* [PATCH 1/4] LoongArch: Fix built-in DTB detection
2024-05-21 20:12 [PATCH 0/4] LoongArch: Bootloader interface fixes Jiaxun Yang
@ 2024-05-21 20:12 ` Jiaxun Yang
2024-05-21 20:12 ` [PATCH 2/4] LoongArch: smp: Add all CPUs enabled by fdt to NUMA node 0 Jiaxun Yang
` (2 subsequent siblings)
3 siblings, 0 replies; 6+ messages in thread
From: Jiaxun Yang @ 2024-05-21 20:12 UTC (permalink / raw)
To: Huacai Chen, Binbin Zhou; +Cc: loongarch, linux-kernel, Jiaxun Yang, stable
fdt_check_header(__dtb_start) will always success because kernel
provided a dummy dtb here, thus fdt passed from firmware will
never be taken.
Fix by trying to utilise __dtb_start only when CONFIG_BUILTIN_DTB
is enabled.
Fixes: 5f346a6e5970 ("LoongArch: Allow device trees be built into the kernel")
Cc: stable@vger.kernel.org
Signed-off-by: Jiaxun Yang <jiaxun.yang@flygoat.com>
---
arch/loongarch/kernel/setup.c | 6 ++++--
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/arch/loongarch/kernel/setup.c b/arch/loongarch/kernel/setup.c
index 60e0fe97f61a..ea6d5db6c878 100644
--- a/arch/loongarch/kernel/setup.c
+++ b/arch/loongarch/kernel/setup.c
@@ -275,16 +275,18 @@ static void __init arch_reserve_crashkernel(void)
static void __init fdt_setup(void)
{
#ifdef CONFIG_OF_EARLY_FLATTREE
- void *fdt_pointer;
+ void *fdt_pointer = NULL;
/* ACPI-based systems do not require parsing fdt */
if (acpi_os_get_root_pointer())
return;
+#ifdef CONFIG_BUILTIN_DTB
/* Prefer to use built-in dtb, checking its legality first. */
if (!fdt_check_header(__dtb_start))
fdt_pointer = __dtb_start;
- else
+#endif
+ if (!fdt_pointer)
fdt_pointer = efi_fdt_pointer(); /* Fallback to firmware dtb */
if (!fdt_pointer || fdt_check_header(fdt_pointer))
--
2.43.0
^ permalink raw reply related [flat|nested] 6+ messages in thread* [PATCH 2/4] LoongArch: smp: Add all CPUs enabled by fdt to NUMA node 0
2024-05-21 20:12 [PATCH 0/4] LoongArch: Bootloader interface fixes Jiaxun Yang
2024-05-21 20:12 ` [PATCH 1/4] LoongArch: Fix built-in DTB detection Jiaxun Yang
@ 2024-05-21 20:12 ` Jiaxun Yang
2024-05-23 22:17 ` kernel test robot
2024-05-21 20:12 ` [PATCH 3/4] LoongArch: Fix entry point in image header Jiaxun Yang
2024-05-21 20:12 ` [PATCH 4/4] LoongArch: Clear higher address bits in JUMP_VIRT_ADDR Jiaxun Yang
3 siblings, 1 reply; 6+ messages in thread
From: Jiaxun Yang @ 2024-05-21 20:12 UTC (permalink / raw)
To: Huacai Chen, Binbin Zhou; +Cc: loongarch, linux-kernel, Jiaxun Yang, stable
NUMA enabled kernel on FDT based machine fails to boot
because CPUs are all in NUMA_NO_NODE and mm subsystem
won't accept that.
Fix by adding them to default NUMA node for now.
