arch: loongarch: replace BPI implementation with a minimal patchset #356

MingcongBai · 2024-08-09T03:52:33Z

Introduce old-world firmware boot support with a more minimal patchset. Also fixes shutdown on systems that do not declare ACPI S5 support:

loongarch/reset: use general efi poweroff
loongarch: fix missing dependency info in DSDT
loongarch: correct missing offset of PCI root controller in DSDT table
loongarch: add MADT ACPI table conversion
loongarch: parse BPI data and add memory mapping
loongarch: basic boot support for legacy firmware
Revert "LoongArch: add kernel setvirtmap for runtime"
Revert "LoongArch: Old BPI compatibility"
Revert "LoongArch: Fix virtual machine startup error"
Revert "LoongArch: Fixed EIOINTC structure members"
Revert "LoongArch: use arch specific phys_to_dma"
Revert "LoongArch: fix KASLR can not be disabled by nokaslr when boot from old BPI"

Tested on ML5A, Lenovo Kaitian M540z, and Excelsior L71 (laptop).

… from old BPI" This reverts commit ff445a1.

This reverts commit 43bef49.

This reverts commit e6e5f92.

This reverts commit 20c6c81.

This reverts commit 59eab58.

This reverts commit d7aad0c.

github-actions · 2024-08-09T03:52:45Z

arch/loongarch/include/asm/acpi.h

@@ -44,6 +44,15 @@ static inline u32 get_acpi_id_for_cpu(unsigned int cpu)
 	return acpi_core_pic[cpu_logical_map(cpu)].processor_id;
 }

+#define ACPI_HAVE_ARCH_TABLE_OVERRIDE
+extern void acpi_arch_os_table_override (struct acpi_table_header *existing_table, struct acpi_table_header **new_table);


WARNING: Prefer a maximum 75 chars per line (possible unwrapped commit description?)

arch/loongarch/kernel/Makefile

github-actions · 2024-08-09T03:52:47Z

arch/loongarch/kernel/Makefile

@@ -11,7 +11,6 @@ obj-y		+= head.o cpu-probe.o cacheinfo.o env.o setup.o entry.o genex.o \
 		   traps.o irq.o idle.o process.o dma.o mem.o io.o reset.o switch.o \
 		   elf.o syscall.o signal.o time.o topology.o inst.o ptrace.o vdso.o \
 		   alternative.o unwind.o


ERROR: patch seems to be corrupt (line wrapped?)

github-actions · 2024-08-09T03:52:48Z

arch/loongarch/kernel/dma.c

-	long nid = (daddr >> node_id_offset) & 0xf;
-
-	return ((nid << node_id_offset) ^ daddr) | (nid << 44);
-}

 void acpi_arch_dma_setup(struct device *dev)
 {
 	int ret;
 	u64 mask, end = 0;
 	const struct bus_dma_region *map = NULL;



WARNING: Prefer a maximum 75 chars per line (possible unwrapped commit description?)

github-actions · 2024-08-09T03:52:49Z

arch/loongarch/kernel/legacy_boot.h

-#define LOONGSON3_BOOT_MEM_MAP_MAX	128
-#define RT_MAP_START			100
-#define FIX_MAP_ENTRY			32
+#define LOONGARCH_BPI_GUID  EFI_GUID(0x4660f721, 0x2ec5, 0x416a,  0x89, 0x9a, 0x43, 0x18, 0x02, 0x50, 0xa0, 0xc9)


WARNING: Prefer a maximum 75 chars per line (possible unwrapped commit description?)

github-actions · 2024-08-09T03:52:50Z

arch/loongarch/kernel/mem.c

 void __init memblock_init(void)
 {
-	u32 i, mem_type;
+	u32 mem_type;
 	u64 mem_start, mem_end, mem_size;
 	efi_memory_desc_t *md;


WARNING: Prefer a maximum 75 chars per line (possible unwrapped commit description?)

github-actions · 2024-08-09T03:52:52Z

drivers/firmware/efi/libstub/loongarch.c

@@ -23,6 +23,8 @@ struct exit_boot_struct {
 	int			runtime_entry_count;
 };

+static int is_oldworld = 0;
+
 static efi_status_t exit_boot_func(struct efi_boot_memmap *map, void *priv)


WARNING: Prefer a maximum 75 chars per line (possible unwrapped commit description?)

github-actions · 2024-08-09T03:52:52Z

drivers/irqchip/irq-loongson-eiointc.c

-	int cores = (cpu_has_hypervisor ? MAX_CORES_PER_EIO_NODE : CORES_PER_EIO_NODE);
-
-	return cpu_logical_map(cpu) / cores;
+	return cpu_logical_map(cpu) / CORES_PER_EIO_NODE;
 }

 static void eiointc_set_irq_route(int pos, unsigned int cpu, unsigned int mnode, nodemask_t *node_map)


WARNING: Prefer a maximum 75 chars per line (possible unwrapped commit description?)

github-actions · 2024-08-09T03:52:53Z

include/linux/acpi.h

@@ -762,6 +762,16 @@ static inline u64 acpi_arch_get_root_pointer(void)
 int acpi_get_local_address(acpi_handle handle, u32 *addr);
 const char *acpi_get_subsystem_id(acpi_handle handle);

+#ifndef ACPI_HAVE_ARCH_TABLE_OVERRIDE
+static inline void acpi_arch_os_table_override (struct acpi_table_header *existing_table, struct acpi_table_header **new_table){


WARNING: Prefer a maximum 75 chars per line (possible unwrapped commit description?)

github-actions · 2024-08-09T03:52:54Z

include/linux/pci-acpi.h

@@ -133,6 +133,10 @@ static inline void pci_acpi_remove_edr_notifier(struct pci_dev *pdev) { }
 int pci_acpi_set_companion_lookup_hook(struct acpi_device *(*func)(struct pci_dev *));
 void pci_acpi_clear_companion_lookup_hook(void);

+#ifndef ACPI_HAVE_ARCH_PCI_ROOT_RES_FILTER
+static inline void acpi_arch_pci_probe_root_dev_filter(struct resource_entry *entry) { }


WARNING: Prefer a maximum 75 chars per line (possible unwrapped commit description?)

