From e33ac68e5e21ec1292490dfe061e75c0dbdd3bd4 Mon Sep 17 00:00:00 2001 From: Pavel Begunkov Date: Thu, 26 Dec 2024 16:49:23 +0000 Subject: [PATCH 01/16] io_uring/sqpoll: fix sqpoll error handling races BUG: KASAN: slab-use-after-free in __lock_acquire+0x370b/0x4a10 kernel/locking/lockdep.c:5089 Call Trace: ... _raw_spin_lock_irqsave+0x3d/0x60 kernel/locking/spinlock.c:162 class_raw_spinlock_irqsave_constructor include/linux/spinlock.h:551 [inline] try_to_wake_up+0xb5/0x23c0 kernel/sched/core.c:4205 io_sq_thread_park+0xac/0xe0 io_uring/sqpoll.c:55 io_sq_thread_finish+0x6b/0x310 io_uring/sqpoll.c:96 io_sq_offload_create+0x162/0x11d0 io_uring/sqpoll.c:497 io_uring_create io_uring/io_uring.c:3724 [inline] io_uring_setup+0x1728/0x3230 io_uring/io_uring.c:3806 ... Kun Hu reports that the SQPOLL creating error path has UAF, which happens if io_uring_alloc_task_context() fails and then io_sq_thread() manages to run and complete before the rest of error handling code, which means io_sq_thread_finish() is looking at already killed task. Note that this is mostly theoretical, requiring fault injection on the allocation side to trigger in practice. Cc: stable@vger.kernel.org Reported-by: Kun Hu Signed-off-by: Pavel Begunkov Link: https://lore.kernel.org/r/0f2f1aa5729332612bd01fe0f2f385fd1f06ce7c.1735231717.git.asml.silence@gmail.com Signed-off-by: Jens Axboe --- io_uring/sqpoll.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/io_uring/sqpoll.c b/io_uring/sqpoll.c index 6df5e649c413..9e5bd79fd2b5 100644 --- a/io_uring/sqpoll.c +++ b/io_uring/sqpoll.c @@ -405,6 +405,7 @@ void io_sqpoll_wait_sq(struct io_ring_ctx *ctx) __cold int io_sq_offload_create(struct io_ring_ctx *ctx, struct io_uring_params *p) { + struct task_struct *task_to_put = NULL; int ret; /* Retain compatibility with failing for an invalid attach attempt */ @@ -480,6 +481,7 @@ __cold int io_sq_offload_create(struct io_ring_ctx *ctx, } sqd->thread = tsk; + task_to_put = get_task_struct(tsk); ret = io_uring_alloc_task_context(tsk, ctx); wake_up_new_task(tsk); if (ret) @@ -490,11 +492,15 @@ __cold int io_sq_offload_create(struct io_ring_ctx *ctx, goto err; } + if (task_to_put) + put_task_struct(task_to_put); return 0; err_sqpoll: complete(&ctx->sq_data->exited); err: io_sq_thread_finish(ctx); + if (task_to_put) + put_task_struct(task_to_put); return ret; } -- 2.51.0 From 6cc45f8c1f898570916044f606be9890d295e129 Mon Sep 17 00:00:00 2001 From: Tomas Glozar Date: Wed, 27 Nov 2024 14:41:30 +0100 Subject: [PATCH 02/16] rtla/timerlat: Fix histogram ALL for zero samples rtla timerlat hist currently computers the minimum, maximum and average latency even in cases when there are zero samples. This leads to nonsensical values being calculated for maximum and minimum, and to divide by zero for average. A similar bug is fixed by 01b05fc0e5f3 ("rtla/timerlat: Fix histogram report when a cpu count is 0") but the bug still remains for printing the sum over all CPUs in timerlat_print_stats_all. The issue can be reproduced with this command: $ rtla timerlat hist -U -d 1s Index over: count: min: avg: max: Floating point exception (core dumped) (There are always no samples with -U unless the user workload is created.) Fix the bug by omitting max/min/avg when sample count is zero, displaying a dash instead, just like we already do for the individual CPUs. The logic is moved into a new function called format_summary_value, which is used for both the individual CPUs and for the overall summary. Cc: stable@vger.kernel.org Link: https://lore.kernel.org/20241127134130.51171-1-tglozar@redhat.com Fixes: 1462501c7a8 ("rtla/timerlat: Add a summary for hist mode") Signed-off-by: Tomas Glozar Signed-off-by: Steven Rostedt (Google) --- tools/tracing/rtla/src/timerlat_hist.c | 177 ++++++++++++++----------- 1 file changed, 96 insertions(+), 81 deletions(-) diff --git a/tools/tracing/rtla/src/timerlat_hist.c b/tools/tracing/rtla/src/timerlat_hist.c index 8b66387e5f35..4403cc4eba30 100644 --- a/tools/tracing/rtla/src/timerlat_hist.c +++ b/tools/tracing/rtla/src/timerlat_hist.c @@ -281,6 +281,21 @@ static void timerlat_hist_header(struct osnoise_tool *tool) trace_seq_reset(s); } +/* + * format_summary_value - format a line of summary value (min, max or avg) + * of hist data + */ +static void format_summary_value(struct trace_seq *seq, + int count, + unsigned long long val, + bool avg) +{ + if (count) + trace_seq_printf(seq, "%9llu ", avg ? val / count : val); + else + trace_seq_printf(seq, "%9c ", '-'); +} + /* * timerlat_print_summary - print the summary of the hist data to the output */ @@ -328,29 +343,23 @@ timerlat_print_summary(struct timerlat_hist_params *params, if (!data->hist[cpu].irq_count && !data->hist[cpu].thread_count) continue; - if (!params->no_irq) { - if (data->hist[cpu].irq_count) - trace_seq_printf(trace->seq, "%9llu ", - data->hist[cpu].min_irq); - else - trace_seq_printf(trace->seq, " - "); - } + if (!params->no_irq) + format_summary_value(trace->seq, + data->hist[cpu].irq_count, + data->hist[cpu].min_irq, + false); - if (!params->no_thread) { - if (data->hist[cpu].thread_count) - trace_seq_printf(trace->seq, "%9llu ", - data->hist[cpu].min_thread); - else - trace_seq_printf(trace->seq, " - "); - } + if (!params->no_thread) + format_summary_value(trace->seq, + data->hist[cpu].thread_count, + data->hist[cpu].min_thread, + false); - if (params->user_hist) { - if (data->hist[cpu].user_count) - trace_seq_printf(trace->seq, "%9llu ", - data->hist[cpu].min_user); - else - trace_seq_printf(trace->seq, " - "); - } + if (params->user_hist) + format_summary_value(trace->seq, + data->hist[cpu].user_count, + data->hist[cpu].min_user, + false); } trace_seq_printf(trace->seq, "\n"); @@ -364,29 +373,23 @@ timerlat_print_summary(struct timerlat_hist_params *params, if (!data->hist[cpu].irq_count && !data->hist[cpu].thread_count) continue; - if (!params->no_irq) { - if (data->hist[cpu].irq_count) - trace_seq_printf(trace->seq, "%9llu ", - data->hist[cpu].sum_irq / data->hist[cpu].irq_count); - else - trace_seq_printf(trace->seq, " - "); - } + if (!params->no_irq) + format_summary_value(trace->seq, + data->hist[cpu].irq_count, + data->hist[cpu].sum_irq, + true); - if (!params->no_thread) { - if (data->hist[cpu].thread_count) - trace_seq_printf(trace->seq, "%9llu ", - data->hist[cpu].sum_thread / data->hist[cpu].thread_count); - else - trace_seq_printf(trace->seq, " - "); - } + if (!params->no_thread) + format_summary_value(trace->seq, + data->hist[cpu].thread_count, + data->hist[cpu].sum_thread, + true); - if (params->user_hist) { - if (data->hist[cpu].user_count) - trace_seq_printf(trace->seq, "%9llu ", - data->hist[cpu].sum_user / data->hist[cpu].user_count); - else - trace_seq_printf(trace->seq, " - "); - } + if (params->user_hist) + format_summary_value(trace->seq, + data->hist[cpu].user_count, + data->hist[cpu].sum_user, + true); } trace_seq_printf(trace->seq, "\n"); @@ -400,29 +403,23 @@ timerlat_print_summary(struct timerlat_hist_params *params, if (!data->hist[cpu].irq_count && !data->hist[cpu].thread_count) continue; - if (!params->no_irq) { - if (data->hist[cpu].irq_count) - trace_seq_printf(trace->seq, "%9llu ", - data->hist[cpu].max_irq); - else - trace_seq_printf(trace->seq, " - "); - } + if (!params->no_irq) + format_summary_value(trace->seq, + data->hist[cpu].irq_count, + data->hist[cpu].max_irq, + false); - if (!params->no_thread) { - if (data->hist[cpu].thread_count) - trace_seq_printf(trace->seq, "%9llu ", - data->hist[cpu].max_thread); - else - trace_seq_printf(trace->seq, " - "); - } + if (!params->no_thread) + format_summary_value(trace->seq, + data->hist[cpu].thread_count, + data->hist[cpu].max_thread, + false); - if (params->user_hist) { - if (data->hist[cpu].user_count) - trace_seq_printf(trace->seq, "%9llu ", - data->hist[cpu].max_user); - else - trace_seq_printf(trace->seq, " - "); - } + if (params->user_hist) + format_summary_value(trace->seq, + data->hist[cpu].user_count, + data->hist[cpu].max_user, + false); } trace_seq_printf(trace->seq, "\n"); trace_seq_do_printf(trace->seq); @@ -506,16 +503,22 @@ timerlat_print_stats_all(struct timerlat_hist_params *params, trace_seq_printf(trace->seq, "min: "); if (!params->no_irq) - trace_seq_printf(trace->seq, "%9llu ", - sum.min_irq); + format_summary_value(trace->seq, + sum.irq_count, + sum.min_irq, + false); if (!params->no_thread) - trace_seq_printf(trace->seq, "%9llu ", - sum.min_thread); + format_summary_value(trace->seq, + sum.thread_count, + sum.min_thread, + false); if (params->user_hist) - trace_seq_printf(trace->seq, "%9llu ", - sum.min_user); + format_summary_value(trace->seq, + sum.user_count, + sum.min_user, + false); trace_seq_printf(trace->seq, "\n"); @@ -523,16 +526,22 @@ timerlat_print_stats_all(struct timerlat_hist_params *params, trace_seq_printf(trace->seq, "avg: "); if (!params->no_irq) - trace_seq_printf(trace->seq, "%9llu ", - sum.sum_irq / sum.irq_count); + format_summary_value(trace->seq, + sum.irq_count, + sum.sum_irq, + true); if (!params->no_thread) - trace_seq_printf(trace->seq, "%9llu ", - sum.sum_thread / sum.thread_count); + format_summary_value(trace->seq, + sum.thread_count, + sum.sum_thread, + true); if (params->user_hist) - trace_seq_printf(trace->seq, "%9llu ", - sum.sum_user / sum.user_count); + format_summary_value(trace->seq, + sum.user_count, + sum.sum_user, + true); trace_seq_printf(trace->seq, "\n"); @@ -540,16 +549,22 @@ timerlat_print_stats_all(struct timerlat_hist_params *params, trace_seq_printf(trace->seq, "max: "); if (!params->no_irq) - trace_seq_printf(trace->seq, "%9llu ", - sum.max_irq); + format_summary_value(trace->seq, + sum.irq_count, + sum.max_irq, + false); if (!params->no_thread) - trace_seq_printf(trace->seq, "%9llu ", - sum.max_thread); + format_summary_value(trace->seq, + sum.thread_count, + sum.max_thread, + false); if (params->user_hist) - trace_seq_printf(trace->seq, "%9llu ", - sum.max_user); + format_summary_value(trace->seq, + sum.user_count, + sum.max_user, + false); trace_seq_printf(trace->seq, "\n"); trace_seq_do_printf(trace->seq); -- 2.51.0 From 31ad36a271290648e7c2288a03d7b933d20254d6 Mon Sep 17 00:00:00 2001 From: chenchangcheng Date: Fri, 20 Dec 2024 15:48:47 +0800 Subject: [PATCH 03/16] objtool: Add bch2_trans_unlocked_error() to bcachefs noreturns Fix the following objtool warning during build time: fs/bcachefs/btree_trans_commit.o: warning: objtool: bch2_trans_commit_write_locked.isra.0() falls through to next function do_bch2_trans_commit.isra.0() fs/bcachefs/btree_trans_commit.o: warning: objtool: .text: unexpected end of section ...... fs/bcachefs/btree_update.o: warning: objtool: bch2_trans_update_get_key_cache() falls through to next function flush_new_cached_update() fs/bcachefs/btree_update.o: warning: objtool: flush_new_cached_update() falls through to next function bch2_trans_update_by_path() bch2_trans_unlocked_error() is an Obviously Correct (tm) panic() wrapper, add it to the list of known noreturns. [ mingo: Improved the changelog ] Fixes: fd104e2967b7 ("bcachefs: bch2_trans_verify_not_unlocked()") Signed-off-by: chenchangcheng Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Ingo Molnar Link: https://lkml.kernel.org/r/20241220074847.3418134-1-ccc194101@163.com --- tools/objtool/noreturns.h | 1 + 1 file changed, 1 insertion(+) diff --git a/tools/objtool/noreturns.h b/tools/objtool/noreturns.h index f37614cc2c1b..b2174894f9f7 100644 --- a/tools/objtool/noreturns.h +++ b/tools/objtool/noreturns.h @@ -19,6 +19,7 @@ NORETURN(__x64_sys_exit_group) NORETURN(arch_cpu_idle_dead) NORETURN(bch2_trans_in_restart_error) NORETURN(bch2_trans_restart_error) +NORETURN(bch2_trans_unlocked_error) NORETURN(cpu_bringup_and_idle) NORETURN(cpu_startup_entry) NORETURN(do_exit) -- 2.51.0 From f718faf3940e95d5d34af9041f279f598396ab7d Mon Sep 17 00:00:00 2001 From: Chen Ridong Date: Tue, 17 Dec 2024 00:48:18 +0000 Subject: [PATCH 04/16] freezer, sched: Report frozen tasks as 'D' instead of 'R' MIME-Version: 1.0 Content-Type: text/plain; charset=utf8 Content-Transfer-Encoding: 8bit Before commit: f5d39b020809 ("freezer,sched: Rewrite core freezer logic") the frozen task stat was reported as 'D' in cgroup v1. However, after rewriting the core freezer logic, the frozen task stat is reported as 'R'. This is confusing, especially when a task with stat of 'S' is frozen. This bug can be reproduced with these steps: $ cd /sys/fs/cgroup/freezer/ $ mkdir test $ sleep 1000 & [1] 739 // task whose stat is 'S' $ echo 739 > test/cgroup.procs $ echo FROZEN > test/freezer.state $ ps -aux | grep 739 root 739 0.1 0.0 8376 1812 pts/0 R 10:56 0:00 sleep 1000 As shown above, a task whose stat is 'S' was changed to 'R' when it was frozen. To solve this regression, simply maintain the same reported state as before the rewrite. [ mingo: Enhanced the changelog and comments ] Fixes: f5d39b020809 ("freezer,sched: Rewrite core freezer logic") Signed-off-by: Chen Ridong Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Ingo Molnar Acked-by: Tejun Heo Acked-by: Michal Koutný Link: https://lore.kernel.org/r/20241217004818.3200515-1-chenridong@huaweicloud.com --- include/linux/sched.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/include/linux/sched.h b/include/linux/sched.h index 66b311fbd5d6..64934e0830af 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1637,8 +1637,9 @@ static inline unsigned int __task_state_index(unsigned int tsk_state, * We're lying here, but rather than expose a completely new task state * to userspace, we can make this appear as if the task has gone through * a regular rt_mutex_lock() call. + * Report frozen tasks as uninterruptible. */ - if (tsk_state & TASK_RTLOCK_WAIT) + if ((tsk_state & TASK_RTLOCK_WAIT) || (tsk_state & TASK_FROZEN)) state = TASK_UNINTERRUPTIBLE; return fls(state); -- 2.51.0 From dc81e556f2a017d681251ace21bf06c126d5a192 Mon Sep 17 00:00:00 2001 From: "Xin Li (Intel)" Date: Wed, 13 Nov 2024 09:59:34 -0800 Subject: [PATCH 05/16] x86/fred: Clear WFE in missing-ENDBRANCH #CPs An indirect branch instruction sets the CPU indirect branch tracker (IBT) into WAIT_FOR_ENDBRANCH (WFE) state and WFE stays asserted across the instruction boundary. When the decoder finds an inappropriate instruction while WFE is set ENDBR, the CPU raises a #CP fault. For the "kernel IBT no ENDBR" selftest where #CPs are deliberately triggered, the WFE state of the interrupted context needs to be cleared to let execution continue. Otherwise when the CPU resumes from the instruction that just caused the previous #CP, another missing-ENDBRANCH #CP is raised and the CPU enters a dead loop. This is not a problem with IDT because it doesn't preserve WFE and IRET doesn't set WFE. But FRED provides space on the entry stack (in an expanded CS area) to save and restore the WFE state, thus the WFE state is no longer clobbered, so software must clear it. Clear WFE to avoid dead looping in ibt_clear_fred_wfe() and the !ibt_fatal code path when execution is allowed to continue. Clobbering WFE in any other circumstance is a security-relevant bug. [ dhansen: changelog rewording ] Fixes: a5f6c2ace997 ("x86/shstk: Add user control-protection fault handler") Signed-off-by: Xin Li (Intel) Signed-off-by: Dave Hansen Signed-off-by: Ingo Molnar Acked-by: Dave Hansen Cc: stable@vger.kernel.org Link: https://lore.kernel.org/all/20241113175934.3897541-1-xin%40zytor.com --- arch/x86/kernel/cet.c | 30 ++++++++++++++++++++++++++++++ 1 file changed, 30 insertions(+) diff --git a/arch/x86/kernel/cet.c b/arch/x86/kernel/cet.c index d2c732a34e5d..303bf74d175b 100644 --- a/arch/x86/kernel/cet.c +++ b/arch/x86/kernel/cet.c @@ -81,6 +81,34 @@ static void do_user_cp_fault(struct pt_regs *regs, unsigned long error_code) static __ro_after_init bool ibt_fatal = true; +/* + * By definition, all missing-ENDBRANCH #CPs are a result of WFE && !ENDBR. + * + * For the kernel IBT no ENDBR selftest where #CPs are deliberately triggered, + * the WFE state of the interrupted context needs to be cleared to let execution + * continue. Otherwise when the CPU resumes from the instruction that just + * caused the previous #CP, another missing-ENDBRANCH #CP is raised and the CPU + * enters a dead loop. + * + * This is not a problem with IDT because it doesn't preserve WFE and IRET doesn't + * set WFE. But FRED provides space on the entry stack (in an expanded CS area) + * to save and restore the WFE state, thus the WFE state is no longer clobbered, + * so software must clear it. + */ +static void ibt_clear_fred_wfe(struct pt_regs *regs) +{ + /* + * No need to do any FRED checks. + * + * For IDT event delivery, the high-order 48 bits of CS are pushed + * as 0s into the stack, and later IRET ignores these bits. + * + * For FRED, a test to check if fred_cs.wfe is set would be dropped + * by compilers. + */ + regs->fred_cs.wfe = 0; +} + static void do_kernel_cp_fault(struct pt_regs *regs, unsigned long error_code) { if ((error_code & CP_EC) != CP_ENDBR) { @@ -90,6 +118,7 @@ static void do_kernel_cp_fault(struct pt_regs *regs, unsigned long error_code) if (unlikely(regs->ip == (unsigned long)&ibt_selftest_noendbr)) { regs->ax = 0; + ibt_clear_fred_wfe(regs); return; } @@ -97,6 +126,7 @@ static void do_kernel_cp_fault(struct pt_regs *regs, unsigned long error_code) if (!ibt_fatal) { printk(KERN_DEFAULT CUT_HERE); __warn(__FILE__, __LINE__, (void *)regs->ip, TAINT_WARN, regs, NULL); + ibt_clear_fred_wfe(regs); return; } BUG(); -- 2.51.0 From 27834971f616c5e154423c578fa95e0444444ce1 Mon Sep 17 00:00:00 2001 From: Li RongQing Date: Wed, 19 Jun 2024 19:18:01 +0800 Subject: [PATCH 06/16] virt: tdx-guest: Just leak decrypted memory on unrecoverable errors In CoCo VMs it is possible for the untrusted host to cause set_memory_decrypted() to fail such that an error is returned and the resulting memory is shared. Callers need to take care to handle these errors to avoid returning decrypted (shared) memory to the page allocator, which could lead to functional or security issues. Leak the decrypted memory when set_memory_decrypted() fails, and don't need to print an error since set_memory_decrypted() will call WARN_ONCE(). Fixes: f4738f56d1dc ("virt: tdx-guest: Add Quote generation support using TSM_REPORTS") Signed-off-by: Li RongQing Signed-off-by: Dave Hansen Signed-off-by: Ingo Molnar Reviewed-by: Rick Edgecombe Reviewed-by: Kirill A. Shutemov Cc: stable@vger.kernel.org Link: https://lore.kernel.org/all/20240619111801.25630-1-lirongqing%40baidu.com --- drivers/virt/coco/tdx-guest/tdx-guest.