[PATCH 1/2] kexec: Default __NR_kexec_file_load is set to undefined

[PATCH 1/2] kexec: Default __NR_kexec_file_load is set to undefined

[Youling Tang]

The initial reason is that after the merger of 29fe5067ed07
("kexec: make -a the default"), kexec cannot be used on LoongArch,
MIPS .etc architectures. We need to add "-c" for normal use. The
current kexec_file_load system call is not implemented in
architectures such as LoongArch, so it needs to pass kexec_load.
So we need to set __NR_kexec_file_load to undefined in unsupported
architectures. This will return EFALLBACK via is_kexec_file_load_implemented,
and then via kexec_load.

Signed-off-by: Youling Tang <tangyouling@loongson.cn>
---
 kexec/kexec-syscall.h | 2 --
 1 file changed, 2 deletions(-)

diff --git a/kexec/kexec-syscall.h b/kexec/kexec-syscall.h
index be6ccd5..ea77936 100644
--- a/kexec/kexec-syscall.h
+++ b/kexec/kexec-syscall.h
@@ -59,9 +59,7 @@
 #endif
 #endif /*ifndef __NR_kexec_load*/
 
-#ifdef __arm__
 #undef __NR_kexec_file_load
-#endif
 
 #ifndef __NR_kexec_file_load
 
-- 
2.37.1


_______________________________________________
kexec mailing list
kexec@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/kexec

[PATCH 2/2] LoongArch: kdump: Set up kernel image segment

[Youling Tang]

On LoongArch, we can use the same kernel image as 1st kernel when
[1] is merged, but we have to modify the entry point as well as
segments' addresses in the kernel's elf header (or pei format
vmlinux.efi) in order to load them into correct places.

[1]: https://lore.kernel.org/loongarch/1677150391-12838-1-git-send-email-tangyouling@loongson.cn/T/#t

Signed-off-by: Youling Tang <tangyouling@loongson.cn>
---
 kexec/arch/loongarch/crashdump-loongarch.c | 22 ++++++++++++++++++++++
 kexec/arch/loongarch/crashdump-loongarch.h |  1 +
 kexec/arch/loongarch/kexec-elf-loongarch.c |  8 ++++++++
 kexec/arch/loongarch/kexec-loongarch.c     |  3 ++-
 kexec/arch/loongarch/kexec-pei-loongarch.c |  7 +++++++
 5 files changed, 40 insertions(+), 1 deletion(-)

diff --git a/kexec/arch/loongarch/crashdump-loongarch.c b/kexec/arch/loongarch/crashdump-loongarch.c
index aaf6cf3..81250e4 100644
--- a/kexec/arch/loongarch/crashdump-loongarch.c
+++ b/kexec/arch/loongarch/crashdump-loongarch.c
@@ -183,6 +183,28 @@ int load_crashdump_segments(struct kexec_info *info)
 	return 0;
 }
 
+/*
+ * e_entry and p_paddr are actually in virtual address space.
+ * Those values will be translated to physcal addresses by using
+ * virt_to_phys() in add_segment().
+ * So let's fix up those values for later use so the memory base will be
+ * correctly replaced with crash_reserved_mem[usablemem_rgns.size - 1].start.
+ */
+void fixup_elf_addrs(struct mem_ehdr *ehdr)
+{
+	struct mem_phdr *phdr;
+	int i;
+
+	ehdr->e_entry += crash_reserved_mem[usablemem_rgns.size - 1].start;
+
+	for (i = 0; i < ehdr->e_phnum; i++) {
+		phdr = &ehdr->e_phdr[i];
+		if (phdr->p_type != PT_LOAD)
+			continue;
+		phdr->p_paddr += crash_reserved_mem[usablemem_rgns.size - 1].start;
+	}
+}
+
 int get_crash_kernel_load_range(uint64_t *start, uint64_t *end)
 {
 	if (!usablemem_rgns.size)
diff --git a/kexec/arch/loongarch/crashdump-loongarch.h b/kexec/arch/loongarch/crashdump-loongarch.h
index 3eb4e0a..25ff24b 100644
--- a/kexec/arch/loongarch/crashdump-loongarch.h
+++ b/kexec/arch/loongarch/crashdump-loongarch.h
@@ -8,6 +8,7 @@ extern struct memory_range elfcorehdr_mem;
 