Currently efi_shutdown_init can register a general sys_off handler, efi_power_off, which will be called during do_kernel_power_off and shut the machine off via efi runtime services. So enable this by providing efi_poweroff_required, like arm and x86, and prevent directly calling efi.reset_system in machine_power_off. Signed-off-by: Miao Wang <shankerwangmiao@gmail.com>

The addresses passed from legacy firmware in efi system tables, the initrd table, the efi system memory mapping table are virtual addresses. This patch converts all that addresses back to physical addresses, to make sure all the following booting processes can run smoothly like on modern firmwares. The conversion happens unconditionally, since physical addresses passed in by modern firmwares remain unchanged after the conversion. In the EFI Stub, the mapping entries given by GetMemoryMap() on legacy firmwares contain virtual addresses in the phys_addr fields and 0x1xxxx addresses in the virt_addr fields, causing the later call to SetVirtualAddressMap() fails. This patch fixes this by correcting the addresses in the virt_addr fields. This patch detects the existence of the legacy firmware by reading DMW1 CSR, as done in the legacy loongarch GRUB port. Only if legacy firmwares detected, this correction happens. With this patch, the linux kernel is basically able to boot.

github-actions · 2024-08-14T09:09:56Z

arch/loongarch/include/asm/acpi.h

@@ -44,6 +44,15 @@ static inline u32 get_acpi_id_for_cpu(unsigned int cpu)
 	return acpi_core_pic[cpu_logical_map(cpu)].processor_id;
 }

+#define ACPI_HAVE_ARCH_TABLE_OVERRIDE
+extern void acpi_arch_os_table_override (struct acpi_table_header *existing_table, struct acpi_table_header **new_table);


WARNING: Prefer a maximum 75 chars per line (possible unwrapped commit description?)

github-actions · 2024-08-14T09:09:58Z

arch/loongarch/kernel/Makefile

@@ -11,7 +11,6 @@ obj-y		+= head.o cpu-probe.o cacheinfo.o env.o setup.o entry.o genex.o \
 		   traps.o irq.o idle.o process.o dma.o mem.o io.o reset.o switch.o \
 		   elf.o syscall.o signal.o time.o topology.o inst.o ptrace.o vdso.o \
 		   alternative.o unwind.o


ERROR: Avoid using diff content in the commit message - patch(1) might not work

github-actions · 2024-08-14T09:09:58Z

arch/loongarch/kernel/Makefile

@@ -11,7 +11,6 @@ obj-y		+= head.o cpu-probe.o cacheinfo.o env.o setup.o entry.o genex.o \
 		   traps.o irq.o idle.o process.o dma.o mem.o io.o reset.o switch.o \
 		   elf.o syscall.o signal.o time.o topology.o inst.o ptrace.o vdso.o \
 		   alternative.o unwind.o


ERROR: patch seems to be corrupt (line wrapped?)

github-actions · 2024-08-14T09:09:59Z

arch/loongarch/kernel/dma.c

-	long nid = (daddr >> node_id_offset) & 0xf;
-
-	return ((nid << node_id_offset) ^ daddr) | (nid << 44);
-}

 void acpi_arch_dma_setup(struct device *dev)
 {
 	int ret;
 	u64 mask, end = 0;
 	const struct bus_dma_region *map = NULL;



WARNING: Prefer a maximum 75 chars per line (possible unwrapped commit description?)

github-actions · 2024-08-14T09:10:00Z

arch/loongarch/kernel/legacy_boot.h

-#define LOONGSON3_BOOT_MEM_MAP_MAX	128
-#define RT_MAP_START			100
-#define FIX_MAP_ENTRY			32
+#define LOONGARCH_BPI_GUID  EFI_GUID(0x4660f721, 0x2ec5, 0x416a,  0x89, 0x9a, 0x43, 0x18, 0x02, 0x50, 0xa0, 0xc9)


WARNING: Prefer a maximum 75 chars per line (possible unwrapped commit description?)

github-actions · 2024-08-14T09:10:01Z

arch/loongarch/kernel/mem.c

 void __init memblock_init(void)
 {
-	u32 i, mem_type;
+	u32 mem_type;
 	u64 mem_start, mem_end, mem_size;
 	efi_memory_desc_t *md;


WARNING: Prefer a maximum 75 chars per line (possible unwrapped commit description?)

github-actions · 2024-08-14T09:10:03Z

drivers/firmware/efi/libstub/loongarch.c

@@ -23,6 +23,8 @@ struct exit_boot_struct {
 	int			runtime_entry_count;
 };

+static int is_oldworld = 0;
+
 static efi_status_t exit_boot_func(struct efi_boot_memmap *map, void *priv)


WARNING: Prefer a maximum 75 chars per line (possible unwrapped commit description?)

On machines with legacy firmware with BPI version BPI01000, i.e. the older version, the content of the MADT ACPI table is not following the later finalized ACPI standard. A conversion of the content is added by this patch. Also the patch generates and adds a MCFG table accordingly. The above behavior is only enabled when BPI data and the BPI01000 version is detected.

On machines with legacy firmware with BPI version BPI01000, the PCI root controller information in the DSDT table lacks AddressTranslation in its WordIO resource defination, causing the failure of the registration for the LIO address space. This patch corrects this issue when the given offset is zero for the PCI root controller. This patch also fixes the lack of the leading 16K, i.e. ISA_IOSIZE, in the defination of WordIO resource. This is because that address range is registered unconditionally on the legacy loongarch linux port. This patch also fixes the start addresses or end addresses of WordIO resource not aligned to the page size by rouding the addresses up to the nearest page starting. The above behavior is only enabled when BPI data and the BPI01000 version is detected.

On machines with legacy firmware with BPI01000 version, the devices on the LIO bus lackes the dependency on the PCI root controller, causing the memory-mapped address of the legacy IO ports read before the setup of the mapping, resulting in kernel panic. Such DSDT can work on the legacy loongarch linux port because the leading 16K is unconditionally registered, before the enumeration of the devices in the DSDT table. This patch addes such dependency info, to order the initialization of the devices on the LIO bus after the initialization of the PCI root controller, fixing this problem. However, the addition should be done on each possible LIO device, and currently the patch only includes the legacy EC device on some laptops located at the path \_SB.PCI0.LPC.EC. Thus, this patch will be improved to include more devices. The above behavior is only enabled when BPI data and the BPI01000 version is detected.

