March 2024 - Linux-stable-mirror

FAILED: Patch "drm/amd/display: set odm_combine_policy based on context in dcn32 resource" failed to apply to 6.6-stable tree

by Sasha Levin

The patch below does not apply to the 6.6-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>. Thanks, Sasha ------------------ original commit in Linus's tree ------------------ From 0a5fd7811a17af708cefdaab93af86838353002d Mon Sep 17 00:00:00 2001 From: Wenjing Liu <wenjing.liu(a)amd.com> Date: Thu, 18 Jan 2024 18:12:15 -0500 Subject: [PATCH] drm/amd/display: set odm_combine_policy based on context in dcn32 resource [why] When populating dml pipes, odm combine policy should be assigned based on the pipe topology of the context passed in. DML pipes could be repopulated multiple times during single validate bandwidth attempt. We need to make sure that whenever we repopulate the dml pipes it is always aligned with the updated context. There is a case where DML pipes get repopulated during FPO optimization after ODM combine policy is changed. Since in the current code we reinitlaize ODM combine policy, even though the current context has ODM combine enabled, we overwrite it despite the pipes are already split. This causes DML to think that MPC combine is used so we mistakenly enable MPC combine because we apply pipe split with ODM combine policy reset. This issue doesn't impact non windowed MPO with ODM case because the legacy policy has restricted use cases. We don't encounter the case where both ODM and FPO optimizations are enabled together. So we decide to leave it as is because it is about to be replaced anyway. Cc: stable(a)vger.kernel.org # 6.6+ Reviewed-by: Chaitanya Dhere <chaitanya.dhere(a)amd.com> Reviewed-by: Alvin Lee <alvin.lee2(a)amd.com> Acked-by: Hamza Mahfooz <hamza.mahfooz(a)amd.com> Signed-off-by: Wenjing Liu <wenjing.liu(a)amd.com> Signed-off-by: Alex Deucher <alexander.deucher(a)amd.com> --- .../drm/amd/display/dc/dml/dcn32/dcn32_fpu.c | 15 ++++++++++---- drivers/gpu/drm/amd/display/dc/inc/resource.h | 20 ++++++++----------- .../dc/resource/dcn32/dcn32_resource.c | 16 ++++++++++++++- 3 files changed, 34 insertions(+), 17 deletions(-) diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c index a7981a0c4158f..4edf7df4c6aad 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c @@ -1289,7 +1289,7 @@ static bool update_pipes_with_split_flags(struct dc *dc, struct dc_state *contex return updated; } -static bool should_allow_odm_power_optimization(struct dc *dc, +static bool should_apply_odm_power_optimization(struct dc *dc, struct dc_state *context, struct vba_vars_st *v, int *split, bool *merge) { @@ -1393,9 +1393,12 @@ static void try_odm_power_optimization_and_revalidate( { int i; unsigned int new_vlevel; + unsigned int cur_policy[MAX_PIPES]; - for (i = 0; i < pipe_cnt; i++) + for (i = 0; i < pipe_cnt; i++) { + cur_policy[i] = pipes[i].pipe.dest.odm_combine_policy; pipes[i].pipe.dest.odm_combine_policy = dm_odm_combine_policy_2to1; + } new_vlevel = dml_get_voltage_level(&context->bw_ctx.dml, pipes, pipe_cnt); @@ -1404,6 +1407,9 @@ static void try_odm_power_optimization_and_revalidate( memset(merge, 0, MAX_PIPES * sizeof(bool)); *vlevel = dcn20_validate_apply_pipe_split_flags(dc, context, new_vlevel, split, merge); context->bw_ctx.dml.vba.VoltageLevel = *vlevel; + } else { + for (i = 0; i < pipe_cnt; i++) + pipes[i].pipe.dest.odm_combine_policy = cur_policy[i]; } } @@ -1581,7 +1587,7 @@ static void dcn32_full_validate_bw_helper(struct dc *dc, } } - if (should_allow_odm_power_optimization(dc, context, vba, split, merge)) + if (should_apply_odm_power_optimization(dc, context, vba, split, merge)) try_odm_power_optimization_and_revalidate( dc, context, pipes, split, merge, vlevel, *pipe_cnt); @@ -2210,7 +2216,8 @@ bool dcn32_internal_validate_bw(struct dc *dc, int i; pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, context, pipes, fast_validate); - dcn32_update_dml_pipes_odm_policy_based_on_context(dc, context, pipes); + if (!dc->config.enable_windowed_mpo_odm) + dcn32_update_dml_pipes_odm_policy_based_on_context(dc, context, pipes); /* repopulate_pipes = 1 means the pipes were either split or merged. In this case * we have to re-calculate the DET allocation and run through DML once more to diff --git a/drivers/gpu/drm/amd/display/dc/inc/resource.h b/drivers/gpu/drm/amd/display/dc/inc/resource.h index 1d51fed12e200..2eae2f3e846d8 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/resource.h +++ b/drivers/gpu/drm/amd/display/dc/inc/resource.h @@ -427,22 +427,18 @@ struct pipe_ctx *resource_get_primary_dpp_pipe(const struct pipe_ctx *dpp_pipe); int resource_get_mpc_slice_index(const struct pipe_ctx *dpp_pipe); /* - * Get number of MPC "cuts" of the plane associated with the pipe. MPC slice - * count is equal to MPC splits + 1. For example if a plane is cut 3 times, it - * will have 4 pieces of slice. - * return - 0 if pipe is not used for a plane with MPCC combine. otherwise - * the number of MPC "cuts" for the plane. + * Get the number of MPC slices associated with the pipe. + * The function returns 0 if the pipe is not associated with an MPC combine + * pipe topology. */ -int resource_get_mpc_slice_count(const struct pipe_ctx *opp_head); +int resource_get_mpc_slice_count(const struct pipe_ctx *pipe); /* - * Get number of ODM "cuts" of the timing associated with the pipe. ODM slice - * count is equal to ODM splits + 1. For example if a timing is cut 3 times, it - * will have 4 pieces of slice. - * return - 0 if pipe is not used for ODM combine. otherwise - * the number of ODM "cuts" for the timing. + * Get the number of ODM slices associated with the pipe. + * The function returns 0 if the pipe is not associated with an ODM combine + * pipe topology. */ -int resource_get_odm_slice_count(const struct pipe_ctx *otg_master); +int resource_get_odm_slice_count(const struct pipe_ctx *pipe); /* Get the ODM slice index counting from 0 from left most slice */ int resource_get_odm_slice_index(const struct pipe_ctx *opp_head); diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c index ac04a9c9a3d86..71cd20618bfe0 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c @@ -1829,7 +1829,21 @@ int dcn32_populate_dml_pipes_from_context( dcn32_zero_pipe_dcc_fraction(pipes, pipe_cnt); DC_FP_END(); pipes[pipe_cnt].pipe.dest.vfront_porch = timing->v_front_porch; - pipes[pipe_cnt].pipe.dest.odm_combine_policy = dm_odm_combine_policy_dal; + if (dc->config.enable_windowed_mpo_odm && + dc->debug.enable_single_display_2to1_odm_policy) { + switch (resource_get_odm_slice_count(pipe)) { + case 2: + pipes[pipe_cnt].pipe.dest.odm_combine_policy = dm_odm_combine_policy_2to1; + break; + case 4: + pipes[pipe_cnt].pipe.dest.odm_combine_policy = dm_odm_combine_policy_4to1; + break; + default: + pipes[pipe_cnt].pipe.dest.odm_combine_policy = dm_odm_combine_policy_dal; + } + } else { + pipes[pipe_cnt].pipe.dest.odm_combine_policy = dm_odm_combine_policy_dal; + } pipes[pipe_cnt].pipe.src.gpuvm_min_page_size_kbytes = 256; // according to spreadsheet pipes[pipe_cnt].pipe.src.unbounded_req_mode = false; pipes[pipe_cnt].pipe.scale_ratio_depth.lb_depth = dm_lb_19; -- 2.43.0

11 months, 2 weeks

1
0
0 0

FAILED: Patch "x86/sev: Fix position dependent variable references in startup code" failed to apply to 6.6-stable tree

