February 2025 - Linux-stable-mirror

[PATCH 2/2] usb: xhci: Fix unassigned variable 'bcdUSB' in xhci_create_usb3x_bos_desc()

by Selvarasu Ganesan

Fix the following smatch error: drivers/usb/host/xhci-hub.c:71 xhci_create_usb3x_bos_desc() error: unassigned variable 'bcdUSB' Fixes: eb02aaf21f29 ("usb: xhci: Rewrite xhci_create_usb3_bos_desc()") Cc: stable(a)vger.kernel.org Signed-off-by: Selvarasu Ganesan <selvarasu.g(a)samsung.com> --- drivers/usb/host/xhci-hub.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/usb/host/xhci-hub.c b/drivers/usb/host/xhci-hub.c index 9693464c0520..5715a8bdda7f 100644 --- a/drivers/usb/host/xhci-hub.c +++ b/drivers/usb/host/xhci-hub.c @@ -39,7 +39,7 @@ static int xhci_create_usb3x_bos_desc(struct xhci_hcd *xhci, char *buf, struct usb_ss_cap_descriptor *ss_cap; struct usb_ssp_cap_descriptor *ssp_cap; struct xhci_port_cap *port_cap = NULL; - u16 bcdUSB; + u16 bcdUSB = 0; u32 reg; u32 min_rate = 0; u8 min_ssid; -- 2.17.1

4 months, 2 weeks

1
0
0 0

[PATCH 1/2] usb: xhci: Fix unassigned variable 'tmp_minor_revision' in xhci_add_in_port()

by Selvarasu Ganesan

Fix the following smatch error: drivers/usb/host/xhci-mem.c:2060 xhci_add_in_port() error: unassigned variable 'tmp_minor_revision' Fixes: d9b0328d0b8b ("xhci: Show ZHAOXIN xHCI root hub speed correctly") Cc: stable(a)vger.kernel.org Signed-off-by: Selvarasu Ganesan <selvarasu.g(a)samsung.com> --- drivers/usb/host/xhci-mem.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/usb/host/xhci-mem.c b/drivers/usb/host/xhci-mem.c index 92703efda1f7..8665893df894 100644 --- a/drivers/usb/host/xhci-mem.c +++ b/drivers/usb/host/xhci-mem.c @@ -1980,7 +1980,7 @@ static void xhci_add_in_port(struct xhci_hcd *xhci, unsigned int num_ports, { u32 temp, port_offset, port_count; int i; - u8 major_revision, minor_revision, tmp_minor_revision; + u8 major_revision, minor_revision, tmp_minor_revision = 0; struct xhci_hub *rhub; struct device *dev = xhci_to_hcd(xhci)->self.sysdev; struct xhci_port_cap *port_cap; -- 2.17.1

4 months, 2 weeks

1
0
0 0

[PATCH 1/2] drm/i915/gvt: Store ->kvm reference in intel_vgpu struct.

