* Re: PROBLEM: monotonic clock going backwards on ppc64
From: Michael Ellerman @ 2019-03-01 13:24 UTC (permalink / raw)
To: Jakub Drnec, linuxppc-dev; +Cc: Stephen Boyd, Thomas Gleixner, John Stultz
In-Reply-To: <HaC.ZfES.62bwlnvAvMP.1STMMj@seznam.cz>
Hi Jakub,
[Cc += Timekeeping maintainers]
"Jakub Drnec" <jaydee@email.cz> writes:
> Hi all,
>
> I think I observed a potential problem, is this the correct place to report it? (CC me, not on list)
>
> [1.] One line summary: monotonic clock can be made to decrease on ppc64
> [2.] Full description:
> Setting the realtime clock can sometimes make the monotonic clock go back by over a hundred years.
> Decreasing the realtime clock across the y2k38 threshold is one reliable way to reproduce.
> Allegedly this can also happen just by running ntpd, I have not managed to reproduce that other
> than booting with rtc at >2038 and then running ntp.
> When this happens, anything with timers (e.g. openjdk) breaks rather badly.
Thanks for the report.
> The problem seems to be in vDSO code in arch/powerpc/kernel/vdso64/gettimeofday.S.
You're right, the wall-to-monotonic offset (wtom_clock_sec) is a signed
32-bit value, so that seems like it's going to have problems.
If I do `date -s 2037-1-1` I see:
[ 26.024061] update_vsyscall: tk->wall_to_monotonic.tv_sec -2114341175
[ 26.042633] update_vsyscall: vdso_data->wtom_clock_sec -2114341175
Which looks sane.
But then 2040-1-1 shows:
[ 32.617020] update_vsyscall: tk->wall_to_monotonic.tv_sec -2208949168
[ 32.632642] update_vsyscall: vdso_data->wtom_clock_sec 2086018128
ie. the larger negative offset has overflowed and become positive.
But then when we go back to 2037 we get a negative offset again and
monotonic time appears to go backward and things are unhappy.
I don't know this code well, but the patch below *appears* to work. I'll
have a closer look on Monday.
cheers
diff --git a/arch/powerpc/include/asm/vdso_datapage.h b/arch/powerpc/include/asm/vdso_datapage.h
index 1afe90ade595..139133ec21d5 100644
--- a/arch/powerpc/include/asm/vdso_datapage.h
+++ b/arch/powerpc/include/asm/vdso_datapage.h
@@ -82,7 +82,7 @@ struct vdso_data {
__u32 icache_block_size; /* L1 i-cache block size */
__u32 dcache_log_block_size; /* L1 d-cache log block size */
__u32 icache_log_block_size; /* L1 i-cache log block size */
- __s32 wtom_clock_sec; /* Wall to monotonic clock */
+ __s64 wtom_clock_sec; /* Wall to monotonic clock */
__s32 wtom_clock_nsec;
struct timespec stamp_xtime; /* xtime as at tb_orig_stamp */
__u32 stamp_sec_fraction; /* fractional seconds of stamp_xtime */
diff --git a/arch/powerpc/kernel/vdso64/gettimeofday.S b/arch/powerpc/kernel/vdso64/gettimeofday.S
index a4ed9edfd5f0..1f324c28705b 100644
--- a/arch/powerpc/kernel/vdso64/gettimeofday.S
+++ b/arch/powerpc/kernel/vdso64/gettimeofday.S
@@ -92,7 +92,7 @@ V_FUNCTION_BEGIN(__kernel_clock_gettime)
* At this point, r4,r5 contain our sec/nsec values.
*/
- lwa r6,WTOM_CLOCK_SEC(r3)
+ ld r6,WTOM_CLOCK_SEC(r3)
lwa r9,WTOM_CLOCK_NSEC(r3)
/* We now have our result in r6,r9. We create a fake dependency
@@ -125,7 +125,7 @@ V_FUNCTION_BEGIN(__kernel_clock_gettime)
bne cr6,75f
/* CLOCK_MONOTONIC_COARSE */
- lwa r6,WTOM_CLOCK_SEC(r3)
+ ld r6,WTOM_CLOCK_SEC(r3)
lwa r9,WTOM_CLOCK_NSEC(r3)
/* check if counter has updated */
> [3.] Keywords: gettimeofday, ppc64, vdso
> [4.] Kernel information
> [4.1.] Kernel version: any (tested on 4.19)
> [4.2.] Kernel .config file: any
> [5.] Most recent kernel version which did not have the bug: not a regression
> [6.] Output of Oops..: not applicable
> [7.] Example program which triggers the problem
> --- testcase.c
> #include <stdio.h>
> #include <time.h>
> #include <stdlib.h>
> #include <unistd.h>
>
> long get_time() {
> struct timespec tp;
> if (clock_gettime(CLOCK_MONOTONIC, &tp) != 0) {
> perror("clock_gettime failed");
> exit(1);
> }
> long result = tp.tv_sec + tp.tv_nsec / 1000000000;
> return result;
> }
>
> int main() {
> printf("monitoring monotonic clock...\n");
> long last = get_time();
> while(1) {
> long now = get_time();
> if (now < last) {
> printf("clock went backwards by %ld seconds!\n",
> last - now);
> }
> last = now;
> sleep(1);
> }
> return 0;
> }
> ---
> when running
> # date -s 2040-1-1
> # date -s 2037-1-1
> program outputs: clock went backwards by 4294967295 seconds!
>
> [8.] Environment: any ppc64, currently reproducing on qemu-system-ppc64le running debian unstable
> [X.] Other notes, patches, fixes, workarounds:
> The problem seems to be in vDSO code in arch/powerpc/kernel/vdso64/gettimeofday.S.