by Sasha Levin

The patch below does not apply to the 6.6-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>. Thanks, Sasha ------------------ original commit in Linus's tree ------------------ From 1c811d403afd73f04bde82b83b24c754011bd0e8 Mon Sep 17 00:00:00 2001 From: Ard Biesheuvel <ardb(a)kernel.org> Date: Sat, 3 Feb 2024 13:53:06 +0100 Subject: [PATCH] x86/sev: Fix position dependent variable references in startup code The early startup code executes from a 1:1 mapping of memory, which differs from the mapping that the code was linked and/or relocated to run at. The latter mapping is not active yet at this point, and so symbol references that rely on it will fault. Given that the core kernel is built without -fPIC, symbol references are typically emitted as absolute, and so any such references occuring in the early startup code will therefore crash the kernel. While an attempt was made to work around this for the early SEV/SME startup code, by forcing RIP-relative addressing for certain global SEV/SME variables via inline assembly (see snp_cpuid_get_table() for example), RIP-relative addressing must be pervasively enforced for SEV/SME global variables when accessed prior to page table fixups. __startup_64() already handles this issue for select non-SEV/SME global variables using fixup_pointer(), which adjusts the pointer relative to a `physaddr` argument. To avoid having to pass around this `physaddr` argument across all functions needing to apply pointer fixups, introduce a macro RIP_RELATIVE_REF() which generates a RIP-relative reference to a given global variable. It is used where necessary to force RIP-relative accesses to global variables. For backporting purposes, this patch makes no attempt at cleaning up other occurrences of this pattern, involving either inline asm or fixup_pointer(). Those will be addressed later. [ bp: Call it "rip_rel_ref" everywhere like other code shortens "rIP-relative reference" and make the asm wrapper __always_inline. ] Co-developed-by: Kevin Loughlin <kevinloughlin(a)google.com> Signed-off-by: Kevin Loughlin <kevinloughlin(a)google.com> Signed-off-by: Ard Biesheuvel <ardb(a)kernel.org> Signed-off-by: Borislav Petkov (AMD) <bp(a)alien8.de> Cc: <stable(a)kernel.org> Link: https://lore.kernel.org/all/20240130220845.1978329-1-kevinloughlin@google.c… --- arch/x86/coco/core.c | 7 +------ arch/x86/include/asm/asm.h | 14 ++++++++++++++ arch/x86/include/asm/coco.h | 8 +++++++- arch/x86/include/asm/mem_encrypt.h | 15 +++++++++------ arch/x86/kernel/sev-shared.c | 12 ++++++------ arch/x86/kernel/sev.c | 4 ++-- arch/x86/mm/mem_encrypt_identity.c | 27 ++++++++++++--------------- 7 files changed, 51 insertions(+), 36 deletions(-) diff --git a/arch/x86/coco/core.c b/arch/x86/coco/core.c index eeec9986570ed..d07be9d05cd03 100644 --- a/arch/x86/coco/core.c +++ b/arch/x86/coco/core.c @@ -14,7 +14,7 @@ #include <asm/processor.h> enum cc_vendor cc_vendor __ro_after_init = CC_VENDOR_NONE; -static u64 cc_mask __ro_after_init; +u64 cc_mask __ro_after_init; static bool noinstr intel_cc_platform_has(enum cc_attr attr) { @@ -148,8 +148,3 @@ u64 cc_mkdec(u64 val) } } EXPORT_SYMBOL_GPL(cc_mkdec); - -__init void cc_set_mask(u64 mask) -{ - cc_mask = mask; -} diff --git a/arch/x86/include/asm/asm.h b/arch/x86/include/asm/asm.h index fbcfec4dc4ccd..ca8eed1d496ab 100644 --- a/arch/x86/include/asm/asm.h +++ b/arch/x86/include/asm/asm.h @@ -113,6 +113,20 @@ #endif +#ifndef __ASSEMBLY__ +#ifndef __pic__ +static __always_inline __pure void *rip_rel_ptr(void *p) +{ + asm("leaq %c1(%%rip), %0" : "=r"(p) : "i"(p)); + + return p; +} +#define RIP_REL_REF(var) (*(typeof(&(var)))rip_rel_ptr(&(var))) +#else +#define RIP_REL_REF(var) (var) +#endif +#endif + /* * Macros to generate condition code outputs from inline assembly, * The output operand must be type "bool". diff --git a/arch/x86/include/asm/coco.h b/arch/x86/include/asm/coco.h index 6ae2d16a7613b..21940ef8d2904 100644 --- a/arch/x86/include/asm/coco.h +++ b/arch/x86/include/asm/coco.h @@ -2,6 +2,7 @@ #ifndef _ASM_X86_COCO_H #define _ASM_X86_COCO_H +#include <asm/asm.h> #include <asm/types.h> enum cc_vendor { @@ -11,9 +12,14 @@ enum cc_vendor { }; extern enum cc_vendor cc_vendor; +extern u64 cc_mask; #ifdef CONFIG_ARCH_HAS_CC_PLATFORM -void cc_set_mask(u64 mask); +static inline void cc_set_mask(u64 mask) +{ + RIP_REL_REF(cc_mask) = mask; +} + u64 cc_mkenc(u64 val); u64 cc_mkdec(u64 val); #else diff --git a/arch/x86/include/asm/mem_encrypt.h b/arch/x86/include/asm/mem_encrypt.h index 359ada486fa92..b31eb9fd59544 100644 --- a/arch/x86/include/asm/mem_encrypt.h +++ b/arch/x86/include/asm/mem_encrypt.h @@ -15,7 +15,8 @@ #include <linux/init.h> #include <linux/cc_platform.h> -#include <asm/bootparam.h> +#include <asm/asm.h> +struct boot_params; #ifdef CONFIG_X86_MEM_ENCRYPT void __init mem_encrypt_init(void); @@ -58,6 +59,11 @@ void __init mem_encrypt_free_decrypted_mem(void); void __init sev_es_init_vc_handling(void); +static inline u64 sme_get_me_mask(void) +{ + return RIP_REL_REF(sme_me_mask); +} + #define __bss_decrypted __section(".bss..decrypted") #else /* !CONFIG_AMD_MEM_ENCRYPT */ @@ -89,6 +95,8 @@ early_set_mem_enc_dec_hypercall(unsigned long vaddr, unsigned long size, bool en static inline void mem_encrypt_free_decrypted_mem(void) { } +static inline u64 sme_get_me_mask(void) { return 0; } + #define __bss_decrypted #endif /* CONFIG_AMD_MEM_ENCRYPT */ @@ -106,11 +114,6 @@ void add_encrypt_protection_map(void); extern char __start_bss_decrypted[], __end_bss_decrypted[], __start_bss_decrypted_unused[]; -static inline u64 sme_get_me_mask(void) -{ - return sme_me_mask; -} - #endif /* __ASSEMBLY__ */ #endif /* __X86_MEM_ENCRYPT_H__ */ diff --git a/arch/x86/kernel/sev-shared.c b/arch/x86/kernel/sev-shared.c index 5db24d0fc557c..ae79f9505298d 100644 --- a/arch/x86/kernel/sev-shared.c +++ b/arch/x86/kernel/sev-shared.c @@ -560,9 +560,9 @@ static int snp_cpuid(struct ghcb *ghcb, struct es_em_ctxt *ctxt, struct cpuid_le leaf->eax = leaf->ebx = leaf->ecx = leaf->edx = 0; /* Skip post-processing for out-of-range zero leafs. */ - if (!(leaf->fn <= cpuid_std_range_max || - (leaf->fn >= 0x40000000 && leaf->fn <= cpuid_hyp_range_max) || - (leaf->fn >= 0x80000000 && leaf->fn <= cpuid_ext_range_max))) + if (!(leaf->fn <= RIP_REL_REF(cpuid_std_range_max) || + (leaf->fn >= 0x40000000 && leaf->fn <= RIP_REL_REF(cpuid_hyp_range_max)) || + (leaf->fn >= 0x80000000 && leaf->fn <= RIP_REL_REF(cpuid_ext_range_max)))) return 0; } @@ -1072,11 +1072,11 @@ static void __init setup_cpuid_table(const struct cc_blob_sev_info *cc_info) const struct snp_cpuid_fn *fn = &cpuid_table->fn[i]; if (fn->eax_in == 0x0) - cpuid_std_range_max = fn->eax; + RIP_REL_REF(cpuid_std_range_max) = fn->eax; else if (fn->eax_in == 0x40000000) - cpuid_hyp_range_max = fn->eax; + RIP_REL_REF(cpuid_hyp_range_max) = fn->eax; else if (fn->eax_in == 0x80000000) - cpuid_ext_range_max = fn->eax; + RIP_REL_REF(cpuid_ext_range_max) = fn->eax; } } diff --git a/arch/x86/kernel/sev.c b/arch/x86/kernel/sev.c index 002af6c30601b..1ef7ae806a01b 100644 --- a/arch/x86/kernel/sev.c +++ b/arch/x86/kernel/sev.c @@ -748,7 +748,7 @@ void __init early_snp_set_memory_private(unsigned long vaddr, unsigned long padd * This eliminates worries about jump tables or checking boot_cpu_data * in the cc_platform_has() function. */ - if (!(sev_status & MSR_AMD64_SEV_SNP_ENABLED)) + if (!(RIP_REL_REF(sev_status) & MSR_AMD64_SEV_SNP_ENABLED)) return; /* @@ -767,7 +767,7 @@ void __init early_snp_set_memory_shared(unsigned long vaddr, unsigned long paddr * This eliminates worries about jump tables or checking boot_cpu_data * in the cc_platform_has() function. */ - if (!(sev_status & MSR_AMD64_SEV_SNP_ENABLED)) + if (!(RIP_REL_REF(sev_status) & MSR_AMD64_SEV_SNP_ENABLED)) return; /* Ask hypervisor to mark the memory pages shared in the RMP table. */ diff --git a/arch/x86/mm/mem_encrypt_identity.c b/arch/x86/mm/mem_encrypt_identity.c index efe9f217fcf99..0166ab1780ccb 100644 --- a/arch/x86/mm/mem_encrypt_identity.c +++ b/arch/x86/mm/mem_encrypt_identity.c @@ -304,7 +304,8 @@ void __init sme_encrypt_kernel(struct boot_params *bp) * instrumentation or checking boot_cpu_data in the cc_platform_has() * function. */ - if (!sme_get_me_mask() || sev_status & MSR_AMD64_SEV_ENABLED) + if (!sme_get_me_mask() || + RIP_REL_REF(sev_status) & MSR_AMD64_SEV_ENABLED) return; /* @@ -541,11 +542,11 @@ void __init sme_enable(struct boot_params *bp) me_mask = 1UL << (ebx & 0x3f); /* Check the SEV MSR whether SEV or SME is enabled */ - sev_status = __rdmsr(MSR_AMD64_SEV); - feature_mask = (sev_status & MSR_AMD64_SEV_ENABLED) ? AMD_SEV_BIT : AMD_SME_BIT; + RIP_REL_REF(sev_status) = msr = __rdmsr(MSR_AMD64_SEV); + feature_mask = (msr & MSR_AMD64_SEV_ENABLED) ? AMD_SEV_BIT : AMD_SME_BIT; /* The SEV-SNP CC blob should never be present unless SEV-SNP is enabled. */ - if (snp && !(sev_status & MSR_AMD64_SEV_SNP_ENABLED)) + if (snp && !(msr & MSR_AMD64_SEV_SNP_ENABLED)) snp_abort(); /* Check if memory encryption is enabled */ @@ -571,7 +572,6 @@ void __init sme_enable(struct boot_params *bp) return; } else { /* SEV state cannot be controlled by a command line option */ - sme_me_mask = me_mask; goto out; } @@ -590,16 +590,13 @@ void __init sme_enable(struct boot_params *bp) cmdline_ptr = (const char *)((u64)bp->hdr.cmd_line_ptr | ((u64)bp->ext_cmd_line_ptr << 32)); - if (cmdline_find_option(cmdline_ptr, cmdline_arg, buffer, sizeof(buffer)) < 0) - goto out; - - if (!strncmp(buffer, cmdline_on, sizeof(buffer))) - sme_me_mask = me_mask; + if (cmdline_find_option(cmdline_ptr, cmdline_arg, buffer, sizeof(buffer)) < 0 || + strncmp(buffer, cmdline_on, sizeof(buffer))) + return; out: - if (sme_me_mask) { - physical_mask &= ~sme_me_mask; - cc_vendor = CC_VENDOR_AMD; - cc_set_mask(sme_me_mask); - } + RIP_REL_REF(sme_me_mask) = me_mask; + physical_mask &= ~me_mask; + cc_vendor = CC_VENDOR_AMD; + cc_set_mask(me_mask); } -- 2.43.0