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/drivers/virt/coco/tdx-guest/tdx-guest.c b/drivers/virt/coco/tdx-guest/tdx-guest.c index d7db6c824e13..224e7dde9cde 100644 --- a/drivers/virt/coco/tdx-guest/tdx-guest.c +++ b/drivers/virt/coco/tdx-guest/tdx-guest.c @@ -124,10 +124,8 @@ static void *alloc_quote_buf(void) if (!addr) return NULL; - if (set_memory_decrypted((unsigned long)addr, count)) { - free_pages_exact(addr, len); + if (set_memory_decrypted((unsigned long)addr, count)) return NULL; - } return addr; } -- 2.51.0 From fc033cf25e612e840e545f8d5ad2edd6ba613ed5 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 29 Dec 2024 13:15:45 -0800 Subject: [PATCH 07/16] Linux 6.13-rc5 --- Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Makefile b/Makefile index 5c9b1d2d59b4..48e89108aa58 100644 --- a/Makefile +++ b/Makefile @@ -2,7 +2,7 @@ VERSION = 6 PATCHLEVEL = 13 SUBLEVEL = 0 -EXTRAVERSION = -rc4 +EXTRAVERSION = -rc5 NAME = Baby Opossum Posse # *DOCUMENTATION* -- 2.51.0 From f0ed39830e6064d62f9c5393505677a26569bb56 Mon Sep 17 00:00:00 2001 From: Umesh Nerlige Ramappa Date: Fri, 20 Dec 2024 09:19:18 -0800 Subject: [PATCH 08/16] xe/oa: Fix query mode of operation for OAR/OAC MIME-Version: 1.0 Content-Type: text/plain; charset=utf8 Content-Transfer-Encoding: 8bit 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 Reviewed-by: Matthew Brost # commit 1 Reviewed-by: Ashutosh Dixit Cc: stable@vger.kernel.org Reviewed-by: Jonathan Cavitt Signed-off-by: Ashutosh Dixit Link: https://patchwork.freedesktop.org/patch/msgid/20241220171919.571528-2-umesh.nerlige.ramappa@intel.com (cherry picked from commit 55039832f98c7e05f1cf9e0d8c12b2490abd0f16) Signed-off-by: Thomas Hellström --- drivers/gpu/drm/xe/xe_oa.c | 134 ++++++++---------------- drivers/gpu/drm/xe/xe_ring_ops.c | 5 +- drivers/gpu/drm/xe/xe_sched_job_types.h | 2 + 3 files changed, 51 insertions(+), 90 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_oa.c b/drivers/gpu/drm/xe/xe_oa.c index 8dd55798ab31..5cc0f6f9bc11 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; @@ -596,19 +590,38 @@ 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++) @@ -623,10 +636,13 @@ 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); } @@ -675,63 +691,19 @@ 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_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) -{ - struct dma_fence *fence; - struct xe_bb *bb; - int err; - - bb = xe_bb_new(stream->gt, 4 * count, 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) +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, 3, false); + bb = xe_bb_new(stream->gt, 2 * count + 1, 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)) { @@ -751,71 +723,55 @@ exit: 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, ®_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, ®_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) @@ -2066,8 +2022,8 @@ 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 0be4f489d3e1..9f327f27c072 100644 --- a/drivers/gpu/drm/xe/xe_ring_ops.c +++ b/drivers/gpu/drm/xe/xe_ring_ops.c @@ -221,7 +221,10 @@ 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[]; }; -- 2.51.0 From 5009628d8509dbb90e1b88e01eda00430fa24b4b Mon Sep 17 00:00:00 2001 From: Alex Hung Date: Tue, 17 Dec 2024 14:03:50 -0700 Subject: [PATCH 09/16] drm/amd/display: Remove unnecessary amdgpu_irq_get/put [WHY & HOW] commit 7fb363c57522 ("drm/amd/display: Let drm_crtc_vblank_on/off manage interrupts") lets drm_crtc_vblank_* to manage interrupts in amdgpu_dm_crtc_set_vblank, and amdgpu_irq_get/put do not need to be called here. Part of that patch got lost somehow, so fix it up. Fixes: 7fb363c57522 ("drm/amd/display: Let drm_crtc_vblank_on/off manage interrupts") Reviewed-by: Alex Deucher Reviewed-by: Leo Li Signed-off-by: Alex Hung Signed-off-by: Alex Deucher (cherry picked from commit 3782305ce5807c18fbf092124b9e8303cf1723ae) Cc: stable@vger.kernel.org --- .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 31 ------------------- 1 file changed, 31 deletions(-) diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c index 48be917e7bc5..4c3b6e6151c5 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -8400,16 +8400,6 @@ static void manage_dm_interrupts(struct amdgpu_device *adev, struct amdgpu_crtc *acrtc, struct dm_crtc_state *acrtc_state) { - /* - * We have no guarantee that the frontend index maps to the same - * backend index - some even map to more than one. - * - * TODO: Use a different interrupt or check DC itself for the mapping. - */ - int irq_type = - amdgpu_display_crtc_idx_to_irq_type( - adev, - acrtc->crtc_id); struct drm_vblank_crtc_config config = {0}; struct dc_crtc_timing *timing; int offdelay; @@ -8435,28 +8425,7 @@ static void manage_dm_interrupts(struct amdgpu_device *adev, drm_crtc_vblank_on_config(&acrtc->base, &config); - - amdgpu_irq_get( - adev, - &adev->pageflip_irq, - irq_type); -#if defined(CONFIG_DRM_AMD_SECURE_DISPLAY) - amdgpu_irq_get( - adev, - &adev->vline0_irq, - irq_type); -#endif } else { -#if defined(CONFIG_DRM_AMD_SECURE_DISPLAY) - amdgpu_irq_put( - adev, - &adev->vline0_irq, - irq_type); -#endif - amdgpu_irq_put( - adev, - &adev->pageflip_irq, - irq_type); drm_crtc_vblank_off(&acrtc->base); } } -- 2.51.0 From 7de8d5c90be9ad9f6575e818a674801db2ada794 Mon Sep 17 00:00:00 2001 From: Melissa Wen Date: Tue, 17 Dec 2024 17:45:03 -0300 Subject: [PATCH 10/16] drm/amd/display: fix page fault due to max surface definition mismatch DC driver is using two different values to define the maximum number of surfaces: MAX_SURFACES and MAX_SURFACE_NUM. Consolidate MAX_SURFACES as the unique definition for surface updates across DC. It fixes page fault faced by Cosmic users on AMD display versions that support two overlay planes, since the introduction of cursor overlay mode. [Nov26 21:33] BUG: unable to handle page fault for address: 0000000051d0f08b [ +0.000015] #PF: supervisor read access in kernel mode [ +0.000006] #PF: error_code(0x0000) - not-present page [ +0.000005] PGD 0 P4D 0 [ +0.000007] Oops: Oops: 0000 [#1] PREEMPT SMP NOPTI [ +0.000006] CPU: 4 PID: 71 Comm: kworker/u32:6 Not tainted 6.10.0+ #300 [ +0.000006] Hardware name: Valve Jupiter/Jupiter, BIOS F7A0131 01/30/2024 [ +0.000007] Workqueue: events_unbound commit_work [drm_kms_helper] [ +0.000040] RIP: 0010:copy_stream_update_to_stream.isra.0+0x30d/0x750 [amdgpu] [ +0.000847] Code: 8b 10 49 89 94 24 f8 00 00 00 48 8b 50 08 49 89 94 24 00 01 00 00 