> (possibly because some values used in the calculation are only 32 bit?)
> Slightly silly workaround:
> nuke the "cmpwi cr1,r3,CLOCK_MONOTONIC" in __kernel_clock_gettime
> Now it always goes through the syscall fallback which does not have the same problem.
>
> Regards,
> Jakub Drnec
^ permalink raw reply related
* Re: [PATCH v4 4/4] hugetlb: allow to free gigantic pages regardless of the configuration
From: Alexandre Ghiti @ 2019-03-01 13:21 UTC (permalink / raw)
To: Mike Kravetz
Cc: Rich Felker, linux-sh, Catalin Marinas, Dave Hansen, Will Deacon,
Dave Hansen, Paul Mackerras, H . Peter Anvin, sparclinux,
linux-s390, Yoshinori Sato, x86, Ingo Molnar, Vlastimil Babka,
Heiko Carstens, Borislav Petkov, Andy Lutomirski, Thomas Gleixner,
linux-arm-kernel, linux-mm, linux-kernel, Martin Schwidefsky,
linuxppc-dev, David S . Miller
In-Reply-To: <1cfaca88-a219-d057-3ab8-37fb1c1687d6@ghiti.fr>
On 03/01/2019 07:25 AM, Alex Ghiti wrote:
> On 2/28/19 5:26 PM, Mike Kravetz wrote:
>> On 2/28/19 12:23 PM, Dave Hansen wrote:
>>> On 2/28/19 11:50 AM, Mike Kravetz wrote:
>>>> On 2/28/19 11:13 AM, Dave Hansen wrote:
>>>>>> + if (hstate_is_gigantic(h) &&
>>>>>> !IS_ENABLED(CONFIG_CONTIG_ALLOC)) {
>>>>>> + spin_lock(&hugetlb_lock);
>>>>>> + if (count > persistent_huge_pages(h)) {
>>>>>> + spin_unlock(&hugetlb_lock);
>>>>>> + return -EINVAL;
>>>>>> + }
>>>>>> + goto decrease_pool;
>>>>>> + }
>>>>> This choice confuses me. The "Decrease the pool size" code already
>>>>> works and the code just falls through to it after skipping all the
>>>>> "Increase the pool size" code.
>>>>>
>>>>> Why did did you need to add this case so early? Why not just let it
>>>>> fall through like before?
>>>> I assume you are questioning the goto, right? You are correct in that
>>>> it is unnecessary and we could just fall through.
>>> Yeah, it just looked odd to me.
>
> (Dave I do not receive your answers, I don't know why).
I collected mistakes here: domain name expired and no mailing list added :)
Really sorry about that, I missed the whole discussion (if any).
Could someone forward it to me (if any) ? Thanks !
> I'd rather avoid useless checks when we already know they won't
> be met and I think that makes the code more understandable.
>
> But that's up to you for the next version.
>
> Thanks
>>>
>>>> However, I wonder if we might want to consider a wacky condition
>>>> that the
>>>> above check would prevent. Consider a system/configuration with 5
>>>> gigantic
>>>> pages allocated at boot time. Also CONFIG_CONTIG_ALLOC is not
>>>> enabled, so
>>>> it is not possible to allocate gigantic pages after boot.
>>>>
>>>> Suppose the admin decreased the number of gigantic pages to 3.
>>>> However, all
>>>> gigantic pages were in use. So, 2 gigantic pages are now 'surplus'.
>>>> h->nr_huge_pages == 5 and h->surplus_huge_pages == 2, so
>>>> persistent_huge_pages() == 3.
>>>>
>>>> Now suppose the admin wanted to increase the number of gigantic
>>>> pages to 5.
>>>> The above check would prevent this. However, we do not need to really
>>>> 'allocate' two gigantic pages. We can simply convert the surplus
>>>> pages.
>>>>
>>>> I admit this is a wacky condition. The ability to 'free' gigantic
>>>> pages
>>>> at runtime if !CONFIG_CONTIG_ALLOC makes it possible. I don't
>>>> necessairly
>>>> think we should consider this. hugetlbfs code just makes me think of
>>>> wacky things. :)
>>> I think you're saying that the newly-added check is overly-restrictive.
>>> If we "fell through" like I was suggesting we would get better
>>> behavior.
>> At first, I did not think it overly restrictive. But, I believe we can
>> just eliminate that check for gigantic pages. If
>> !CONFIG_CONTIG_ALLOC and
>> this is a request to allocate more gigantic pages,
>> alloc_pool_huge_page()
>> should return NULL.
>>
>> The only potential issue I see is that in the past we have returned
>> EINVAL
>> if !CONFIG_CONTIG_ALLOC and someone attempted to increase the pool size.
>> Now, we will not increase the pool and will not return an error. Not
>> sure
>> if that is an acceptable change in user behavior.
>
> If I may, I think that this is the kind of info the user wants to have
> and we should
> return an error when it is not possible to allocate runtime huge pages.
> I already noticed that if someone asks for 10 huge pages, and only 5
> are allocated,
> no error is returned to the user and I found that surprising.
>
>>
>> If we go down this path, then we could remove this change as well:
>
> I agree that in that path, we do not need the following change neither.
>
>>
>>> @@ -2428,7 +2442,9 @@ static ssize_t
>>> __nr_hugepages_store_common(bool obey_mempolicy,
>>> } else
>>> nodes_allowed = &node_states[N_MEMORY];
>>> - h->max_huge_pages = set_max_huge_pages(h, count, nodes_allowed);
>>> + err = set_max_huge_pages(h, count, nodes_allowed);
>>> + if (err)
>>> + goto out;
>>> if (nodes_allowed != &node_states[N_MEMORY])
>>> NODEMASK_FREE(nodes_allowed);
>> Do note that I beleive there is a bug the above change. The code after
>> the out label is:
>>
>> out:
>> NODEMASK_FREE(nodes_allowed);
>> return err;
>> }
>>
>> With the new goto, we need the same
>> if (nodes_allowed != &node_states[N_MEMORY]) before NODEMASK_FREE().