by Andrey Ryabinin

'vfio_device' keeps the ->kvm pointer with elevated counter from the first open of the device up until the last close(). So the kvm struct and its dependencies (kvm kthreads, cgroups ...) kept alive even for VFIO device that don't need ->kvm. Copy ->kvm pointer from the vfio_device struct and store it in the 'intel_vgpu'. Note that kvm_page_track_[un]register_notifier() already does get/put calls, keeping the kvm struct alive. This will allow to release ->kvm from the vfio_device righ after the first open call, so that devices not using kvm not keeping it alive. Devices that are using kvm (like intel_vgpu) will be expected to mange the lifetime of the kvm struct by themselves. Fixes: 2b48f52f2bff ("vfio: fix deadlock between group lock and kvm lock") Cc: <stable(a)vger.kernel.org> Signed-off-by: Andrey Ryabinin <arbn(a)yandex-team.com> --- drivers/gpu/drm/i915/gvt/gvt.h | 1 + drivers/gpu/drm/i915/gvt/kvmgt.c | 14 +++++++------- 2 files changed, 8 insertions(+), 7 deletions(-) diff --git a/drivers/gpu/drm/i915/gvt/gvt.h b/drivers/gpu/drm/i915/gvt/gvt.h index 2c95aeef4e41..6c62467df22c 100644 --- a/drivers/gpu/drm/i915/gvt/gvt.h +++ b/drivers/gpu/drm/i915/gvt/gvt.h @@ -232,6 +232,7 @@ struct intel_vgpu { unsigned long nr_cache_entries; struct mutex cache_lock; + struct kvm *kvm; struct kvm_page_track_notifier_node track_node; #define NR_BKT (1 << 18) struct hlist_head ptable[NR_BKT]; diff --git a/drivers/gpu/drm/i915/gvt/kvmgt.c b/drivers/gpu/drm/i915/gvt/kvmgt.c index b27ff77bfb50..cf418e2c560d 100644 --- a/drivers/gpu/drm/i915/gvt/kvmgt.c +++ b/drivers/gpu/drm/i915/gvt/kvmgt.c @@ -36,6 +36,7 @@ #include <linux/init.h> #include <linux/mm.h> #include <linux/kthread.h> +#include <linux/kvm_host.h> #include <linux/sched/mm.h> #include <linux/types.h> #include <linux/list.h> @@ -649,7 +650,7 @@ static bool __kvmgt_vgpu_exist(struct intel_vgpu *vgpu) if (!test_bit(INTEL_VGPU_STATUS_ATTACHED, itr->status)) continue; - if (vgpu->vfio_device.kvm == itr->vfio_device.kvm) { + if (vgpu->kvm == itr->kvm) { ret = true; goto out; } @@ -664,13 +665,13 @@ static int intel_vgpu_open_device(struct vfio_device *vfio_dev) struct intel_vgpu *vgpu = vfio_dev_to_vgpu(vfio_dev); int ret; + vgpu->kvm = vgpu->vfio_device.kvm; if (__kvmgt_vgpu_exist(vgpu)) return -EEXIST; vgpu->track_node.track_write = kvmgt_page_track_write; vgpu->track_node.track_remove_region = kvmgt_page_track_remove_region; - ret = kvm_page_track_register_notifier(vgpu->vfio_device.kvm, - &vgpu->track_node); + ret = kvm_page_track_register_notifier(vgpu->kvm, &vgpu->track_node); if (ret) { gvt_vgpu_err("KVM is required to use Intel vGPU\n"); return ret; @@ -707,8 +708,7 @@ static void intel_vgpu_close_device(struct vfio_device *vfio_dev) debugfs_lookup_and_remove(KVMGT_DEBUGFS_FILENAME, vgpu->debugfs); - kvm_page_track_unregister_notifier(vgpu->vfio_device.kvm, - &vgpu->track_node); + kvm_page_track_unregister_notifier(vgpu->kvm, &vgpu->track_node); kvmgt_protect_table_destroy(vgpu); gvt_cache_destroy(vgpu); @@ -1560,7 +1560,7 @@ int intel_gvt_page_track_add(struct intel_vgpu *info, u64 gfn) if (kvmgt_gfn_is_write_protected(info, gfn)) return 0; - r = kvm_write_track_add_gfn(info->vfio_device.kvm, gfn); + r = kvm_write_track_add_gfn(info->kvm, gfn); if (r) return r; @@ -1578,7 +1578,7 @@ int intel_gvt_page_track_remove(struct intel_vgpu *info, u64 gfn) if (!kvmgt_gfn_is_write_protected(info, gfn)) return 0; - r = kvm_write_track_remove_gfn(info->vfio_device.kvm, gfn); + r = kvm_write_track_remove_gfn(info->kvm, gfn); if (r) return r; -- 2.45.3

4 months, 2 weeks

1
1
0 0

Let me know

by Martyn Beasley

Hi there! I hope you can get back to me soon. I'm currently in need of help with purchasing some raw materials from China. Thank you! Martyn Beasley

4 months, 2 weeks

1
0
0 0

FAILED: patch "[PATCH] statmount: let unset strings be empty" 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 e52e97f09fb66fd868260d05bd6b74a9a3db39ee # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2025021110-demeaning-mushroom-9922@gregkh' --subject-prefix 'PATCH 6.12.y' HEAD^.. Possible dependencies: thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From e52e97f09fb66fd868260d05bd6b74a9a3db39ee Mon Sep 17 00:00:00 2001 From: Miklos Szeredi <mszeredi(a)redhat.com> Date: Thu, 30 Jan 2025 13:15:00 +0100 Subject: [PATCH] statmount: let unset strings be empty Just like it's normal for unset values to be zero, unset strings should be empty instead of containing random values. It seems to be a typical mistake that the mask returned by statmount is not checked, which can result in various bugs. With this fix, these bugs are prevented, since it is highly likely that userspace would just want to turn the missing mask case into an empty string anyway (most of the recently found cases are of this type). Link: https://lore.kernel.org/all/CAJfpegsVCPfCn2DpM8iiYSS5DpMsLB8QBUCHecoj6s0Vxf… Fixes: 68385d77c05b ("statmount: simplify string option retrieval") Fixes: 46eae99ef733 ("add statmount(2) syscall") Cc: stable(a)vger.kernel.org # v6.8 Signed-off-by: Miklos Szeredi <mszeredi(a)redhat.com> Link: https://lore.kernel.org/r/20250130121500.113446-1-mszeredi@redhat.com Reviewed-by: Jeff Layton <jlayton(a)kernel.org> Signed-off-by: Christian Brauner <brauner(a)kernel.org> diff --git a/fs/namespace.c b/fs/namespace.c index a3ed3f2980cb..9c4d307a82cd 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -5191,39 +5191,45 @@ static int statmount_string(struct kstatmount *s, u64 flag) size_t kbufsize; struct seq_file *seq = &s->seq; struct statmount *sm = &s->sm; - u32 start = seq->count; + u32 start, *offp; + + /* Reserve an empty string at the beginning for any unset offsets */ + if (!seq->count) + seq_putc(seq, 0); + + start = seq->count; switch (flag) { case STATMOUNT_FS_TYPE: - sm->fs_type = start; + offp = &sm->fs_type; ret = statmount_fs_type(s, seq); break; case STATMOUNT_MNT_ROOT: - sm->mnt_root = start; + offp = &sm->mnt_root; ret = statmount_mnt_root(s, seq); break; case STATMOUNT_MNT_POINT: - sm->mnt_point = start; + offp = &sm->mnt_point; ret = statmount_mnt_point(s, seq); break; case STATMOUNT_MNT_OPTS: - sm->mnt_opts = start; + offp = &sm->mnt_opts; ret = statmount_mnt_opts(s, seq); break; case STATMOUNT_OPT_ARRAY: - sm->opt_array = start; + offp = &sm->opt_array; ret = statmount_opt_array(s, seq); break; case STATMOUNT_OPT_SEC_ARRAY: - sm->opt_sec_array = start; + offp = &sm->opt_sec_array; ret = statmount_opt_sec_array(s, seq); break; case STATMOUNT_FS_SUBTYPE: - sm->fs_subtype = start; + offp = &sm->fs_subtype; statmount_fs_subtype(s, seq); break; case STATMOUNT_SB_SOURCE: - sm->sb_source = start; + offp = &sm->sb_source; ret = statmount_sb_source(s, seq); break; default: @@ -5251,6 +5257,7 @@ static int statmount_string(struct kstatmount *s, u64 flag) seq->buf[seq->count++] = '\0'; sm->mask |= flag; + *offp = start; return 0; }

4 months, 2 weeks

2
1
0 0

Let me know

by Martyn Beasley

Hi there! I hope you can get back to me soon. I'm currently in need of help with purchasing some raw materials from China. Thank you! Martyn Beasley

4 months, 2 weeks

1
0
0 0

FAILED: patch "[PATCH] fs: fix adding security options to statmount.mnt_opt" 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 5eb987105357cb7cfa7cf3b1e2f66d5c0977e412 # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2025021142-spearmint-explain-70ac@gregkh' --subject-prefix 'PATCH 6.12.y' HEAD^.. Possible dependencies: thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 5eb987105357cb7cfa7cf3b1e2f66d5c0977e412 Mon Sep 17 00:00:00 2001 From: Miklos Szeredi <mszeredi(a)redhat.com> Date: Wed, 29 Jan 2025 16:12:53 +0100 Subject: [PATCH] fs: fix adding security options to statmount.mnt_opt Prepending security options was made conditional on sb->s_op->show_options, but security options are independent of sb options. Fixes: 056d33137bf9 ("fs: prepend statmount.mnt_opts string with security_sb_mnt_opts()") Fixes: f9af549d1fd3 ("fs: export mount options via statmount()") Cc: stable(a)vger.kernel.org # v6.11 Signed-off-by: Miklos Szeredi <mszeredi(a)redhat.com> Link: https://lore.kernel.org/r/20250129151253.33241-1-mszeredi@redhat.com Signed-off-by: Christian Brauner <brauner(a)kernel.org> diff --git a/fs/namespace.c b/fs/namespace.c index 9c4d307a82cd..8f1000f9f3df 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -5087,31 +5087,30 @@ static int statmount_mnt_opts(struct kstatmount *s, struct seq_file *seq) { struct vfsmount *mnt = s->mnt; struct super_block *sb = mnt->mnt_sb; + size_t start = seq->count; int err; + err = security_sb_show_options(seq, sb); + if (err) + return err; + if (sb->s_op->show_options) { - size_t start = seq->count; - - err = security_sb_show_options(seq, sb); - if (err) - return err; - err = sb->s_op->show_options(seq, mnt->mnt_root); if (err) return err; - - if (unlikely(seq_has_overflowed(seq))) - return -EAGAIN; - - if (seq->count == start) - return 0; - - /* skip leading comma */ - memmove(seq->buf + start, seq->buf + start + 1, - seq->count - start - 1); - seq->count--; } + if (unlikely(seq_has_overflowed(seq))) + return -EAGAIN; + + if (seq->count == start) + return 0; + + /* skip leading comma */ + memmove(seq->buf + start, seq->buf + start + 1, + seq->count - start - 1); + seq->count--; + return 0; }

4 months, 2 weeks

2
1
0 0

[PATCH v3] sched/topology: Enable topology_span_sane check only for debug builds

by Naman Jain

From: Saurabh Sengar <ssengar(a)linux.microsoft.com> On a x86 system under test with 1780 CPUs, topology_span_sane() takes around 8 seconds cumulatively for all the iterations. It is an expensive operation which does the sanity of non-NUMA topology masks. CPU topology is not something which changes very frequently hence make this check optional for the systems where the topology is trusted and need faster bootup. Restrict this to sched_verbose kernel cmdline option so that this penalty can be avoided for the systems who want to avoid it. Cc: stable(a)vger.kernel.org Fixes: ccf74128d66c ("sched/topology: Assert non-NUMA topology masks don't (partially) overlap") Signed-off-by: Saurabh Sengar <ssengar(a)linux.microsoft.com> Co-developed-by: Naman Jain <namjain(a)linux.microsoft.com> Signed-off-by: Naman Jain <namjain(a)linux.microsoft.com> --- Changes since v2: https://lore.kernel.org/all/1731922777-7121-1-git-send-email-ssengar@linux.… - Use sched_debug() instead of using sched_debug_verbose variable directly (addressing Prateek's comment) Changes since v1: https://lore.kernel.org/all/1729619853-2597-1-git-send-email-ssengar@linux.… - Use kernel cmdline param instead of compile time flag. Adding a link to the other patch which is under review. https://lore.kernel.org/lkml/20241031200431.182443-1-steve.wahl@hpe.com/ Above patch tries to optimize the topology sanity check, whereas this patch makes it optional. We believe both patches can coexist, as even with optimization, there will still be some performance overhead for this check. --- kernel/sched/topology.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/kernel/sched/topology.c b/kernel/sched/topology.c index c49aea8c1025..b030c1a2121f 100644 --- a/kernel/sched/topology.c +++ b/kernel/sched/topology.c @@ -2359,6 +2359,13 @@ static bool topology_span_sane(struct sched_domain_topology_level *tl, { int i = cpu + 1; + /* Skip the topology sanity check for non-debug, as it is a time-consuming operatin */ + if (!sched_debug()) { + pr_info_once("%s: Skipping topology span sanity check. Use `sched_verbose` boot parameter to enable it.\n", + __func__); + return true; + } + /* NUMA levels are allowed to overlap */ if (tl->flags & SDTL_OVERLAP) return true; base-commit: 00f3246adeeacbda0bd0b303604e46eb59c32e6e -- 2.43.0