Loongarch CPUs are using HT bus interface, which is only 40-bit wide, to communicate with the bridge chipset. However, the actual memory bus of the CPUs is 48-bit wide, and the available address space of DMA requests of PCI devices is also larger than the 40-bit address space. The address space of loongarch systems is not continuous, the bits [47:44] of which are used to denote the belonging node id, which is far beyond the space provided by the 40-bit wide HT bus. As a result, a translation on the both LS7A and CPU sides is needed. The translation happens on the LS7A side is controlled by the higher half of the register, HT_ROUTE. Bits [12:8] denotes dma_node_id_offset and bits[15:13] denotes dma_node_id_offset_mapped. The behavior of the translation is that the chip extracts the node id from bit dma_node_id_offset + 36 of a DMA address, places it at bit dma_node_id_offset_mapped + 32, and generates the address on the HT bus. On the CPU side, an alike translation happens, to convert the address on the HT bus back to a proper memory address. On machines with legacy firmware with BPI01000 version, dma_node_id_offset is configured with 0, resulting the address which should be used by the DMA engine of a PCI device differs from the actural physical memeory address, which requires a pair of arch-specific phys_to_dma and dma_to_phys functions or setting up the whole mapping in the _DMA method of the PCI root device in the DSDT table. The former method requires we add back the prevoiusly removed functions, and the latter method degrades the performance since when the translation happens, the mapping table is scanned linearly. This patch addresses this issue by directly setting dma_node_id_offset to 8, like what is done by modern firmwares, making the address used by the DMA engine just the same as the actual physical memory address, and eliminating the need of DMA address translation on the kernel side.

On machines with legacy firmware with BPI01000 version, the HT_RX_INT_TRANS register on the node 5, i.e. the node connected with the second LS7A bridge chip, is not initialized correctly, causing the failure of the delivery of interrupts from PCIe devices connected to the second LS7A bridge chip. This patch fixes this by correctly pointing the HT_RX_INT_TRANS register to the EXT_IOI_SEND_OFF register on the corresponding node.

github-actions · 2024-08-14T11:13:57Z

arch/loongarch/include/asm/acpi.h

@@ -44,6 +44,15 @@ static inline u32 get_acpi_id_for_cpu(unsigned int cpu)
 	return acpi_core_pic[cpu_logical_map(cpu)].processor_id;
 }

+#define ACPI_HAVE_ARCH_TABLE_OVERRIDE
+extern void acpi_arch_os_table_override (struct acpi_table_header *existing_table, struct acpi_table_header **new_table);


WARNING: Prefer a maximum 75 chars per line (possible unwrapped commit description?)

github-actions · 2024-08-14T11:13:58Z

arch/loongarch/kernel/Makefile

@@ -11,7 +11,6 @@ obj-y		+= head.o cpu-probe.o cacheinfo.o env.o setup.o entry.o genex.o \
 		   traps.o irq.o idle.o process.o dma.o mem.o io.o reset.o switch.o \
 		   elf.o syscall.o signal.o time.o topology.o inst.o ptrace.o vdso.o \
 		   alternative.o unwind.o


ERROR: Avoid using diff content in the commit message - patch(1) might not work

github-actions · 2024-08-14T11:13:59Z

arch/loongarch/kernel/Makefile

@@ -11,7 +11,6 @@ obj-y		+= head.o cpu-probe.o cacheinfo.o env.o setup.o entry.o genex.o \
 		   traps.o irq.o idle.o process.o dma.o mem.o io.o reset.o switch.o \
 		   elf.o syscall.o signal.o time.o topology.o inst.o ptrace.o vdso.o \
 		   alternative.o unwind.o


ERROR: patch seems to be corrupt (line wrapped?)

github-actions · 2024-08-14T11:14:00Z

arch/loongarch/kernel/dma.c

-	long nid = (daddr >> node_id_offset) & 0xf;
-
-	return ((nid << node_id_offset) ^ daddr) | (nid << 44);
-}

 void acpi_arch_dma_setup(struct device *dev)
 {
 	int ret;
 	u64 mask, end = 0;
 	const struct bus_dma_region *map = NULL;



WARNING: Prefer a maximum 75 chars per line (possible unwrapped commit description?)

github-actions · 2024-08-14T11:14:01Z

arch/loongarch/kernel/legacy_boot.h

-#define LOONGSON3_BOOT_MEM_MAP_MAX	128
-#define RT_MAP_START			100
-#define FIX_MAP_ENTRY			32
+#define LOONGARCH_BPI_GUID  EFI_GUID(0x4660f721, 0x2ec5, 0x416a,  0x89, 0x9a, 0x43, 0x18, 0x02, 0x50, 0xa0, 0xc9)


WARNING: Prefer a maximum 75 chars per line (possible unwrapped commit description?)

github-actions · 2024-08-14T11:14:02Z

arch/loongarch/kernel/mem.c

 void __init memblock_init(void)
 {
-	u32 i, mem_type;
+	u32 mem_type;
 	u64 mem_start, mem_end, mem_size;
 	efi_memory_desc_t *md;


WARNING: Prefer a maximum 75 chars per line (possible unwrapped commit description?)

deepin-ci-robot · 2024-08-14T11:14:04Z

deepin pr auto review

ACPI: Fix the detection of old world firmware

This patch fixes the detection of old world firmware.

Signed-off-by: Alvaro Saurin ossa-git@ossa.org

github-actions · 2024-08-14T11:14:04Z

drivers/firmware/efi/libstub/loongarch.c

@@ -23,6 +23,8 @@ struct exit_boot_struct {
 	int			runtime_entry_count;
 };

+static int is_oldworld = 0;
+
 static efi_status_t exit_boot_func(struct efi_boot_memmap *map, void *priv)


WARNING: Prefer a maximum 75 chars per line (possible unwrapped commit description?)

github-actions · 2024-08-14T11:14:04Z

drivers/irqchip/irq-loongson-eiointc.c

-	int cores = (cpu_has_hypervisor ? MAX_CORES_PER_EIO_NODE : CORES_PER_EIO_NODE);
-
-	return cpu_logical_map(cpu) / cores;
+	return cpu_logical_map(cpu) / CORES_PER_EIO_NODE;
 }

 static void eiointc_set_irq_route(int pos, unsigned int cpu, unsigned int mnode, nodemask_t *node_map)


