public inbox for linux-arm-kernel@lists.infradead.org 
 help / color / mirror / Atom feed
From: cyril@ti•com (Cyril Chemparathy)
To: linux-arm-kernel@lists•infradead.org
Subject: [PATCH 02/22] ARM: use late patch framework for phys-virt patching
Date: Sun, 5 Aug 2012 10:03:45 -0400	[thread overview]
Message-ID: <501E7D41.8070303@ti.com> (raw)
In-Reply-To: <alpine.LFD.2.02.1208040138560.5231@xanadu.home>

Hi Nicolas,

On 8/4/2012 2:15 AM, Nicolas Pitre wrote:
> On Tue, 31 Jul 2012, Cyril Chemparathy wrote:
>
>> This patch replaces the original physical offset patching implementation
>> with one that uses the newly added patching framework.  In the process, we now
>> unconditionally initialize the __pv_phys_offset and __pv_offset globals in the
>> head.S code.
>
> Why unconditionally initializing those?  There is no reason for that.
>

We could keep this conditional on LPAE, but do you see any specific need 
for keeping it conditional?

>> Signed-off-by: Cyril Chemparathy <cyril@ti•com>
>
> Comments below.
>
>> diff --git a/arch/arm/kernel/head.S b/arch/arm/kernel/head.S
>> index 835898e..d165896 100644
>> --- a/arch/arm/kernel/head.S
>> +++ b/arch/arm/kernel/head.S
> [...]
>>   	.data
>>   	.globl	__pv_phys_offset
>>   	.type	__pv_phys_offset, %object
>>   __pv_phys_offset:
>>   	.long	0
>>   	.size	__pv_phys_offset, . - __pv_phys_offset
>> +
>> +	.globl	__pv_offset
>> +	.type	__pv_offset, %object
>>   __pv_offset:
>>   	.long	0
>> -#endif
>> +	.size	__pv_offset, . - __pv_offset
>
> Please move those to C code.  They aren't of much use in this file
> anymore.  This will allow you to use pphys_addr_t for them as well in
> your subsequent patch. And more importantly get rid of that ugly
> pv_offset_high that you introduced iin another patch.
>

Moving it to C-code caused problems because these get filled in prior to 
BSS being cleared.

We could potentially have this initialized in C with a mystery dummy 
value to prevent it from landing in BSS.  Would that be acceptable?

>> diff --git a/arch/arm/kernel/module.c b/arch/arm/kernel/module.c
>> index df5e897..39f8fce 100644
>> --- a/arch/arm/kernel/module.c
>> +++ b/arch/arm/kernel/module.c
>> @@ -317,11 +317,6 @@ int module_finalize(const Elf32_Ehdr *hdr, const Elf_Shdr *sechdrs,
>>   					         maps[i].txt_sec->sh_addr,
>>   					         maps[i].txt_sec->sh_size);
>>   #endif
>> -#ifdef CONFIG_ARM_PATCH_PHYS_VIRT
>> -	s = find_mod_section(hdr, sechdrs, ".pv_table");
>> -	if (s)
>> -		fixup_pv_table((void *)s->sh_addr, s->sh_size);
>> -#endif
>>   	s = find_mod_section(hdr, sechdrs, ".patch.table");
>>   	if (s)
>>   		patch_kernel((void *)s->sh_addr, s->sh_size);
>
> The patch_kernel code and its invokation should still be conditional on
> CONFIG_ARM_PATCH_PHYS_VIRT.  This ability may still be configured out
> irrespective of the implementation used.
>

Maybe CONFIG_ARM_PATCH_PHYS_VIRT is not quite appropriate if this is 
used to patch up other things in addition to phys-virt stuff?

I could have this dependent on CONFIG_ARM_INIT_PATCH (or whatever 
nomenclature we chose for this) and have CONFIG_ARM_PATCH_PHYS_VIRT 
depend on it.

>> diff --git a/arch/arm/kernel/vmlinux.lds.S b/arch/arm/kernel/vmlinux.lds.S
>> index bacb275..13731e3 100644
>> --- a/arch/arm/kernel/vmlinux.lds.S
>> +++ b/arch/arm/kernel/vmlinux.lds.S
>> @@ -162,11 +162,6 @@ SECTIONS
>>   		__smpalt_end = .;
>>   	}
>>   #endif
>> -	.init.pv_table : {
>> -		__pv_table_begin = .;
>> -		*(.pv_table)
>> -		__pv_table_end = .;
>> -	}
>>   	.init.patch_table : {
>>   		__patch_table_begin = .;
>>   		*(.patch.table)
>
> Since you're changing the module ABI,it is important to also modify the
> module vermagic string in asm/module.h to prevent the loading of
> incompatible kernel modules.
>

Absolutely.  Thanks.