11 months, 2 weeks

1
0
0 0

FAILED: Patch "tracing/ring-buffer: Fix wait_on_pipe() race" failed to apply to 6.6-stable tree

by Sasha Levin

The patch below does not apply to the 6.6-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>. Thanks, Sasha ------------------ original commit in Linus's tree ------------------ From 2aa043a55b9a764c9cbde5a8c654eeaaffe224cf Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" <rostedt(a)goodmis.org> Date: Tue, 12 Mar 2024 08:15:08 -0400 Subject: [PATCH] tracing/ring-buffer: Fix wait_on_pipe() race When the trace_pipe_raw file is closed, there should be no new readers on the file descriptor. This is mostly handled with the waking and wait_index fields of the iterator. But there's still a slight race. CPU 0 CPU 1 ----- ----- wait_index++; index = wait_index; ring_buffer_wake_waiters(); wait_on_pipe() ring_buffer_wait(); The ring_buffer_wait() will miss the wakeup from CPU 1. The problem is that the ring_buffer_wait() needs the logic of: prepare_to_wait(); if (!condition) schedule(); Where the missing condition check is the iter->wait_index update. Have the ring_buffer_wait() take a conditional callback function and a data parameter that can be used within the wait_event_interruptible() of the ring_buffer_wait() function. In wait_on_pipe(), pass a condition function that will check if the wait_index has been updated, if it has, it will return true to break out of the wait_event_interruptible() loop. Create a new field "closed" in the trace_iterator and set it in the .flush() callback before calling ring_buffer_wake_waiters(). This will keep any new readers from waiting on a closed file descriptor. Have the wait_on_pipe() condition callback also check the closed field. Change the wait_index field of the trace_iterator to atomic_t. There's no reason it needs to be 'long' and making it atomic and using atomic_read_acquire() and atomic_fetch_inc_release() will provide the necessary memory barriers. Add a "woken" flag to tracing_buffers_splice_read() to exit the loop after one more try to fetch data. That is, if it waited for data and something woke it up, it should try to collect any new data and then exit back to user space. Link: https://lore.kernel.org/linux-trace-kernel/CAHk-=wgsNgewHFxZAJiAQznwPMqEtQm… Link: https://lore.kernel.org/linux-trace-kernel/20240312121703.557950713@goodmis… Cc: stable(a)vger.kernel.org Cc: Masami Hiramatsu <mhiramat(a)kernel.org> Cc: Mark Rutland <mark.rutland(a)arm.com> Cc: Mathieu Desnoyers <mathieu.desnoyers(a)efficios.com> Cc: Andrew Morton <akpm(a)linux-foundation.org> Cc: Linus Torvalds <torvalds(a)linux-foundation.org> Cc: linke li <lilinke99(a)qq.com> Cc: Rabin Vincent <rabin(a)rab.in> Fixes: f3ddb74ad0790 ("tracing: Wake up ring buffer waiters on closing of the file") Signed-off-by: Steven Rostedt (Google) <rostedt(a)goodmis.org> --- include/linux/ring_buffer.h | 3 ++- include/linux/trace_events.h | 5 ++++- kernel/trace/ring_buffer.c | 13 ++++++----- kernel/trace/trace.c | 43 ++++++++++++++++++++++++++---------- 4 files changed, 45 insertions(+), 19 deletions(-) diff --git a/include/linux/ring_buffer.h b/include/linux/ring_buffer.h index 338a33db1577e..dc5ae4e96aee0 100644 --- a/include/linux/ring_buffer.h +++ b/include/linux/ring_buffer.h @@ -99,7 +99,8 @@ __ring_buffer_alloc(unsigned long size, unsigned flags, struct lock_class_key *k }) typedef bool (*ring_buffer_cond_fn)(void *data); -int ring_buffer_wait(struct trace_buffer *buffer, int cpu, int full); +int ring_buffer_wait(struct trace_buffer *buffer, int cpu, int full, + ring_buffer_cond_fn cond, void *data); __poll_t ring_buffer_poll_wait(struct trace_buffer *buffer, int cpu, struct file *filp, poll_table *poll_table, int full); void ring_buffer_wake_waiters(struct trace_buffer *buffer, int cpu); diff --git a/include/linux/trace_events.h b/include/linux/trace_events.h index d68ff9b1247f9..fc6d0af56bb17 100644 --- a/include/linux/trace_events.h +++ b/include/linux/trace_events.h @@ -103,13 +103,16 @@ struct trace_iterator { unsigned int temp_size; char *fmt; /* modified format holder */ unsigned int fmt_size; - long wait_index; + atomic_t wait_index; /* trace_seq for __print_flags() and __print_symbolic() etc. */ struct trace_seq tmp_seq; cpumask_var_t started; + /* Set when the file is closed to prevent new waiters */ + bool closed; + /* it's true when current open file is snapshot */ bool snapshot; diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index f4c34b7c7e1e7..350607cce8694 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -902,23 +902,26 @@ static bool rb_wait_once(void *data) * @buffer: buffer to wait on * @cpu: the cpu buffer to wait on * @full: wait until the percentage of pages are available, if @cpu != RING_BUFFER_ALL_CPUS + * @cond: condition function to break out of wait (NULL to run once) + * @data: the data to pass to @cond. * * If @cpu == RING_BUFFER_ALL_CPUS then the task will wake up as soon * as data is added to any of the @buffer's cpu buffers. Otherwise * it will wait for data to be added to a specific cpu buffer. */ -int ring_buffer_wait(struct trace_buffer *buffer, int cpu, int full) +int ring_buffer_wait(struct trace_buffer *buffer, int cpu, int full, + ring_buffer_cond_fn cond, void *data) { struct ring_buffer_per_cpu *cpu_buffer; struct wait_queue_head *waitq; - ring_buffer_cond_fn cond; struct rb_irq_work *rbwork; - void *data; long once = 0; int ret = 0; - cond = rb_wait_once; - data = &once; + if (!cond) { + cond = rb_wait_once; + data = &once; + } /* * Depending on what the caller is waiting for, either any diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index c9c8983073485..d390fea3a6a52 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -1955,15 +1955,36 @@ update_max_tr_single(struct trace_array *tr, struct task_struct *tsk, int cpu) #endif /* CONFIG_TRACER_MAX_TRACE */ +struct pipe_wait { + struct trace_iterator *iter; + int wait_index; +}; + +static bool wait_pipe_cond(void *data) +{ + struct pipe_wait *pwait = data; + struct trace_iterator *iter = pwait->iter; + + if (atomic_read_acquire(&iter->wait_index) != pwait->wait_index) + return true; + + return iter->closed; +} + static int wait_on_pipe(struct trace_iterator *iter, int full) { + struct pipe_wait pwait; int ret; /* Iterators are static, they should be filled or empty */ if (trace_buffer_iter(iter, iter->cpu_file)) return 0; - ret = ring_buffer_wait(iter->array_buffer->buffer, iter->cpu_file, full); + pwait.wait_index = atomic_read_acquire(&iter->wait_index); + pwait.iter = iter; + + ret = ring_buffer_wait(iter->array_buffer->buffer, iter->cpu_file, full, + wait_pipe_cond, &pwait); #ifdef CONFIG_TRACER_MAX_TRACE /* @@ -8398,9 +8419,9 @@ static int tracing_buffers_flush(struct file *file, fl_owner_t id) struct ftrace_buffer_info *info = file->private_data; struct trace_iterator *iter = &info->iter; - iter->wait_index++; + iter->closed = true; /* Make sure the waiters see the new wait_index */ - smp_wmb(); + (void)atomic_fetch_inc_release(&iter->wait_index); ring_buffer_wake_waiters(iter->array_buffer->buffer, iter->cpu_file); @@ -8500,6 +8521,7 @@ tracing_buffers_splice_read(struct file *file, loff_t *ppos, .spd_release = buffer_spd_release, }; struct buffer_ref *ref; + bool woken = false; int page_size; int entries, i; ssize_t ret = 0; @@ -8573,17 +8595,17 @@ tracing_buffers_splice_read(struct file *file, loff_t *ppos, /* did we read anything? */ if (!spd.nr_pages) { - long wait_index; if (ret) goto out; + if (woken) + goto out; + ret = -EAGAIN; if ((file->f_flags & O_NONBLOCK) || (flags & SPLICE_F_NONBLOCK)) goto out; - wait_index = READ_ONCE(iter->wait_index); - ret = wait_on_pipe(iter, iter->snapshot ? 0 : iter->tr->buffer_percent); if (ret) goto out; @@ -8592,10 +8614,8 @@ tracing_buffers_splice_read(struct file *file, loff_t *ppos, if (!tracer_tracing_is_on(iter->tr)) goto out; - /* Make sure we see the new wait_index */ - smp_rmb(); - if (wait_index != iter->wait_index) - goto out; + /* Iterate one more time to collect any new data then exit */ + woken = true; goto again; } @@ -8618,9 +8638,8 @@ static long tracing_buffers_ioctl(struct file *file, unsigned int cmd, unsigned mutex_lock(&trace_types_lock); - iter->wait_index++; /* Make sure the waiters see the new wait_index */ - smp_wmb(); + (void)atomic_fetch_inc_release(&iter->wait_index); ring_buffer_wake_waiters(iter->array_buffer->buffer, iter->cpu_file); -- 2.43.0

