Date:   Wed, 10 Jun 2020 11:44:30 -0700
From:   Vijay Balakrishna <vijayb@...ux.microsoft.com>
To:     Baoquan He <bhe@...hat.com>
Cc:     Dave Young <dyoung@...hat.com>, clang-built-linux@...glegroups.com,
        Tyler Hicks <tyhicks@...ux.microsoft.com>,
        kexec@...ts.infradead.org, linux-kernel@...r.kernel.org,
        Vivek Goyal <vgoyal@...hat.com>
Subject: Re: [PATCH][RFC] kdump: append kernel build-id string to VMCOREINFO

Thanks Baoquan.  See inline.

On 6/9/2020 7:20 PM, Baoquan He wrote:
> On 06/04/20 at 05:01pm, Vijay Balakrishna wrote:
>> Make kernel GNU build-id available in VMCOREINFO.  Having
>> build-id in VMCOREINFO facilitates presenting appropriate kernel
>> namelist image with debug information file to kernel crash dump
>> analysis tools.  Currently VMCOREINFO lacks uniquely identifiable
>> key for crash analysis automation.
> 
> Looks like a good idea. I have several concerns, please check below inline
> comments.
> 
>>
>> Signed-off-by: Vijay Balakrishna <vijayb@...ux.microsoft.com>
>> ---
>> Regarding if this patch is necessary or matching of linux_banner
>> and OSRELEASE in VMCOREINFO employed by crash(8) meets the
>> need -- IMO, build-id approach more foolproof, in most instances it
>> is a cryptographic hash generated using internal code/ELF bits unlike
>> kernel version string upon which linux_banner is based that is
>> external to the code.  I feel each is intended for a different purpose.
>> Also OSRELEASE is not suitable when two different kernel builds
>> from same version with different features enabled.
>>
>> Currently for most linux (and non-linux) systems build-id can be
>> extracted using standard methods for file types such as user mode crash
>> dumps, shared libraries, loadable kernel modules etc.  This is an
>> exception for linux kernel dump.  Having build-id in VMCOREINFO brings
>> some uniformity for automation tools.
> 
> These two paragraphs can be added to log too?
Sure, I will move it too commit log in my next version.

> 
>>
>>   include/linux/crash_core.h |  6 +++++
>>   kernel/crash_core.c        | 50 ++++++++++++++++++++++++++++++++++++++
>>   2 files changed, 56 insertions(+)
>>
>> diff --git a/include/linux/crash_core.h b/include/linux/crash_core.h
>> index 525510a9f965..6594dbc34a37 100644
>> --- a/include/linux/crash_core.h
>> +++ b/include/linux/crash_core.h
>> @@ -38,6 +38,8 @@ phys_addr_t paddr_vmcoreinfo_note(void);
>>   
>>   #define VMCOREINFO_OSRELEASE(value) \
>>   	vmcoreinfo_append_str("OSRELEASE=%s\n", value)
>> +#define VMCOREINFO_BUILD_ID(value) \
>> +	vmcoreinfo_append_str("BUILD-ID=%s\n", value)
>>   #define VMCOREINFO_PAGESIZE(value) \
>>   	vmcoreinfo_append_str("PAGESIZE=%ld\n", value)
>>   #define VMCOREINFO_SYMBOL(name) \
>> @@ -64,6 +66,10 @@ extern unsigned char *vmcoreinfo_data;
>>   extern size_t vmcoreinfo_size;
>>   extern u32 *vmcoreinfo_note;
>>   
>> +/* raw contents of kernel .notes section */
>> +extern const void __start_notes __weak;
>> +extern const void __stop_notes __weak;
>> +
>>   Elf_Word *append_elf_note(Elf_Word *buf, char *name, unsigned int type,
>>   			  void *data, size_t data_len);
>>   void final_note(Elf_Word *buf);
>> diff --git a/kernel/crash_core.c b/kernel/crash_core.c
>> index 9f1557b98468..c094db478803 100644
>> --- a/kernel/crash_core.c
>> +++ b/kernel/crash_core.c
>> @@ -11,6 +11,8 @@
>>   #include <asm/page.h>
>>   #include <asm/sections.h>
>>   
>> +#include <crypto/sha.h>
>> +
>>   /* vmcoreinfo stuff */
>>   unsigned char *vmcoreinfo_data;
>>   size_t vmcoreinfo_size;
>> @@ -376,6 +378,53 @@ phys_addr_t __weak paddr_vmcoreinfo_note(void)
>>   }
>>   EXPORT_SYMBOL(paddr_vmcoreinfo_note);
>>   
>> +#define NOTES_SIZE (&__stop_notes - &__start_notes)
>> +#define BUILD_ID_MAX SHA1_DIGEST_SIZE
>> +#define NT_GNU_BUILD_ID 3
>> +
>> +struct elf_note_section {
>> +	struct elf_note	n_hdr;
>> +	u8 n_data[];
>> +};
>> +
>> +/*
>> + * Add build ID from .notes section as generated by the GNU ld(1)
>> + * or LLVM lld(1) --build-id option.
>> + */
>> +static void add_build_id_vmcoreinfo(void)
>> +{
>> +	char build_id[BUILD_ID_MAX * 2 + 1];
> 
> Here, could you tell why we need two times the build id length?

GNU build-id is binary data, it is converted to hex (string) 
representation here.

> 
>> +	int n_remain = NOTES_SIZE;
>> +
>> +	while (n_remain >= sizeof(struct elf_note)) {
>> +		const struct elf_note_section *g_build_id =
> 
> Here naming of 'g_build_id' looks a little weird. We usually define
> global variable with 'g_xxxx'. Maybe we can change it to 'note_sec' or
> something else, because not all iterated elf_note_section is build_id
> related.

I will rename it in next version.

Vijay
> 
>> +			&__start_notes + NOTES_SIZE - n_remain;
>> +		const u32 n_namesz = g_build_id->n_hdr.n_namesz;
>> +
>> +		if (g_build_id->n_hdr.n_type == NT_GNU_BUILD_ID &&
>> +		    n_namesz != 0 &&
>> +		    !strcmp((char *)&g_build_id->n_data[0], "GNU")) {
>> +			if (g_build_id->n_hdr.n_descsz <= BUILD_ID_MAX) {
>> +				const u32 n_descsz = g_build_id->n_hdr.n_descsz;
>> +				const u8 *s = &g_build_id->n_data[n_namesz];
>> +
>> +				s = PTR_ALIGN(s, 4);
>> +				bin2hex(build_id, s, n_descsz);
>> +				build_id[2 * n_descsz] = '\0';
>> +				VMCOREINFO_BUILD_ID(build_id);
>> +				return;
>> +			}
>> +			pr_warn("Build ID is too large to include in vmcoreinfo: %lu > %lu\n",
>> +				g_build_id->n_hdr.n_descsz,
>> +				BUILD_ID_MAX);
>> +			return;
>> +		}
>> +		n_remain -= sizeof(struct elf_note) +
>> +			ALIGN(g_build_id->n_hdr.n_namesz, 4) +
>> +			ALIGN(g_build_id->n_hdr.n_descsz, 4);
>> +	}
>> +}
>> +
>>   static int __init crash_save_vmcoreinfo_init(void)
>>   {
>>   	vmcoreinfo_data = (unsigned char *)get_zeroed_page(GFP_KERNEL);
>> @@ -394,6 +443,7 @@ static int __init crash_save_vmcoreinfo_init(void)
>>   	}
>>   
>>   	VMCOREINFO_OSRELEASE(init_uts_ns.name.release);
>> +	add_build_id_vmcoreinfo();
>>   	VMCOREINFO_PAGESIZE(PAGE_SIZE);
>>   
>>   	VMCOREINFO_SYMBOL(init_uts_ns);
>> -- 
>> 2.26.2
>>
>>
>> _______________________________________________
>> kexec mailing list
>> kexec@...ts.infradead.org
>> http://lists.infradead.org/mailman/listinfo/kexec
>>

Powered by blists - more mailing lists