>>
>> Sorry, I missed this in previous versions.
>
> Oh right, I'm really sorry I missed that, thank you for noticing.
>
^ permalink raw reply
* [PATCH] powerpc/64s: Fix unrelocated interrupt trampoline address test
From: Nicholas Piggin @ 2019-03-01 12:56 UTC (permalink / raw)
To: linuxppc-dev; +Cc: Nicholas Piggin
The recent commit got this test wrong, it declared the assembler
symbols the wrong way, and also used the wrong symbol name
(xxx_start rather than start_xxx, see asm/head-64.h).
Fixes: ccd477028a ("powerpc/64s: Fix HV NMI vs HV interrupt recoverability test")
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/include/asm/asm-prototypes.h | 8 --------
arch/powerpc/include/asm/sections.h | 7 +++++++
arch/powerpc/kernel/exceptions-64s.S | 8 ++++----
arch/powerpc/kernel/traps.c | 9 +++++----
4 files changed, 16 insertions(+), 16 deletions(-)
diff --git a/arch/powerpc/include/asm/asm-prototypes.h b/arch/powerpc/include/asm/asm-prototypes.h
index 0f8326644fa4..1d911f68a23b 100644
--- a/arch/powerpc/include/asm/asm-prototypes.h
+++ b/arch/powerpc/include/asm/asm-prototypes.h
@@ -51,14 +51,6 @@ int exit_vmx_usercopy(void);
int enter_vmx_ops(void);
void *exit_vmx_ops(void *dest);
-/* Exceptions */
-#ifdef CONFIG_PPC_POWERNV
-extern unsigned long real_trampolines_start;
-extern unsigned long real_trampolines_end;
-extern unsigned long virt_trampolines_start;
-extern unsigned long virt_trampolines_end;
-#endif
-
/* Traps */
long machine_check_early(struct pt_regs *regs);
long hmi_exception_realmode(struct pt_regs *regs);
diff --git a/arch/powerpc/include/asm/sections.h b/arch/powerpc/include/asm/sections.h
index e335a8f846af..4a1664a8658d 100644
--- a/arch/powerpc/include/asm/sections.h
+++ b/arch/powerpc/include/asm/sections.h
@@ -17,6 +17,13 @@ extern char __end_interrupts[];
extern char __prom_init_toc_start[];
extern char __prom_init_toc_end[];
+#ifdef CONFIG_PPC_POWERNV
+extern char start_real_trampolines[];
+extern char end_real_trampolines[];
+extern char start_virt_trampolines[];
+extern char end_virt_trampolines[];
+#endif
+
static inline int in_kernel_text(unsigned long addr)
{
if (addr >= (unsigned long)_stext && addr < (unsigned long)__init_end)
diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S
index d2e9fc968655..6f0057566a80 100644
--- a/arch/powerpc/kernel/exceptions-64s.S
+++ b/arch/powerpc/kernel/exceptions-64s.S
@@ -70,10 +70,10 @@ OPEN_FIXED_SECTION(virt_vectors, 0x4000, 0x5900)
OPEN_FIXED_SECTION(virt_trampolines, 0x5900, 0x7000)
#ifdef CONFIG_PPC_POWERNV
- .globl real_trampolines_start
- .globl real_trampolines_end
- .globl virt_trampolines_start
- .globl virt_trampolines_end
+ .globl start_real_trampolines
+ .globl end_real_trampolines
+ .globl start_virt_trampolines
+ .globl end_virt_trampolines
#endif
#if defined(CONFIG_PPC_PSERIES) || defined(CONFIG_PPC_POWERNV)
diff --git a/arch/powerpc/kernel/traps.c b/arch/powerpc/kernel/traps.c
index 12b54908c15d..ee7c6cb876ba 100644
--- a/arch/powerpc/kernel/traps.c
+++ b/arch/powerpc/kernel/traps.c
@@ -426,12 +426,13 @@ void hv_nmi_check_nonrecoverable(struct pt_regs *regs)
goto nonrecoverable;
if ((nip >= 0xf80 && nip < 0xfa0) || (nip >= 0x4f80 && nip < 0x4fa0))
goto nonrecoverable;
+
/* Trampoline code runs un-relocated so subtract kbase. */
- if (nip >= real_trampolines_start - kbase &&
- nip < real_trampolines_end - kbase)
+ if (nip >= (unsigned long)(start_real_trampolines - kbase) &&
+ nip < (unsigned long)(end_real_trampolines - kbase))
goto nonrecoverable;
- if (nip >= virt_trampolines_start - kbase &&
- nip < virt_trampolines_end - kbase)
+ if (nip >= (unsigned long)(start_virt_trampolines - kbase) &&
+ nip < (unsigned long)(end_virt_trampolines - kbase))
goto nonrecoverable;
return;
--
2.18.0
^ permalink raw reply related
* Re: [PATCH] ASoC: fsl_asrc: add protection for the asrc of older version
From: Mark Brown @ 2019-03-01 12:38 UTC (permalink / raw)
To: S.j. Wang
Cc: alsa-devel@alsa-project.org, timur@kernel.org,
Xiubo.Lee@gmail.com, festevam@gmail.com, Nicolin Chen,
linuxppc-dev@lists.ozlabs.org
In-Reply-To: <VI1PR0402MB339254C31997242633511D12E3760@VI1PR0402MB3392.eurprd04.prod.outlook.com>
[-- Attachment #1: Type: text/plain, Size: 1017 bytes --]
On Fri, Mar 01, 2019 at 06:55:25AM +0000, S.j. Wang wrote:
> > Alternatively, I feel instead of error-out at here, should we add a HW
> > constraint or at least fence it off at the beginning of the hw_params()? This
> > is actually nothing specific to the pair-request function but a hardware
> > constraint.
> How about add constraint in startup?
> static int fsl_asrc_dai_startup(struct snd_pcm_substream *substream,
> struct snd_soc_dai *dai)
> {
> struct fsl_asrc *asrc_priv = snd_soc_dai_get_drvdata(dai);
>
> if (asrc_priv->channel_bits == 3) {
> snd_pcm_hw_constraint_step(substream->runtime, 0,
> SNDRV_PCM_HW_PARAM_CHANNELS, 2);
> }
>
> return 0;
> }
Yes, that's definitely good - the general idea is that we should never
need to return an error from hw_params() as the constraints code will
have filtered out any invalid configurations before they get that far.
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]
^ permalink raw reply
* [PATCH v9 11/11] powerpc/32s: set up an early static hash table for KASAN.
From: Christophe Leroy @ 2019-03-01 12:33 UTC (permalink / raw)
To: Benjamin Herrenschmidt, Paul Mackerras, Michael Ellerman,
Nicholas Piggin, Aneesh Kumar K.V, Andrey Ryabinin,
Alexander Potapenko, Dmitry Vyukov, Daniel Axtens
Cc: linux-mm, linuxppc-dev, linux-kernel, kasan-dev
In-Reply-To: <cover.1551443452.git.christophe.leroy@c-s.fr>
KASAN requires early activation of hash table, before memblock()
functions are available.
This patch implements an early hash_table statically defined in
__initdata.
During early boot, a single page table is used.
For hash32, when doing the final init, one page table is allocated
for each PGD entry because of the _PAGE_HASHPTE flag which can't be
common to several virt pages. This is done after memblock get
available but before switching to the final hash table, otherwise
there are issues with TLB flushing due to the shared entries.
For hash32, the zero shadow page gets mapped with PAGE_READONLY instead
of PAGE_KERNEL_RO, because the PP bits don't provide a RO kernel, so
PAGE_KERNEL_RO is equivalent to PAGE_KERNEL. By using PAGE_READONLY,
the page is RO for both kernel and user, but this is not a security issue
as it contains only zeroes.
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
---
arch/powerpc/include/asm/kasan.h | 1 +
arch/powerpc/kernel/head_32.S | 43 ++++++++++++++++++++++-------
arch/powerpc/mm/kasan/kasan_init_32.c | 51 ++++++++++++++++++++++++++++++-----
arch/powerpc/mm/mmu_decl.h | 1 +
4 files changed, 79 insertions(+), 17 deletions(-)
diff --git a/arch/powerpc/include/asm/kasan.h b/arch/powerpc/include/asm/kasan.h
index 74a4ba9fb8a3..c9fe0369a8fc 100644
--- a/arch/powerpc/include/asm/kasan.h
+++ b/arch/powerpc/include/asm/kasan.h
@@ -29,6 +29,7 @@
#ifdef CONFIG_KASAN
void kasan_early_init(void);
+void kasan_mmu_init(void);
void kasan_init(void);
#else
static inline void kasan_init(void) { }
diff --git a/arch/powerpc/kernel/head_32.S b/arch/powerpc/kernel/head_32.S
index e644aab2cf5b..7f7fbdd73b79 100644
--- a/arch/powerpc/kernel/head_32.S
+++ b/arch/powerpc/kernel/head_32.S
@@ -160,6 +160,10 @@ __after_mmu_off:
bl flush_tlbs
bl initial_bats
+ bl load_segment_registers
+#ifdef CONFIG_KASAN
+ bl early_hash_table
+#endif
#if defined(CONFIG_BOOTX_TEXT)
bl setup_disp_bat
#endif
@@ -205,7 +209,7 @@ __after_mmu_off:
*/
turn_on_mmu:
mfmsr r0
- ori r0,r0,MSR_DR|MSR_IR
+ ori r0,r0,MSR_DR|MSR_IR|MSR_RI
mtspr SPRN_SRR1,r0
lis r0,start_here@h
ori r0,r0,start_here@l
@@ -881,11 +885,24 @@ _ENTRY(__restore_cpu_setup)
blr
#endif /* !defined(CONFIG_PPC_BOOK3S_32) */
-
/*
* Load stuff into the MMU. Intended to be called with
* IR=0 and DR=0.
*/
+#ifdef CONFIG_KASAN
+early_hash_table:
+ sync /* Force all PTE updates to finish */
+ isync
+ tlbia /* Clear all TLB entries */
+ sync /* wait for tlbia/tlbie to finish */
+ TLBSYNC /* ... on all CPUs */
+ /* Load the SDR1 register (hash table base & size) */
+ lis r6, early_hash - PAGE_OFFSET@h
+ ori r6, r6, 3 /* 256kB table */
+ mtspr SPRN_SDR1, r6
+ blr
+#endif
+
load_up_mmu:
sync /* Force all PTE updates to finish */
isync
@@ -897,14 +914,6 @@ load_up_mmu:
tophys(r6,r6)
lwz r6,_SDR1@l(r6)
mtspr SPRN_SDR1,r6
- li r0,16 /* load up segment register values */
- mtctr r0 /* for context 0 */
- lis r3,0x2000 /* Ku = 1, VSID = 0 */
- li r4,0
-3: mtsrin r3,r4
- addi r3,r3,0x111 /* increment VSID */
- addis r4,r4,0x1000 /* address of next segment */
- bdnz 3b
/* Load the BAT registers with the values set up by MMU_init.
MMU_init takes care of whether we're on a 601 or not. */
@@ -926,6 +935,17 @@ BEGIN_MMU_FTR_SECTION
END_MMU_FTR_SECTION_IFSET(MMU_FTR_USE_HIGH_BATS)
blr
+load_segment_registers:
+ li r0, 16 /* load up segment register values */
+ mtctr r0 /* for context 0 */
+ lis r3, 0x2000 /* Ku = 1, VSID = 0 */
+ li r4, 0
+3: mtsrin r3, r4
+ addi r3, r3, 0x111 /* increment VSID */
+ addis r4, r4, 0x1000 /* address of next segment */
+ bdnz 3b
+ blr
+
/*
* This is where the main kernel code starts.
*/
@@ -961,6 +981,9 @@ start_here:
bl __save_cpu_setup
bl MMU_init
BEGIN_MMU_FTR_SECTION
+#ifdef CONFIG_KASAN
+ bl kasan_mmu_init
+#endif
bl MMU_init_hw_patch
END_MMU_FTR_SECTION_IFSET(MMU_FTR_HPTE_TABLE)
diff --git a/arch/powerpc/mm/kasan/kasan_init_32.c b/arch/powerpc/mm/kasan/kasan_init_32.c
index cc788917ce38..f6dbc537c051 100644
--- a/arch/powerpc/mm/kasan/kasan_init_32.c
+++ b/arch/powerpc/mm/kasan/kasan_init_32.c
@@ -39,7 +39,10 @@ static int kasan_init_shadow_page_tables(unsigned long k_start, unsigned long k_
if (!new)
return -ENOMEM;
- kasan_populate_pte(new, PAGE_KERNEL_RO);
+ if (early_mmu_has_feature(MMU_FTR_HPTE_TABLE))
+ kasan_populate_pte(new, PAGE_READONLY);
+ else
+ kasan_populate_pte(new, PAGE_KERNEL_RO);
pmd_populate_kernel(&init_mm, pmd, new);
}
return 0;
@@ -60,10 +63,13 @@ static int __ref kasan_init_region(void *start, size_t size)
unsigned long k_cur;
pmd_t *pmd;
void *block = NULL;
- int ret = kasan_init_shadow_page_tables(k_start, k_end);
- if (ret)
- return ret;
+ if (!early_mmu_has_feature(MMU_FTR_HPTE_TABLE)) {
+ int ret = kasan_init_shadow_page_tables(k_start, k_end);
+
+ if (ret)
+ return ret;
+ }
if (!slab_is_available())
block = memblock_alloc(k_end - k_start, PAGE_SIZE);
@@ -84,15 +90,26 @@ static int __ref kasan_init_region(void *start, size_t size)
static void __init kasan_remap_early_shadow_ro(void)
{
- kasan_populate_pte(kasan_early_shadow_pte, PAGE_KERNEL_RO);
+ if (early_mmu_has_feature(MMU_FTR_HPTE_TABLE))
+ kasan_populate_pte(kasan_early_shadow_pte, PAGE_READONLY);
+ else
+ kasan_populate_pte(kasan_early_shadow_pte, PAGE_KERNEL_RO);
+
flush_tlb_kernel_range(KASAN_SHADOW_START, KASAN_SHADOW_END);
}
-void __init kasan_init(void)
+void __init kasan_mmu_init(void)
{
int ret;
struct memblock_region *reg;
+ if (early_mmu_has_feature(MMU_FTR_HPTE_TABLE)) {
+ ret = kasan_init_shadow_page_tables(KASAN_SHADOW_START, KASAN_SHADOW_END);
+
+ if (ret)
+ panic("kasan: kasan_init_shadow_page_tables() failed");
+ }
+
for_each_memblock(memory, reg) {
phys_addr_t base = reg->base;
phys_addr_t top = min(base + reg->size, total_lowmem);
@@ -104,6 +121,12 @@ void __init kasan_init(void)
if (ret)
panic("kasan: kasan_init_region() failed");
}
+}
+
+void __init kasan_init(void)
+{
+ if (!early_mmu_has_feature(MMU_FTR_HPTE_TABLE))
+ kasan_mmu_init();
kasan_remap_early_shadow_ro();
@@ -131,6 +154,20 @@ void *module_alloc(unsigned long size)
}
#endif
+#ifdef CONFIG_PPC_BOOK3S_32
+u8 __initdata early_hash[256 << 10] __aligned(256 << 10) = {0};
+
+static void __init kasan_early_hash_table(void)
+{
+ modify_instruction_site(&patch__hash_page_A0, 0xffff, __pa(early_hash) >> 16);
+ modify_instruction_site(&patch__flush_hash_A0, 0xffff, __pa(early_hash) >> 16);
+
+ Hash = (struct hash_pte *)early_hash;
+}
+#else
+static void __init kasan_early_hash_table(void) {}
+#endif
+
void __init kasan_early_init(void)
{
unsigned long addr = KASAN_SHADOW_START;
@@ -148,5 +185,5 @@ void __init kasan_early_init(void)
} while (pmd++, addr = next, addr != end);
if (early_mmu_has_feature(MMU_FTR_HPTE_TABLE))
- WARN(1, "KASAN not supported on hash 6xx");
+ kasan_early_hash_table();
}
diff --git a/arch/powerpc/mm/mmu_decl.h b/arch/powerpc/mm/mmu_decl.h
index d726ff776054..31fce3914ddc 100644
--- a/arch/powerpc/mm/mmu_decl.h
+++ b/arch/powerpc/mm/mmu_decl.h
@@ -106,6 +106,7 @@ extern unsigned int rtas_data, rtas_size;
struct hash_pte;
extern struct hash_pte *Hash, *Hash_end;
extern unsigned long Hash_size, Hash_mask;
+extern u8 early_hash[];
#endif /* CONFIG_PPC32 */
--
2.13.3
^ permalink raw reply related
* [PATCH v9 10/11] powerpc/32s: move hash code patching out of MMU_init_hw()
From: Christophe Leroy @ 2019-03-01 12:33 UTC (permalink / raw)
To: Benjamin Herrenschmidt, Paul Mackerras, Michael Ellerman,
Nicholas Piggin, Aneesh Kumar K.V, Andrey Ryabinin,
Alexander Potapenko, Dmitry Vyukov, Daniel Axtens
Cc: linux-mm, linuxppc-dev, linux-kernel, kasan-dev
In-Reply-To: <cover.1551443452.git.christophe.leroy@c-s.fr>
For KASAN, hash table handling will be activated early for
accessing to KASAN shadow areas.
In order to avoid any modification of the hash functions while
they are still used with the early hash table, the code patching
is moved out of MMU_init_hw() and put close to the big-bang switch
to the final hash table.
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
---
arch/powerpc/kernel/head_32.S | 3 +++
arch/powerpc/mm/mmu_decl.h | 1 +
arch/powerpc/mm/ppc_mmu_32.c | 36 ++++++++++++++++++++++--------------
3 files changed, 26 insertions(+), 14 deletions(-)
diff --git a/arch/powerpc/kernel/head_32.S b/arch/powerpc/kernel/head_32.S
index 02229c005853..e644aab2cf5b 100644
--- a/arch/powerpc/kernel/head_32.S
+++ b/arch/powerpc/kernel/head_32.S
@@ -960,6 +960,9 @@ start_here:
bl machine_init
bl __save_cpu_setup
bl MMU_init
+BEGIN_MMU_FTR_SECTION
+ bl MMU_init_hw_patch
+END_MMU_FTR_SECTION_IFSET(MMU_FTR_HPTE_TABLE)
/*
* Go back to running unmapped so we can load up new values
diff --git a/arch/powerpc/mm/mmu_decl.h b/arch/powerpc/mm/mmu_decl.h
index 74ff61dabcb1..d726ff776054 100644
--- a/arch/powerpc/mm/mmu_decl.h
+++ b/arch/powerpc/mm/mmu_decl.h
@@ -130,6 +130,7 @@ extern void wii_memory_fixups(void);
*/
#ifdef CONFIG_PPC32
extern void MMU_init_hw(void);
+void MMU_init_hw_patch(void);
unsigned long mmu_mapin_ram(unsigned long base, unsigned long top);
#endif
diff --git a/arch/powerpc/mm/ppc_mmu_32.c b/arch/powerpc/mm/ppc_mmu_32.c
index 2d5b0d50fb31..38c0e28c21e1 100644
--- a/arch/powerpc/mm/ppc_mmu_32.c
+++ b/arch/powerpc/mm/ppc_mmu_32.c
@@ -39,6 +39,7 @@
struct hash_pte *Hash, *Hash_end;
unsigned long Hash_size, Hash_mask;
unsigned long _SDR1;
+static unsigned int hash_mb, hash_mb2;
struct ppc_bat BATS[8][2]; /* 8 pairs of IBAT, DBAT */
@@ -308,7 +309,6 @@ void hash_preload(struct mm_struct *mm, unsigned long ea,
*/
void __init MMU_init_hw(void)
{
- unsigned int hmask, mb, mb2;
unsigned int n_hpteg, lg_n_hpteg;
if (!mmu_has_feature(MMU_FTR_HPTE_TABLE))
@@ -349,20 +349,30 @@ void __init MMU_init_hw(void)
(unsigned long long)(total_memory >> 20), Hash_size >> 10, Hash);
- /*
- * Patch up the instructions in hashtable.S:create_hpte
- */
- if ( ppc_md.progress ) ppc_md.progress("hash:patch", 0x345);
Hash_mask = n_hpteg - 1;
- hmask = Hash_mask >> (16 - LG_HPTEG_SIZE);
- mb2 = mb = 32 - LG_HPTEG_SIZE - lg_n_hpteg;
+ hash_mb2 = hash_mb = 32 - LG_HPTEG_SIZE - lg_n_hpteg;
if (lg_n_hpteg > 16)
- mb2 = 16 - LG_HPTEG_SIZE;
+ hash_mb2 = 16 - LG_HPTEG_SIZE;
+}
+
+void __init MMU_init_hw_patch(void)
+{
+ unsigned int hmask = Hash_mask >> (16 - LG_HPTEG_SIZE);
+ if (ppc_md.progress)
+ ppc_md.progress("hash:patch", 0x345);
+ if (ppc_md.progress)
+ ppc_md.progress("hash:done", 0x205);
+
+ /* WARNING: Make sure nothing can trigger a KASAN check past this point */
+
+ /*
+ * Patch up the instructions in hashtable.S:create_hpte
+ */
modify_instruction_site(&patch__hash_page_A0, 0xffff,
((unsigned int)Hash - PAGE_OFFSET) >> 16);
- modify_instruction_site(&patch__hash_page_A1, 0x7c0, mb << 6);
- modify_instruction_site(&patch__hash_page_A2, 0x7c0, mb2 << 6);
+ modify_instruction_site(&patch__hash_page_A1, 0x7c0, hash_mb << 6);
+ modify_instruction_site(&patch__hash_page_A2, 0x7c0, hash_mb2 << 6);
modify_instruction_site(&patch__hash_page_B, 0xffff, hmask);
modify_instruction_site(&patch__hash_page_C, 0xffff, hmask);
@@ -371,11 +381,9 @@ void __init MMU_init_hw(void)
*/
modify_instruction_site(&patch__flush_hash_A0, 0xffff,
((unsigned int)Hash - PAGE_OFFSET) >> 16);
- modify_instruction_site(&patch__flush_hash_A1, 0x7c0, mb << 6);
- modify_instruction_site(&patch__flush_hash_A2, 0x7c0, mb2 << 6);
+ modify_instruction_site(&patch__flush_hash_A1, 0x7c0, hash_mb << 6);
+ modify_instruction_site(&patch__flush_hash_A2, 0x7c0, hash_mb2 << 6);
modify_instruction_site(&patch__flush_hash_B, 0xffff, hmask);
-
- if ( ppc_md.progress ) ppc_md.progress("hash:done", 0x205);
}
void setup_initial_memory_limit(phys_addr_t first_memblock_base,
--
2.13.3
^ permalink raw reply related
* [PATCH v9 07/11] powerpc/32: prepare shadow area for KASAN
From: Christophe Leroy @ 2019-03-01 12:33 UTC (permalink / raw)
To: Benjamin Herrenschmidt, Paul Mackerras, Michael Ellerman,
Nicholas Piggin, Aneesh Kumar K.V, Andrey Ryabinin,
Alexander Potapenko, Dmitry Vyukov, Daniel Axtens
Cc: linux-mm, linuxppc-dev, linux-kernel, kasan-dev
In-Reply-To: <cover.1551443452.git.christophe.leroy@c-s.fr>
This patch prepares a shadow area for KASAN.