>
> Nicolas
>

-- 
Thanks
- Cyril

  reply	other threads:[~2012-08-05 14:03 UTC|newest]

Thread overview: 64+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-07-31 23:04 [PATCH 00/22] Introducing the TI Keystone platform Cyril Chemparathy
2012-07-31 23:04 ` [PATCH 01/22] ARM: add mechanism for late code patching Cyril Chemparathy
2012-08-04  5:38   ` Nicolas Pitre
2012-08-05 13:56     ` Cyril Chemparathy
2012-08-07 22:52     ` Cyril Chemparathy
2012-08-08  5:56       ` Nicolas Pitre
2012-08-08 13:18         ` Cyril Chemparathy
2012-08-08 13:55           ` Nicolas Pitre
2012-08-08 16:05             ` Russell King - ARM Linux
2012-08-08 16:56               ` Nicolas Pitre
2012-08-09  6:59                 ` Tixy
2012-08-06 11:12   ` Russell King - ARM Linux
2012-08-06 13:19     ` Cyril Chemparathy
2012-08-06 13:26       ` Russell King - ARM Linux
2012-08-06 13:38         ` Cyril Chemparathy
2012-08-06 18:02         ` Nicolas Pitre
2012-07-31 23:04 ` [PATCH 02/22] ARM: use late patch framework for phys-virt patching Cyril Chemparathy
2012-08-04  6:15   ` Nicolas Pitre
2012-08-05 14:03     ` Cyril Chemparathy [this message]
2012-08-06  2:06       ` Nicolas Pitre
2012-07-31 23:04 ` [PATCH 03/22] ARM: LPAE: use phys_addr_t on virt <--> phys conversion Cyril Chemparathy
2012-08-04  6:24   ` Nicolas Pitre
2012-08-05 14:05     ` Cyril Chemparathy
2012-08-06 11:14   ` Russell King - ARM Linux
2012-08-06 13:30     ` Cyril Chemparathy
2012-08-09 14:10     ` Cyril Chemparathy
2012-07-31 23:04 ` [PATCH 04/22] ARM: LPAE: support 64-bit virt/phys patching Cyril Chemparathy
2012-08-04  6:49   ` Nicolas Pitre
2012-08-05 14:21     ` Cyril Chemparathy
2012-08-06  2:19       ` Nicolas Pitre
2012-07-31 23:04 ` [PATCH 05/22] ARM: LPAE: use signed arithmetic for mask definitions Cyril Chemparathy
2012-07-31 23:04 ` [PATCH 06/22] ARM: LPAE: use phys_addr_t in alloc_init_pud() Cyril Chemparathy
2012-08-01 12:08   ` Sergei Shtylyov
2012-08-01 15:42     ` Cyril Chemparathy
2012-08-04  6:51   ` Nicolas Pitre
2012-07-31 23:04 ` [PATCH 07/22] ARM: LPAE: use phys_addr_t in free_memmap() Cyril Chemparathy
2012-08-04  6:54   ` Nicolas Pitre
2012-07-31 23:04 ` [PATCH 08/22] ARM: LPAE: use phys_addr_t for initrd location and size Cyril Chemparathy
2012-08-04  6:57   ` Nicolas Pitre
2012-08-05 14:23     ` Cyril Chemparathy
2012-07-31 23:04 ` [PATCH 09/22] ARM: LPAE: use 64-bit pgd physical address in switch_mm() Cyril Chemparathy
2012-08-04  7:04   ` Nicolas Pitre
2012-08-05 14:29     ` Cyril Chemparathy
2012-08-06  2:35       ` Nicolas Pitre
2012-07-31 23:04 ` [PATCH 10/22] ARM: LPAE: use 64-bit accessors for TTBR registers Cyril Chemparathy
2012-07-31 23:04 ` [PATCH 11/22] ARM: LPAE: define ARCH_LOW_ADDRESS_LIMIT for bootmem Cyril Chemparathy
2012-07-31 23:04 ` [PATCH 12/22] ARM: LPAE: factor out T1SZ and TTBR1 computations Cyril Chemparathy
2012-07-31 23:04 ` [PATCH 13/22] ARM: LPAE: allow proc override of TTB setup Cyril Chemparathy
2012-07-31 23:04 ` [PATCH 14/22] ARM: LPAE: accomodate >32-bit addresses for page table base Cyril Chemparathy
2012-07-31 23:04 ` [PATCH 15/22] ARM: mm: use physical addresses in highmem sanity checks Cyril Chemparathy
2012-07-31 23:04 ` [PATCH 16/22] ARM: mm: cleanup checks for membank overlap with vmalloc area Cyril Chemparathy
2012-07-31 23:04 ` [PATCH 17/22] ARM: mm: clean up membank size limit checks Cyril Chemparathy
2012-07-31 23:04 ` [PATCH 18/22] ARM: add virt_to_idmap for interconnect aliasing Cyril Chemparathy
2012-07-31 23:04 ` [PATCH 19/22] ARM: recreate kernel mappings in early_paging_init() Cyril Chemparathy
2012-07-31 23:05 ` [RFC 21/22] ARM: keystone: enable SMP on Keystone machines Cyril Chemparathy
2012-07-31 23:05 ` [RFC 22/22] ARM: keystone: add switch over to high physical address range Cyril Chemparathy
2012-07-31 23:06 ` [RFC 20/22] ARM: keystone: introducing TI Keystone platform Cyril Chemparathy
2012-07-31 23:16   ` Arnd Bergmann
2012-08-01 15:41     ` Cyril Chemparathy
2012-08-01 17:20       ` Arnd Bergmann
2012-08-04  8:39 ` [PATCH 00/22] Introducing the " Russell King - ARM Linux
2012-08-05 15:10   ` Cyril Chemparathy
2012-08-08 15:43     ` Catalin Marinas
2012-08-08 13:57 ` Will Deacon

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=501E7D41.8070303@ti.com \
    --to=cyril@ti$(echo .)com \
    --cc=linux-arm-kernel@lists$(echo .)infradead.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox