subject-prefix 'PATCH 6.12.y' HEAD^.. Possible dependencies: thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 847748fc66d08a89135a74e29362a66ba4e3ab15 Mon Sep 17 00:00:00 2001 From: Hangbin Liu <liuhangbin(a)gmail.com> Date: Fri, 5 Sep 2025 09:15:33 +0000 Subject: [PATCH] hsr: hold rcu and dev lock for hsr_get_port_ndev hsr_get_port_ndev calls hsr_for_each_port, which need to hold rcu lock. On the other hand, before return the port device, we need to hold the device reference to avoid UaF in the caller function. Suggested-by: Paolo Abeni <pabeni(a)redhat.com> Fixes: 9c10dd8eed74 ("net: hsr: Create and export hsr_get_port_ndev()") Signed-off-by: Hangbin Liu <liuhangbin(a)gmail.com> Reviewed-by: Simon Horman <horms(a)kernel.org> Link: https://patch.msgid.link/20250905091533.377443-4-liuhangbin@gmail.com Signed-off-by: Paolo Abeni <pabeni(a)redhat.com> diff --git a/drivers/net/ethernet/ti/icssg/icssg_prueth.c b/drivers/net/ethernet/ti/icssg/icssg_prueth.c index dadce6009791..e42d0fdefee1 100644 --- a/drivers/net/ethernet/ti/icssg/icssg_prueth.c +++ b/drivers/net/ethernet/ti/icssg/icssg_prueth.c @@ -654,7 +654,7 @@ static void icssg_prueth_hsr_fdb_add_del(struct prueth_emac *emac, static int icssg_prueth_hsr_add_mcast(struct net_device *ndev, const u8 *addr) { - struct net_device *real_dev; + struct net_device *real_dev, *port_dev; struct prueth_emac *emac; u8 vlan_id, i; @@ -663,11 +663,15 @@ static int icssg_prueth_hsr_add_mcast(struct net_device *ndev, const u8 *addr) if (is_hsr_master(real_dev)) { for (i = HSR_PT_SLAVE_A; i < HSR_PT_INTERLINK; i++) { - emac = netdev_priv(hsr_get_port_ndev(real_dev, i)); - if (!emac) + port_dev = hsr_get_port_ndev(real_dev, i); + emac = netdev_priv(port_dev); + if (!emac) { + dev_put(port_dev); return -EINVAL; + } icssg_prueth_hsr_fdb_add_del(emac, addr, vlan_id, true); + dev_put(port_dev); } } else { emac = netdev_priv(real_dev); @@ -679,7 +683,7 @@ static int icssg_prueth_hsr_add_mcast(struct net_device *ndev, const u8 *addr) static int icssg_prueth_hsr_del_mcast(struct net_device *ndev, const u8 *addr) { - struct net_device *real_dev; + struct net_device *real_dev, *port_dev; struct prueth_emac *emac; u8 vlan_id, i; @@ -688,11 +692,15 @@ static int icssg_prueth_hsr_del_mcast(struct net_device *ndev, const u8 *addr) if (is_hsr_master(real_dev)) { for (i = HSR_PT_SLAVE_A; i < HSR_PT_INTERLINK; i++) { - emac = netdev_priv(hsr_get_port_ndev(real_dev, i)); - if (!emac) + port_dev = hsr_get_port_ndev(real_dev, i); + emac = netdev_priv(port_dev); + if (!emac) { + dev_put(port_dev); return -EINVAL; + } icssg_prueth_hsr_fdb_add_del(emac, addr, vlan_id, false); + dev_put(port_dev); } } else { emac = netdev_priv(real_dev); diff --git a/net/hsr/hsr_device.c b/net/hsr/hsr_device.c index 702da1f9aaa9..fbbc3ccf9df6 100644 --- a/net/hsr/hsr_device.c +++ b/net/hsr/hsr_device.c @@ -675,9 +675,14 @@ struct net_device *hsr_get_port_ndev(struct net_device *ndev, struct hsr_priv *hsr = netdev_priv(ndev); struct hsr_port *port; + rcu_read_lock(); hsr_for_each_port(hsr, port) - if (port->type == pt) + if (port->type == pt) { + dev_hold(port->dev); + rcu_read_unlock(); return port->dev; + } + rcu_read_unlock(); return NULL; } EXPORT_SYMBOL(hsr_get_port_ndev);

1 week, 5 days

3
2
0 0

[PATCH v2 3/5] usb: ohci-nxp: fix device leak on probe failure

by Johan Hovold

Make sure to drop the reference taken when looking up the PHY I2C device during probe on probe failure (e.g. probe deferral) and on driver unbind. Fixes: 73108aa90cbf ("USB: ohci-nxp: Use isp1301 driver") Cc: stable(a)vger.kernel.org # 3.5 Reported-by: Ma Ke <make24(a)iscas.ac.cn> Link: https://lore.kernel.org/lkml/20251117013428.21840-1-make24@iscas.ac.cn/ Signed-off-by: Johan Hovold <johan(a)kernel.org> --- drivers/usb/host/ohci-nxp.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/usb/host/ohci-nxp.c b/drivers/usb/host/ohci-nxp.c index 24d5a1dc5056..509ca7d8d513 100644 --- a/drivers/usb/host/ohci-nxp.c +++ b/drivers/usb/host/ohci-nxp.c @@ -223,6 +223,7 @@ static int ohci_hcd_nxp_probe(struct platform_device *pdev) fail_resource: usb_put_hcd(hcd); fail_disable: + put_device(&isp1301_i2c_client->dev); isp1301_i2c_client = NULL; return ret; } @@ -234,6 +235,7 @@ static void ohci_hcd_nxp_remove(struct platform_device *pdev) usb_remove_hcd(hcd); ohci_nxp_stop_hc(); usb_put_hcd(hcd); + put_device(&isp1301_i2c_client->dev); isp1301_i2c_client = NULL; } -- 2.51.2

1 week, 5 days

3
2
0 0

[PATCH v2 1/5] usb: gadget: lpc32xx_udc: fix clock imbalance in error path

by Johan Hovold

A recent change fixing a device reference leak introduced a clock imbalance by reusing an error path so that the clock may be disabled before having been enabled. Note that the clock framework allows for passing in NULL clocks so there is no risk for a NULL pointer dereference. Also drop the bogus I2C client NULL check added by the offending commit as the pointer has already been verified to be non-NULL. Fixes: c84117912bdd ("USB: lpc32xx_udc: Fix error handling in probe") Cc: stable(a)vger.kernel.org Cc: Ma Ke <make24(a)iscas.ac.cn> Signed-off-by: Johan Hovold <johan(a)kernel.org> --- drivers/usb/gadget/udc/lpc32xx_udc.c | 20 ++++++++++---------- 1 file changed, 10 insertions(+), 10 deletions(-) diff --git a/drivers/usb/gadget/udc/lpc32xx_udc.c b/drivers/usb/gadget/udc/lpc32xx_udc.c index 73c0f28a8585..a962d4294fbe 100644 --- a/drivers/usb/gadget/udc/lpc32xx_udc.c +++ b/drivers/usb/gadget/udc/lpc32xx_udc.c @@ -3020,7 +3020,7 @@ static int lpc32xx_udc_probe(struct platform_device *pdev) pdev->dev.dma_mask = &lpc32xx_usbd_dmamask; retval = dma_set_coherent_mask(&pdev->dev, DMA_BIT_MASK(32)); if (retval) - goto i2c_fail; + goto err_put_client; udc->board = &lpc32xx_usbddata; @@ -3040,7 +3040,7 @@ static int lpc32xx_udc_probe(struct platform_device *pdev) udc->udp_irq[i] = platform_get_irq(pdev, i); if (udc->udp_irq[i] < 0) { retval = udc->udp_irq[i]; - goto i2c_fail; + goto err_put_client; } } @@ -3048,7 +3048,7 @@ static int lpc32xx_udc_probe(struct platform_device *pdev) if (IS_ERR(udc->udp_baseaddr)) { dev_err(udc->dev, "IO map failure\n"); retval = PTR_ERR(udc->udp_baseaddr); - goto i2c_fail; + goto err_put_client; } /* Get USB device clock */ @@ -3056,14 +3056,14 @@ static int lpc32xx_udc_probe(struct platform_device *pdev) if (IS_ERR(udc->usb_slv_clk)) { dev_err(udc->dev, "failed to acquire USB device clock\n"); retval = PTR_ERR(udc->usb_slv_clk); - goto i2c_fail; + goto err_put_client; } /* Enable USB device clock */ retval = clk_prepare_enable(udc->usb_slv_clk); if (retval < 0) { dev_err(udc->dev, "failed to start USB device clock\n"); - goto i2c_fail; + goto err_put_client; } /* Setup deferred workqueue data */ @@ -3165,9 +3165,10 @@ static int lpc32xx_udc_probe(struct platform_device *pdev) dma_free_coherent(&pdev->dev, UDCA_BUFF_SIZE, udc->udca_v_base, udc->udca_p_base); i2c_fail: - if (udc->isp1301_i2c_client) - put_device(&udc->isp1301_i2c_client->dev); clk_disable_unprepare(udc->usb_slv_clk); +err_put_client: + put_device(&udc->isp1301_i2c_client->dev); + dev_err(udc->dev, "%s probe failed, %d\n", driver_name, retval); return retval; @@ -3195,10 +3196,9 @@ static void lpc32xx_udc_remove(struct platform_device *pdev) dma_free_coherent(&pdev->dev, UDCA_BUFF_SIZE, udc->udca_v_base, udc->udca_p_base); - if (udc->isp1301_i2c_client) - put_device(&udc->isp1301_i2c_client->dev); - clk_disable_unprepare(udc->usb_slv_clk); + + put_device(&udc->isp1301_i2c_client->dev); } #ifdef CONFIG_PM -- 2.51.2

1 week, 5 days

2
1
0 0

[PATCH v6] drm/pagemap, drm/xe: Ensure that the devmem allocation is idle before use

by Thomas Hellström

In situations where no system memory is migrated to devmem, and in upcoming patches where another GPU is performing the migration to the newly allocated devmem buffer, there is nothing to ensure any ongoing clear to the devmem allocation or async eviction from the devmem allocation is complete. Address that by passing a struct dma_fence down to the copy functions, and ensure it is waited for before migration is marked complete. v3: - New patch. v4: - Update the logic used for determining when to wait for the pre_migrate_fence. - Update the logic used for determining when to warn for the pre_migrate_fence since the scheduler fences apparently can signal out-of-order. v5: - Fix a UAF (CI) - Remove references to source P2P migration (Himal) - Put the pre_migrate_fence after migration. v6: - Pipeline the pre_migrate_fence dependency (Matt Brost) Fixes: c5b3eb5a906c ("drm/xe: Add GPUSVM device memory copy vfunc functions") Cc: Matthew Brost <matthew.brost(a)intel.com> Cc: <stable(a)vger.kernel.org> # v6.15+ Signed-off-by: Thomas Hellström <thomas.hellstrom(a)linux.intel.com> --- drivers/gpu/drm/drm_pagemap.c | 17 +++++++++--- drivers/gpu/drm/xe/xe_migrate.c | 25 +++++++++++++---- drivers/gpu/drm/xe/xe_migrate.h | 6 ++-- drivers/gpu/drm/xe/xe_svm.c | 49 +++++++++++++++++++++++++-------- include/drm/drm_pagemap.h | 17 ++++++++++-- 5 files changed, 88 insertions(+), 26 deletions(-) diff --git a/drivers/gpu/drm/drm_pagemap.c b/drivers/gpu/drm/drm_pagemap.c index 37d7cfbbb3e8..06c1bd8fc4d1 100644 --- a/drivers/gpu/drm/drm_pagemap.c +++ b/drivers/gpu/drm/drm_pagemap.c @@ -3,6 +3,7 @@ * Copyright © 2024-2025 Intel Corporation */ +#include <linux/dma-fence.h> #include <linux/dma-mapping.h> #include <linux/migrate.h> #include <linux/pagemap.h> @@ -408,10 +409,14 @@ int drm_pagemap_migrate_to_devmem(struct drm_pagemap_devmem *devmem_allocation, drm_pagemap_get_devmem_page(page, zdd); } - err = ops->copy_to_devmem(pages, pagemap_addr, npages); + err = ops->copy_to_devmem(pages, pagemap_addr, npages, + devmem_allocation->pre_migrate_fence); if (err) goto err_finalize; + dma_fence_put(devmem_allocation->pre_migrate_fence); + devmem_allocation->pre_migrate_fence = NULL; + /* Upon success bind devmem allocation to range and zdd */ devmem_allocation->timeslice_expiration = get_jiffies_64() + msecs_to_jiffies(timeslice_ms); @@ -596,7 +601,7 @@ int drm_pagemap_evict_to_ram(struct drm_pagemap_devmem *devmem_allocation) for (i = 0; i < npages; ++i) pages[i] = migrate_pfn_to_page(src[i]); - err = ops->copy_to_ram(pages, pagemap_addr, npages); + err = ops->copy_to_ram(pages, pagemap_addr, npages, NULL); if (err) goto err_finalize; @@ -732,7 +737,7 @@ static int __drm_pagemap_migrate_to_ram(struct vm_area_struct *vas, for (i = 0; i < npages; ++i) pages[i] = migrate_pfn_to_page(migrate.src[i]); - err = ops->copy_to_ram(pages, pagemap_addr, npages); + err = ops->copy_to_ram(pages, pagemap_addr, npages, NULL); if (err) goto err_finalize; @@ -813,11 +818,14 @@ EXPORT_SYMBOL_GPL(drm_pagemap_pagemap_ops_get); * @ops: Pointer to the operations structure for GPU SVM device memory * @dpagemap: The struct drm_pagemap we're allocating from. * @size: Size of device memory allocation + * @pre_migrate_fence: Fence to wait for or pipeline behind before migration starts. + * (May be NULL). */ void drm_pagemap_devmem_init(struct drm_pagemap_devmem *devmem_allocation, struct device *dev, struct mm_struct *mm, const struct drm_pagemap_devmem_ops *ops, - struct drm_pagemap *dpagemap, size_t size) + struct drm_pagemap *dpagemap, size_t size, + struct dma_fence *pre_migrate_fence) { init_completion(&devmem_allocation->detached); devmem_allocation->dev = dev; @@ -825,6 +833,7 @@ void drm_pagemap_devmem_init(struct drm_pagemap_devmem *devmem_allocation, devmem_allocation->ops = ops; devmem_allocation->dpagemap = dpagemap; devmem_allocation->size = size; + devmem_allocation->pre_migrate_fence = pre_migrate_fence; } EXPORT_SYMBOL_GPL(drm_pagemap_devmem_init); diff --git a/drivers/gpu/drm/xe/xe_migrate.c b/drivers/gpu/drm/xe/xe_migrate.c index f3b66b55acfb..284b575346ee 100644 --- a/drivers/gpu/drm/xe/xe_migrate.c +++ b/drivers/gpu/drm/xe/xe_migrate.c @@ -2113,6 +2113,7 @@ static struct dma_fence *xe_migrate_vram(struct xe_migrate *m, unsigned long sram_offset, struct drm_pagemap_addr *sram_addr, u64 vram_addr, + struct dma_fence *deps, const enum xe_migrate_copy_dir dir) { struct xe_gt *gt = m->tile->primary_gt; @@ -2201,6 +2202,14 @@ static struct dma_fence *xe_migrate_vram(struct xe_migrate *m, xe_sched_job_add_migrate_flush(job, MI_INVALIDATE_TLB); + if (deps && !dma_fence_is_signaled(deps)) { + dma_fence_get(deps); + err = drm_sched_job_add_dependency(&job->drm, deps); + if (err) + dma_fence_wait(deps, false); + err = 0; + } + mutex_lock(&m->job_mutex); xe_sched_job_arm(job); fence = dma_fence_get(&job->drm.s_fence->finished); @@ -2226,6 +2235,8 @@ static struct dma_fence *xe_migrate_vram(struct xe_migrate *m, * @npages: Number of pages to migrate. * @src_addr: Array of DMA information (source of migrate) * @dst_addr: Device physical address of VRAM (destination of migrate) + * @deps: struct dma_fence representing the dependencies that need + * to be signaled before migration. * * Copy from an array dma addresses to a VRAM device physical address * @@ -2235,10 +2246,11 @@ static struct dma_fence *xe_migrate_vram(struct xe_migrate *m, struct dma_fence *xe_migrate_to_vram(struct xe_migrate *m, unsigned long npages, struct drm_pagemap_addr *src_addr, - u64 dst_addr) + u64 dst_addr, + struct dma_fence *deps) { return xe_migrate_vram(m, npages * PAGE_SIZE, 0, src_addr, dst_addr, - XE_MIGRATE_COPY_TO_VRAM); + deps, XE_MIGRATE_COPY_TO_VRAM); } /** @@ -2247,6 +2259,8 @@ struct dma_fence *xe_migrate_to_vram(struct xe_migrate *m, * @npages: Number of pages to migrate. * @src_addr: Device physical address of VRAM (source of migrate) * @dst_addr: Array of DMA information (destination of migrate) + * @deps: struct dma_fence representing the dependencies that need + * to be signaled before migration. * * Copy from a VRAM device physical address to an array dma addresses * @@ -2256,10 +2270,11 @@ struct dma_fence *xe_migrate_to_vram(struct xe_migrate *m, struct dma_fence *xe_migrate_from_vram(struct xe_migrate *m, unsigned long npages, u64 src_addr, - struct drm_pagemap_addr *dst_addr) + struct drm_pagemap_addr *dst_addr, + struct dma_fence *deps) { return xe_migrate_vram(m, npages * PAGE_SIZE, 0, dst_addr, src_addr, - XE_MIGRATE_COPY_TO_SRAM); + deps, XE_MIGRATE_COPY_TO_SRAM); } static void xe_migrate_dma_unmap(struct xe_device *xe, @@ -2435,7 +2450,7 @@ int xe_migrate_access_memory(struct xe_migrate *m, struct xe_bo *bo, __fence = xe_migrate_vram(m, current_bytes, (unsigned long)buf & ~PAGE_MASK, &pagemap_addr[current_page], - vram_addr, write ? + vram_addr, NULL, write ? XE_MIGRATE_COPY_TO_VRAM : XE_MIGRATE_COPY_TO_SRAM); if (IS_ERR(__fence)) { diff --git a/drivers/gpu/drm/xe/xe_migrate.h b/drivers/gpu/drm/xe/xe_migrate.h index 464c05dde1ba..1522afb37dcf 100644 --- a/drivers/gpu/drm/xe/xe_migrate.h +++ b/drivers/gpu/drm/xe/xe_migrate.h @@ -116,12 +116,14 @@ int xe_migrate_init(struct xe_migrate *m); struct dma_fence *xe_migrate_to_vram(struct xe_migrate *m, unsigned long npages, struct drm_pagemap_addr *src_addr, - u64 dst_addr); + u64 dst_addr, + struct dma_fence *deps); struct dma_fence *xe_migrate_from_vram(struct xe_migrate *m, unsigned long npages, u64 src_addr, - struct drm_pagemap_addr *dst_addr); + struct drm_pagemap_addr *dst_addr, + struct dma_fence *deps); struct dma_fence *xe_migrate_copy(struct xe_migrate *m, struct xe_bo *src_bo, diff --git a/drivers/gpu/drm/xe/xe_svm.c b/drivers/gpu/drm/xe/xe_svm.c index 93550c7c84ac..a985148b6820 100644 --- a/drivers/gpu/drm/xe/xe_svm.c +++ b/drivers/gpu/drm/xe/xe_svm.c @@ -472,7 +472,8 @@ static void xe_svm_copy_us_stats_incr(struct xe_gt *gt, static int xe_svm_copy(struct page **pages, struct drm_pagemap_addr *pagemap_addr, - unsigned long npages, const enum xe_svm_copy_dir dir) + unsigned long npages, const enum xe_svm_copy_dir dir, + struct dma_fence *pre_migrate_fence) { struct xe_vram_region *vr = NULL; struct xe_gt *gt = NULL; @@ -561,7 +562,8 @@ static int xe_svm_copy(struct page **pages, __fence = xe_migrate_from_vram(vr->migrate, i - pos + incr, vram_addr, - &pagemap_addr[pos]); + &pagemap_addr[pos], + pre_migrate_fence); } else { vm_dbg(&xe->drm, "COPY TO VRAM - 0x%016llx -> 0x%016llx, NPAGES=%ld", @@ -570,13 +572,14 @@ static int xe_svm_copy(struct page **pages, __fence = xe_migrate_to_vram(vr->migrate, i - pos + incr, &pagemap_addr[pos], - vram_addr); + vram_addr, + pre_migrate_fence); } if (IS_ERR(__fence)) { err = PTR_ERR(__fence); goto err_out; } - + pre_migrate_fence = NULL; dma_fence_put(fence); fence = __fence; } @@ -599,20 +602,22 @@ static int xe_svm_copy(struct page **pages, vram_addr, (u64)pagemap_addr[pos].addr, 1); __fence = xe_migrate_from_vram(vr->migrate, 1, vram_addr, - &pagemap_addr[pos]); + &pagemap_addr[pos], + pre_migrate_fence); } else { vm_dbg(&xe->drm, "COPY TO VRAM - 0x%016llx -> 0x%016llx, NPAGES=%d", (u64)pagemap_addr[pos].addr, vram_addr, 1); __fence = xe_migrate_to_vram(vr->migrate, 1, &pagemap_addr[pos], - vram_addr); + vram_addr, + pre_migrate_fence); } if (IS_ERR(__fence)) { err = PTR_ERR(__fence); goto err_out; } - + pre_migrate_fence = NULL; dma_fence_put(fence); fence = __fence; } @@ -625,6 +630,8 @@ static int xe_svm_copy(struct page **pages, dma_fence_wait(fence, false); dma_fence_put(fence); } + if (pre_migrate_fence) + dma_fence_wait(pre_migrate_fence, false); /* * XXX: We can't derive the GT here (or anywhere in this functions, but @@ -641,16 +648,20 @@ static int xe_svm_copy(struct page **pages, static int xe_svm_copy_to_devmem(struct page **pages, struct drm_pagemap_addr *pagemap_addr, - unsigned long npages) + unsigned long npages, + struct dma_fence *pre_migrate_fence) { - return xe_svm_copy(pages, pagemap_addr, npages, XE_SVM_COPY_TO_VRAM); + return xe_svm_copy(pages, pagemap_addr, npages, XE_SVM_COPY_TO_VRAM, + pre_migrate_fence); } static int xe_svm_copy_to_ram(struct page **pages, struct drm_pagemap_addr *pagemap_addr, - unsigned long npages) + unsigned long npages, + struct dma_fence *pre_migrate_fence) { - return xe_svm_copy(pages, pagemap_addr, npages, XE_SVM_COPY_TO_SRAM); + return xe_svm_copy(pages, pagemap_addr, npages, XE_SVM_COPY_TO_SRAM, + pre_migrate_fence); } static struct xe_bo *to_xe_bo(struct drm_pagemap_devmem *devmem_allocation) @@ -663,6 +674,7 @@ static void xe_svm_devmem_release(struct drm_pagemap_devmem *devmem_allocation) struct xe_bo *bo = to_xe_bo(devmem_allocation); struct xe_device *xe = xe_bo_device(bo); + dma_fence_put(devmem_allocation->pre_migrate_fence); xe_bo_put_async(bo); xe_pm_runtime_put(xe); } @@ -857,6 +869,7 @@ static int xe_drm_pagemap_populate_mm(struct drm_pagemap *dpagemap, unsigned long timeslice_ms) { struct xe_vram_region *vr = container_of(dpagemap, typeof(*vr), dpagemap); + struct dma_fence *pre_migrate_fence = NULL; struct xe_device *xe = vr->xe; struct device *dev = xe->drm.dev; struct drm_buddy_block *block; @@ -883,8 +896,20 @@ static int xe_drm_pagemap_populate_mm(struct drm_pagemap *dpagemap, break; } + /* Ensure that any clearing or async eviction will complete before migration. */ + if (!dma_resv_test_signaled(bo->ttm.base.resv, DMA_RESV_USAGE_KERNEL)) { + err = dma_resv_get_singleton(bo->ttm.base.resv, DMA_RESV_USAGE_KERNEL, + &pre_migrate_fence); + if (err) + dma_resv_wait_timeout(bo->ttm.base.resv, DMA_RESV_USAGE_KERNEL, + false, MAX_SCHEDULE_TIMEOUT); + else if (pre_migrate_fence) + dma_fence_enable_sw_signaling(pre_migrate_fence); + } + drm_pagemap_devmem_init(&bo->devmem_allocation, dev, mm, - &dpagemap_devmem_ops, dpagemap, end - start); + &dpagemap_devmem_ops, dpagemap, end - start, + pre_migrate_fence); blocks = &to_xe_ttm_vram_mgr_resource(bo->ttm.resource)->blocks; list_for_each_entry(block, blocks, link) diff --git a/include/drm/drm_pagemap.h b/include/drm/drm_pagemap.h index f6e7e234c089..70a7991f784f 100644 --- a/include/drm/drm_pagemap.h +++ b/include/drm/drm_pagemap.h @@ -8,6 +8,7 @@ #define NR_PAGES(order) (1U << (order)) +struct dma_fence; struct drm_pagemap; struct drm_pagemap_zdd; struct device; @@ -174,6 +175,8 @@ struct drm_pagemap_devmem_ops { * @pages: Pointer to array of device memory pages (destination) * @pagemap_addr: Pointer to array of DMA information (source) * @npages: Number of pages to copy + * @pre_migrate_fence: dma-fence to wait for before migration start. + * May be NULL. * * Copy pages to device memory. If the order of a @pagemap_addr entry * is greater than 0, the entry is populated but subsequent entries @@ -183,13 +186,16 @@ struct drm_pagemap_devmem_ops { */ int (*copy_to_devmem)(struct page **pages, struct drm_pagemap_addr *pagemap_addr, - unsigned long npages); + unsigned long npages, + struct dma_fence *pre_migrate_fence); /** * @copy_to_ram: Copy to system RAM (required for migration) * @pages: Pointer to array of device memory pages (source) * @pagemap_addr: Pointer to array of DMA information (destination) * @npages: Number of pages to copy + * @pre_migrate_fence: dma-fence to wait for before migration start. + * May be NULL. * * Copy pages to system RAM. If the order of a @pagemap_addr entry * is greater than 0, the entry is populated but subsequent entries @@ -199,7 +205,8 @@ struct drm_pagemap_devmem_ops { */ int (*copy_to_ram)(struct page **pages, struct drm_pagemap_addr *pagemap_addr, - unsigned long npages); + unsigned long npages, + struct dma_fence *pre_migrate_fence); }; /** @@ -212,6 +219,8 @@ struct drm_pagemap_devmem_ops { * @dpagemap: The struct drm_pagemap of the pages this allocation belongs to. * @size: Size of device memory allocation * @timeslice_expiration: Timeslice expiration in jiffies + * @pre_migrate_fence: Fence to wait for or pipeline behind before migration starts. + * (May be NULL). */ struct drm_pagemap_devmem { struct device *dev; @@ -221,6 +230,7 @@ struct drm_pagemap_devmem { struct drm_pagemap *dpagemap; size_t size; u64 timeslice_expiration; + struct dma_fence *pre_migrate_fence; }; int drm_pagemap_migrate_to_devmem(struct drm_pagemap_devmem *devmem_allocation, @@ -238,7 +248,8 @@ struct drm_pagemap *drm_pagemap_page_to_dpagemap(struct page *page); void drm_pagemap_devmem_init(struct drm_pagemap_devmem *devmem_allocation, struct device *dev, struct mm_struct *mm, const struct drm_pagemap_devmem_ops *ops, - struct drm_pagemap *dpagemap, size_t size); + struct drm_pagemap *dpagemap, size_t size, + struct dma_fence *pre_migrate_fence); int drm_pagemap_populate_mm(struct drm_pagemap *dpagemap, unsigned long start, unsigned long end, -- 2.51.1

1 week, 5 days

2
1
0 0

[to-be-updated] mm-memory-failure-fix-missing-mf_stats-count-in-hugetlb-poison.patch removed from -mm tree

by Andrew Morton

The quilt patch titled Subject: mm/memory-failure: fix missing ->mf_stats count in hugetlb poison has been removed from the -mm tree. Its filename was mm-memory-failure-fix-missing-mf_stats-count-in-hugetlb-poison.patch This patch was dropped because an updated version will be issued ------------------------------------------------------ From: Jane Chu <jane.chu(a)oracle.com> Subject: mm/memory-failure: fix missing ->mf_stats count in hugetlb poison Date: Tue, 16 Dec 2025 14:56:21 -0700 When a newly poisoned subpage ends up in an already poisoned hugetlb folio, 'num_poisoned_pages' is incremented, but the per node ->mf_stats is not. Fix the inconsistency by designating action_result() to update them both. Link: https://lkml.kernel.org/r/20251216215621.920093-1-jane.chu@oracle.com Fixes: 18f41fa616ee ("mm: memory-failure: bump memory failure stats to pglist_data") Signed-off-by: Jane Chu <jane.chu(a)oracle.com> Cc: David Rientjes <rientjes(a)google.com> Cc: Jiaqi Yan <jiaqiyan(a)google.com> Cc: Liam Howlett <liam.howlett(a)oracle.com> Cc: Lorenzo Stoakes <lorenzo.stoakes(a)oracle.com> Cc: Miaohe Lin <linmiaohe(a)huawei.com> Cc: Michal Hocko <mhocko(a)suse.com> Cc: Mike Rapoport <rppt(a)kernel.org> Cc: Muchun Song <muchun.song(a)linux.dev> Cc: Oscar Salvador <osalvador(a)suse.de> Cc: Suren Baghdasaryan <surenb(a)google.com> Cc: William Roche <william.roche(a)oracle.com> Cc: Liam Howlett <liam.howlett(a)oracle.com> Cc: "David Hildenbrand (Red Hat)" <david(a)kernel.org> Cc: <stable(a)vger.kernel.org> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> --- include/linux/hugetlb.h | 4 ++-- include/linux/mm.h | 4 ++-- mm/hugetlb.c | 4 ++-- mm/memory-failure.c | 22 +++++++++++++--------- 4 files changed, 19 insertions(+), 15 deletions(-) --- a/include/linux/hugetlb.h~mm-memory-failure-fix-missing-mf_stats-count-in-hugetlb-poison +++ a/include/linux/hugetlb.h @@ -156,7 +156,7 @@ long hugetlb_unreserve_pages(struct inod bool folio_isolate_hugetlb(struct folio *folio, struct list_head *list); int get_hwpoison_hugetlb_folio(struct folio *folio, bool *hugetlb, bool unpoison); int get_huge_page_for_hwpoison(unsigned long pfn, int flags, - bool *migratable_cleared); + bool *migratable_cleared, bool *samepg); void folio_putback_hugetlb(struct folio *folio); void move_hugetlb_state(struct folio *old_folio, struct folio *new_folio, int reason); void hugetlb_fix_reserve_counts(struct inode *inode); @@ -418,7 +418,7 @@ static inline int get_hwpoison_hugetlb_f } static inline int get_huge_page_for_hwpoison(unsigned long pfn, int flags, - bool *migratable_cleared) + bool *migratable_cleared, bool *samepg) { return 0; } --- a/include/linux/mm.h~mm-memory-failure-fix-missing-mf_stats-count-in-hugetlb-poison +++ a/include/linux/mm.h @@ -4351,7 +4351,7 @@ extern int soft_offline_page(unsigned lo extern const struct attribute_group memory_failure_attr_group; extern void memory_failure_queue(unsigned long pfn, int flags); extern int __get_huge_page_for_hwpoison(unsigned long pfn, int flags, - bool *migratable_cleared); + bool *migratable_cleared, bool *samepg); void num_poisoned_pages_inc(unsigned long pfn); void num_poisoned_pages_sub(unsigned long pfn, long i); #else @@ -4360,7 +4360,7 @@ static inline void memory_failure_queue( } static inline int __get_huge_page_for_hwpoison(unsigned long pfn, int flags, - bool *migratable_cleared) + bool *migratable_cleared, bool *samepg) { return 0; } --- a/mm/hugetlb.c~mm-memory-failure-fix-missing-mf_stats-count-in-hugetlb-poison +++ a/mm/hugetlb.c @@ -7132,12 +7132,12 @@ int get_hwpoison_hugetlb_folio(struct fo } int get_huge_page_for_hwpoison(unsigned long pfn, int flags, - bool *migratable_cleared) + bool *migratable_cleared, bool *samepg) { int ret; spin_lock_irq(&hugetlb_lock); - ret = __get_huge_page_for_hwpoison(pfn, flags, migratable_cleared); + ret = __get_huge_page_for_hwpoison(pfn, flags, migratable_cleared, samepg); spin_unlock_irq(&hugetlb_lock); return ret; } --- a/mm/memory-failure.c~mm-memory-failure-fix-missing-mf_stats-count-in-hugetlb-poison +++ a/mm/memory-failure.c @@ -1883,7 +1883,8 @@ static unsigned long __folio_free_raw_hw return count; } -static int folio_set_hugetlb_hwpoison(struct folio *folio, struct page *page) +static int folio_set_hugetlb_hwpoison(struct folio *folio, struct page *page, + bool *samepg) { struct llist_head *head; struct raw_hwp_page *raw_hwp; @@ -1899,17 +1900,16 @@ static int folio_set_hugetlb_hwpoison(st return -EHWPOISON; head = raw_hwp_list_head(folio); llist_for_each_entry(p, head->first, node) { - if (p->page == page) + if (p->page == page) { + *samepg = true; return -EHWPOISON; + } } raw_hwp = kmalloc(sizeof(struct raw_hwp_page), GFP_ATOMIC); if (raw_hwp) { raw_hwp->page = page; llist_add(&raw_hwp->node, head); - /* the first error event will be counted in action_result(). */ - if (ret) - num_poisoned_pages_inc(page_to_pfn(page)); } else { /* * Failed to save raw error info. We no longer trace all @@ -1966,7 +1966,7 @@ void folio_clear_hugetlb_hwpoison(struct * -EHWPOISON - the hugepage is already hwpoisoned */ int __get_huge_page_for_hwpoison(unsigned long pfn, int flags, - bool *migratable_cleared) + bool *migratable_cleared, bool *samepg) { struct page *page = pfn_to_page(pfn); struct folio *folio = page_folio(page); @@ -1991,7 +1991,7 @@ int __get_huge_page_for_hwpoison(unsigne goto out; } - if (folio_set_hugetlb_hwpoison(folio, page)) { + if (folio_set_hugetlb_hwpoison(folio, page, samepg)) { ret = -EHWPOISON; goto out; } @@ -2024,11 +2024,12 @@ static int try_memory_failure_hugetlb(un struct page *p = pfn_to_page(pfn); struct folio *folio; unsigned long page_flags; + bool samepg = false; bool migratable_cleared = false; *hugetlb = 1; retry: - res = get_huge_page_for_hwpoison(pfn, flags, &migratable_cleared); + res = get_huge_page_for_hwpoison(pfn, flags, &migratable_cleared, &samepg); if (res == 2) { /* fallback to normal page handling */ *hugetlb = 0; return 0; @@ -2037,7 +2038,10 @@ retry: folio = page_folio(p); res = kill_accessing_process(current, folio_pfn(folio), flags); } - action_result(pfn, MF_MSG_ALREADY_POISONED, MF_FAILED); + if (samepg) + action_result(pfn, MF_MSG_ALREADY_POISONED, MF_FAILED); + else + action_result(pfn, MF_MSG_HUGE, MF_FAILED); return res; } else if (res == -EBUSY) { if (!(flags & MF_NO_RETRY)) { _ Patches currently in -mm which might be from jane.chu(a)oracle.com are

1 week, 5 days

1
0
0 0

+ lib-buildid-use-__kernel_read-for-sleepable-context.patch added to mm-nonmm-unstable branch

by Andrew Morton

The patch titled Subject: lib/buildid: use __kernel_read() for sleepable context has been added to the -mm mm-nonmm-unstable branch. Its filename is lib-buildid-use-__kernel_read-for-sleepable-context.patch This patch will shortly appear at https://git.kernel.org/pub/scm/linux/kernel/git/akpm/25-new.git/tree/patche… This patch will later appear in the mm-nonmm-unstable branch at git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm Before you just go and hit "reply", please: a) Consider who else should be cc'ed b) Prefer to cc a suitable mailing list as well c) Ideally: find the original patch on the mailing list and do a reply-to-all to that, adding suitable additional cc's *** Remember to use Documentation/process/submit-checklist.rst when testing your code *** The -mm tree is included into linux-next via various branches at git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm and is updated there most days ------------------------------------------------------ From: Shakeel Butt <shakeel.butt(a)linux.dev> Subject: lib/buildid: use __kernel_read() for sleepable context Date: Thu, 18 Dec 2025 12:55:05 -0800 For the sleepable context, convert freader to use __kernel_read() instead of direct page cache access via read_cache_folio(). This simplifies the faultable code path by using the standard kernel file reading interface which handles all the complexity of reading file data. At the moment we are not changing the code for non-sleepable context which uses filemap_get_folio() and only succeeds if the target folios are already in memory and up-to-date. The reason is to keep the patch simple and easier to backport to stable kernels. Syzbot repro does not crash the kernel anymore and the selftests run successfully. In the follow up we will make __kernel_read() with IOCB_NOWAIT work for non-sleepable contexts. In addition, I would like to replace the secretmem check with a more generic approach and will add fstest for the buildid code. Link: https://lkml.kernel.org/r/20251218205505.2415840-1-shakeel.butt@linux.dev Fixes: ad41251c290d ("lib/buildid: implement sleepable build_id_parse() API") Signed-off-by: Shakeel Butt <shakeel.butt(a)linux.dev>Reported-by: syzbot+09b7d050e4806540153d(a)syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=09b7d050e4806540153d Reviewed-by: Christoph Hellwig <hch(a)lst.de> Cc: Alexei Starovoitov <ast(a)kernel.org> Cc: Andrii Nakryiko <andrii(a)kernel.org> Cc: Daniel Borkman <daniel(a)iogearbox.net> Cc: "Darrick J. Wong" <djwong(a)kernel.org> Cc: Matthew Wilcox (Oracle) <willy(a)infradead.org> Cc: <stable(a)vger.kernel.org> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> --- lib/buildid.c | 47 +++++++++++++++++++++++++++++++++++------------ 1 file changed, 35 insertions(+), 12 deletions(-) --- a/lib/buildid.c~lib-buildid-use-__kernel_read-for-sleepable-context +++ a/lib/buildid.c @@ -5,6 +5,7 @@ #include <linux/elf.h> #include <linux/kernel.h> #include <linux/pagemap.h> +#include <linux/fs.h> #include <linux/secretmem.h> #define BUILD_ID 3 @@ -37,6 +38,29 @@ static void freader_put_folio(struct fre r->folio = NULL; } +/* + * Data is read directly into r->buf. Returns pointer to the buffer + * on success, NULL on failure with r->err set. + */ +static const void *freader_fetch_sync(struct freader *r, loff_t file_off, size_t sz) +{ + ssize_t ret; + loff_t pos = file_off; + char *buf = r->buf; + + do { + ret = __kernel_read(r->file, buf, sz, &pos); + if (ret <= 0) { + r->err = ret ?: -EIO; + return NULL; + } + buf += ret; + sz -= ret; + } while (sz > 0); + + return r->buf; +} + static int freader_get_folio(struct freader *r, loff_t file_off) { /* check if we can just reuse current folio */ @@ -46,20 +70,9 @@ static int freader_get_folio(struct frea freader_put_folio(r); - /* reject secretmem folios created with memfd_secret() */ - if (secretmem_mapping(r->file->f_mapping)) - return -EFAULT; - + /* only use page cache lookup - fail if not already cached */ r->folio = filemap_get_folio(r->file->f_mapping, file_off >> PAGE_SHIFT); - /* if sleeping is allowed, wait for the page, if necessary */ - if (r->may_fault && (IS_ERR(r->folio) || !folio_test_uptodate(r->folio))) { - filemap_invalidate_lock_shared(r->file->f_mapping); - r->folio = read_cache_folio(r->file->f_mapping, file_off >> PAGE_SHIFT, - NULL, r->file); - filemap_invalidate_unlock_shared(r->file->f_mapping); - } - if (IS_ERR(r->folio) || !folio_test_uptodate(r->folio)) { if (!IS_ERR(r->folio)) folio_put(r->folio); @@ -97,6 +110,16 @@ const void *freader_fetch(struct freader return r->data + file_off; } + /* reject secretmem folios created with memfd_secret() */ + if (secretmem_mapping(r->file->f_mapping)) { + r->err = -EFAULT; + return NULL; + } + + /* use __kernel_read() for sleepable context */ + if (r->may_fault) + return freader_fetch_sync(r, file_off, sz); + /* fetch or reuse folio for given file offset */ r->err = freader_get_folio(r, file_off); if (r->err) _ Patches currently in -mm which might be from shakeel.butt(a)linux.dev are mm-memcg-fix-unit-conversion-for-k-macro-in-oom-log.patch lib-buildid-use-__kernel_read-for-sleepable-context.patch

1 week, 5 days

1
0
0 0

[PATCH 1/2] af_unix: don't post cmsg for SO_INQ unless explicitly asked for

by Jens Axboe

A previous commit added SO_INQ support for AF_UNIX (SOCK_STREAM), but it posts a SCM_INQ cmsg even if just msg->msg_get_inq is set. This is incorrect, as ->msg_get_inq is just the caller asking for the remainder to be passed back in msg->msg_inq, it has nothing to do with cmsg. The original commit states that this is done to make sockets io_uring-friendly", but it's actually incorrect as io_uring doesn't use cmsg headers internally at all, and it's actively wrong as this means that cmsg's are always posted if someone does recvmsg via io_uring. Fix that up by only posting cmsg if u->recvmsg_inq is set. Cc: stable(a)vger.kernel.org Fixes: df30285b3670 ("af_unix: Introduce SO_INQ.") Reported-by: Julian Orth <ju.orth(a)gmail.com> Link: https://github.com/axboe/liburing/issues/1509 Signed-off-by: Jens Axboe <axboe(a)kernel.dk> --- net/unix/af_unix.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index 55cdebfa0da0..110d716087b5 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -3086,12 +3086,16 @@ static int unix_stream_read_generic(struct unix_stream_read_state *state, mutex_unlock(&u->iolock); if (msg) { + bool do_cmsg; + scm_recv_unix(sock, msg, &scm, flags); - if (READ_ONCE(u->recvmsg_inq) || msg->msg_get_inq) { + do_cmsg = READ_ONCE(u->recvmsg_inq); + if (do_cmsg || msg->msg_get_inq) { msg->msg_inq = READ_ONCE(u->inq_len); - put_cmsg(msg, SOL_SOCKET, SCM_INQ, - sizeof(msg->msg_inq), &msg->msg_inq); + if (do_cmsg) + put_cmsg(msg, SOL_SOCKET, SCM_INQ, + sizeof(msg->msg_inq), &msg->msg_inq); } } else { scm_destroy(&scm); -- 2.51.0