The shadow area will be at the top of the kernel virtual
memory space above the fixmap area and will occupy one
eighth of the total kernel virtual memory space.
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
---
arch/powerpc/Kconfig.debug | 5 +++++
arch/powerpc/include/asm/fixmap.h | 5 +++++
arch/powerpc/include/asm/kasan.h | 16 ++++++++++++++++
arch/powerpc/mm/mem.c | 4 ++++
arch/powerpc/mm/ptdump/ptdump.c | 8 ++++++++
5 files changed, 38 insertions(+)
diff --git a/arch/powerpc/Kconfig.debug b/arch/powerpc/Kconfig.debug
index 4e00cb0a5464..61febbbdd02b 100644
--- a/arch/powerpc/Kconfig.debug
+++ b/arch/powerpc/Kconfig.debug
@@ -366,3 +366,8 @@ config PPC_FAST_ENDIAN_SWITCH
depends on DEBUG_KERNEL && PPC_BOOK3S_64
help
If you're unsure what this is, say N.
+
+config KASAN_SHADOW_OFFSET
+ hex
+ depends on KASAN
+ default 0xe0000000
diff --git a/arch/powerpc/include/asm/fixmap.h b/arch/powerpc/include/asm/fixmap.h
index b9fbed84ddca..0cfc365d814b 100644
--- a/arch/powerpc/include/asm/fixmap.h
+++ b/arch/powerpc/include/asm/fixmap.h
@@ -22,7 +22,12 @@
#include <asm/kmap_types.h>
#endif
+#ifdef CONFIG_KASAN
+#include <asm/kasan.h>
+#define FIXADDR_TOP (KASAN_SHADOW_START - PAGE_SIZE)
+#else
#define FIXADDR_TOP ((unsigned long)(-PAGE_SIZE))
+#endif
/*
* Here we define all the compile-time 'special' virtual
diff --git a/arch/powerpc/include/asm/kasan.h b/arch/powerpc/include/asm/kasan.h
index c3161b8fc017..8dc1e3819171 100644
--- a/arch/powerpc/include/asm/kasan.h
+++ b/arch/powerpc/include/asm/kasan.h
@@ -12,4 +12,20 @@
#define EXPORT_SYMBOL_KASAN(fn) EXPORT_SYMBOL(fn)
#endif
+#ifndef __ASSEMBLY__
+
+#include <asm/page.h>
+
+#define KASAN_SHADOW_SCALE_SHIFT 3
+
+#define KASAN_SHADOW_OFFSET ASM_CONST(CONFIG_KASAN_SHADOW_OFFSET)
+
+#define KASAN_SHADOW_START (KASAN_SHADOW_OFFSET + \
+ (PAGE_OFFSET >> KASAN_SHADOW_SCALE_SHIFT))
+
+#define KASAN_SHADOW_END 0UL
+
+#define KASAN_SHADOW_SIZE (KASAN_SHADOW_END - KASAN_SHADOW_START)
+
+#endif /* __ASSEMBLY */
#endif
diff --git a/arch/powerpc/mm/mem.c b/arch/powerpc/mm/mem.c
index f6787f90e158..4e7fa4eb2dd3 100644
--- a/arch/powerpc/mm/mem.c
+++ b/arch/powerpc/mm/mem.c
@@ -309,6 +309,10 @@ void __init mem_init(void)
mem_init_print_info(NULL);
#ifdef CONFIG_PPC32
pr_info("Kernel virtual memory layout:\n");
+#ifdef CONFIG_KASAN
+ pr_info(" * 0x%08lx..0x%08lx : kasan shadow mem\n",
+ KASAN_SHADOW_START, KASAN_SHADOW_END);
+#endif
pr_info(" * 0x%08lx..0x%08lx : fixmap\n", FIXADDR_START, FIXADDR_TOP);
#ifdef CONFIG_HIGHMEM
pr_info(" * 0x%08lx..0x%08lx : highmem PTEs\n",
diff --git a/arch/powerpc/mm/ptdump/ptdump.c b/arch/powerpc/mm/ptdump/ptdump.c
index 37138428ab55..812ed680024f 100644
--- a/arch/powerpc/mm/ptdump/ptdump.c
+++ b/arch/powerpc/mm/ptdump/ptdump.c
@@ -101,6 +101,10 @@ static struct addr_marker address_markers[] = {
{ 0, "Fixmap start" },
{ 0, "Fixmap end" },
#endif
+#ifdef CONFIG_KASAN
+ { 0, "kasan shadow mem start" },
+ { 0, "kasan shadow mem end" },
+#endif
{ -1, NULL },
};
@@ -322,6 +326,10 @@ static void populate_markers(void)
#endif
address_markers[i++].start_address = FIXADDR_START;
address_markers[i++].start_address = FIXADDR_TOP;
+#ifdef CONFIG_KASAN
+ address_markers[i++].start_address = KASAN_SHADOW_START;
+ address_markers[i++].start_address = KASAN_SHADOW_END;
+#endif
#endif /* CONFIG_PPC64 */
}
--
2.13.3
^ permalink raw reply related
* [PATCH v9 09/11] powerpc/32: Add KASAN support