WARNING: Prefer a maximum 75 chars per line (possible unwrapped commit description?)

github-actions · 2024-08-14T11:14:05Z

include/linux/acpi.h

@@ -762,6 +762,16 @@ static inline u64 acpi_arch_get_root_pointer(void)
 int acpi_get_local_address(acpi_handle handle, u32 *addr);
 const char *acpi_get_subsystem_id(acpi_handle handle);

+#ifndef ACPI_HAVE_ARCH_TABLE_OVERRIDE
+static inline void acpi_arch_os_table_override (struct acpi_table_header *existing_table, struct acpi_table_header **new_table){


WARNING: Prefer a maximum 75 chars per line (possible unwrapped commit description?)

github-actions · 2024-08-14T11:14:06Z

include/linux/pci-acpi.h

@@ -133,6 +133,10 @@ static inline void pci_acpi_remove_edr_notifier(struct pci_dev *pdev) { }
 int pci_acpi_set_companion_lookup_hook(struct acpi_device *(*func)(struct pci_dev *));
 void pci_acpi_clear_companion_lookup_hook(void);

+#ifndef ACPI_HAVE_ARCH_PCI_ROOT_RES_FILTER
+static inline void acpi_arch_pci_probe_root_dev_filter(struct resource_entry *entry) { }


WARNING: Prefer a maximum 75 chars per line (possible unwrapped commit description?)

deepin-ci-robot · 2024-08-14T11:14:09Z

[APPROVALNOTIFIER] This PR is NOT APPROVED

This pull-request has been approved by:
Once this PR has been reviewed and has the lgtm label, please ask for approval from mingcongbai. For more information see the Code Review Process.

The full list of commands accepted by this bot can be found here.

Needs approval from an approver in each of these files:

deepin/OWNERS

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

MingcongBai · 2024-08-14T13:55:41Z

Tested good on the following:

Lenovo Kaitian M540z
Excelsior L71
706 ML5A
4C5LG (4-way 3C5000L/7A1000)

Also backported to rolling-stable (6.9.6).

Avenger-285714 · 2024-08-14T13:58:03Z

Great!

commit 4aa923a6e6406b43566ef6ac35a3d9a3197fa3e8 upstream. KASAN reports that the GPU metrics table allocated in vangogh_tables_init() is not large enough for the memset done in smu_cmn_init_soft_gpu_metrics(). Condensed report follows: [ 33.861314] BUG: KASAN: slab-out-of-bounds in smu_cmn_init_soft_gpu_metrics+0x73/0x200 [amdgpu] [ 33.861799] Write of size 168 at addr ffff888129f59500 by task mangoapp/1067 ... [ 33.861808] CPU: 6 UID: 1000 PID: 1067 Comm: mangoapp Tainted: G W 6.12.0-rc4 deepin-community#356 1a56f59a8b5182eeaf67eb7cb8b13594dd23b544 [ 33.861816] Tainted: [W]=WARN [ 33.861818] Hardware name: Valve Galileo/Galileo, BIOS F7G0107 12/01/2023 [ 33.861822] Call Trace: [ 33.861826] <TASK> [ 33.861829] dump_stack_lvl+0x66/0x90 [ 33.861838] print_report+0xce/0x620 [ 33.861853] kasan_report+0xda/0x110 [ 33.862794] kasan_check_range+0xfd/0x1a0 [ 33.862799] __asan_memset+0x23/0x40 [ 33.862803] smu_cmn_init_soft_gpu_metrics+0x73/0x200 [amdgpu 13b1bc364ec578808f676eba412c20eaab792779] [ 33.863306] vangogh_get_gpu_metrics_v2_4+0x123/0xad0 [amdgpu 13b1bc364ec578808f676eba412c20eaab792779] [ 33.864257] vangogh_common_get_gpu_metrics+0xb0c/0xbc0 [amdgpu 13b1bc364ec578808f676eba412c20eaab792779] [ 33.865682] amdgpu_dpm_get_gpu_metrics+0xcc/0x110 [amdgpu 13b1bc364ec578808f676eba412c20eaab792779] [ 33.866160] amdgpu_get_gpu_metrics+0x154/0x2d0 [amdgpu 13b1bc364ec578808f676eba412c20eaab792779] [ 33.867135] dev_attr_show+0x43/0xc0 [ 33.867147] sysfs_kf_seq_show+0x1f1/0x3b0 [ 33.867155] seq_read_iter+0x3f8/0x1140 [ 33.867173] vfs_read+0x76c/0xc50 [ 33.867198] ksys_read+0xfb/0x1d0 [ 33.867214] do_syscall_64+0x90/0x160 ... [ 33.867353] Allocated by task 378 on cpu 7 at 22.794876s: [ 33.867358] kasan_save_stack+0x33/0x50 [ 33.867364] kasan_save_track+0x17/0x60 [ 33.867367] __kasan_kmalloc+0x87/0x90 [ 33.867371] vangogh_init_smc_tables+0x3f9/0x840 [amdgpu] [ 33.867835] smu_sw_init+0xa32/0x1850 [amdgpu] [ 33.868299] amdgpu_device_init+0x467b/0x8d90 [amdgpu] [ 33.868733] amdgpu_driver_load_kms+0x19/0xf0 [amdgpu] [ 33.869167] amdgpu_pci_probe+0x2d6/0xcd0 [amdgpu] [ 33.869608] local_pci_probe+0xda/0x180 [ 33.869614] pci_device_probe+0x43f/0x6b0 Empirically we can confirm that the former allocates 152 bytes for the table, while the latter memsets the 168 large block. Root cause appears that when GPU metrics tables for v2_4 parts were added it was not considered to enlarge the table to fit. The fix in this patch is rather "brute force" and perhaps later should be done in a smarter way, by extracting and consolidating the part version to size logic to a common helper, instead of brute forcing the largest possible allocation. Nevertheless, for now this works and fixes the out of bounds write. v2: * Drop impossible v3_0 case. (Mario) Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com> Fixes: 41cec40 ("drm/amd/pm: Vangogh: Add new gpu_metrics_v2_4 to acquire gpu_metrics") Cc: Mario Limonciello <mario.limonciello@amd.com> Cc: Evan Quan <evan.quan@amd.com> Cc: Wenyou Yang <WenYou.Yang@amd.com> Cc: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Mario Limonciello <mario.limonciello@amd.com> Link: https://lore.kernel.org/r/20241025145639.19124-1-tursulin@igalia.com Signed-off-by: Mario Limonciello <mario.limonciello@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit 0880f58f9609f0200483a49429af0f050d281703) Cc: stable@vger.kernel.org # v6.6+ Signed-off-by: Bin Lan <bin.lan.cn@windriver.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