Cc: stable@vger.kernel.org
Fixes: 88d4d957edc7 ("LoongArch: Add FDT booting support from efi system table")
Signed-off-by: Jiaxun Yang <jiaxun.yang@flygoat.com>
---
arch/loongarch/kernel/smp.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/arch/loongarch/kernel/smp.c b/arch/loongarch/kernel/smp.c
index 0dfe2388ef41..866757b76ecb 100644
--- a/arch/loongarch/kernel/smp.c
+++ b/arch/loongarch/kernel/smp.c
@@ -273,7 +273,6 @@ static void __init fdt_smp_setup(void)
if (cpuid == loongson_sysconf.boot_cpu_id) {
cpu = 0;
- numa_add_cpu(cpu);
} else {
cpu = cpumask_next_zero(-1, cpu_present_mask);
}
@@ -283,6 +282,10 @@ static void __init fdt_smp_setup(void)
set_cpu_present(cpu, true);
__cpu_number_map[cpuid] = cpu;
__cpu_logical_map[cpu] = cpuid;
+
+ early_numa_add_cpu(cpu, 0);
+ set_cpuid_to_node(cpuid, 0);
+ numa_add_cpu(cpu);
}
loongson_sysconf.nr_cpus = num_processors;
--
2.43.0
^ permalink raw reply related [flat|nested] 6+ messages in thread* Re: [PATCH 2/4] LoongArch: smp: Add all CPUs enabled by fdt to NUMA node 0
2024-05-21 20:12 ` [PATCH 2/4] LoongArch: smp: Add all CPUs enabled by fdt to NUMA node 0 Jiaxun Yang
@ 2024-05-23 22:17 ` kernel test robot
0 siblings, 0 replies; 6+ messages in thread
From: kernel test robot @ 2024-05-23 22:17 UTC (permalink / raw)
To: Jiaxun Yang, Huacai Chen, Binbin Zhou
Cc: oe-kbuild-all, loongarch, linux-kernel, Jiaxun Yang, stable
Hi Jiaxun,
kernel test robot noticed the following build errors:
[auto build test ERROR on 124cfbcd6d185d4f50be02d5f5afe61578916773]
url: https://github.com/intel-lab-lkp/linux/commits/Jiaxun-Yang/LoongArch-Fix-built-in-DTB-detection/20240522-041534
base: 124cfbcd6d185d4f50be02d5f5afe61578916773
patch link: https://lore.kernel.org/r/20240521-loongarch-booting-fixes-v1-2-659c201c0370%40flygoat.com
patch subject: [PATCH 2/4] LoongArch: smp: Add all CPUs enabled by fdt to NUMA node 0
config: loongarch-randconfig-001-20240524 (https://download.01.org/0day-ci/archive/20240524/202405240627.5Ey7vV8v-lkp@intel.com/config)
compiler: loongarch64-linux-gcc (GCC) 13.2.0
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20240524/202405240627.5Ey7vV8v-lkp@intel.com/reproduce)
If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <lkp@intel.com>
| Closes: https://lore.kernel.org/oe-kbuild-all/202405240627.5Ey7vV8v-lkp@intel.com/
All errors (new ones prefixed by >>):
arch/loongarch/kernel/smp.c: In function 'fdt_smp_setup':
>> arch/loongarch/kernel/smp.c:287:17: error: implicit declaration of function 'set_cpuid_to_node'; did you mean 'set_cpu_numa_node'? [-Werror=implicit-function-declaration]
287 | set_cpuid_to_node(cpuid, 0);
| ^~~~~~~~~~~~~~~~~
| set_cpu_numa_node
cc1: some warnings being treated as errors
vim +287 arch/loongarch/kernel/smp.c
259
260 static void __init fdt_smp_setup(void)
261 {
262 #ifdef CONFIG_OF
263 unsigned int cpu, cpuid;
264 struct device_node *node = NULL;
265
266 for_each_of_cpu_node(node) {
267 if (!of_device_is_available(node))
268 continue;
269
270 cpuid = of_get_cpu_hwid(node, 0);
271 if (cpuid >= nr_cpu_ids)
272 continue;
273
274 if (cpuid == loongson_sysconf.boot_cpu_id) {
275 cpu = 0;
276 } else {
277 cpu = cpumask_next_zero(-1, cpu_present_mask);
278 }
279
280 num_processors++;
281 set_cpu_possible(cpu, true);
282 set_cpu_present(cpu, true);
283 __cpu_number_map[cpuid] = cpu;
284 __cpu_logical_map[cpu] = cpuid;
285
286 early_numa_add_cpu(cpu, 0);
> 287 set_cpuid_to_node(cpuid, 0);
288 numa_add_cpu(cpu);
289 }
290
291 loongson_sysconf.nr_cpus = num_processors;
292 set_bit(0, loongson_sysconf.cores_io_master);
293 #endif
294 }
295
--
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki
^ permalink raw reply [flat|nested] 6+ messages in thread
* [PATCH 3/4] LoongArch: Fix entry point in image header
2024-05-21 20:12 [PATCH 0/4] LoongArch: Bootloader interface fixes Jiaxun Yang
2024-05-21 20:12 ` [PATCH 1/4] LoongArch: Fix built-in DTB detection Jiaxun Yang
2024-05-21 20:12 ` [PATCH 2/4] LoongArch: smp: Add all CPUs enabled by fdt to NUMA node 0 Jiaxun Yang
@ 2024-05-21 20:12 ` Jiaxun Yang
2024-05-21 20:12 ` [PATCH 4/4] LoongArch: Clear higher address bits in JUMP_VIRT_ADDR Jiaxun Yang
3 siblings, 0 replies; 6+ messages in thread
From: Jiaxun Yang @ 2024-05-21 20:12 UTC (permalink / raw)
To: Huacai Chen, Binbin Zhou; +Cc: loongarch, linux-kernel, Jiaxun Yang, stable
Currently kernel entry in head.S is in DMW address range,
firmware is instructed to jump to this address after loading
the image.
However kernel should not make any assumption on firmware's
DMW setting, thus the entry point should be a physical address
falls into direct translation region.
Fix by applying a calculation to the entry. Note that due
to relocation restriction TO_PHYS can't be used, we can
only do plus and minus here.
Cc: stable@vger.kernel.org
Signed-off-by: Jiaxun Yang <jiaxun.yang@flygoat.com>
---
arch/loongarch/kernel/head.S | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/arch/loongarch/kernel/head.S b/arch/loongarch/kernel/head.S
index c4f7de2e2805..1a83564023e1 100644
--- a/arch/loongarch/kernel/head.S
+++ b/arch/loongarch/kernel/head.S
@@ -22,7 +22,7 @@
_head:
.word MZ_MAGIC /* "MZ", MS-DOS header */
.org 0x8
- .dword kernel_entry /* Kernel entry point */
+ .dword PHYS_LINK_KADDR + (kernel_entry - _head) /* Kernel entry point */
.dword _kernel_asize /* Kernel image effective size */
.quad PHYS_LINK_KADDR /* Kernel image load offset from start of RAM */
.org 0x38 /* 0x20 ~ 0x37 reserved */
--
2.43.0
^ permalink raw reply related [flat|nested] 6+ messages in thread
* [PATCH 4/4] LoongArch: Clear higher address bits in JUMP_VIRT_ADDR
2024-05-21 20:12 [PATCH 0/4] LoongArch: Bootloader interface fixes Jiaxun Yang
` (2 preceding siblings ...)
2024-05-21 20:12 ` [PATCH 3/4] LoongArch: Fix entry point in image header Jiaxun Yang
@ 2024-05-21 20:12 ` Jiaxun Yang
3 siblings, 0 replies; 6+ messages in thread
From: Jiaxun Yang @ 2024-05-21 20:12 UTC (permalink / raw)
To: Huacai Chen, Binbin Zhou; +Cc: loongarch, linux-kernel, Jiaxun Yang, stable
In JUMP_VIRT_ADDR we are performing an or calculation on
address value directly from pcaddi.
This will only work if we are currently running from direct
translation addresses or firmware's DMW is configured exactly
same as kernel. Still, we should not rely on such assumption.
Fix by clearing higher bits in address comes from pcaddi,
so we can get real physcal address before applying or
operator.
Cc: stable@vger.kernel.org
Signed-off-by: Jiaxun Yang <jiaxun.yang@flygoat.com>
---
arch/loongarch/include/asm/stackframe.h | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/arch/loongarch/include/asm/stackframe.h b/arch/loongarch/include/asm/stackframe.h
index 45b507a7b06f..a325d20a4503 100644
--- a/arch/loongarch/include/asm/stackframe.h
+++ b/arch/loongarch/include/asm/stackframe.h
@@ -42,8 +42,10 @@
.macro JUMP_VIRT_ADDR temp1 temp2
li.d \temp1, CACHE_BASE
pcaddi \temp2, 0
+ PTR_SLL \temp2, \temp2, (BITS_PER_LONG - DMW_PABITS)
+ PTR_SRL \temp2, \temp2, (BITS_PER_LONG - DMW_PABITS)
or \temp1, \temp1, \temp2
- jirl zero, \temp1, 0xc
+ jirl zero, \temp1, 0x14
.endm
.macro BACKUP_T0T1
--
2.43.0
^ permalink raw reply related [flat|nested] 6+ messages in thread