8b 40 10 41 89 84 24 08 01 00 00 49 8b 45 78 48 85 c0 74 0b <0f> b6 00 41 88 84 24 90 64 00 00 49 8b 45 60 48 85 c0 74 3b 48 8b [ +0.000010] RSP: 0018:ffffc203802f79a0 EFLAGS: 00010206 [ +0.000009] RAX: 0000000051d0f08b RBX: 0000000000000004 RCX: ffff9f964f0a8070 [ +0.000004] RDX: ffff9f9710f90e40 RSI: ffff9f96600c8000 RDI: ffff9f964f000000 [ +0.000004] RBP: ffffc203802f79f8 R08: 0000000000000000 R09: 0000000000000000 [ +0.000005] R10: 0000000000000000 R11: 0000000000000000 R12: ffff9f96600c8000 [ +0.000004] R13: ffff9f9710f90e40 R14: ffff9f964f000000 R15: ffff9f96600c8000 [ +0.000004] FS: 0000000000000000(0000) GS:ffff9f9970000000(0000) knlGS:0000000000000000 [ +0.000005] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ +0.000005] CR2: 0000000051d0f08b CR3: 00000002e6a20000 CR4: 0000000000350ef0 [ +0.000005] Call Trace: [ +0.000011] [ +0.000010] ? __die_body.cold+0x19/0x27 [ +0.000012] ? page_fault_oops+0x15a/0x2d0 [ +0.000014] ? exc_page_fault+0x7e/0x180 [ +0.000009] ? asm_exc_page_fault+0x26/0x30 [ +0.000013] ? copy_stream_update_to_stream.isra.0+0x30d/0x750 [amdgpu] [ +0.000739] ? dc_commit_state_no_check+0xd6c/0xe70 [amdgpu] [ +0.000470] update_planes_and_stream_state+0x49b/0x4f0 [amdgpu] [ +0.000450] ? srso_return_thunk+0x5/0x5f [ +0.000009] ? commit_minimal_transition_state+0x239/0x3d0 [amdgpu] [ +0.000446] update_planes_and_stream_v2+0x24a/0x590 [amdgpu] [ +0.000464] ? srso_return_thunk+0x5/0x5f [ +0.000009] ? sort+0x31/0x50 [ +0.000007] ? amdgpu_dm_atomic_commit_tail+0x159f/0x3a30 [amdgpu] [ +0.000508] ? srso_return_thunk+0x5/0x5f [ +0.000009] ? amdgpu_crtc_get_scanout_position+0x28/0x40 [amdgpu] [ +0.000377] ? srso_return_thunk+0x5/0x5f [ +0.000009] ? drm_crtc_vblank_helper_get_vblank_timestamp_internal+0x160/0x390 [drm] [ +0.000058] ? srso_return_thunk+0x5/0x5f [ +0.000005] ? dma_fence_default_wait+0x8c/0x260 [ +0.000010] ? srso_return_thunk+0x5/0x5f [ +0.000005] ? wait_for_completion_timeout+0x13b/0x170 [ +0.000006] ? srso_return_thunk+0x5/0x5f [ +0.000005] ? dma_fence_wait_timeout+0x108/0x140 [ +0.000010] ? commit_tail+0x94/0x130 [drm_kms_helper] [ +0.000024] ? process_one_work+0x177/0x330 [ +0.000008] ? worker_thread+0x266/0x3a0 [ +0.000006] ? __pfx_worker_thread+0x10/0x10 [ +0.000004] ? kthread+0xd2/0x100 [ +0.000006] ? __pfx_kthread+0x10/0x10 [ +0.000006] ? ret_from_fork+0x34/0x50 [ +0.000004] ? __pfx_kthread+0x10/0x10 [ +0.000005] ? ret_from_fork_asm+0x1a/0x30 [ +0.000011] Fixes: 1b04dcca4fb1 ("drm/amd/display: Introduce overlay cursor mode") Suggested-by: Leo Li Link: https://gitlab.freedesktop.org/drm/amd/-/issues/3693 Signed-off-by: Melissa Wen Reviewed-by: Rodrigo Siqueira Signed-off-by: Rodrigo Siqueira Signed-off-by: Alex Deucher (cherry picked from commit 1c86c81a86c60f9b15d3e3f43af0363cf56063e7) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/display/dc/core/dc.c | 2 +- drivers/gpu/drm/amd/display/dc/core/dc_state.c | 8 ++++---- drivers/gpu/drm/amd/display/dc/dc.h | 2 +- drivers/gpu/drm/amd/display/dc/dc_stream.h | 2 +- drivers/gpu/drm/amd/display/dc/dc_types.h | 1 - drivers/gpu/drm/amd/display/dc/dml2/dml2_mall_phantom.c | 2 +- 6 files changed, 8 insertions(+), 9 deletions(-) diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c index 49fe7dcf9372..2723558049d6 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc.c @@ -4510,7 +4510,7 @@ static bool commit_minimal_transition_based_on_current_context(struct dc *dc, struct pipe_split_policy_backup policy; struct dc_state *intermediate_context; struct dc_state *old_current_state = dc->current_state; - struct dc_surface_update srf_updates[MAX_SURFACE_NUM] = {0}; + struct dc_surface_update srf_updates[MAX_SURFACES] = {0}; int surface_count; /* diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_state.c b/drivers/gpu/drm/amd/display/dc/core/dc_state.c index e006f816ff2f..1b2cce127981 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_state.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_state.c @@ -483,9 +483,9 @@ bool dc_state_add_plane( if (stream_status == NULL) { dm_error("Existing stream not found; failed to attach surface!\n"); goto out; - } else if (stream_status->plane_count == MAX_SURFACE_NUM) { + } else if (stream_status->plane_count == MAX_SURFACES) { dm_error("Surface: can not attach plane_state %p! Maximum is: %d\n", - plane_state, MAX_SURFACE_NUM); + plane_state, MAX_SURFACES); goto out; } else if (!otg_master_pipe) { goto out; @@ -600,7 +600,7 @@ bool dc_state_rem_all_planes_for_stream( { int i, old_plane_count; struct dc_stream_status *stream_status = NULL; - struct dc_plane_state *del_planes[MAX_SURFACE_NUM] = { 0 }; + struct dc_plane_state *del_planes[MAX_SURFACES] = { 0 }; for (i = 0; i < state->stream_count; i++) if (state->streams[i] == stream) { @@ -875,7 +875,7 @@ bool dc_state_rem_all_phantom_planes_for_stream( { int i, old_plane_count; struct dc_stream_status *stream_status = NULL; - struct dc_plane_state *del_planes[MAX_SURFACE_NUM] = { 0 }; + struct dc_plane_state *del_planes[MAX_SURFACES] = { 0 }; for (i = 0; i < state->stream_count; i++) if (state->streams[i] == phantom_stream) { diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h index e9b9126c0401..1341a6ebd262 100644 --- a/drivers/gpu/drm/amd/display/dc/dc.h +++ b/drivers/gpu/drm/amd/display/dc/dc.h @@ -1398,7 +1398,7 @@ struct dc_scratch_space { * store current value in plane states so we can still recover * a valid current state during dc update. */ - struct dc_plane_state plane_states[MAX_SURFACE_NUM]; + struct dc_plane_state plane_states[MAX_SURFACES]; struct dc_stream_state stream_state; }; diff --git a/drivers/gpu/drm/amd/display/dc/dc_stream.h b/drivers/gpu/drm/amd/display/dc/dc_stream.h index 413970588a26..860506c6bda4 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_stream.h +++ b/drivers/gpu/drm/amd/display/dc/dc_stream.h @@ -56,7 +56,7 @@ struct dc_stream_status { int plane_count; int audio_inst; struct timing_sync_info timing_sync_info; - struct dc_plane_state *plane_states[MAX_SURFACE_NUM]; + struct dc_plane_state *plane_states[MAX_SURFACES]; bool is_abm_supported; struct mall_stream_config mall_stream_config; bool fpo_in_use; diff --git a/drivers/gpu/drm/amd/display/dc/dc_types.h b/drivers/gpu/drm/amd/display/dc/dc_types.h index edf4df1d03b5..9466b63644d5 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_types.h @@ -76,7 +76,6 @@ struct dc_perf_trace { unsigned long last_entry_write; }; -#define MAX_SURFACE_NUM 6 #define NUM_PIXEL_FORMATS 10 enum tiling_mode { diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml2_mall_phantom.c b/drivers/gpu/drm/amd/display/dc/dml2/dml2_mall_phantom.c index 3d29169dd6bb..6b3b8803e0ae 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml2_mall_phantom.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml2_mall_phantom.c @@ -813,7 +813,7 @@ static bool remove_all_phantom_planes_for_stream(struct dml2_context *ctx, struc { int i, old_plane_count; struct dc_stream_status *stream_status = NULL; - struct dc_plane_state *del_planes[MAX_SURFACE_NUM] = { 0 }; + struct dc_plane_state *del_planes[MAX_SURFACES] = { 0 }; for (i = 0; i < context->stream_count; i++) if (context->streams[i] == stream) { -- 2.51.0 From 21541bc6b44241e3f791f9e552352d8440b2b29e Mon Sep 17 00:00:00 2001 From: Melissa Wen Date: Tue, 17 Dec 2024 17:45:04 -0300 Subject: [PATCH 11/16] drm/amd/display: increase MAX_SURFACES to the value supported by hw As the hw supports up to 4 surfaces, increase the maximum number of surfaces to prevent the DC error when trying to use more than three planes. [drm:dc_state_add_plane [amdgpu]] *ERROR* Surface: can not attach plane_state 000000003e2cb82c! Maximum is: 3 Link: https://gitlab.freedesktop.org/drm/amd/-/issues/3693 Signed-off-by: Melissa Wen Reviewed-by: Rodrigo Siqueira Signed-off-by: Rodrigo Siqueira Signed-off-by: Alex Deucher (cherry picked from commit b8d6daffc871a42026c3c20bff7b8fa0302298c1) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/display/dc/dc.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h index 1341a6ebd262..08c5a315b3a6 100644 --- a/drivers/gpu/drm/amd/display/dc/dc.h +++ b/drivers/gpu/drm/amd/display/dc/dc.h @@ -57,7 +57,7 @@ struct dmub_notification; #define DC_VER "3.2.310" -#define MAX_SURFACES 3 +#define MAX_SURFACES 4 #define MAX_PLANES 6 #define MAX_STREAMS 6 #define MIN_VIEWPORT_SIZE 12 -- 2.51.0 From 5225fd2a26211d012533acf98a6ad3f983885817 Mon Sep 17 00:00:00 2001 From: Melissa Wen Date: Tue, 17 Dec 2024 17:45:05 -0300 Subject: [PATCH 12/16] drm/amd/display: fix divide error in DM plane scale calcs dm_get_plane_scale doesn't take into account plane scaled size equal to zero, leading to a kernel oops due to division by zero. Fix by setting out-scale size as zero when the dst size is zero, similar to what is done by drm_calc_scale(). This issue started with the introduction of cursor ovelay mode that uses this function to assess cursor mode changes via dm_crtc_get_cursor_mode() before checking plane state. [Dec17 17:14] Oops: divide error: 0000 [#1] PREEMPT SMP NOPTI [ +0.000018] CPU: 5 PID: 1660 Comm: surface-DP-1 Not tainted 6.10.0+ #231 [ +0.000007] Hardware name: Valve Jupiter/Jupiter, BIOS F7A0131 01/30/2024 [ +0.000004] RIP: 0010:dm_get_plane_scale+0x3f/0x60 [amdgpu] [ +0.000553] Code: 44 0f b7 41 3a 44 0f b7 49 3e 83 e0 0f 48 0f a3 c2 73 21 69 41 28 e8 03 00 00 31 d2 41 f7 f1 31 d2 89 06 69 41 2c e8 03 00 00 <41> f7 f0 89 07 e9 d7 d8 7e e9 44 89 c8 45 89 c1 41 89 c0 eb d4 66 [ +0.000005] RSP: 0018:ffffa8df0de6b8a0 EFLAGS: 00010246 [ +0.000006] RAX: 00000000000003e8 RBX: ffff9ac65c1f6e00 RCX: ffff9ac65d055500 [ +0.000003] RDX: 0000000000000000 RSI: ffffa8df0de6b8b0 RDI: ffffa8df0de6b8b4 [ +0.000004] RBP: ffff9ac64e7a5800 R08: 0000000000000000 R09: 0000000000000a00 [ +0.000003] R10: 00000000000000ff R11: 0000000000000054 R12: ffff9ac6d0700010 [ +0.000003] R13: ffff9ac65d054f00 R14: ffff9ac65d055500 R15: ffff9ac64e7a60a0 [ +0.000004] FS: 00007f869ea00640(0000) GS:ffff9ac970080000(0000) knlGS:0000000000000000 [ +0.000004] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ +0.000003] CR2: 000055ca701becd0 CR3: 000000010e7f2000 CR4: 0000000000350ef0 [ +0.000004] Call Trace: [ +0.000007] [ +0.000006] ? __die_body.cold+0x19/0x27 [ +0.000009] ? die+0x2e/0x50 [ +0.000007] ? do_trap+0xca/0x110 [ +0.000007] ? do_error_trap+0x6a/0x90 [ +0.000006] ? dm_get_plane_scale+0x3f/0x60 [amdgpu] [ +0.000504] ? exc_divide_error+0x38/0x50 [ +0.000005] ? dm_get_plane_scale+0x3f/0x60 [amdgpu] [ +0.000488] ? asm_exc_divide_error+0x1a/0x20 [ +0.000011] ? dm_get_plane_scale+0x3f/0x60 [amdgpu] [ +0.000593] dm_crtc_get_cursor_mode+0x33f/0x430 [amdgpu] [ +0.000562] amdgpu_dm_atomic_check+0x2ef/0x1770 [amdgpu] [ +0.000501] drm_atomic_check_only+0x5e1/0xa30 [drm] [ +0.000047] drm_mode_atomic_ioctl+0x832/0xcb0 [drm] [ +0.000050] ? __pfx_drm_mode_atomic_ioctl+0x10/0x10 [drm] [ +0.000047] drm_ioctl_kernel+0xb3/0x100 [drm] [ +0.000062] drm_ioctl+0x27a/0x4f0 [drm] [ +0.000049] ? __pfx_drm_mode_atomic_ioctl+0x10/0x10 [drm] [ +0.000055] amdgpu_drm_ioctl+0x4e/0x90 [amdgpu] [ +0.000360] __x64_sys_ioctl+0x97/0xd0 [ +0.000010] do_syscall_64+0x82/0x190 [ +0.000008] ? __pfx_drm_mode_createblob_ioctl+0x10/0x10 [drm] [ +0.000044] ? srso_return_thunk+0x5/0x5f [ +0.000006] ? drm_ioctl_kernel+0xb3/0x100 [drm] [ +0.000040] ? srso_return_thunk+0x5/0x5f [ +0.000005] ? __check_object_size+0x50/0x220 [ +0.000007] ? srso_return_thunk+0x5/0x5f [ +0.000005] ? srso_return_thunk+0x5/0x5f [ +0.000005] ? drm_ioctl+0x2a4/0x4f0 [drm] [ +0.000039] ? __pfx_drm_mode_createblob_ioctl+0x10/0x10 [drm] [ +0.000043] ? srso_return_thunk+0x5/0x5f [ +0.000005] ? srso_return_thunk+0x5/0x5f [ +0.000005] ? __pm_runtime_suspend+0x69/0xc0 [ +0.000006] ? srso_return_thunk+0x5/0x5f [ +0.000005] ? amdgpu_drm_ioctl+0x71/0x90 [amdgpu] [ +0.000366] ? srso_return_thunk+0x5/0x5f [ +0.000006] ? syscall_exit_to_user_mode+0x77/0x210 [ +0.000007] ? srso_return_thunk+0x5/0x5f [ +0.000005] ? do_syscall_64+0x8e/0x190 [ +0.000006] ? srso_return_thunk+0x5/0x5f [ +0.000006] ? do_syscall_64+0x8e/0x190 [ +0.000006] ? srso_return_thunk+0x5/0x5f [ +0.000007] entry_SYSCALL_64_after_hwframe+0x76/0x7e [ +0.000008] RIP: 0033:0x55bb7cd962bc [ +0.000007] Code: 4c 89 6c 24 18 4c 89 64 24 20 4c 89 74 24 28 0f 57 c0 0f 11 44 24 30 89 c7 48 8d 54 24 08 b8 10 00 00 00 be bc 64 38 c0 0f 05 <49> 89 c7 48 83 3b 00 74 09 4c 89 c7 ff 15 62 64 99 00 48 83 7b 18 [ +0.000005] RSP: 002b:00007f869e9f4da0 EFLAGS: 00000217 ORIG_RAX: 0000000000000010 [ +0.000007] RAX: ffffffffffffffda RBX: 00007f869e9f4fb8 RCX: 000055bb7cd962bc [ +0.000004] RDX: 00007f869e9f4da8 RSI: 00000000c03864bc RDI: 000000000000003b [ +0.000003] RBP: 000055bb9ddcbcc0 R08: 00007f86541b9920 R09: 0000000000000009 [ +0.000004] R10: 0000000000000004 R11: 0000000000000217 R12: 00007f865406c6b0 [ +0.000003] R13: 00007f86541b5290 R14: 00007f865410b700 R15: 000055bb9ddcbc18 [ +0.000009] Fixes: 1b04dcca4fb1 ("drm/amd/display: Introduce overlay cursor mode") Link: https://gitlab.freedesktop.org/drm/amd/-/issues/3729 Reported-by: Fabio Scaccabarozzi Co-developed-by: Fabio Scaccabarozzi Signed-off-by: Fabio Scaccabarozzi Signed-off-by: Melissa Wen Reviewed-by: Rodrigo Siqueira Signed-off-by: Rodrigo Siqueira Signed-off-by: Alex Deucher (cherry picked from commit ab75a0d2e07942ae15d32c0a5092fd336451378c) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c index 4c3b6e6151c5..cd16dae534dc 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -11124,8 +11124,8 @@ dm_get_plane_scale(struct drm_plane_state *plane_state, int plane_src_w, plane_src_h; dm_get_oriented_plane_size(plane_state, &plane_src_w, &plane_src_h); - *out_plane_scale_w = plane_state->crtc_w * 1000 / plane_src_w; - *out_plane_scale_h = plane_state->crtc_h * 1000 / plane_src_h; + *out_plane_scale_w = plane_src_w ? plane_state->crtc_w * 1000 / plane_src_w : 0; + *out_plane_scale_h = plane_src_h ? plane_state->crtc_h * 1000 / plane_src_h : 0; } /* -- 2.51.0 From 9738609449c3e44d1afb73eecab4763362b57930 Mon Sep 17 00:00:00 2001 From: "Jesse.zhang@amd.com" Date: Wed, 18 Dec 2024 18:23:52 +0800 Subject: [PATCH 13/16] drm/amdkfd: fixed page fault when enable MES shader debugger Initialize the process context address before setting the shader debugger. [ 260.781212] amdgpu 0000:03:00.0: amdgpu: [gfxhub] page fault (src_id:0 ring:32 vmid:0 pasid:0) [ 260.781236] amdgpu 0000:03:00.0: amdgpu: in page starting at address 0x0000000000000000 from client 10 [ 260.781255] amdgpu 0000:03:00.0: amdgpu: GCVM_L2_PROTECTION_FAULT_STATUS:0x00040A40 [ 260.781270] amdgpu 0000:03:00.0: amdgpu: Faulty UTCL2 client ID: CPC (0x5) [ 260.781284] amdgpu 0000:03:00.0: amdgpu: MORE_FAULTS: 0x0 [ 260.781296] amdgpu 0000:03:00.0: amdgpu: WALKER_ERROR: 0x0 [ 260.781308] amdgpu 0000:03:00.0: amdgpu: PERMISSION_FAULTS: 0x4 [ 260.781320] amdgpu 0000:03:00.0: amdgpu: MAPPING_ERROR: 0x0 [ 260.781332] amdgpu 0000:03:00.0: amdgpu: RW: 0x1 [ 260.782017] amdgpu 0000:03:00.0: amdgpu: [gfxhub] page fault (src_id:0 ring:32 vmid:0 pasid:0) [ 260.782039] amdgpu 0000:03:00.0: amdgpu: in page starting at address 0x0000000000000000 from client 10 [ 260.782058] amdgpu 0000:03:00.0: amdgpu: GCVM_L2_PROTECTION_FAULT_STATUS:0x00040A41 [ 260.782073] amdgpu 0000:03:00.0: amdgpu: Faulty UTCL2 client ID: CPC (0x5) [ 260.782087] amdgpu 0000:03:00.0: amdgpu: MORE_FAULTS: 0x1 [ 260.782098] amdgpu 0000:03:00.0: amdgpu: WALKER_ERROR: 0x0 [ 260.782110] amdgpu 0000:03:00.0: amdgpu: PERMISSION_FAULTS: 0x4 [ 260.782122] amdgpu 0000:03:00.0: amdgpu: MAPPING_ERROR: 0x0 [ 260.782137] amdgpu 0000:03:00.0: amdgpu: RW: 0x1 [ 260.782155] amdgpu 0000:03:00.0: amdgpu: [gfxhub] page fault (src_id:0 ring:32 vmid:0 pasid:0) [ 260.782166] amdgpu 0000:03:00.0: amdgpu: in page starting at address 0x0000000000000000 from client 10 Fixes: 438b39ac74e2 ("drm/amdkfd: pause autosuspend when creating pdd") Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3849 Signed-off-by: Jesse Zhang Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher (cherry picked from commit 5b231f5bc9ff02ec5737f2ec95cdf15ac95088e9) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/amdkfd/kfd_debug.c | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c index 312dfa84f29f..a8abc3091801 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c @@ -350,10 +350,27 @@ int kfd_dbg_set_mes_debug_mode(struct kfd_process_device *pdd, bool sq_trap_en) { uint32_t spi_dbg_cntl = pdd->spi_dbg_override | pdd->spi_dbg_launch_mode; uint32_t flags = pdd->process->dbg_flags; + struct amdgpu_device *adev = pdd->dev->adev; + int r; if (!kfd_dbg_is_per_vmid_supported(pdd->dev)) return 0; + if (!pdd->proc_ctx_cpu_ptr) { + r = amdgpu_amdkfd_alloc_gtt_mem(adev, + AMDGPU_MES_PROC_CTX_SIZE, + &pdd->proc_ctx_bo, + &pdd->proc_ctx_gpu_addr, + &pdd->proc_ctx_cpu_ptr, + false); + if (r) { + dev_err(adev->dev, + "failed to allocate process context bo\n"); + return r; + } + memset(pdd->proc_ctx_cpu_ptr, 0, AMDGPU_MES_PROC_CTX_SIZE); + } + return amdgpu_mes_set_shader_debugger(pdd->dev->adev, pdd->proc_ctx_gpu_addr, spi_dbg_cntl, pdd->watch_points, flags, sq_trap_en); } -- 2.51.0 From 0881fbc4fd62e00a2b8e102725f76d10351b2ea8 Mon Sep 17 00:00:00 2001 From: Roman Li Date: Fri, 13 Dec 2024 13:51:07 -0500 Subject: [PATCH 14/16] drm/amd/display: Add check for granularity in dml ceil/floor helpers [Why] Wrapper functions for dcn_bw_ceil2() and dcn_bw_floor2() should check for granularity is non zero to avoid assert and divide-by-zero error in dcn_bw_ functions. [How] Add check for granularity 0. Cc: Mario Limonciello Reviewed-by: Alvin Lee Signed-off-by: Roman Li Tested-by: Daniel Wheeler Signed-off-by: Alex Deucher (cherry picked from commit f6e09701c3eb2ccb8cb0518e0b67f1c69742a4ec) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/display/dc/dml/dml_inline_defs.h | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/drivers/gpu/drm/amd/display/dc/dml/dml_inline_defs.h b/drivers/gpu/drm/amd/display/dc/dml/dml_inline_defs.h index 072bd0539605..6b2ab4ec2b5f 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dml_inline_defs.h +++ b/drivers/gpu/drm/amd/display/dc/dml/dml_inline_defs.h @@ -66,11 +66,15 @@ static inline double dml_max5(double a, double b, double c, double d, double e) static inline double dml_ceil(double a, double granularity) { + if (granularity == 0) + return 0; return (double) dcn_bw_ceil2(a, granularity); } static inline double dml_floor(double a, double granularity) { + if (granularity == 0) + return 0; return (double) dcn_bw_floor2(a, granularity); } @@ -114,11 +118,15 @@ static inline double dml_ceil_2(double f) static inline double dml_ceil_ex(double x, double granularity) { + if (granularity == 0) + return 0; return (double) dcn_bw_ceil2(x, granularity); } static inline double dml_floor_ex(double x, double