4 months, 2 weeks

6
18
0 0

[PATCH v3 0/8] KVM: arm64: FPSIMD/SVE/SME fixes

by Mark Rutland

These patches fix some issues with the way KVM manages FPSIMD/SVE/SME state. The series supersedes my earlier attempt at fixing the host SVE state corruption issue: https://lore.kernel.org/linux-arm-kernel/20250121100026.3974971-1-mark.rutl… Patch 1 addresses the host SVE state corruption issue by always saving and unbinding the host state when loading a vCPU, as discussed on the earlier patch: https://lore.kernel.org/linux-arm-kernel/Z4--YuG5SWrP_pW7@J2N7QTR9R3/ https://lore.kernel.org/linux-arm-kernel/86plkful48.wl-maz@kernel.org/ Patches 2 to 4 remove code made redundant by patch 1. These probably warrant backporting along with patch 1 as there is some historical brokenness in the code they remove. Patches 5 to 7 are preparatory refactoring for patch 8, and are not intended to have any functional impact. Patch 8 addresses some mismanagement of ZCR_EL{1,2} which can result in the host VMM unexpectedly receiving a SIGKILL. To fix this, we eagerly switch ZCR_EL{1,2} at guest<->host transitions, as discussed on another series: https://lore.kernel.org/linux-arm-kernel/Z4pAMaEYvdLpmbg2@J2N7QTR9R3/ https://lore.kernel.org/linux-arm-kernel/86o6zzukwr.wl-maz@kernel.org/ https://lore.kernel.org/linux-arm-kernel/Z5Dc-WMu2azhTuMn@J2N7QTR9R3/ The end result is that KVM loses ~100 lines of code, and becomes a bit simpler to reason about. I've pushed these patches to the arm64-kvm-fpsimd-fixes-20250210 tag on my kernel.org repo: https://git.kernel.org/pub/scm/linux/kernel/git/mark/linux.git/ git://git.kernel.org/pub/scm/linux/kernel/git/mark/linux.git The (unstable) arm64/kvm/fpsimd-fixes branch in that repo contains the fixes plus additional debug patches I've used for testing. I've given this some basic testing on a virtual platform, booting a host and a guest with and without constraining the guest's max SVE VL, with: * kvm_arm.mode=vhe * kvm_arm.mode=nvhe * kvm_arm.mode=protected (IIUC this will default to hVHE) Since v1 [1]: * Address some additional compiler warnings in patch 7 * Use ZCR_EL1 alias in VHE code * Fold in Tested-by and Reviewed-by tags * Fix typos Since v2 [2]: * Ensure context synchronization in patch 8 * Fold in Tested-by and Reviewed-by tags * Fix typos [1] https://lore.kernel.org/linux-arm-kernel/20250204152100.705610-1-mark.rutla… [2] https://lore.kernel.org/linux-arm-kernel/20250206141102.954688-1-mark.rutla… Mark. Mark Rutland (8): KVM: arm64: Unconditionally save+flush host FPSIMD/SVE/SME state KVM: arm64: Remove host FPSIMD saving for non-protected KVM KVM: arm64: Remove VHE host restore of CPACR_EL1.ZEN KVM: arm64: Remove VHE host restore of CPACR_EL1.SMEN KVM: arm64: Refactor CPTR trap deactivation KVM: arm64: Refactor exit handlers KVM: arm64: Mark some header functions as inline KVM: arm64: Eagerly switch ZCR_EL{1,2} arch/arm64/include/asm/kvm_emulate.h | 42 -------- arch/arm64/include/asm/kvm_host.h | 22 +--- arch/arm64/kernel/fpsimd.c | 25 ----- arch/arm64/kvm/arm.c | 8 -- arch/arm64/kvm/fpsimd.c | 100 ++---------------- arch/arm64/kvm/hyp/entry.S | 5 + arch/arm64/kvm/hyp/include/hyp/switch.h | 133 +++++++++++++++++------- arch/arm64/kvm/hyp/nvhe/hyp-main.c | 15 ++- arch/arm64/kvm/hyp/nvhe/switch.c | 91 ++++++++-------- arch/arm64/kvm/hyp/vhe/switch.c | 33 +++--- 10 files changed, 187 insertions(+), 287 deletions(-) -- 2.30.2

4 months, 2 weeks

4
11
0 0

FAILED: patch "[PATCH] xe/oa: Fix query mode of operation for OAR/OAC" 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 55039832f98c7e05f1cf9e0d8c12b2490abd0f16 # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2025021013-cavalry-unsightly-0671@gregkh' --subject-prefix 'PATCH 6.12.y' HEAD^.. Possible dependencies: thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 55039832f98c7e05f1cf9e0d8c12b2490abd0f16 Mon Sep 17 00:00:00 2001 From: Umesh Nerlige Ramappa <umesh.nerlige.ramappa(a)intel.com> Date: Fri, 20 Dec 2024 09:19:18 -0800 Subject: [PATCH] xe/oa: Fix query mode of operation for OAR/OAC This is a set of squashed commits to facilitate smooth applying to stable. Each commit message is retained for reference. 1) Allow a GGTT mapped batch to be submitted to user exec queue For a OA use case, one of the HW registers needs to be modified by submitting an MI_LOAD_REGISTER_IMM command to the users exec queue, so that the register is modified in the user's hardware context. In order to do this a batch that is mapped in GGTT, needs to be submitted to the user exec queue. Since all user submissions use q->vm and hence PPGTT, add some plumbing to enable submission of batches mapped in GGTT. v2: ggtt is zero-initialized, so no need to set it false (Matt Brost) 2) xe/oa: Use MI_LOAD_REGISTER_IMMEDIATE to enable OAR/OAC To enable OAR/OAC, a bit in RING_CONTEXT_CONTROL needs to be set. Setting this bit cause the context image size to change and if not done correct, can cause undesired hangs. Current code uses a separate exec_queue to modify this bit and is error-prone. As per HW recommendation, submit MI_LOAD_REGISTER_IMM to the target hardware context to modify the relevant bit. In v2 version, an attempt to submit everything to the user-queue was made, but it failed the unprivileged-single-ctx-counters test. It appears that the OACTXCONTROL must be modified from a remote context. In v3 version, all context specific register configurations were moved to use LOAD_REGISTER_IMMEDIATE and that seems to work well. This is a cleaner way, since we can now submit all configuration to user exec_queue and the fence handling is simplified. v2: (Matt) - set job->ggtt to true if create job is successful - unlock vm on job error (Ashutosh) - don't wait on job submission - use kernel exec queue where possible v3: (Ashutosh) - Fix checkpatch issues - Remove extra spaces/new-lines - Add Fixes: and Cc: tags - Reset context control bit when OA stream is closed - Submit all config via MI_LOAD_REGISTER_IMMEDIATE (Umesh) - Update commit message for v3 experiment - Squash patches for easier port to stable v4: (Ashutosh) - No need to pass q to xe_oa_submit_bb - Do not support exec queues with width > 1 - Fix disabling of CTX_CTRL_OAC_CONTEXT_ENABLE v5: (Ashutosh) - Drop reg_lri related comments - Use XE_OA_SUBMIT_NO_DEPS in xe_oa_load_with_lri Fixes: 8135f1c09dd2 ("drm/xe/oa: Don't reset OAC_CONTEXT_ENABLE on OA stream close") Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa(a)intel.com> Reviewed-by: Matthew Brost <matthew.brost(a)intel.com> # commit 1 Reviewed-by: Ashutosh Dixit <ashutosh.dixit(a)intel.com> Cc: stable(a)vger.kernel.org Reviewed-by: Jonathan Cavitt <jonathan.cavitt(a)intel.com> Signed-off-by: Ashutosh Dixit <ashutosh.dixit(a)intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20241220171919.571528-2-umesh… diff --git a/drivers/gpu/drm/xe/xe_oa.c b/drivers/gpu/drm/xe/xe_oa.c index ae94490b0eac..9add60097ab5 100644 --- a/drivers/gpu/drm/xe/xe_oa.c +++ b/drivers/gpu/drm/xe/xe_oa.c @@ -74,12 +74,6 @@ struct xe_oa_config { struct rcu_head rcu; }; -struct flex { - struct xe_reg reg; - u32 offset; - u32 