11 months, 2 weeks

1
0
0 0

FAILED: Patch "drm/amd/display: Implement wait_for_odm_update_pending_complete" failed to apply to 6.6-stable tree

by Sasha Levin

The patch below does not apply to the 6.6-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>. Thanks, Sasha ------------------ original commit in Linus's tree ------------------ From 2d7f3d1a5866705be2393150e1ffdf67030ab88d Mon Sep 17 00:00:00 2001 From: Wenjing Liu <wenjing.liu(a)amd.com> Date: Fri, 23 Feb 2024 15:38:40 -0500 Subject: [PATCH] drm/amd/display: Implement wait_for_odm_update_pending_complete [WHY] Odm update is doubled buffered. We need to wait for ODM update to be completed before optimizing bandwidth or programming new udpates. [HOW] implement wait_for_odm_update_pending_complete function to wait for: 1. odm configuration update is no longer pending in timing generator. 2. no pending dpg pattern update for each active OPP. Cc: Mario Limonciello <mario.limonciello(a)amd.com> Cc: Alex Deucher <alexander.deucher(a)amd.com> Cc: stable(a)vger.kernel.org Reviewed-by: Alvin Lee <alvin.lee2(a)amd.com> Acked-by: Alex Hung <alex.hung(a)amd.com> Signed-off-by: Wenjing Liu <wenjing.liu(a)amd.com> Tested-by: Daniel Wheeler <daniel.wheeler(a)amd.com> Signed-off-by: Alex Deucher <alexander.deucher(a)amd.com> --- drivers/gpu/drm/amd/display/dc/core/dc.c | 56 ++++++++++++++++++- .../gpu/drm/amd/display/dc/dcn10/dcn10_opp.c | 1 + .../gpu/drm/amd/display/dc/dcn20/dcn20_opp.c | 14 +++++ .../gpu/drm/amd/display/dc/dcn20/dcn20_opp.h | 2 + .../drm/amd/display/dc/dcn201/dcn201_opp.c | 1 + .../amd/display/dc/hwss/dcn20/dcn20_hwseq.c | 4 +- drivers/gpu/drm/amd/display/dc/inc/hw/opp.h | 3 + .../amd/display/dc/inc/hw/timing_generator.h | 1 + .../amd/display/dc/optc/dcn10/dcn10_optc.h | 3 +- .../amd/display/dc/optc/dcn32/dcn32_optc.c | 8 +++ .../amd/display/dc/optc/dcn32/dcn32_optc.h | 1 + 11 files changed, 90 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c index 958552a8605ff..e7dc128f6284b 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc.c @@ -1302,6 +1302,54 @@ static void disable_vbios_mode_if_required( } } +/** + * wait_for_blank_complete - wait for all active OPPs to finish pending blank + * pattern updates + * + * @dc: [in] dc reference + * @context: [in] hardware context in use + */ +static void wait_for_blank_complete(struct dc *dc, + struct dc_state *context) +{ + struct pipe_ctx *opp_head; + struct dce_hwseq *hws = dc->hwseq; + int i; + + if (!hws->funcs.wait_for_blank_complete) + return; + + for (i = 0; i < MAX_PIPES; i++) { + opp_head = &context->res_ctx.pipe_ctx[i]; + + if (!resource_is_pipe_type(opp_head, OPP_HEAD) || + dc_state_get_pipe_subvp_type(context, opp_head) == SUBVP_PHANTOM) + continue; + + hws->funcs.wait_for_blank_complete(opp_head->stream_res.opp); + } +} + +static void wait_for_odm_update_pending_complete(struct dc *dc, struct dc_state *context) +{ + struct pipe_ctx *otg_master; + struct timing_generator *tg; + int i; + + for (i = 0; i < MAX_PIPES; i++) { + otg_master = &context->res_ctx.pipe_ctx[i]; + if (!resource_is_pipe_type(otg_master, OTG_MASTER) || + dc_state_get_pipe_subvp_type(context, otg_master) == SUBVP_PHANTOM) + continue; + tg = otg_master->stream_res.tg; + if (tg->funcs->wait_odm_doublebuffer_pending_clear) + tg->funcs->wait_odm_doublebuffer_pending_clear(tg); + } + + /* ODM update may require to reprogram blank pattern for each OPP */ + wait_for_blank_complete(dc, context); +} + static void wait_for_no_pipes_pending(struct dc *dc, struct dc_state *context) { int i; @@ -1993,6 +2041,11 @@ static enum dc_status dc_commit_state_no_check(struct dc *dc, struct dc_state *c context->stream_count == 0) { /* Must wait for no flips to be pending before doing optimize bw */ wait_for_no_pipes_pending(dc, context); + /* + * optimized dispclk depends on ODM setup. Need to wait for ODM + * update pending complete before optimizing bandwidth. + */ + wait_for_odm_update_pending_complete(dc, context); /* pplib is notified if disp_num changed */ dc->hwss.optimize_bandwidth(dc, context); /* Need to do otg sync again as otg could be out of sync due to otg @@ -3496,7 +3549,7 @@ static void commit_planes_for_stream_fast(struct dc *dc, top_pipe_to_program->stream->update_flags.raw = 0; } -static void wait_for_outstanding_hw_updates(struct dc *dc, const struct dc_state *dc_context) +static void wait_for_outstanding_hw_updates(struct dc *dc, struct dc_state *dc_context) { /* * This function calls HWSS to wait for any potentially double buffered @@ -3534,6 +3587,7 @@ static void wait_for_outstanding_hw_updates(struct dc *dc, const struct dc_state } } } + wait_for_odm_update_pending_complete(dc, dc_context); } static void commit_planes_for_stream(struct dc *dc, diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_opp.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_opp.c index 48a40dcc7050b..5838a11efd00c 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_opp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_opp.c @@ -384,6 +384,7 @@ static const struct opp_funcs dcn10_opp_funcs = { .opp_set_disp_pattern_generator = NULL, .opp_program_dpg_dimensions = NULL, .dpg_is_blanked = NULL, + .dpg_is_pending = NULL, .opp_destroy = opp1_destroy }; diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_opp.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_opp.c index 0784d01986610..fbf1b6370eb23 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_opp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_opp.c @@ -337,6 +337,19 @@ bool opp2_dpg_is_blanked(struct output_pixel_processor *opp) (double_buffer_pending == 0); } +bool opp2_dpg_is_pending(struct output_pixel_processor *opp) +{ + struct dcn20_opp *oppn20 = TO_DCN20_OPP(opp); + uint32_t double_buffer_pending; + uint32_t dpg_en; + + REG_GET(DPG_CONTROL, DPG_EN, &dpg_en); + + REG_GET(DPG_STATUS, DPG_DOUBLE_BUFFER_PENDING, &double_buffer_pending); + + return (dpg_en == 1 && double_buffer_pending == 1); +} + void opp2_program_left_edge_extra_pixel ( struct output_pixel_processor *opp, bool count) @@ -363,6 +376,7 @@ static struct opp_funcs dcn20_opp_funcs = { .opp_set_disp_pattern_generator = opp2_set_disp_pattern_generator, .opp_program_dpg_dimensions = opp2_program_dpg_dimensions, .dpg_is_blanked = opp2_dpg_is_blanked, + .dpg_is_pending = opp2_dpg_is_pending, .opp_dpg_set_blank_color = opp2_dpg_set_blank_color, .opp_destroy = opp1_destroy, .opp_program_left_edge_extra_pixel = opp2_program_left_edge_extra_pixel, diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_opp.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_opp.h index 3ab221bdd27dd..8f186abd558db 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_opp.h +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_opp.h @@ -159,6 +159,8 @@ void opp2_program_dpg_dimensions( bool opp2_dpg_is_blanked(struct output_pixel_processor *opp); +bool opp2_dpg_is_pending(struct output_pixel_processor *opp); + void opp2_dpg_set_blank_color( struct output_pixel_processor *opp, const struct tg_color *color); diff --git a/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_opp.c b/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_opp.c index 8e77db46a4090..6a71ba3dfc632 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_opp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_opp.c @@ -50,6 +50,7 @@ static struct opp_funcs dcn201_opp_funcs = { .opp_set_disp_pattern_generator = opp2_set_disp_pattern_generator, .opp_program_dpg_dimensions = opp2_program_dpg_dimensions, .dpg_is_blanked = opp2_dpg_is_blanked, + .dpg_is_pending = opp2_dpg_is_pending, .opp_dpg_set_blank_color = opp2_dpg_set_blank_color, .opp_destroy = opp1_destroy, .opp_program_left_edge_extra_pixel = opp2_program_left_edge_extra_pixel, diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c index 40098d9f70cbc..8b3536c380b8d 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c @@ -2452,7 +2452,7 @@ bool dcn20_wait_for_blank_complete( int counter; for (counter = 0; counter < 1000; counter++) { - if (opp->funcs->dpg_is_blanked(opp)) + if (!opp->funcs->dpg_is_pending(opp)) break; udelay(100); @@ -2463,7 +2463,7 @@ bool dcn20_wait_for_blank_complete( return false; } - return true; + return opp->funcs->dpg_is_blanked(opp); } bool dcn20_dmdata_status_done(struct pipe_ctx *pipe_ctx) diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/opp.h b/drivers/gpu/drm/amd/display/dc/inc/hw/opp.h index aee5372e292c5..d89c92370d5b3 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/opp.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/opp.h @@ -337,6 +337,9 @@ struct opp_funcs { bool (*dpg_is_blanked)( struct output_pixel_processor *opp); + bool (*dpg_is_pending)(struct output_pixel_processor *opp); + + void (*opp_dpg_set_blank_color)( struct output_pixel_processor *opp, const struct tg_color *color); diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h b/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h index d98d72f35be5b..ffad8fe16c54d 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h @@ -331,6 +331,7 @@ struct timing_generator_funcs { void (*init_odm)(struct timing_generator *tg); void (*wait_drr_doublebuffer_pending_clear)(struct timing_generator *tg); + void (*wait_odm_doublebuffer_pending_clear)(struct timing_generator *tg); }; #endif diff --git a/drivers/gpu/drm/amd/display/dc/optc/dcn10/dcn10_optc.h b/drivers/gpu/drm/amd/display/dc/optc/dcn10/dcn10_optc.h index ab81594a7fadc..6c2e84d3967fc 100644 --- a/drivers/gpu/drm/amd/display/dc/optc/dcn10/dcn10_optc.h +++ b/drivers/gpu/drm/amd/display/dc/optc/dcn10/dcn10_optc.h @@ -557,7 +557,8 @@ struct dcn_optc_registers { type OTG_CRC_DATA_STREAM_SPLIT_MODE;\ type OTG_CRC_DATA_FORMAT;\ type OTG_V_TOTAL_LAST_USED_BY_DRR;\ - type OTG_DRR_TIMING_DBUF_UPDATE_PENDING; + type OTG_DRR_TIMING_DBUF_UPDATE_PENDING;\ + type OTG_H_TIMING_DIV_MODE_DB_UPDATE_PENDING; #define TG_REG_FIELD_LIST_DCN3_2(type) \ type OTG_H_TIMING_DIV_MODE_MANUAL; diff --git a/drivers/gpu/drm/amd/display/dc/optc/dcn32/dcn32_optc.c b/drivers/gpu/drm/amd/display/dc/optc/dcn32/dcn32_optc.c index 8234935433254..f07a4c7e48bc2 100644 --- a/drivers/gpu/drm/amd/display/dc/optc/dcn32/dcn32_optc.c +++ b/drivers/gpu/drm/amd/display/dc/optc/dcn32/dcn32_optc.c @@ -122,6 +122,13 @@ void optc32_get_odm_combine_segments(struct timing_generator *tg, int *odm_combi } } +void optc32_wait_odm_doublebuffer_pending_clear(struct timing_generator *tg) +{ + struct optc *optc1 = DCN10TG_FROM_TG(tg); + + REG_WAIT(OTG_DOUBLE_BUFFER_CONTROL, OTG_H_TIMING_DIV_MODE_DB_UPDATE_PENDING, 0, 2, 50000); +} + void optc32_set_h_timing_div_manual_mode(struct timing_generator *optc, bool manual_mode) { struct optc *optc1 = DCN10TG_FROM_TG(optc); @@ -345,6 +352,7 @@ static struct timing_generator_funcs dcn32_tg_funcs = { .set_odm_bypass = optc32_set_odm_bypass, .set_odm_combine = optc32_set_odm_combine, .get_odm_combine_segments = optc32_get_odm_combine_segments, + .wait_odm_doublebuffer_pending_clear = optc32_wait_odm_doublebuffer_pending_clear, .set_h_timing_div_manual_mode = optc32_set_h_timing_div_manual_mode, .get_optc_source = optc2_get_optc_source, .set_out_mux = optc3_set_out_mux, diff --git a/drivers/gpu/drm/amd/display/dc/optc/dcn32/dcn32_optc.h b/drivers/gpu/drm/amd/display/dc/optc/dcn32/dcn32_optc.h index 8ce3b178cab06..0c2c146955619 100644 --- a/drivers/gpu/drm/amd/display/dc/optc/dcn32/dcn32_optc.h +++ b/drivers/gpu/drm/amd/display/dc/optc/dcn32/dcn32_optc.h @@ -183,5 +183,6 @@ void optc32_set_h_timing_div_manual_mode(struct timing_generator *optc, bool man void optc32_get_odm_combine_segments(struct timing_generator *tg, int *odm_combine_segments); void optc32_set_odm_bypass(struct timing_generator *optc, const struct dc_crtc_timing *dc_crtc_timing); +void optc32_wait_odm_doublebuffer_pending_clear(struct timing_generator *tg); #endif /* __DC_OPTC_DCN32_H__ */ -- 2.43.0