1 week, 5 days

2
4
0 0

[PATCH v5 03/24] drm/pagemap, drm/xe: Ensure that the devmem allocation is idle before use

by Thomas Hellström

In situations where no system memory is migrated to devmem, and in upcoming patches where another GPU is performing the migration to the newly allocated devmem buffer, there is nothing to ensure any ongoing clear to the devmem allocation or async eviction from the devmem allocation is complete. Address that by passing a struct dma_fence down to the copy functions, and ensure it is waited for before migration is marked complete. v3: - New patch. v4: - Update the logic used for determining when to wait for the pre_migrate_fence. - Update the logic used for determining when to warn for the pre_migrate_fence since the scheduler fences apparently can signal out-of-order. v5: - Fix a UAF (CI) - Remove references to source P2P migration (Himal) - Put the pre_migrate_fence after migration. Fixes: c5b3eb5a906c ("drm/xe: Add GPUSVM device memory copy vfunc functions") Cc: Matthew Brost <matthew.brost(a)intel.com> Cc: <stable(a)vger.kernel.org> # v6.15+ Signed-off-by: Thomas Hellström <thomas.hellstrom(a)linux.intel.com> --- drivers/gpu/drm/drm_pagemap.c | 17 ++++++--- drivers/gpu/drm/xe/xe_svm.c | 65 ++++++++++++++++++++++++++++++----- include/drm/drm_pagemap.h | 17 +++++++-- 3 files changed, 83 insertions(+), 16 deletions(-) diff --git a/drivers/gpu/drm/drm_pagemap.c b/drivers/gpu/drm/drm_pagemap.c index 4cf8f54e5a27..ac3832f85190 100644 --- a/drivers/gpu/drm/drm_pagemap.c +++ b/drivers/gpu/drm/drm_pagemap.c @@ -3,6 +3,7 @@ * Copyright © 2024-2025 Intel Corporation */ +#include <linux/dma-fence.h> #include <linux/dma-mapping.h> #include <linux/migrate.h> #include <linux/pagemap.h> @@ -408,10 +409,14 @@ int drm_pagemap_migrate_to_devmem(struct drm_pagemap_devmem *devmem_allocation, drm_pagemap_get_devmem_page(page, zdd); } - err = ops->copy_to_devmem(pages, pagemap_addr, npages); + err = ops->copy_to_devmem(pages, pagemap_addr, npages, + devmem_allocation->pre_migrate_fence); if (err) goto err_finalize; + dma_fence_put(devmem_allocation->pre_migrate_fence); + devmem_allocation->pre_migrate_fence = NULL; + /* Upon success bind devmem allocation to range and zdd */ devmem_allocation->timeslice_expiration = get_jiffies_64() + msecs_to_jiffies(timeslice_ms); @@ -596,7 +601,7 @@ int drm_pagemap_evict_to_ram(struct drm_pagemap_devmem *devmem_allocation) for (i = 0; i < npages; ++i) pages[i] = migrate_pfn_to_page(src[i]); - err = ops->copy_to_ram(pages, pagemap_addr, npages); + err = ops->copy_to_ram(pages, pagemap_addr, npages, NULL); if (err) goto err_finalize; @@ -719,7 +724,7 @@ static int __drm_pagemap_migrate_to_ram(struct vm_area_struct *vas, for (i = 0; i < npages; ++i) pages[i] = migrate_pfn_to_page(migrate.src[i]); - err = ops->copy_to_ram(pages, pagemap_addr, npages); + err = ops->copy_to_ram(pages, pagemap_addr, npages, NULL); if (err) goto err_finalize; @@ -800,11 +805,14 @@ EXPORT_SYMBOL_GPL(drm_pagemap_pagemap_ops_get); * @ops: Pointer to the operations structure for GPU SVM device memory * @dpagemap: The struct drm_pagemap we're allocating from. * @size: Size of device memory allocation + * @pre_migrate_fence: Fence to wait for or pipeline behind before migration starts. + * (May be NULL). */ void drm_pagemap_devmem_init(struct drm_pagemap_devmem *devmem_allocation, struct device *dev, struct mm_struct *mm, const struct drm_pagemap_devmem_ops *ops, - struct drm_pagemap *dpagemap, size_t size) + struct drm_pagemap *dpagemap, size_t size, + struct dma_fence *pre_migrate_fence) { init_completion(&devmem_allocation->detached); devmem_allocation->dev = dev; @@ -812,6 +820,7 @@ void drm_pagemap_devmem_init(struct drm_pagemap_devmem *devmem_allocation, devmem_allocation->ops = ops; devmem_allocation->dpagemap = dpagemap; devmem_allocation->size = size; + devmem_allocation->pre_migrate_fence = pre_migrate_fence; } EXPORT_SYMBOL_GPL(drm_pagemap_devmem_init); diff --git a/drivers/gpu/drm/xe/xe_svm.c b/drivers/gpu/drm/xe/xe_svm.c index bab8e6cbe53d..b806a1fce188 100644 --- a/drivers/gpu/drm/xe/xe_svm.c +++ b/drivers/gpu/drm/xe/xe_svm.c @@ -472,11 +472,12 @@ static void xe_svm_copy_us_stats_incr(struct xe_gt *gt, static int xe_svm_copy(struct page **pages, struct drm_pagemap_addr *pagemap_addr, - unsigned long npages, const enum xe_svm_copy_dir dir) + unsigned long npages, const enum xe_svm_copy_dir dir, + struct dma_fence *pre_migrate_fence) { struct xe_vram_region *vr = NULL; struct xe_gt *gt = NULL; - struct xe_device *xe; + struct xe_device *xe = NULL; struct dma_fence *fence = NULL; unsigned long i; #define XE_VRAM_ADDR_INVALID ~0x0ull @@ -485,6 +486,16 @@ static int xe_svm_copy(struct page **pages, bool sram = dir == XE_SVM_COPY_TO_SRAM; ktime_t start = xe_gt_stats_ktime_get(); + if (pre_migrate_fence && dma_fence_is_container(pre_migrate_fence)) { + /* + * This would typically be a composite fence operation on the destination memory. + * Ensure that the other GPU operation on the destination is complete. + */ + err = dma_fence_wait(pre_migrate_fence, true); + if (err) + return err; + } + /* * This flow is complex: it locates physically contiguous device pages, * derives the starting physical address, and performs a single GPU copy @@ -621,10 +632,28 @@ static int xe_svm_copy(struct page **pages, err_out: /* Wait for all copies to complete */ - if (fence) { + if (fence) dma_fence_wait(fence, false); - dma_fence_put(fence); + + /* + * If migrating to devmem, we should have pipelined the migration behind + * the pre_migrate_fence. Verify that this is indeed likely. If we + * didn't perform any copying, just wait for the pre_migrate_fence. + */ + if (pre_migrate_fence && !dma_fence_is_signaled(pre_migrate_fence)) { + if (xe && fence && + (pre_migrate_fence->context != fence->context || + dma_fence_is_later(pre_migrate_fence, fence))) { + drm_WARN(&xe->drm, true, "Unsignaled pre-migrate fence"); + drm_warn(&xe->drm, "fence contexts: %llu %llu. container %d\n", + (unsigned long long)fence->context, + (unsigned long long)pre_migrate_fence->context, + dma_fence_is_container(pre_migrate_fence)); + } + + dma_fence_wait(pre_migrate_fence, false); } + dma_fence_put(fence); /* * XXX: We can't derive the GT here (or anywhere in this functions, but @@ -641,16 +670,20 @@ static int xe_svm_copy(struct page **pages, static int xe_svm_copy_to_devmem(struct page **pages, struct drm_pagemap_addr *pagemap_addr, - unsigned long npages) + unsigned long npages, + struct dma_fence *pre_migrate_fence) { - return xe_svm_copy(pages, pagemap_addr, npages, XE_SVM_COPY_TO_VRAM); + return xe_svm_copy(pages, pagemap_addr, npages, XE_SVM_COPY_TO_VRAM, + pre_migrate_fence); } static int xe_svm_copy_to_ram(struct page **pages, struct drm_pagemap_addr *pagemap_addr, - unsigned long npages) + unsigned long npages, + struct dma_fence *pre_migrate_fence) { - return xe_svm_copy(pages, pagemap_addr, npages, XE_SVM_COPY_TO_SRAM); + return xe_svm_copy(pages, pagemap_addr, npages, XE_SVM_COPY_TO_SRAM, + pre_migrate_fence); } static struct xe_bo *to_xe_bo(struct drm_pagemap_devmem *devmem_allocation) @@ -663,6 +696,7 @@ static void xe_svm_devmem_release(struct drm_pagemap_devmem *devmem_allocation) struct xe_bo *bo = to_xe_bo(devmem_allocation); struct xe_device *xe = xe_bo_device(bo); + dma_fence_put(devmem_allocation->pre_migrate_fence); xe_bo_put_async(bo); xe_pm_runtime_put(xe); } @@ -857,6 +891,7 @@ static int xe_drm_pagemap_populate_mm(struct drm_pagemap *dpagemap, unsigned long timeslice_ms) { struct xe_vram_region *vr = container_of(dpagemap, typeof(*vr), dpagemap); + struct dma_fence *pre_migrate_fence = NULL; struct xe_device *xe = vr->xe; struct device *dev = xe->drm.dev; struct drm_buddy_block *block; @@ -883,8 +918,20 @@ static int xe_drm_pagemap_populate_mm(struct drm_pagemap *dpagemap, break; } + /* Ensure that any clearing or async eviction will complete before migration. */ + if (!dma_resv_test_signaled(bo->ttm.base.resv, DMA_RESV_USAGE_KERNEL)) { + err = dma_resv_get_singleton(bo->ttm.base.resv, DMA_RESV_USAGE_KERNEL, + &pre_migrate_fence); + if (err) + dma_resv_wait_timeout(bo->ttm.base.resv, DMA_RESV_USAGE_KERNEL, + false, MAX_SCHEDULE_TIMEOUT); + else if (pre_migrate_fence) + dma_fence_enable_sw_signaling(pre_migrate_fence); + } + drm_pagemap_devmem_init(&bo->devmem_allocation, dev, mm, - &dpagemap_devmem_ops, dpagemap, end - start); + &dpagemap_devmem_ops, dpagemap, end - start, + pre_migrate_fence); blocks = &to_xe_ttm_vram_mgr_resource(bo->ttm.resource)->blocks; list_for_each_entry(block, blocks, link) diff --git a/include/drm/drm_pagemap.h b/include/drm/drm_pagemap.h index f6e7e234c089..70a7991f784f 100644 --- a/include/drm/drm_pagemap.h +++ b/include/drm/drm_pagemap.h @@ -8,6 +8,7 @@ #define NR_PAGES(order) (1U << (order)) +struct dma_fence; struct drm_pagemap; struct drm_pagemap_zdd; struct device; @@ -174,6 +175,8 @@ struct drm_pagemap_devmem_ops { * @pages: Pointer to array of device memory pages (destination) * @pagemap_addr: Pointer to array of DMA information (source) * @npages: Number of pages to copy + * @pre_migrate_fence: dma-fence to wait for before migration start. + * May be NULL. * * Copy pages to device memory. If the order of a @pagemap_addr entry * is greater than 0, the entry is populated but subsequent entries @@ -183,13 +186,16 @@ struct drm_pagemap_devmem_ops { */ int (*copy_to_devmem)(struct page **pages, struct drm_pagemap_addr *pagemap_addr, - unsigned long npages); + unsigned long npages, + struct dma_fence *pre_migrate_fence); /** * @copy_to_ram: Copy to system RAM (required for migration) * @pages: Pointer to array of device memory pages (source) * @pagemap_addr: Pointer to array of DMA information (destination) * @npages: Number of pages to copy + * @pre_migrate_fence: dma-fence to wait for before migration start. + * May be NULL. * * Copy pages to system RAM. If the order of a @pagemap_addr entry * is greater than 0, the entry is populated but subsequent entries @@ -199,7 +205,8 @@ struct drm_pagemap_devmem_ops { */ int (*copy_to_ram)(struct page **pages, struct drm_pagemap_addr *pagemap_addr, - unsigned long npages); + unsigned long npages, + struct dma_fence *pre_migrate_fence); }; /** @@ -212,6 +219,8 @@ struct drm_pagemap_devmem_ops { * @dpagemap: The struct drm_pagemap of the pages this allocation belongs to. * @size: Size of device memory allocation * @timeslice_expiration: Timeslice expiration in jiffies + * @pre_migrate_fence: Fence to wait for or pipeline behind before migration starts. + * (May be NULL). */ struct drm_pagemap_devmem { struct device *dev; @@ -221,6 +230,7 @@ struct drm_pagemap_devmem { struct drm_pagemap *dpagemap; size_t size; u64 timeslice_expiration; + struct dma_fence *pre_migrate_fence; }; int drm_pagemap_migrate_to_devmem(struct drm_pagemap_devmem *devmem_allocation, @@ -238,7 +248,8 @@ struct drm_pagemap *drm_pagemap_page_to_dpagemap(struct page *page); void drm_pagemap_devmem_init(struct drm_pagemap_devmem *devmem_allocation, struct device *dev, struct mm_struct *mm, const struct drm_pagemap_devmem_ops *ops, - struct drm_pagemap *dpagemap, size_t size); + struct drm_pagemap *dpagemap, size_t size, + struct dma_fence *pre_migrate_fence); int drm_pagemap_populate_mm(struct drm_pagemap *dpagemap, unsigned long start, unsigned long end, -- 2.51.1

1 week, 5 days

2
3
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror December 2025