commit 4aa923a6e6406b43566ef6ac35a3d9a3197fa3e8 upstream. KASAN reports that the GPU metrics table allocated in vangogh_tables_init() is not large enough for the memset done in smu_cmn_init_soft_gpu_metrics(). Condensed report follows: [ 33.861314] BUG: KASAN: slab-out-of-bounds in smu_cmn_init_soft_gpu_metrics+0x73/0x200 [amdgpu] [ 33.861799] Write of size 168 at addr ffff888129f59500 by task mangoapp/1067 ... [ 33.861808] CPU: 6 UID: 1000 PID: 1067 Comm: mangoapp Tainted: G W 6.12.0-rc4 #356 1a56f59a8b5182eeaf67eb7cb8b13594dd23b544 [ 33.861816] Tainted: [W]=WARN [ 33.861818] Hardware name: Valve Galileo/Galileo, BIOS F7G0107 12/01/2023 [ 33.861822] Call Trace: [ 33.861826] <TASK> [ 33.861829] dump_stack_lvl+0x66/0x90 [ 33.861838] print_report+0xce/0x620 [ 33.861853] kasan_report+0xda/0x110 [ 33.862794] kasan_check_range+0xfd/0x1a0 [ 33.862799] __asan_memset+0x23/0x40 [ 33.862803] smu_cmn_init_soft_gpu_metrics+0x73/0x200 [amdgpu 13b1bc364ec578808f676eba412c20eaab792779] [ 33.863306] vangogh_get_gpu_metrics_v2_4+0x123/0xad0 [amdgpu 13b1bc364ec578808f676eba412c20eaab792779] [ 33.864257] vangogh_common_get_gpu_metrics+0xb0c/0xbc0 [amdgpu 13b1bc364ec578808f676eba412c20eaab792779] [ 33.865682] amdgpu_dpm_get_gpu_metrics+0xcc/0x110 [amdgpu 13b1bc364ec578808f676eba412c20eaab792779] [ 33.866160] amdgpu_get_gpu_metrics+0x154/0x2d0 [amdgpu 13b1bc364ec578808f676eba412c20eaab792779] [ 33.867135] dev_attr_show+0x43/0xc0 [ 33.867147] sysfs_kf_seq_show+0x1f1/0x3b0 [ 33.867155] seq_read_iter+0x3f8/0x1140 [ 33.867173] vfs_read+0x76c/0xc50 [ 33.867198] ksys_read+0xfb/0x1d0 [ 33.867214] do_syscall_64+0x90/0x160 ... [ 33.867353] Allocated by task 378 on cpu 7 at 22.794876s: [ 33.867358] kasan_save_stack+0x33/0x50 [ 33.867364] kasan_save_track+0x17/0x60 [ 33.867367] __kasan_kmalloc+0x87/0x90 [ 33.867371] vangogh_init_smc_tables+0x3f9/0x840 [amdgpu] [ 33.867835] smu_sw_init+0xa32/0x1850 [amdgpu] [ 33.868299] amdgpu_device_init+0x467b/0x8d90 [amdgpu] [ 33.868733] amdgpu_driver_load_kms+0x19/0xf0 [amdgpu] [ 33.869167] amdgpu_pci_probe+0x2d6/0xcd0 [amdgpu] [ 33.869608] local_pci_probe+0xda/0x180 [ 33.869614] pci_device_probe+0x43f/0x6b0 Empirically we can confirm that the former allocates 152 bytes for the table, while the latter memsets the 168 large block. Root cause appears that when GPU metrics tables for v2_4 parts were added it was not considered to enlarge the table to fit. The fix in this patch is rather "brute force" and perhaps later should be done in a smarter way, by extracting and consolidating the part version to size logic to a common helper, instead of brute forcing the largest possible allocation. Nevertheless, for now this works and fixes the out of bounds write. v2: * Drop impossible v3_0 case. (Mario) Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com> Fixes: 41cec40 ("drm/amd/pm: Vangogh: Add new gpu_metrics_v2_4 to acquire gpu_metrics") Cc: Mario Limonciello <mario.limonciello@amd.com> Cc: Evan Quan <evan.quan@amd.com> Cc: Wenyou Yang <WenYou.Yang@amd.com> Cc: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Mario Limonciello <mario.limonciello@amd.com> Link: https://lore.kernel.org/r/20241025145639.19124-1-tursulin@igalia.com Signed-off-by: Mario Limonciello <mario.limonciello@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit 0880f58f9609f0200483a49429af0f050d281703) Cc: stable@vger.kernel.org # v6.6+ Signed-off-by: Bin Lan <bin.lan.cn@windriver.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> (cherry picked from commit f111de0f010308949254ee1cc45df8e6b8e1d7d4)