11 months, 2 weeks

1
0
0 0

FAILED: Patch "Revert "drm/amd/display: Send DTBCLK disable message on first commit"" failed to apply to 6.6-stable tree

by Sasha Levin

The patch below does not apply to the 6.6-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>. Thanks, Sasha ------------------ original commit in Linus's tree ------------------ From 3a6a32b31a111f6e66526fb2d3cb13a876465076 Mon Sep 17 00:00:00 2001 From: Gabe Teeger <gabe.teeger(a)amd.com> Date: Mon, 29 Jan 2024 13:31:44 -0500 Subject: [PATCH] Revert "drm/amd/display: Send DTBCLK disable message on first commit" This reverts commit f341055b10bd8be55c3c995dff5f770b236b8ca9. System hang observed, this commit is thought to be the regression point. Cc: Mario Limonciello <mario.limonciello(a)amd.com> Cc: Alex Deucher <alexander.deucher(a)amd.com> Cc: stable(a)vger.kernel.org Reviewed-by: Ovidiu Bunea <ovidiu.bunea(a)amd.com> Acked-by: Aurabindo Pillai <aurabindo.pillai(a)amd.com> Signed-off-by: Gabe Teeger <gabe.teeger(a)amd.com> Tested-by: Daniel Wheeler <daniel.wheeler(a)amd.com> Signed-off-by: Alex Deucher <alexander.deucher(a)amd.com> --- drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_clk_mgr.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_clk_mgr.c index 06edca50a8fa1..36e5bb611fb10 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_clk_mgr.c @@ -414,7 +414,6 @@ static void init_clk_states(struct clk_mgr *clk_mgr) uint32_t ref_dtbclk = clk_mgr->clks.ref_dtbclk_khz; memset(&(clk_mgr->clks), 0, sizeof(struct dc_clocks)); - clk_mgr->clks.dtbclk_en = true; clk_mgr->clks.ref_dtbclk_khz = ref_dtbclk; // restore ref_dtbclk clk_mgr->clks.p_state_change_support = true; clk_mgr->clks.prev_p_state_change_support = true; -- 2.43.0

11 months, 2 weeks

1
0
0 0

FAILED: Patch "drm/amd/display: Fix idle check for shared firmware state" failed to apply to 6.6-stable tree

by Sasha Levin

The patch below does not apply to the 6.6-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>. Thanks, Sasha ------------------ original commit in Linus's tree ------------------ From 3d066f9547dd58329b526db44f42c487a7974703 Mon Sep 17 00:00:00 2001 From: Nicholas Kazlauskas <nicholas.kazlauskas(a)amd.com> Date: Wed, 21 Feb 2024 12:27:31 -0500 Subject: [PATCH] drm/amd/display: Fix idle check for shared firmware state [WHY] We still had an instance of get_idle_state checking the PMFW scratch register instead of the actual idle allow signal. [HOW] Replace it with the SW state check for whether we had allowed idle through notify_idle. Cc: Mario Limonciello <mario.limonciello(a)amd.com> Cc: Alex Deucher <alexander.deucher(a)amd.com> Cc: stable(a)vger.kernel.org Reviewed-by: Duncan Ma <duncan.ma(a)amd.com> Acked-by: Alex Hung <alex.hung(a)amd.com> Signed-off-by: Nicholas Kazlauskas <nicholas.kazlauskas(a)amd.com> Tested-by: Daniel Wheeler <daniel.wheeler(a)amd.com> Signed-off-by: Alex Deucher <alexander.deucher(a)amd.com> --- drivers/gpu/drm/amd/display/dc/core/dc.c | 12 +++--------- 1 file changed, 3 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 613d09c42f3b9..958552a8605ff 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc.c @@ -4847,22 +4847,16 @@ void dc_exit_ips_for_hw_access(struct dc *dc) bool dc_dmub_is_ips_idle_state(struct dc *dc) { - uint32_t idle_state = 0; - if (dc->debug.disable_idle_power_optimizations) return false; if (!dc->caps.ips_support || (dc->config.disable_ips == DMUB_IPS_DISABLE_ALL)) return false; - if (dc->hwss.get_idle_state) - idle_state = dc->hwss.get_idle_state(dc); - - if (!(idle_state & DMUB_IPS1_ALLOW_MASK) || - !(idle_state & DMUB_IPS2_ALLOW_MASK)) - return true; + if (!dc->ctx->dmub_srv) + return false; - return false; + return dc->ctx->dmub_srv->idle_allowed; } /* set min and max memory clock to lowest and highest DPM level, respectively */ -- 2.43.0

11 months, 2 weeks

1
0
0 0

FAILED: Patch "drm/amd/display: Only allow dig mapping to pwrseq in new asic" failed to apply to 6.6-stable tree

by Sasha Levin

The patch below does not apply to the 6.6-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>. Thanks, Sasha ------------------ original commit in Linus's tree ------------------ From 4af4d2c275aeb667bc2bca0d2135b825e931a55a Mon Sep 17 00:00:00 2001 From: Lewis Huang <lewis.huang(a)amd.com> Date: Wed, 31 Jan 2024 17:20:17 +0800 Subject: [PATCH] drm/amd/display: Only allow dig mapping to pwrseq in new asic [Why] The old asic only have 1 pwrseq hw. We don't need to map the diginst to pwrseq inst in old asic. [How] 1. Only mapping dig to pwrseq for new asic. 2. Move mapping function into dcn specific panel control component Cc: Stable <stable(a)vger.kernel.org> # v6.6+ Cc: Mario Limonciello <mario.limonciello(a)amd.com> Link: https://gitlab.freedesktop.org/drm/amd/-/issues/3122 Reviewed-by: Anthony Koo <anthony.koo(a)amd.com> Acked-by: Rodrigo Siqueira <rodrigo.siqueira(a)amd.com> Tested-by: Daniel Wheeler <daniel.wheeler(a)amd.com> Signed-off-by: Lewis Huang <lewis.huang(a)amd.com> Signed-off-by: Alex Deucher <alexander.deucher(a)amd.com> --- .../drm/amd/display/dc/dce/dce_panel_cntl.c | 1 + .../amd/display/dc/dcn301/dcn301_panel_cntl.c | 1 + .../amd/display/dc/dcn31/dcn31_panel_cntl.c | 18 ++++++++++++- .../drm/amd/display/dc/inc/hw/panel_cntl.h | 2 +- .../drm/amd/display/dc/link/link_factory.c | 26 +------------------ 5 files changed, 21 insertions(+), 27 deletions(-) diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_panel_cntl.c b/drivers/gpu/drm/amd/display/dc/dce/dce_panel_cntl.c index e8570060d007b..5bca67407c5b1 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_panel_cntl.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_panel_cntl.c @@ -290,4 +290,5 @@ void dce_panel_cntl_construct( dce_panel_cntl->base.funcs = &dce_link_panel_cntl_funcs; dce_panel_cntl->base.ctx = init_data->ctx; dce_panel_cntl->base.inst = init_data->inst; + dce_panel_cntl->base.pwrseq_inst = 0; } diff --git a/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_panel_cntl.c b/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_panel_cntl.c index ad0df1a72a90a..9e96a3ace2077 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_panel_cntl.c +++ b/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_panel_cntl.c @@ -215,4 +215,5 @@ void dcn301_panel_cntl_construct( dcn301_panel_cntl->base.funcs = &dcn301_link_panel_cntl_funcs; dcn301_panel_cntl->base.ctx = init_data->ctx; dcn301_panel_cntl->base.inst = init_data->inst; + dcn301_panel_cntl->base.pwrseq_inst = 0; } diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_panel_cntl.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_panel_cntl.c index 03248422d6ffd..281be20b1a107 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_panel_cntl.c +++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_panel_cntl.c @@ -154,8 +154,24 @@ void dcn31_panel_cntl_construct( struct dcn31_panel_cntl *dcn31_panel_cntl, const struct panel_cntl_init_data *init_data) { + uint8_t pwrseq_inst = 0xF; + dcn31_panel_cntl->base.funcs = &dcn31_link_panel_cntl_funcs; dcn31_panel_cntl->base.ctx = init_data->ctx; dcn31_panel_cntl->base.inst = init_data->inst; - dcn31_panel_cntl->base.pwrseq_inst = init_data->pwrseq_inst; + + switch (init_data->eng_id) { + case ENGINE_ID_DIGA: + pwrseq_inst = 0; + break; + case ENGINE_ID_DIGB: + pwrseq_inst = 1; + break; + default: + DC_LOG_WARNING("Unsupported pwrseq engine id: %d!\n", init_data->eng_id); + ASSERT(false); + break; + } + + dcn31_panel_cntl->base.pwrseq_inst = pwrseq_inst; } diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/panel_cntl.h b/drivers/gpu/drm/amd/display/dc/inc/hw/panel_cntl.h index 5dcbaa2db964a..e97d964a1791c 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/panel_cntl.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/panel_cntl.h @@ -57,7 +57,7 @@ struct panel_cntl_funcs { struct panel_cntl_init_data { struct dc_context *ctx; uint32_t inst; - uint32_t pwrseq_inst; + uint32_t eng_id; }; struct panel_cntl { diff --git a/drivers/gpu/drm/amd/display/dc/link/link_factory.c b/drivers/gpu/drm/amd/display/dc/link/link_factory.c index 37d3027c32dcb..cf22b8f28ba6c 100644 --- a/drivers/gpu/drm/amd/display/dc/link/link_factory.c +++ b/drivers/gpu/drm/amd/display/dc/link/link_factory.c @@ -370,30 +370,6 @@ static enum transmitter translate_encoder_to_transmitter( } } -static uint8_t translate_dig_inst_to_pwrseq_inst(struct dc_link *link) -{ - uint8_t pwrseq_inst = 0xF; - struct dc_context *dc_ctx = link->dc->ctx; - - DC_LOGGER_INIT(dc_ctx->logger); - - switch (link->eng_id) { - case ENGINE_ID_DIGA: - pwrseq_inst = 0; - break; - case ENGINE_ID_DIGB: - pwrseq_inst = 1; - break; - default: - DC_LOG_WARNING("Unsupported pwrseq engine id: %d!\n", link->eng_id); - ASSERT(false); - break; - } - - return pwrseq_inst; -} - - static void link_destruct(struct dc_link *link) { int i; @@ -657,7 +633,7 @@ static bool construct_phy(struct dc_link *link, link->link_id.id == CONNECTOR_ID_LVDS)) { panel_cntl_init_data.ctx = dc_ctx; panel_cntl_init_data.inst = panel_cntl_init_data.ctx->dc_edp_id_count; - panel_cntl_init_data.pwrseq_inst = translate_dig_inst_to_pwrseq_inst(link); + panel_cntl_init_data.eng_id = link->eng_id; link->panel_cntl = link->dc->res_pool->funcs->panel_cntl_create( &panel_cntl_init_data); -- 2.43.0

11 months, 2 weeks

1
0
0 0

FAILED: Patch "drm/amd/display: Fix dcn35 8k30 Underflow/Corruption Issue" failed to apply to 6.6-stable tree

by Sasha Levin

The patch below does not apply to the 6.6-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>. Thanks, Sasha ------------------ original commit in Linus's tree ------------------ From 4ba9ca63e696f7bdc91293aeb70c22203b7089be Mon Sep 17 00:00:00 2001 From: Fangzhi Zuo <jerry.zuo(a)amd.com> Date: Thu, 11 Jan 2024 14:46:01 -0500 Subject: [PATCH] drm/amd/display: Fix dcn35 8k30 Underflow/Corruption Issue [why] odm calculation is missing for pipe split policy determination and cause Underflow/Corruption issue. [how] Add the odm calculation. Cc: Mario Limonciello <mario.limonciello(a)amd.com> Cc: Alex Deucher <alexander.deucher(a)amd.com> Cc: stable(a)vger.kernel.org Reviewed-by: Charlene Liu <charlene.liu(a)amd.com> Acked-by: Tom Chung <chiahsuan.chung(a)amd.com> Signed-off-by: Fangzhi Zuo <jerry.zuo(a)amd.com> Tested-by: Daniel Wheeler <daniel.wheeler(a)amd.com> Signed-off-by: Alex Deucher <alexander.deucher(a)amd.com> --- .../display/dc/dml2/dml2_translation_helper.c | 29 +++++++------------ .../gpu/drm/amd/display/dc/inc/core_types.h | 2 ++ 2 files changed, 13 insertions(+), 18 deletions(-) diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml2_translation_helper.c b/drivers/gpu/drm/amd/display/dc/dml2/dml2_translation_helper.c index 8b0f930be5ae1..23a608274096f 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml2_translation_helper.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml2_translation_helper.c @@ -791,35 +791,28 @@ static void populate_dml_surface_cfg_from_plane_state(enum dml_project_id dml2_p } } -/*TODO no support for mpc combine, need rework - should calculate scaling params based on plane+stream*/ -static struct scaler_data get_scaler_data_for_plane(const struct dc_plane_state *in, const struct dc_state *context) +static struct scaler_data get_scaler_data_for_plane(const struct dc_plane_state *in, struct dc_state *context) { int i; - struct scaler_data data = { 0 }; + struct pipe_ctx *temp_pipe = &context->res_ctx.temp_pipe; + + memset(temp_pipe, 0, sizeof(struct pipe_ctx)); for (i = 0; i < MAX_PIPES; i++) { const struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; if (pipe->plane_state == in && !pipe->prev_odm_pipe) { - const struct pipe_ctx *next_pipe = pipe->next_odm_pipe; - - data = context->res_ctx.pipe_ctx[i].plane_res.scl_data; - while (next_pipe) { - data.h_active += next_pipe->plane_res.scl_data.h_active; - data.recout.width += next_pipe->plane_res.scl_data.recout.width; - if (in->rotation == ROTATION_ANGLE_0 || in->rotation == ROTATION_ANGLE_180) { - data.viewport.width += next_pipe->plane_res.scl_data.viewport.width; - } else { - data.viewport.height += next_pipe->plane_res.scl_data.viewport.height; - } - next_pipe = next_pipe->next_odm_pipe; - } + temp_pipe->stream = pipe->stream; + temp_pipe->plane_state = pipe->plane_state; + temp_pipe->plane_res.scl_data.taps = pipe->plane_res.scl_data.taps; + + resource_build_scaling_params(temp_pipe); break; } } ASSERT(i < MAX_PIPES); - return data; + return temp_pipe->plane_res.scl_data; } static void populate_dummy_dml_plane_cfg(struct dml_plane_cfg_st *out, unsigned int location, const struct dc_stream_state *in) @@ -864,7 +857,7 @@ static void populate_dummy_dml_plane_cfg(struct dml_plane_cfg_st *out, unsigned out->ScalerEnabled[location] = false; } -static void populate_dml_plane_cfg_from_plane_state(struct dml_plane_cfg_st *out, unsigned int location, const struct dc_plane_state *in, const struct dc_state *context) +static void populate_dml_plane_cfg_from_plane_state(struct dml_plane_cfg_st *out, unsigned int location, const struct dc_plane_state *in, struct dc_state *context) { const struct scaler_data scaler_data = get_scaler_data_for_plane(in, context); diff --git a/drivers/gpu/drm/amd/display/dc/inc/core_types.h b/drivers/gpu/drm/amd/display/dc/inc/core_types.h index f74ae0d41d3c4..3a6bf77a68732 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/core_types.h +++ b/drivers/gpu/drm/amd/display/dc/inc/core_types.h @@ -469,6 +469,8 @@ struct resource_context { unsigned int hpo_dp_link_enc_to_link_idx[MAX_HPO_DP2_LINK_ENCODERS]; int hpo_dp_link_enc_ref_cnts[MAX_HPO_DP2_LINK_ENCODERS]; bool is_mpc_3dlut_acquired[MAX_PIPES]; + /* solely used for build scalar data in dml2 */ + struct pipe_ctx temp_pipe; }; struct dce_bw_output { -- 2.43.0

11 months, 2 weeks

1
0
0 0

FAILED: Patch "drm/amd/display: Init DPPCLK from SMU on dcn32" failed to apply to 6.6-stable tree

by Sasha Levin

The patch below does not apply to the 6.6-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>. Thanks, Sasha ------------------ original commit in Linus's tree ------------------ From 4f5b8d78ca43fcc695ba16c83ebfabbfe09506d6 Mon Sep 17 00:00:00 2001 From: Dillon Varone <dillon.varone(a)amd.com> Date: Wed, 21 Feb 2024 13:21:20 -0500 Subject: [PATCH] drm/amd/display: Init DPPCLK from SMU on dcn32 [WHY & HOW] DPPCLK ranges should be obtained from the SMU when available. Cc: Mario Limonciello <mario.limonciello(a)amd.com> Cc: Alex Deucher <alexander.deucher(a)amd.com> Cc: stable(a)vger.kernel.org Reviewed-by: Chaitanya Dhere <chaitanya.dhere(a)amd.com> Acked-by: Alex Hung <alex.hung(a)amd.com> Signed-off-by: Dillon Varone <dillon.varone(a)amd.com> Tested-by: Daniel Wheeler <daniel.wheeler(a)amd.com> Signed-off-by: Alex Deucher <alexander.deucher(a)amd.com> --- .../display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c | 14 ++++++++++ .../drm/amd/display/dc/dml2/dml2_wrapper.c | 28 +++++++++++++------ .../drm/amd/display/dc/dml2/dml2_wrapper.h | 3 ++ .../dc/resource/dcn32/dcn32_resource.c | 2 ++ .../dc/resource/dcn321/dcn321_resource.c | 2 ++ 5 files changed, 41 insertions(+), 8 deletions(-) diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c index 668f05c8654ef..bec252e1dd27a 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c @@ -216,6 +216,16 @@ void dcn32_init_clocks(struct clk_mgr *clk_mgr_base) if (clk_mgr_base->bw_params->dc_mode_limit.dispclk_mhz > 1950) clk_mgr_base->bw_params->dc_mode_limit.dispclk_mhz = 1950; + /* DPPCLK */ + dcn32_init_single_clock(clk_mgr, PPCLK_DPPCLK, + &clk_mgr_base->bw_params->clk_table.entries[0].dppclk_mhz, + &num_entries_per_clk->num_dppclk_levels); + num_levels = num_entries_per_clk->num_dppclk_levels; + clk_mgr_base->bw_params->dc_mode_limit.dppclk_mhz = dcn30_smu_get_dc_mode_max_dpm_freq(clk_mgr, PPCLK_DPPCLK); + //HW recommends limit of 1950 MHz in display clock for all DCN3.2.x + if (clk_mgr_base->bw_params->dc_mode_limit.dppclk_mhz > 1950) + clk_mgr_base->bw_params->dc_mode_limit.dppclk_mhz = 1950; + if (num_entries_per_clk->num_dcfclk_levels && num_entries_per_clk->num_dtbclk_levels && num_entries_per_clk->num_dispclk_levels) @@ -240,6 +250,10 @@ void dcn32_init_clocks(struct clk_mgr *clk_mgr_base) = khz_to_mhz_ceil(clk_mgr_base->ctx->dc->debug.min_dpp_clk_khz); } + for (i = 0; i < num_levels; i++) + if (clk_mgr_base->bw_params->clk_table.entries[i].dppclk_mhz > 1950) + clk_mgr_base->bw_params->clk_table.entries[i].dppclk_mhz = 1950; + /* Get UCLK, update bounding box */ clk_mgr_base->funcs->get_memclk_states_from_smu(clk_mgr_base); diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.c b/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.c index 2a58a7687bdb5..72cca367062e1 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.c @@ -703,13 +703,8 @@ static inline struct dml2_context *dml2_allocate_memory(void) return (struct dml2_context *) kzalloc(sizeof(struct dml2_context), GFP_KERNEL); } -bool dml2_create(const struct dc *in_dc, const struct dml2_configuration_options *config, struct dml2_context **dml2) +static void dml2_init(const struct dc *in_dc, const struct dml2_configuration_options *config, struct dml2_context **dml2) { - // Allocate Mode Lib Ctx - *dml2 = dml2_allocate_memory(); - - if (!(*dml2)) - return false; // Store config options (*dml2)->config = *config; @@ -737,9 +732,18 @@ bool dml2_create(const struct dc *in_dc, const struct dml2_configuration_options initialize_dml2_soc_bbox(*dml2, in_dc, &(*dml2)->v20.dml_core_ctx.soc); initialize_dml2_soc_states(*dml2, in_dc, &(*dml2)->v20.dml_core_ctx.soc, &(*dml2)->v20.dml_core_ctx.states); +} + +bool dml2_create(const struct dc *in_dc, const struct dml2_configuration_options *config, struct dml2_context **dml2) +{ + // Allocate Mode Lib Ctx + *dml2 = dml2_allocate_memory(); + + if (!(*dml2)) + return false; + + dml2_init(in_dc, config, dml2); - /*Initialize DML20 instance which calls dml2_core_create, and core_dcn3_populate_informative*/ - //dml2_initialize_instance(&(*dml_ctx)->v20.dml_init); return true; } @@ -779,3 +783,11 @@ bool dml2_create_copy(struct dml2_context **dst_dml2, return true; } + +void dml2_reinit(const struct dc *in_dc, + const struct dml2_configuration_options *config, + struct dml2_context **dml2) +{ + + dml2_init(in_dc, config, dml2); +} diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.h b/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.h index ee0eb184eb6d7..cc662d682fd4d 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.h +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.h @@ -214,6 +214,9 @@ void dml2_copy(struct dml2_context *dst_dml2, struct dml2_context *src_dml2); bool dml2_create_copy(struct dml2_context **dst_dml2, struct dml2_context *src_dml2); +void dml2_reinit(const struct dc *in_dc, + const struct dml2_configuration_options *config, + struct dml2_context **dml2); /* * dml2_validate - Determines if a display configuration is supported or not. diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c index f844f57ecc49b..ce1754cc1f463 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c @@ -1931,6 +1931,8 @@ static void dcn32_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw { DC_FP_START(); dcn32_update_bw_bounding_box_fpu(dc, bw_params); + if (dc->debug.using_dml2 && dc->current_state && dc->current_state->bw_ctx.dml2) + dml2_reinit(dc, &dc->dml2_options, &dc->current_state->bw_ctx.dml2); DC_FP_END(); } diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn321/dcn321_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn321/dcn321_resource.c index b356fed1726d9..296a0a8e71459 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn321/dcn321_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn321/dcn321_resource.c @@ -1581,6 +1581,8 @@ static void dcn321_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *b { DC_FP_START(); dcn321_update_bw_bounding_box_fpu(dc, bw_params); + if (dc->debug.using_dml2 && dc->current_state && dc->current_state->bw_ctx.dml2) + dml2_reinit(dc, &dc->dml2_options, &dc->current_state->bw_ctx.dml2); DC_FP_END(); } -- 2.43.0

11 months, 2 weeks

1
0
0 0

FAILED: Patch "drm/amd/display: Update phantom pipe enable / disable sequence" failed to apply to 6.6-stable tree