 int load_crashdump_segments(struct kexec_info *info);
 int is_crashkernel_mem_reserved(void);
+void fixup_elf_addrs(struct mem_ehdr *ehdr);
 int get_crash_kernel_load_range(uint64_t *start, uint64_t *end);
 
 #define PAGE_OFFSET	0x9000000000000000ULL
diff --git a/kexec/arch/loongarch/kexec-elf-loongarch.c b/kexec/arch/loongarch/kexec-elf-loongarch.c
index 2bf128f..45387ca 100644
--- a/kexec/arch/loongarch/kexec-elf-loongarch.c
+++ b/kexec/arch/loongarch/kexec-elf-loongarch.c
@@ -90,6 +90,14 @@ int elf_loongarch_load(int argc, char **argv, const char *kernel_buf,
 		}
 	}
 
+	/* load the kernel */
+	if (info->kexec_flags & KEXEC_ON_CRASH)
+		/*
+		 * offset addresses in elf header in order to load
+		 * vmlinux (elf_exec) into crash kernel's memory.
+		 */
+		fixup_elf_addrs(&ehdr);
+
 	info->entry = (void *)virt_to_phys(ehdr.e_entry);
 
 	result = elf_exec_load(&ehdr, info);
diff --git a/kexec/arch/loongarch/kexec-loongarch.c b/kexec/arch/loongarch/kexec-loongarch.c
index 4c7361c..f47c998 100644
--- a/kexec/arch/loongarch/kexec-loongarch.c
+++ b/kexec/arch/loongarch/kexec-loongarch.c
@@ -253,7 +253,8 @@ unsigned long loongarch_locate_kernel_segment(struct kexec_info *info)
 		unsigned long hole_end;
 
 		hole = (crash_reserved_mem[usablemem_rgns.size - 1].start < mem_min ?
-				mem_min : crash_reserved_mem[usablemem_rgns.size - 1].start);
+				mem_min : crash_reserved_mem[usablemem_rgns.size - 1].start) +
+				loongarch_mem.text_offset;
 		hole = _ALIGN_UP(hole, MiB(1));
 		hole_end = hole + loongarch_mem.text_offset + loongarch_mem.image_size;
 
diff --git a/kexec/arch/loongarch/kexec-pei-loongarch.c b/kexec/arch/loongarch/kexec-pei-loongarch.c
index f86ac61..1a11103 100644
--- a/kexec/arch/loongarch/kexec-pei-loongarch.c
+++ b/kexec/arch/loongarch/kexec-pei-loongarch.c
@@ -66,6 +66,13 @@ int pei_loongarch_load(int argc, char **argv, const char *buf,
 
 	kernel_entry = virt_to_phys(loongarch_header_kernel_entry(header));
 
+	if (info->kexec_flags & KEXEC_ON_CRASH)
+		/*
+		 * offset addresses in order to load vmlinux.efi into
+		 * crash kernel's memory.
+		 */
+		kernel_entry += crash_reserved_mem[usablemem_rgns.size - 1].start;
+
 	dbgprintf("%s: kernel_segment: %016lx\n", __func__, kernel_segment);
 	dbgprintf("%s: kernel_entry:   %016lx\n", __func__, kernel_entry);
 	dbgprintf("%s: image_size:     %016lx\n", __func__,
-- 
2.37.1


_______________________________________________
kexec mailing list
kexec@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/kexec

Re: [PATCH 1/2] kexec: Default __NR_kexec_file_load is set to undefined

[Philipp Rudo]

Hi Youling,

On Fri, 24 Feb 2023 17:51:07 +0800
Youling Tang <tangyouling@loongson.cn> wrote:

> The initial reason is that after the merger of 29fe5067ed07
> ("kexec: make -a the default"), kexec cannot be used on LoongArch,
> MIPS .etc architectures. We need to add "-c" for normal use. The
> current kexec_file_load system call is not implemented in
> architectures such as LoongArch, so it needs to pass kexec_load.
> So we need to set __NR_kexec_file_load to undefined in unsupported
> architectures. This will return EFALLBACK via is_kexec_file_load_implemented,
> and then via kexec_load.
> 
> Signed-off-by: Youling Tang <tangyouling@loongson.cn>
> ---
>  kexec/kexec-syscall.h | 2 --
>  1 file changed, 2 deletions(-)
> 
> diff --git a/kexec/kexec-syscall.h b/kexec/kexec-syscall.h
> index be6ccd5..ea77936 100644
> --- a/kexec/kexec-syscall.h
> +++ b/kexec/kexec-syscall.h
> @@ -59,9 +59,7 @@
>  #endif
>  #endif /*ifndef __NR_kexec_load*/
>  
> -#ifdef __arm__
>  #undef __NR_kexec_file_load
> -#endif
>  
>  #ifndef __NR_kexec_file_load

I don't think this will work as intended. 

On the top of the file sys/syscall.h gets included. In there
architectures that support kexec_file_load define __NR_kexec_file_load.
This also means that if an architecture doesn't support kexec_file_load
__NR_kexec_file_load shouldn't be defined in the first place. Thus I
suggest that you find out why sys/syscall.h defines
__NR_kexec_file_load for LoongArch even when the system call is not
supported and fix it there.

Thanks
Philipp 


_______________________________________________
kexec mailing list
kexec@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/kexec

Re: [PATCH 1/2] kexec: Default __NR_kexec_file_load is set to undefined

[Baoquan He]

Hi Philipp,

On 02/27/23 at 04:19pm, Philipp Rudo wrote:
...... 
> > diff --git a/kexec/kexec-syscall.h b/kexec/kexec-syscall.h
> > index be6ccd5..ea77936 100644
> > --- a/kexec/kexec-syscall.h
> > +++ b/kexec/kexec-syscall.h
> > @@ -59,9 +59,7 @@
> >  #endif
> >  #endif /*ifndef __NR_kexec_load*/
> >  
> > -#ifdef __arm__
> >  #undef __NR_kexec_file_load
> > -#endif
> >  
> >  #ifndef __NR_kexec_file_load
> 
> I don't think this will work as intended. 
> 
> On the top of the file sys/syscall.h gets included. In there
> architectures that support kexec_file_load define __NR_kexec_file_load.
> This also means that if an architecture doesn't support kexec_file_load
> __NR_kexec_file_load shouldn't be defined in the first place. Thus I
> suggest that you find out why sys/syscall.h defines
> __NR_kexec_file_load for LoongArch even when the system call is not
> supported and fix it there.

Checking whether LoongArch has defined __NR_kexec_file_load sounds a
good suggestion. Wondering why we still need add __NR_kexec_file_load
definition in kexec-tools. E.g below s390 kexec_file support you added.
<sys/syscall.h> sometime won't be found or __NR_kexec_file_load could be
not defined yet?

commit d4a948c268272cf37c71be820fb02bf40e56292b
Author: Philipp Rudo <prudo@linux.ibm.com>
Date:   Wed May 16 14:27:18 2018 +0200

    kexec/s390: Add support for kexec_file_load


_______________________________________________
kexec mailing list
kexec@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/kexec

Re: [PATCH 1/2] kexec: Default __NR_kexec_file_load is set to undefined

[Philipp Rudo]

Hi Youling,

On Tue, 28 Feb 2023 10:02:16 +0800
Youling Tang <tangyouling@loongson.cn> wrote:

> Hi, Philipp
> 
> On 02/27/2023 11:19 PM, Philipp Rudo wrote:
> > Hi Youling,
> >
> > On Fri, 24 Feb 2023 17:51:07 +0800
> > Youling Tang <tangyouling@loongson.cn> wrote:
> >  
> >> The initial reason is that after the merger of 29fe5067ed07
> >> ("kexec: make -a the default"), kexec cannot be used on LoongArch,
> >> MIPS .etc architectures. We need to add "-c" for normal use. The
> >> current kexec_file_load system call is not implemented in
> >> architectures such as LoongArch, so it needs to pass kexec_load.
> >> So we need to set __NR_kexec_file_load to undefined in unsupported
> >> architectures. This will return EFALLBACK via is_kexec_file_load_implemented,
> >> and then via kexec_load.
> >>
> >> Signed-off-by: Youling Tang <tangyouling@loongson.cn>
> >> ---
> >>  kexec/kexec-syscall.h | 2 --
> >>  1 file changed, 2 deletions(-)
> >>
> >> diff --git a/kexec/kexec-syscall.h b/kexec/kexec-syscall.h
> >> index be6ccd5..ea77936 100644
> >> --- a/kexec/kexec-syscall.h
> >> +++ b/kexec/kexec-syscall.h
> >> @@ -59,9 +59,7 @@
> >>  #endif
> >>  #endif /*ifndef __NR_kexec_load*/
> >>
> >> -#ifdef __arm__
> >>  #undef __NR_kexec_file_load
> >> -#endif
> >>
> >>  #ifndef __NR_kexec_file_load  
> >
> > I don't think this will work as intended.  
> 
> Works fine in LoongArch after applying this patch.

I believe you. The problem is that it changes the behavior on other
architectures.

> >
> > On the top of the file sys/syscall.h gets included. In there
> > architectures that support kexec_file_load define __NR_kexec_file_load.
> > This also means that if an architecture doesn't support kexec_file_load
> > __NR_kexec_file_load shouldn't be defined in the first place. Thus I
> > suggest that you find out why sys/syscall.h defines
> > __NR_kexec_file_load for LoongArch even when the system call is not
> > supported and fix it there.  
> Yes, in the kernel, LoongArch uses the generic syscall number (in
> include/uapi/asm-generic/unistd.h), so __NR_kexec_file_load is
> defined as 294.

Ok, I see. I have expected that it is wrapped in some ifdef
CONFIG_KEXEC_FILE. But apparently that is not the case. Thanks for the
clarification.

> I think the simpler way is to modify it in kexec-tools as follows,
> #ifdef __loongarch__
> #undef __NR_kexec_file_load
> #endif

Yeah, looks like this is the best solution. Only thing you could do is
to merge it with the arm special handling. I.e. so it reads

#if defined(__arm__) || defined(__longarch__)
#undef __NR_kexec_file_load
#endif

But that is more a matter of taste. So you and Simon need to agree what
you like more.

Thanks
Philipp


_______________________________________________
kexec mailing list
kexec@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/kexec

Re: [PATCH 1/2] kexec: Default __NR_kexec_file_load is set to undefined

[Philipp Rudo]

Hi Baoquan,

On Tue, 28 Feb 2023 21:32:26 +0800
Baoquan He <bhe@redhat.com> wrote:

> Hi Philipp,
> 
> On 02/27/23 at 04:19pm, Philipp Rudo wrote:
> ...... 
> > > diff --git a/kexec/kexec-syscall.h b/kexec/kexec-syscall.h
> > > index be6ccd5..ea77936 100644
> > > --- a/kexec/kexec-syscall.h
> > > +++ b/kexec/kexec-syscall.h
> > > @@ -59,9 +59,7 @@
> > >  #endif
> > >  #endif /*ifndef __NR_kexec_load*/
> > >  
> > > -#ifdef __arm__
> > >  #undef __NR_kexec_file_load
> > > -#endif
> > >  
> > >  #ifndef __NR_kexec_file_load  
> > 
> > I don't think this will work as intended. 
> > 
> > On the top of the file sys/syscall.h gets included. In there
> > architectures that support kexec_file_load define __NR_kexec_file_load.
> > This also means that if an architecture doesn't support kexec_file_load
> > __NR_kexec_file_load shouldn't be defined in the first place. Thus I
> > suggest that you find out why sys/syscall.h defines
> > __NR_kexec_file_load for LoongArch even when the system call is not
> > supported and fix it there.  
> 
> Checking whether LoongArch has defined __NR_kexec_file_load sounds a
> good suggestion. Wondering why we still need add __NR_kexec_file_load
> definition in kexec-tools. E.g below s390 kexec_file support you added.
> <sys/syscall.h> sometime won't be found or __NR_kexec_file_load could be
> not defined yet?
> 
> commit d4a948c268272cf37c71be820fb02bf40e56292b
> Author: Philipp Rudo <prudo@linux.ibm.com>
> Date:   Wed May 16 14:27:18 2018 +0200
> 
>     kexec/s390: Add support for kexec_file_load
> 

To be honest I don't remember why I have added it back then. Most
likely I simply copied what others have done before.

The benefit in having the extra definition I see is that you don't need
to rebuild glibc when you implement the syscall and don't use the
generic unistd.h. But that is something only few people should ever
encounter.

Thanks
Philipp


_______________________________________________
kexec mailing list
kexec@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/kexec

Re: [PATCH 1/2] kexec: Default __NR_kexec_file_load is set to undefined

[Baoquan He]

On 03/03/23 at 12:03pm, Philipp Rudo wrote:
> Hi Baoquan,
> 
> On Tue, 28 Feb 2023 21:32:26 +0800
> Baoquan He <bhe@redhat.com> wrote:
> 
> > Hi Philipp,
> > 
> > On 02/27/23 at 04:19pm, Philipp Rudo wrote:
> > ...... 
> > > > diff --git a/kexec/kexec-syscall.h b/kexec/kexec-syscall.h
> > > > index be6ccd5..ea77936 100644
> > > > --- a/kexec/kexec-syscall.h
> > > > +++ b/kexec/kexec-syscall.h
> > > > @@ -59,9 +59,7 @@
> > > >  #endif
> > > >  #endif /*ifndef __NR_kexec_load*/
> > > >  
> > > > -#ifdef __arm__
> > > >  #undef __NR_kexec_file_load
> > > > -#endif
> > > >  
> > > >  #ifndef __NR_kexec_file_load  
> > > 
> > > I don't think this will work as intended. 
> > > 
> > > On the top of the file sys/syscall.h gets included. In there
> > > architectures that support kexec_file_load define __NR_kexec_file_load.
> > > This also means that if an architecture doesn't support kexec_file_load
> > > __NR_kexec_file_load shouldn't be defined in the first place. Thus I
> > > suggest that you find out why sys/syscall.h defines
> > > __NR_kexec_file_load for LoongArch even when the system call is not
> > > supported and fix it there.  
> > 
> > Checking whether LoongArch has defined __NR_kexec_file_load sounds a
> > good suggestion. Wondering why we still need add __NR_kexec_file_load
> > definition in kexec-tools. E.g below s390 kexec_file support you added.
> > <sys/syscall.h> sometime won't be found or __NR_kexec_file_load could be
> > not defined yet?
> > 
> > commit d4a948c268272cf37c71be820fb02bf40e56292b
> > Author: Philipp Rudo <prudo@linux.ibm.com>
> > Date:   Wed May 16 14:27:18 2018 +0200
> > 
> >     kexec/s390: Add support for kexec_file_load
> > 
> 
> To be honest I don't remember why I have added it back then. Most
> likely I simply copied what others have done before.
> 
> The benefit in having the extra definition I see is that you don't need
> to rebuild glibc when you implement the syscall and don't use the
> generic unistd.h. But that is something only few people should ever
> encounter.

OK, got it. From the code, the adding won't cuase issue, but a double
insurance, not sure if it's redundant. Someone interested can
investigate and clean up if needed.

Thanks for the information.


_______________________________________________
kexec mailing list
kexec@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/kexec