commit 4aa923a6e6406b43566ef6ac35a3d9a3197fa3e8 upstream. KASAN reports that the GPU metrics table allocated in vangogh_tables_init() is not large enough for the memset done in smu_cmn_init_soft_gpu_metrics(). Condensed report follows: [ 33.861314] BUG: KASAN: slab-out-of-bounds in smu_cmn_init_soft_gpu_metrics+0x73/0x200 [amdgpu] [ 33.861799] Write of size 168 at addr ffff888129f59500 by task mangoapp/1067 ... [ 33.861808] CPU: 6 UID: 1000 PID: 1067 Comm: mangoapp Tainted: G W 6.12.0-rc4 deepin-community#356 1a56f59a8b5182eeaf67eb7cb8b13594dd23b544 [ 33.861816] Tainted: [W]=WARN [ 33.861818] Hardware name: Valve Galileo/Galileo, BIOS F7G0107 12/01/2023 [ 33.861822] Call Trace: [ 33.861826] <TASK> [ 33.861829] dump_stack_lvl+0x66/0x90 [ 33.861838] print_report+0xce/0x620 [ 33.861853] kasan_report+0xda/0x110 [ 33.862794] kasan_check_range+0xfd/0x1a0 [ 33.862799] __asan_memset+0x23/0x40 [ 33.862803] smu_cmn_init_soft_gpu_metrics+0x73/0x200 [amdgpu 13b1bc364ec578808f676eba412c20eaab792779] [ 33.863306] vangogh_get_gpu_metrics_v2_4+0x123/0xad0 [amdgpu 13b1bc364ec578808f676eba412c20eaab792779] [ 33.864257] vangogh_common_get_gpu_metrics+0xb0c/0xbc0 [amdgpu 13b1bc364ec578808f676eba412c20eaab792779] [ 33.865682] amdgpu_dpm_get_gpu_metrics+0xcc/0x110 [amdgpu 13b1bc364ec578808f676eba412c20eaab792779] [ 33.866160] amdgpu_get_gpu_metrics+0x154/0x2d0 [amdgpu 13b1bc364ec578808f676eba412c20eaab792779] [ 33.867135] dev_attr_show+0x43/0xc0 [ 33.867147] sysfs_kf_seq_show+0x1f1/0x3b0 [ 33.867155] seq_read_iter+0x3f8/0x1140 [ 33.867173] vfs_read+0x76c/0xc50 [ 33.867198] ksys_read+0xfb/0x1d0 [ 33.867214] do_syscall_64+0x90/0x160 ... [ 33.867353] Allocated by task 378 on cpu 7 at 22.794876s: [ 33.867358] kasan_save_stack+0x33/0x50 [ 33.867364] kasan_save_track+0x17/0x60 [ 33.867367] __kasan_kmalloc+0x87/0x90 [ 33.867371] vangogh_init_smc_tables+0x3f9/0x840 [amdgpu] [ 33.867835] smu_sw_init+0xa32/0x1850 [amdgpu] [ 33.868299] amdgpu_device_init+0x467b/0x8d90 [amdgpu] [ 33.868733] amdgpu_driver_load_kms+0x19/0xf0 [amdgpu] [ 33.869167] amdgpu_pci_probe+0x2d6/0xcd0 [amdgpu] [ 33.869608] local_pci_probe+0xda/0x180 [ 33.869614] pci_device_probe+0x43f/0x6b0 Empirically we can confirm that the former allocates 152 bytes for the table, while the latter memsets the 168 large block. Root cause appears that when GPU metrics tables for v2_4 parts were added it was not considered to enlarge the table to fit. The fix in this patch is rather "brute force" and perhaps later should be done in a smarter way, by extracting and consolidating the part version to size logic to a common helper, instead of brute forcing the largest possible allocation. Nevertheless, for now this works and fixes the out of bounds write. v2: * Drop impossible v3_0 case. (Mario) Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com> Fixes: 41cec40 ("drm/amd/pm: Vangogh: Add new gpu_metrics_v2_4 to acquire gpu_metrics") Cc: Mario Limonciello <mario.limonciello@amd.com> Cc: Evan Quan <evan.quan@amd.com> Cc: Wenyou Yang <WenYou.Yang@amd.com> Cc: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Mario Limonciello <mario.limonciello@amd.com> Link: https://lore.kernel.org/r/20241025145639.19124-1-tursulin@igalia.com Signed-off-by: Mario Limonciello <mario.limonciello@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit 0880f58f9609f0200483a49429af0f050d281703) Cc: stable@vger.kernel.org # v6.6+ Signed-off-by: Bin Lan <bin.lan.cn@windriver.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

commit 4aa923a6e6406b43566ef6ac35a3d9a3197fa3e8 upstream. KASAN reports that the GPU metrics table allocated in vangogh_tables_init() is not large enough for the memset done in smu_cmn_init_soft_gpu_metrics(). Condensed report follows: [ 33.861314] BUG: KASAN: slab-out-of-bounds in smu_cmn_init_soft_gpu_metrics+0x73/0x200 [amdgpu] [ 33.861799] Write of size 168 at addr ffff888129f59500 by task mangoapp/1067 ... [ 33.861808] CPU: 6 UID: 1000 PID: 1067 Comm: mangoapp Tainted: G W 6.12.0-rc4 #356 1a56f59a8b5182eeaf67eb7cb8b13594dd23b544 [ 33.861816] Tainted: [W]=WARN [ 33.861818] Hardware name: Valve Galileo/Galileo, BIOS F7G0107 12/01/2023 [ 33.861822] Call Trace: [ 33.861826] <TASK> [ 33.861829] dump_stack_lvl+0x66/0x90 [ 33.861838] print_report+0xce/0x620 [ 33.861853] kasan_report+0xda/0x110 [ 33.862794] kasan_check_range+0xfd/0x1a0 [ 33.862799] __asan_memset+0x23/0x40 [ 33.862803] smu_cmn_init_soft_gpu_metrics+0x73/0x200 [amdgpu 13b1bc364ec578808f676eba412c20eaab792779] [ 33.863306] vangogh_get_gpu_metrics_v2_4+0x123/0xad0 [amdgpu 13b1bc364ec578808f676eba412c20eaab792779] [ 33.864257] vangogh_common_get_gpu_metrics+0xb0c/0xbc0 [amdgpu 13b1bc364ec578808f676eba412c20eaab792779] [ 33.865682] amdgpu_dpm_get_gpu_metrics+0xcc/0x110 [amdgpu 13b1bc364ec578808f676eba412c20eaab792779] [ 33.866160] amdgpu_get_gpu_metrics+0x154/0x2d0 [amdgpu 13b1bc364ec578808f676eba412c20eaab792779] [ 33.867135] dev_attr_show+0x43/0xc0 [ 33.867147] sysfs_kf_seq_show+0x1f1/0x3b0 [ 33.867155] seq_read_iter+0x3f8/0x1140 [ 33.867173] vfs_read+0x76c/0xc50 [ 33.867198] ksys_read+0xfb/0x1d0 [ 33.867214] do_syscall_64+0x90/0x160 ... [ 33.867353] Allocated by task 378 on cpu 7 at 22.794876s: [ 33.867358] kasan_save_stack+0x33/0x50 [ 33.867364] kasan_save_track+0x17/0x60 [ 33.867367] __kasan_kmalloc+0x87/0x90 [ 33.867371] vangogh_init_smc_tables+0x3f9/0x840 [amdgpu] [ 33.867835] smu_sw_init+0xa32/0x1850 [amdgpu] [ 33.868299] amdgpu_device_init+0x467b/0x8d90 [amdgpu] [ 33.868733] amdgpu_driver_load_kms+0x19/0xf0 [amdgpu] [ 33.869167] amdgpu_pci_probe+0x2d6/0xcd0 [amdgpu] [ 33.869608] local_pci_probe+0xda/0x180 [ 33.869614] pci_device_probe+0x43f/0x6b0 Empirically we can confirm that the former allocates 152 bytes for the table, while the latter memsets the 168 large block. Root cause appears that when GPU metrics tables for v2_4 parts were added it was not considered to enlarge the table to fit. The fix in this patch is rather "brute force" and perhaps later should be done in a smarter way, by extracting and consolidating the part version to size logic to a common helper, instead of brute forcing the largest possible allocation. Nevertheless, for now this works and fixes the out of bounds write. v2: * Drop impossible v3_0 case. (Mario) Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com> Fixes: 41cec40 ("drm/amd/pm: Vangogh: Add new gpu_metrics_v2_4 to acquire gpu_metrics") Cc: Mario Limonciello <mario.limonciello@amd.com> Cc: Evan Quan <evan.quan@amd.com> Cc: Wenyou Yang <WenYou.Yang@amd.com> Cc: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Mario Limonciello <mario.limonciello@amd.com> Link: https://lore.kernel.org/r/20241025145639.19124-1-tursulin@igalia.com Signed-off-by: Mario Limonciello <mario.limonciello@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit 0880f58f9609f0200483a49429af0f050d281703) Cc: stable@vger.kernel.org # v6.6+ Signed-off-by: Bin Lan <bin.lan.cn@windriver.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

commit 4aa923a6e6406b43566ef6ac35a3d9a3197fa3e8 upstream. KASAN reports that the GPU metrics table allocated in vangogh_tables_init() is not large enough for the memset done in smu_cmn_init_soft_gpu_metrics(). Condensed report follows: [ 33.861314] BUG: KASAN: slab-out-of-bounds in smu_cmn_init_soft_gpu_metrics+0x73/0x200 [amdgpu] [ 33.861799] Write of size 168 at addr ffff888129f59500 by task mangoapp/1067 ... [ 33.861808] CPU: 6 UID: 1000 PID: 1067 Comm: mangoapp Tainted: G W 6.12.0-rc4 deepin-community#356 1a56f59a8b5182eeaf67eb7cb8b13594dd23b544 [ 33.861816] Tainted: [W]=WARN [ 33.861818] Hardware name: Valve Galileo/Galileo, BIOS F7G0107 12/01/2023 [ 33.861822] Call Trace: [ 33.861826] <TASK> [ 33.861829] dump_stack_lvl+0x66/0x90 [ 33.861838] print_report+0xce/0x620 [ 33.861853] kasan_report+0xda/0x110 [ 33.862794] kasan_check_range+0xfd/0x1a0 [ 33.862799] __asan_memset+0x23/0x40 [ 33.862803] smu_cmn_init_soft_gpu_metrics+0x73/0x200 [amdgpu 13b1bc364ec578808f676eba412c20eaab792779] [ 33.863306] vangogh_get_gpu_metrics_v2_4+0x123/0xad0 [amdgpu 13b1bc364ec578808f676eba412c20eaab792779] [ 33.864257] vangogh_common_get_gpu_metrics+0xb0c/0xbc0 [amdgpu 13b1bc364ec578808f676eba412c20eaab792779] [ 33.865682] amdgpu_dpm_get_gpu_metrics+0xcc/0x110 [amdgpu 13b1bc364ec578808f676eba412c20eaab792779] [ 33.866160] amdgpu_get_gpu_metrics+0x154/0x2d0 [amdgpu 13b1bc364ec578808f676eba412c20eaab792779] [ 33.867135] dev_attr_show+0x43/0xc0 [ 33.867147] sysfs_kf_seq_show+0x1f1/0x3b0 [ 33.867155] seq_read_iter+0x3f8/0x1140 [ 33.867173] vfs_read+0x76c/0xc50 [ 33.867198] ksys_read+0xfb/0x1d0 [ 33.867214] do_syscall_64+0x90/0x160 ... [ 33.867353] Allocated by task 378 on cpu 7 at 22.794876s: [ 33.867358] kasan_save_stack+0x33/0x50 [ 33.867364] kasan_save_track+0x17/0x60 [ 33.867367] __kasan_kmalloc+0x87/0x90 [ 33.867371] vangogh_init_smc_tables+0x3f9/0x840 [amdgpu] [ 33.867835] smu_sw_init+0xa32/0x1850 [amdgpu] [ 33.868299] amdgpu_device_init+0x467b/0x8d90 [amdgpu] [ 33.868733] amdgpu_driver_load_kms+0x19/0xf0 [amdgpu] [ 33.869167] amdgpu_pci_probe+0x2d6/0xcd0 [amdgpu] [ 33.869608] local_pci_probe+0xda/0x180 [ 33.869614] pci_device_probe+0x43f/0x6b0 Empirically we can confirm that the former allocates 152 bytes for the table, while the latter memsets the 168 large block. Root cause appears that when GPU metrics tables for v2_4 parts were added it was not considered to enlarge the table to fit. The fix in this patch is rather "brute force" and perhaps later should be done in a smarter way, by extracting and consolidating the part version to size logic to a common helper, instead of brute forcing the largest possible allocation. Nevertheless, for now this works and fixes the out of bounds write. v2: * Drop impossible v3_0 case. (Mario) Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com> Fixes: 41cec40 ("drm/amd/pm: Vangogh: Add new gpu_metrics_v2_4 to acquire gpu_metrics") Cc: Mario Limonciello <mario.limonciello@amd.com> Cc: Evan Quan <evan.quan@amd.com> Cc: Wenyou Yang <WenYou.Yang@amd.com> Cc: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Mario Limonciello <mario.limonciello@amd.com> Link: https://lore.kernel.org/r/20241025145639.19124-1-tursulin@igalia.com Signed-off-by: Mario Limonciello <mario.limonciello@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit 0880f58f9609f0200483a49429af0f050d281703) Cc: stable@vger.kernel.org # v6.6+ Signed-off-by: Bin Lan <bin.lan.cn@windriver.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

MingcongBai added 6 commits August 8, 2024 12:29

Revert "LoongArch: fix KASLR can not be disabled by nokaslr when boot…

59b08d8

… from old BPI" This reverts commit ff445a1.

Revert "LoongArch: use arch specific phys_to_dma"

ce4b0fe

This reverts commit 43bef49.

Revert "LoongArch: Fixed EIOINTC structure members"

16768b2

This reverts commit e6e5f92.

Revert "LoongArch: Fix virtual machine startup error"

b3a58a0

This reverts commit 20c6c81.

Revert "LoongArch: Old BPI compatibility"

3d7be00

This reverts commit 59eab58.

Revert "LoongArch: add kernel setvirtmap for runtime"

7a8e4b9

This reverts commit d7aad0c.

deepin-ci-robot requested review from justforlxz and myml August 9, 2024 03:52

github-actions bot reviewed Aug 9, 2024

View reviewed changes

arch/loongarch/kernel/Makefile Show resolved Hide resolved

github-actions bot reviewed Aug 9, 2024

View reviewed changes

Avenger-285714 changed the title ~~arch: loongarch: replace BPI implementation with a minimal patchset~~ [WIP] arch: loongarch: replace BPI implementation with a minimal patchset Aug 9, 2024

deepin-ci-robot added the do-not-merge/work-in-progress label Aug 9, 2024

shankerwangmiao added 2 commits August 14, 2024 17:08

MingcongBai force-pushed the bai/loong64-clean-bpi-support branch from 860c024 to 3fb1481 Compare August 14, 2024 09:09

github-actions bot reviewed Aug 14, 2024

View reviewed changes

shankerwangmiao added 5 commits August 14, 2024 19:13

MingcongBai force-pushed the bai/loong64-clean-bpi-support branch from 3fb1481 to df06a5a Compare August 14, 2024 11:13

github-actions bot reviewed Aug 14, 2024

View reviewed changes

MingcongBai changed the title ~~[WIP] arch: loongarch: replace BPI implementation with a minimal patchset~~ arch: loongarch: replace BPI implementation with a minimal patchset Aug 14, 2024

deepin-ci-robot removed the do-not-merge/work-in-progress label Aug 14, 2024

Avenger-285714 merged commit 409da4e into linux-6.6.y Aug 14, 2024
5 of 11 checks passed

opsiff deleted the bai/loong64-clean-bpi-support branch September 21, 2024 09:15

opsiff restored the bai/loong64-clean-bpi-support branch September 21, 2024 09:15

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

arch: loongarch: replace BPI implementation with a minimal patchset #356

arch: loongarch: replace BPI implementation with a minimal patchset #356

MingcongBai commented Aug 9, 2024

github-actions bot Aug 9, 2024

github-actions bot Aug 9, 2024

github-actions bot Aug 9, 2024

github-actions bot Aug 9, 2024

github-actions bot Aug 9, 2024

github-actions bot Aug 9, 2024

github-actions bot Aug 9, 2024

github-actions bot Aug 9, 2024

github-actions bot Aug 9, 2024

github-actions bot Aug 14, 2024

github-actions bot Aug 14, 2024

github-actions bot Aug 14, 2024

github-actions bot Aug 14, 2024

github-actions bot Aug 14, 2024

github-actions bot Aug 14, 2024

github-actions bot Aug 14, 2024

github-actions bot Aug 14, 2024

github-actions bot Aug 14, 2024

github-actions bot Aug 14, 2024

github-actions bot Aug 14, 2024

github-actions bot Aug 14, 2024

github-actions bot Aug 14, 2024

deepin-ci-robot commented Aug 14, 2024

github-actions bot Aug 14, 2024

github-actions bot Aug 14, 2024

github-actions bot Aug 14, 2024

github-actions bot Aug 14, 2024

deepin-ci-robot commented Aug 14, 2024

MingcongBai commented Aug 14, 2024

Avenger-285714 commented Aug 14, 2024

arch: loongarch: replace BPI implementation with a minimal patchset #356

arch: loongarch: replace BPI implementation with a minimal patchset #356

Conversation

MingcongBai commented Aug 9, 2024

github-actions bot Aug 9, 2024

Choose a reason for hiding this comment

github-actions bot Aug 9, 2024

Choose a reason for hiding this comment

github-actions bot Aug 9, 2024

Choose a reason for hiding this comment

github-actions bot Aug 9, 2024

Choose a reason for hiding this comment

github-actions bot Aug 9, 2024

Choose a reason for hiding this comment

github-actions bot Aug 9, 2024

Choose a reason for hiding this comment

github-actions bot Aug 9, 2024

Choose a reason for hiding this comment

github-actions bot Aug 9, 2024

Choose a reason for hiding this comment

github-actions bot Aug 9, 2024

Choose a reason for hiding this comment

github-actions bot Aug 14, 2024

Choose a reason for hiding this comment

github-actions bot Aug 14, 2024

Choose a reason for hiding this comment

github-actions bot Aug 14, 2024

Choose a reason for hiding this comment

github-actions bot Aug 14, 2024

Choose a reason for hiding this comment

github-actions bot Aug 14, 2024

Choose a reason for hiding this comment

github-actions bot Aug 14, 2024

Choose a reason for hiding this comment

github-actions bot Aug 14, 2024

Choose a reason for hiding this comment

github-actions bot Aug 14, 2024

Choose a reason for hiding this comment

github-actions bot Aug 14, 2024

Choose a reason for hiding this comment

github-actions bot Aug 14, 2024

Choose a reason for hiding this comment

github-actions bot Aug 14, 2024

Choose a reason for hiding this comment

github-actions bot Aug 14, 2024

Choose a reason for hiding this comment

github-actions bot Aug 14, 2024

Choose a reason for hiding this comment

deepin-ci-robot commented Aug 14, 2024

deepin pr auto review

github-actions bot Aug 14, 2024

Choose a reason for hiding this comment

github-actions bot Aug 14, 2024

Choose a reason for hiding this comment

github-actions bot Aug 14, 2024

Choose a reason for hiding this comment

github-actions bot Aug 14, 2024

Choose a reason for hiding this comment

deepin-ci-robot commented Aug 14, 2024

MingcongBai commented Aug 14, 2024

Avenger-285714 commented Aug 14, 2024