by Sasha Levin

The patch below does not apply to the 6.6-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>. Thanks, Sasha ------------------ original commit in Linus's tree ------------------ From 6a068e64fb25dbc81256fc03db0d4579d222bccd Mon Sep 17 00:00:00 2001 From: Alvin Lee <alvin.lee2(a)amd.com> Date: Fri, 26 Jan 2024 16:47:20 -0500 Subject: [PATCH] drm/amd/display: Update phantom pipe enable / disable sequence Previously we would call apply_ctx_to_hw to enable and disable phantom pipes. However, apply_ctx_to_hw can potentially update non-phantom pipes as well which is undesired. Instead of calling apply_ctx_to_hw as a whole, call the relevant helpers for each phantom pipe when enabling / disabling which will avoid us modifying hardware state for non-phantom pipes unknowingly. The use case is for an FRL display where FRL_Update is requested by the display. In this case link_state_valid flag is cleared in a passive callback thread and should be handled in the next stream / link update. However, due to the call to apply_ctx_to_hw for the phantom pipes during a flip, the main pipes were modified outside of the desired sequence (driver does not handle link_state_valid = 0 on flips). Cc: stable(a)vger.kernel.org # 6.6+ Reviewed-by: Samson Tam <samson.tam(a)amd.com> Acked-by: Hamza Mahfooz <hamza.mahfooz(a)amd.com> Signed-off-by: Alvin Lee <alvin.lee2(a)amd.com> Signed-off-by: Alex Deucher <alexander.deucher(a)amd.com> --- drivers/gpu/drm/amd/display/dc/core/dc.c | 4 +- .../amd/display/dc/hwss/dce110/dce110_hwseq.c | 4 +- .../amd/display/dc/hwss/dce110/dce110_hwseq.h | 4 + .../amd/display/dc/hwss/dcn20/dcn20_hwseq.c | 2 +- .../amd/display/dc/hwss/dcn20/dcn20_hwseq.h | 4 + .../amd/display/dc/hwss/dcn32/dcn32_hwseq.c | 76 ++++++++++++++++--- .../amd/display/dc/hwss/dcn32/dcn32_hwseq.h | 2 + .../amd/display/dc/hwss/dcn32/dcn32_init.c | 3 + .../drm/amd/display/dc/hwss/hw_sequencer.h | 1 + .../display/dc/hwss/hw_sequencer_private.h | 7 ++ 10 files changed, 94 insertions(+), 13 deletions(-) diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c index c81f8af5d3741..72512903f88f7 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc.c @@ -3851,7 +3851,9 @@ static void commit_planes_for_stream(struct dc *dc, * programming has completed (we turn on phantom OTG in order * to complete the plane disable for phantom pipes). */ - dc->hwss.apply_ctx_to_hw(dc, context); + + if (dc->hwss.disable_phantom_streams) + dc->hwss.disable_phantom_streams(dc, context); } if (update_type != UPDATE_TYPE_FAST) diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c index 3255034749e2c..9d5df4c0da597 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c @@ -1529,7 +1529,7 @@ static enum dc_status dce110_enable_stream_timing( return DC_OK; } -static enum dc_status apply_single_controller_ctx_to_hw( +enum dc_status dce110_apply_single_controller_ctx_to_hw( struct pipe_ctx *pipe_ctx, struct dc_state *context, struct dc *dc) @@ -2356,7 +2356,7 @@ enum dc_status dce110_apply_ctx_to_hw( if (pipe_ctx->top_pipe || pipe_ctx->prev_odm_pipe) continue; - status = apply_single_controller_ctx_to_hw( + status = dce110_apply_single_controller_ctx_to_hw( pipe_ctx, context, dc); diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.h b/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.h index 08028a1779ae8..ed3cc3648e8e2 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.h @@ -39,6 +39,10 @@ enum dc_status dce110_apply_ctx_to_hw( struct dc *dc, struct dc_state *context); +enum dc_status dce110_apply_single_controller_ctx_to_hw( + struct pipe_ctx *pipe_ctx, + struct dc_state *context, + struct dc *dc); void dce110_enable_stream(struct pipe_ctx *pipe_ctx); diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c index 7557e58f58b24..bc0a21957e33f 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c @@ -2671,7 +2671,7 @@ void dcn20_setup_vupdate_interrupt(struct dc *dc, struct pipe_ctx *pipe_ctx) tg->funcs->setup_vertical_interrupt2(tg, start_line); } -static void dcn20_reset_back_end_for_pipe( +void dcn20_reset_back_end_for_pipe( struct dc *dc, struct pipe_ctx *pipe_ctx, struct dc_state *context) diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.h b/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.h index 90316327e6fc5..5c874f7b0683e 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.h @@ -86,6 +86,10 @@ enum dc_status dcn20_enable_stream_timing( void dcn20_disable_stream_gating(struct dc *dc, struct pipe_ctx *pipe_ctx); void dcn20_enable_stream_gating(struct dc *dc, struct pipe_ctx *pipe_ctx); void dcn20_setup_vupdate_interrupt(struct dc *dc, struct pipe_ctx *pipe_ctx); +void dcn20_reset_back_end_for_pipe( + struct dc *dc, + struct pipe_ctx *pipe_ctx, + struct dc_state *context); void dcn20_init_blank( struct dc *dc, struct timing_generator *tg); diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.c index 6c9299c7683df..aa36d7a56ca8c 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.c @@ -1474,9 +1474,44 @@ void dcn32_update_dsc_pg(struct dc *dc, } } +void dcn32_disable_phantom_streams(struct dc *dc, struct dc_state *context) +{ + struct dce_hwseq *hws = dc->hwseq; + int i; + + for (i = dc->res_pool->pipe_count - 1; i >= 0 ; i--) { + struct pipe_ctx *pipe_ctx_old = + &dc->current_state->res_ctx.pipe_ctx[i]; + struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; + + if (!pipe_ctx_old->stream) + continue; + + if (dc_state_get_pipe_subvp_type(dc->current_state, pipe_ctx_old) != SUBVP_PHANTOM) + continue; + + if (pipe_ctx_old->top_pipe || pipe_ctx_old->prev_odm_pipe) + continue; + + if (!pipe_ctx->stream || pipe_need_reprogram(pipe_ctx_old, pipe_ctx) || + (pipe_ctx->stream && dc_state_get_pipe_subvp_type(context, pipe_ctx) != SUBVP_PHANTOM)) { + struct clock_source *old_clk = pipe_ctx_old->clock_source; + + if (hws->funcs.reset_back_end_for_pipe) + hws->funcs.reset_back_end_for_pipe(dc, pipe_ctx_old, dc->current_state); + if (hws->funcs.enable_stream_gating) + hws->funcs.enable_stream_gating(dc, pipe_ctx_old); + if (old_clk) + old_clk->funcs->cs_power_down(old_clk); + } + } +} + void dcn32_enable_phantom_streams(struct dc *dc, struct dc_state *context) { unsigned int i; + enum dc_status status = DC_OK; + struct dce_hwseq *hws = dc->hwseq; for (i = 0; i < dc->res_pool->pipe_count; i++) { struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; @@ -1497,16 +1532,39 @@ void dcn32_enable_phantom_streams(struct dc *dc, struct dc_state *context) } } for (i = 0; i < dc->res_pool->pipe_count; i++) { - struct pipe_ctx *new_pipe = &context->res_ctx.pipe_ctx[i]; - - if (new_pipe->stream && dc_state_get_pipe_subvp_type(context, new_pipe) == SUBVP_PHANTOM) { - // If old context or new context has phantom pipes, apply - // the phantom timings now. We can't change the phantom - // pipe configuration safely without driver acquiring - // the DMCUB lock first. - dc->hwss.apply_ctx_to_hw(dc, context); - break; + struct pipe_ctx *pipe_ctx_old = + &dc->current_state->res_ctx.pipe_ctx[i]; + struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; + + if (pipe_ctx->stream == NULL) + continue; + + if (dc_state_get_pipe_subvp_type(context, pipe_ctx) != SUBVP_PHANTOM) + continue; + + if (pipe_ctx->stream == pipe_ctx_old->stream && + pipe_ctx->stream->link->link_state_valid) { + continue; } + + if (pipe_ctx_old->stream && !pipe_need_reprogram(pipe_ctx_old, pipe_ctx)) + continue; + + if (pipe_ctx->top_pipe || pipe_ctx->prev_odm_pipe) + continue; + + if (hws->funcs.apply_single_controller_ctx_to_hw) + status = hws->funcs.apply_single_controller_ctx_to_hw( + pipe_ctx, + context, + dc); + + ASSERT(status == DC_OK); + +#ifdef CONFIG_DRM_AMD_DC_FP + if (hws->funcs.resync_fifo_dccg_dio) + hws->funcs.resync_fifo_dccg_dio(hws, dc, context); +#endif } } diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.h b/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.h index cecf7f0f56719..069e20bc87c0a 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.h @@ -111,6 +111,8 @@ void dcn32_update_dsc_pg(struct dc *dc, void dcn32_enable_phantom_streams(struct dc *dc, struct dc_state *context); +void dcn32_disable_phantom_streams(struct dc *dc, struct dc_state *context); + void dcn32_init_blank( struct dc *dc, struct timing_generator *tg); diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_init.c index 0980df6c65ea4..2b073123d3ede 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_init.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_init.c @@ -109,6 +109,7 @@ static const struct hw_sequencer_funcs dcn32_funcs = { .get_dcc_en_bits = dcn10_get_dcc_en_bits, .commit_subvp_config = dcn32_commit_subvp_config, .enable_phantom_streams = dcn32_enable_phantom_streams, + .disable_phantom_streams = dcn32_disable_phantom_streams, .subvp_pipe_control_lock = dcn32_subvp_pipe_control_lock, .update_visual_confirm_color = dcn10_update_visual_confirm_color, .subvp_pipe_control_lock_fast = dcn32_subvp_pipe_control_lock_fast, @@ -159,6 +160,8 @@ static const struct hwseq_private_funcs dcn32_private_funcs = { .set_pixels_per_cycle = dcn32_set_pixels_per_cycle, .resync_fifo_dccg_dio = dcn32_resync_fifo_dccg_dio, .is_dp_dig_pixel_rate_div_policy = dcn32_is_dp_dig_pixel_rate_div_policy, + .apply_single_controller_ctx_to_hw = dce110_apply_single_controller_ctx_to_hw, + .reset_back_end_for_pipe = dcn20_reset_back_end_for_pipe, }; void dcn32_hw_sequencer_init_functions(struct dc *dc) diff --git a/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer.h index 91b1b43a728fd..f89f205e42a1a 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer.h @@ -381,6 +381,7 @@ struct hw_sequencer_funcs { struct dc_cursor_attributes *cursor_attr); void (*commit_subvp_config)(struct dc *dc, struct dc_state *context); void (*enable_phantom_streams)(struct dc *dc, struct dc_state *context); + void (*disable_phantom_streams)(struct dc *dc, struct dc_state *context); void (*subvp_pipe_control_lock)(struct dc *dc, struct dc_state *context, bool lock, diff --git a/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer_private.h b/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer_private.h index 6137cf09aa54d..b3c62a82cb1cf 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer_private.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer_private.h @@ -165,8 +165,15 @@ struct hwseq_private_funcs { void (*set_pixels_per_cycle)(struct pipe_ctx *pipe_ctx); void (*resync_fifo_dccg_dio)(struct dce_hwseq *hws, struct dc *dc, struct dc_state *context); + enum dc_status (*apply_single_controller_ctx_to_hw)( + struct pipe_ctx *pipe_ctx, + struct dc_state *context, + struct dc *dc); bool (*is_dp_dig_pixel_rate_div_policy)(struct pipe_ctx *pipe_ctx); #endif + void (*reset_back_end_for_pipe)(struct dc *dc, + struct pipe_ctx *pipe_ctx, + struct dc_state *context); }; struct dce_hwseq { -- 2.43.0

11 months, 2 weeks

1
0
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror March 2024