granularity) { + if (granularity == 0) + return 0; return (double) dcn_bw_floor2(x, granularity); } -- 2.51.0 From a993d319aebb7cce8a10c6e685344b7c2ad5c4c2 Mon Sep 17 00:00:00 2001 From: Zhu Lingshan Date: Wed, 11 Dec 2024 11:51:13 +0800 Subject: [PATCH 15/16] drm/amdkfd: wq_release signals dma_fence only when available kfd_process_wq_release() signals eviction fence by dma_fence_signal() which wanrs if dma_fence is NULL. kfd_process->ef is initialized by kfd_process_device_init_vm() through ioctl. That means the fence is NULL for a new created kfd_process, and close a kfd_process right after open it will trigger the warning. This commit conditionally signals the eviction fence in kfd_process_wq_release() only when it is available. [ 503.660882] WARNING: CPU: 0 PID: 9 at drivers/dma-buf/dma-fence.c:467 dma_fence_signal+0x74/0xa0 [ 503.782940] Workqueue: kfd_process_wq kfd_process_wq_release [amdgpu] [ 503.789640] RIP: 0010:dma_fence_signal+0x74/0xa0 [ 503.877620] Call Trace: [ 503.880066] [ 503.882168] ? __warn+0xcd/0x260 [ 503.885407] ? dma_fence_signal+0x74/0xa0 [ 503.889416] ? report_bug+0x288/0x2d0 [ 503.893089] ? handle_bug+0x53/0xa0 [ 503.896587] ? exc_invalid_op+0x14/0x50 [ 503.900424] ? asm_exc_invalid_op+0x16/0x20 [ 503.904616] ? dma_fence_signal+0x74/0xa0 [ 503.908626] kfd_process_wq_release+0x6b/0x370 [amdgpu] [ 503.914081] process_one_work+0x654/0x10a0 [ 503.918186] worker_thread+0x6c3/0xe70 [ 503.921943] ? srso_alias_return_thunk+0x5/0xfbef5 [ 503.926735] ? srso_alias_return_thunk+0x5/0xfbef5 [ 503.931527] ? __kthread_parkme+0x82/0x140 [ 503.935631] ? __pfx_worker_thread+0x10/0x10 [ 503.939904] kthread+0x2a8/0x380 [ 503.943132] ? __pfx_kthread+0x10/0x10 [ 503.946882] ret_from_fork+0x2d/0x70 [ 503.950458] ? __pfx_kthread+0x10/0x10 [ 503.954210] ret_from_fork_asm+0x1a/0x30 [ 503.958142] [ 503.960328] ---[ end trace 0000000000000000 ]--- Fixes: 967d226eaae8 ("dma-buf: add WARN_ON() illegal dma-fence signaling") Signed-off-by: Zhu Lingshan Reviewed-by: Felix Kuehling Signed-off-by: Alex Deucher (cherry picked from commit 2774ef7625adb5fb9e9265c26a59dca7b8fd171e) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/amdkfd/kfd_process.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c index d0ee173acf82..edfe0b4788f4 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c @@ -1160,7 +1160,8 @@ static void kfd_process_wq_release(struct work_struct *work) */ synchronize_rcu(); ef = rcu_access_pointer(p->ef); - dma_fence_signal(ef); + if (ef) + dma_fence_signal(ef); kfd_process_remove_sysfs(p); -- 2.51.0 From 2a238b09bfd04e8155a7a323364bce1c38b28c0f Mon Sep 17 00:00:00 2001 From: Kun Liu Date: Fri, 27 Dec 2024 11:43:22 +0800 Subject: [PATCH 16/16] drm/amd/pm: fix BUG: scheduling while atomic atomic scheduling will be triggered in interrupt handler for AC/DC mode switch as following backtrace. Call Trace: dump_stack_lvl __schedule_bug __schedule schedule schedule_preempt_disabled __mutex_lock smu_cmn_send_smc_msg_with_param smu_v13_0_irq_process amdgpu_irq_dispatch amdgpu_ih_process amdgpu_irq_handler __handle_irq_event_percpu handle_irq_event handle_edge_irq __common_interrupt common_interrupt asm_common_interrupt Reviewed-by: Lijo Lazar Reviewed-by: Kenneth Feng Signed-off-by: Kun Liu Signed-off-by: Alex Deucher (cherry picked from commit 03cc84b102d1a832e8dfc59344346dedcebcdf42) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h | 2 ++ drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c | 12 ++++++------ drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c | 1 + drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c | 1 + 4 files changed, 10 insertions(+), 6 deletions(-) diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h index ae3563d71fa0..356d9422b411 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h @@ -303,5 +303,7 @@ int smu_v13_0_set_wbrf_exclusion_ranges(struct smu_context *smu, int smu_v13_0_get_boot_freq_by_index(struct smu_context *smu, enum smu_clk_type clk_type, uint32_t *value); + +void smu_v13_0_interrupt_work(struct smu_context *smu); #endif #endif diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c index 2bfea740dace..2d1e7ebd1bac 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c @@ -1320,11 +1320,11 @@ static int smu_v13_0_set_irq_state(struct amdgpu_device *adev, return 0; } -static int smu_v13_0_ack_ac_dc_interrupt(struct smu_context *smu) +void smu_v13_0_interrupt_work(struct smu_context *smu) { - return smu_cmn_send_smc_msg(smu, - SMU_MSG_ReenableAcDcInterrupt, - NULL); + smu_cmn_send_smc_msg(smu, + SMU_MSG_ReenableAcDcInterrupt, + NULL); } #define THM_11_0__SRCID__THM_DIG_THERM_L2H 0 /* ASIC_TEMP > CG_THERMAL_INT.DIG_THERM_INTH */ @@ -1377,12 +1377,12 @@ static int smu_v13_0_irq_process(struct amdgpu_device *adev, switch (ctxid) { case SMU_IH_INTERRUPT_CONTEXT_ID_AC: dev_dbg(adev->dev, "Switched to AC mode!\n"); - smu_v13_0_ack_ac_dc_interrupt(smu); + schedule_work(&smu->interrupt_work); adev->pm.ac_power = true; break; case SMU_IH_INTERRUPT_CONTEXT_ID_DC: dev_dbg(adev->dev, "Switched to DC mode!\n"); - smu_v13_0_ack_ac_dc_interrupt(smu); + schedule_work(&smu->interrupt_work); adev->pm.ac_power = false; break; case SMU_IH_INTERRUPT_CONTEXT_ID_THERMAL_THROTTLING: diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c index 3aa705aae4c0..2a09b27788e8 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c @@ -3219,6 +3219,7 @@ static const struct pptable_funcs smu_v13_0_0_ppt_funcs = { .is_asic_wbrf_supported = smu_v13_0_0_wbrf_support_check, .enable_uclk_shadow = smu_v13_0_enable_uclk_shadow, .set_wbrf_exclusion_ranges = smu_v13_0_set_wbrf_exclusion_ranges, + .interrupt_work = smu_v13_0_interrupt_work, }; void smu_v13_0_0_set_ppt_funcs(struct smu_context *smu) diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c index aabb94796005..55ef18517b0f 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c @@ -2797,6 +2797,7 @@ static const struct pptable_funcs smu_v13_0_7_ppt_funcs = { .is_asic_wbrf_supported = smu_v13_0_7_wbrf_support_check, .enable_uclk_shadow = smu_v13_0_enable_uclk_shadow, .set_wbrf_exclusion_ranges = smu_v13_0_set_wbrf_exclusion_ranges, + .interrupt_work = smu_v13_0_interrupt_work, }; void smu_v13_0_7_set_ppt_funcs(struct smu_context *smu) -- 2.51.0