value; -}; - struct xe_oa_open_param { struct xe_file *xef; u32 oa_unit_id; @@ -605,19 +599,38 @@ static __poll_t xe_oa_poll(struct file *file, poll_table *wait) return ret; } +static void xe_oa_lock_vma(struct xe_exec_queue *q) +{ + if (q->vm) { + down_read(&q->vm->lock); + xe_vm_lock(q->vm, false); + } +} + +static void xe_oa_unlock_vma(struct xe_exec_queue *q) +{ + if (q->vm) { + xe_vm_unlock(q->vm); + up_read(&q->vm->lock); + } +} + static struct dma_fence *xe_oa_submit_bb(struct xe_oa_stream *stream, enum xe_oa_submit_deps deps, struct xe_bb *bb) { + struct xe_exec_queue *q = stream->exec_q ?: stream->k_exec_q; struct xe_sched_job *job; struct dma_fence *fence; int err = 0; - /* Kernel configuration is issued on stream->k_exec_q, not stream->exec_q */ - job = xe_bb_create_job(stream->k_exec_q, bb); + xe_oa_lock_vma(q); + + job = xe_bb_create_job(q, bb); if (IS_ERR(job)) { err = PTR_ERR(job); goto exit; } + job->ggtt = true; if (deps == XE_OA_SUBMIT_ADD_DEPS) { for (int i = 0; i < stream->num_syncs && !err; i++) @@ -632,10 +645,13 @@ static struct dma_fence *xe_oa_submit_bb(struct xe_oa_stream *stream, enum xe_oa fence = dma_fence_get(&job->drm.s_fence->finished); xe_sched_job_push(job); + xe_oa_unlock_vma(q); + return fence; err_put_job: xe_sched_job_put(job); exit: + xe_oa_unlock_vma(q); return ERR_PTR(err); } @@ -684,65 +700,19 @@ static void xe_oa_free_configs(struct xe_oa_stream *stream) dma_fence_put(stream->last_fence); } -static void xe_oa_store_flex(struct xe_oa_stream *stream, struct xe_lrc *lrc, - struct xe_bb *bb, const struct flex *flex, u32 count) -{ - u32 offset = xe_bo_ggtt_addr(lrc->bo); - - do { - bb->cs[bb->len++] = MI_STORE_DATA_IMM | MI_SDI_GGTT | - MI_FORCE_WRITE_COMPLETION_CHECK | - MI_SDI_NUM_DW(1); - bb->cs[bb->len++] = offset + flex->offset * sizeof(u32); - bb->cs[bb->len++] = 0; - bb->cs[bb->len++] = flex->value; - - } while (flex++, --count); -} - -static int xe_oa_modify_ctx_image(struct xe_oa_stream *stream, struct xe_lrc *lrc, - const struct flex *flex, u32 count) +static int xe_oa_load_with_lri(struct xe_oa_stream *stream, struct xe_oa_reg *reg_lri, u32 count) { struct dma_fence *fence; struct xe_bb *bb; int err; - bb = xe_bb_new(stream->gt, 4 * count, false); + bb = xe_bb_new(stream->gt, 2 * count + 1, false); if (IS_ERR(bb)) { err = PTR_ERR(bb); goto exit; } - xe_oa_store_flex(stream, lrc, bb, flex, count); - - fence = xe_oa_submit_bb(stream, XE_OA_SUBMIT_NO_DEPS, bb); - if (IS_ERR(fence)) { - err = PTR_ERR(fence); - goto free_bb; - } - xe_bb_free(bb, fence); - dma_fence_put(fence); - - return 0; -free_bb: - xe_bb_free(bb, NULL); -exit: - return err; -} - -static int xe_oa_load_with_lri(struct xe_oa_stream *stream, struct xe_oa_reg *reg_lri) -{ - struct dma_fence *fence; - struct xe_bb *bb; - int err; - - bb = xe_bb_new(stream->gt, 3, false); - if (IS_ERR(bb)) { - err = PTR_ERR(bb); - goto exit; - } - - write_cs_mi_lri(bb, reg_lri, 1); + write_cs_mi_lri(bb, reg_lri, count); fence = xe_oa_submit_bb(stream, XE_OA_SUBMIT_NO_DEPS, bb); if (IS_ERR(fence)) { @@ -762,71 +732,55 @@ static int xe_oa_load_with_lri(struct xe_oa_stream *stream, struct xe_oa_reg *re static int xe_oa_configure_oar_context(struct xe_oa_stream *stream, bool enable) { const struct xe_oa_format *format = stream->oa_buffer.format; - struct xe_lrc *lrc = stream->exec_q->lrc[0]; - u32 regs_offset = xe_lrc_regs_offset(lrc) / sizeof(u32); u32 oacontrol = __format_to_oactrl(format, OAR_OACONTROL_COUNTER_SEL_MASK) | (enable ? OAR_OACONTROL_COUNTER_ENABLE : 0); - struct flex regs_context[] = { + struct xe_oa_reg reg_lri[] = { { OACTXCONTROL(stream->hwe->mmio_base), - stream->oa->ctx_oactxctrl_offset[stream->hwe->class] + 1, enable ? OA_COUNTER_RESUME : 0, }, + { + OAR_OACONTROL, + oacontrol, + }, { RING_CONTEXT_CONTROL(stream->hwe->mmio_base), - regs_offset + CTX_CONTEXT_CONTROL, - _MASKED_BIT_ENABLE(CTX_CTRL_OAC_CONTEXT_ENABLE), + _MASKED_FIELD(CTX_CTRL_OAC_CONTEXT_ENABLE, + enable ? CTX_CTRL_OAC_CONTEXT_ENABLE : 0) }, }; - struct xe_oa_reg reg_lri = { OAR_OACONTROL, oacontrol }; - int err; - /* Modify stream hwe context image with regs_context */ - err = xe_oa_modify_ctx_image(stream, stream->exec_q->lrc[0], - regs_context, ARRAY_SIZE(regs_context)); - if (err) - return err; - - /* Apply reg_lri using LRI */ - return xe_oa_load_with_lri(stream, &reg_lri); + return xe_oa_load_with_lri(stream, reg_lri, ARRAY_SIZE(reg_lri)); } static int xe_oa_configure_oac_context(struct xe_oa_stream *stream, bool enable) { const struct xe_oa_format *format = stream->oa_buffer.format; - struct xe_lrc *lrc = stream->exec_q->lrc[0]; - u32 regs_offset = xe_lrc_regs_offset(lrc) / sizeof(u32); u32 oacontrol = __format_to_oactrl(format, OAR_OACONTROL_COUNTER_SEL_MASK) | (enable ? OAR_OACONTROL_COUNTER_ENABLE : 0); - struct flex regs_context[] = { + struct xe_oa_reg reg_lri[] = { { OACTXCONTROL(stream->hwe->mmio_base), - stream->oa->ctx_oactxctrl_offset[stream->hwe->class] + 1, enable ? OA_COUNTER_RESUME : 0, }, + { + OAC_OACONTROL, + oacontrol + }, { RING_CONTEXT_CONTROL(stream->hwe->mmio_base), - regs_offset + CTX_CONTEXT_CONTROL, - _MASKED_BIT_ENABLE(CTX_CTRL_OAC_CONTEXT_ENABLE) | + _MASKED_FIELD(CTX_CTRL_OAC_CONTEXT_ENABLE, + enable ? CTX_CTRL_OAC_CONTEXT_ENABLE : 0) | _MASKED_FIELD(CTX_CTRL_RUN_ALONE, enable ? CTX_CTRL_RUN_ALONE : 0), }, }; - struct xe_oa_reg reg_lri = { OAC_OACONTROL, oacontrol }; - int err; /* Set ccs select to enable programming of OAC_OACONTROL */ xe_mmio_write32(&stream->gt->mmio, __oa_regs(stream)->oa_ctrl, __oa_ccs_select(stream)); - /* Modify stream hwe context image with regs_context */ - err = xe_oa_modify_ctx_image(stream, stream->exec_q->lrc[0], - regs_context, ARRAY_SIZE(regs_context)); - if (err) - return err; - - /* Apply reg_lri using LRI */ - return xe_oa_load_with_lri(stream, &reg_lri); + return xe_oa_load_with_lri(stream, reg_lri, ARRAY_SIZE(reg_lri)); } static int xe_oa_configure_oa_context(struct xe_oa_stream *stream, bool enable) @@ -2110,8 +2064,8 @@ int xe_oa_stream_open_ioctl(struct drm_device *dev, u64 data, struct drm_file *f if (XE_IOCTL_DBG(oa->xe, !param.exec_q)) return -ENOENT; - if (param.exec_q->width > 1) - drm_dbg(&oa->xe->drm, "exec_q->width > 1, programming only exec_q->lrc[0]\n"); + if (XE_IOCTL_DBG(oa->xe, param.exec_q->width > 1)) + return -EOPNOTSUPP; } /* diff --git a/drivers/gpu/drm/xe/xe_ring_ops.c b/drivers/gpu/drm/xe/xe_ring_ops.c index 3a75a08b6be9..c8ab37fa0d19 100644 --- a/drivers/gpu/drm/xe/xe_ring_ops.c +++ b/drivers/gpu/drm/xe/xe_ring_ops.c @@ -223,7 +223,10 @@ static int emit_pipe_imm_ggtt(u32 addr, u32 value, bool stall_only, u32 *dw, static u32 get_ppgtt_flag(struct xe_sched_job *job) { - return job->q->vm ? BIT(8) : 0; + if (job->q->vm && !job->ggtt) + return BIT(8); + + return 0; } static int emit_copy_timestamp(struct xe_lrc *lrc, u32 *dw, int i) diff --git a/drivers/gpu/drm/xe/xe_sched_job_types.h b/drivers/gpu/drm/xe/xe_sched_job_types.h index f13f333f00be..d942b20a9f29 100644 --- a/drivers/gpu/drm/xe/xe_sched_job_types.h +++ b/drivers/gpu/drm/xe/xe_sched_job_types.h @@ -56,6 +56,8 @@ struct xe_sched_job { u32 migrate_flush_flags; /** @ring_ops_flush_tlb: The ring ops need to flush TLB before payload. */ bool ring_ops_flush_tlb; + /** @ggtt: mapped in ggtt. */ + bool ggtt; /** @ptrs: per instance pointers. */ struct xe_job_ptrs ptrs[]; };

4 months, 2 weeks

3
2
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror February 2025