From: Christophe Leroy @ 2019-03-01 12:33 UTC (permalink / raw)
To: Benjamin Herrenschmidt, Paul Mackerras, Michael Ellerman,
Nicholas Piggin, Aneesh Kumar K.V, Andrey Ryabinin,
Alexander Potapenko, Dmitry Vyukov, Daniel Axtens
Cc: linux-mm, linuxppc-dev, linux-kernel, kasan-dev
In-Reply-To: <cover.1551443452.git.christophe.leroy@c-s.fr>
This patch adds KASAN support for PPC32. The following patch
will add an early activation of hash table for book3s. Until
then, a warning will be raised if trying to use KASAN on an
hash 6xx.
To support KASAN, this patch initialises that MMU mapings for
accessing to the KASAN shadow area defined in a previous patch.
An early mapping is set as soon as the kernel code has been
relocated at its definitive place.
Then the definitive mapping is set once paging is initialised.
For modules, the shadow area is allocated at module_alloc().
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
---
arch/powerpc/Kconfig | 1 +
arch/powerpc/include/asm/kasan.h | 7 ++
arch/powerpc/kernel/head_32.S | 3 +
arch/powerpc/kernel/head_40x.S | 3 +
arch/powerpc/kernel/head_44x.S | 3 +
arch/powerpc/kernel/head_8xx.S | 3 +
arch/powerpc/kernel/head_fsl_booke.S | 3 +
arch/powerpc/kernel/setup-common.c | 3 +
arch/powerpc/mm/Makefile | 1 +
arch/powerpc/mm/kasan/Makefile | 5 ++
arch/powerpc/mm/kasan/kasan_init_32.c | 152 ++++++++++++++++++++++++++++++++++
11 files changed, 184 insertions(+)
create mode 100644 arch/powerpc/mm/kasan/Makefile
create mode 100644 arch/powerpc/mm/kasan/kasan_init_32.c
diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
index 652c25260838..8d6108c83299 100644
--- a/arch/powerpc/Kconfig
+++ b/arch/powerpc/Kconfig
@@ -172,6 +172,7 @@ config PPC
select GENERIC_TIME_VSYSCALL
select HAVE_ARCH_AUDITSYSCALL
select HAVE_ARCH_JUMP_LABEL
+ select HAVE_ARCH_KASAN if PPC32
select HAVE_ARCH_KGDB
select HAVE_ARCH_MMAP_RND_BITS
select HAVE_ARCH_MMAP_RND_COMPAT_BITS if COMPAT
diff --git a/arch/powerpc/include/asm/kasan.h b/arch/powerpc/include/asm/kasan.h
index 8dc1e3819171..74a4ba9fb8a3 100644
--- a/arch/powerpc/include/asm/kasan.h
+++ b/arch/powerpc/include/asm/kasan.h
@@ -27,5 +27,12 @@
#define KASAN_SHADOW_SIZE (KASAN_SHADOW_END - KASAN_SHADOW_START)
+#ifdef CONFIG_KASAN
+void kasan_early_init(void);
+void kasan_init(void);
+#else
+static inline void kasan_init(void) { }
+#endif
+
#endif /* __ASSEMBLY */
#endif
diff --git a/arch/powerpc/kernel/head_32.S b/arch/powerpc/kernel/head_32.S
index ce6a972f2584..02229c005853 100644
--- a/arch/powerpc/kernel/head_32.S
+++ b/arch/powerpc/kernel/head_32.S
@@ -952,6 +952,9 @@ start_here:
* Do early platform-specific initialization,
* and set up the MMU.
*/
+#ifdef CONFIG_KASAN
+ bl kasan_early_init
+#endif
li r3,0
mr r4,r31
bl machine_init
diff --git a/arch/powerpc/kernel/head_40x.S b/arch/powerpc/kernel/head_40x.S
index a9c934f2319b..efa219d2136e 100644
--- a/arch/powerpc/kernel/head_40x.S
+++ b/arch/powerpc/kernel/head_40x.S
@@ -848,6 +848,9 @@ start_here:
/*
* Decide what sort of machine this is and initialize the MMU.
*/
+#ifdef CONFIG_KASAN
+ bl kasan_early_init
+#endif
li r3,0
mr r4,r31
bl machine_init
diff --git a/arch/powerpc/kernel/head_44x.S b/arch/powerpc/kernel/head_44x.S
index 37117ab11584..34a5df827b38 100644
--- a/arch/powerpc/kernel/head_44x.S
+++ b/arch/powerpc/kernel/head_44x.S
@@ -203,6 +203,9 @@ _ENTRY(_start);
/*
* Decide what sort of machine this is and initialize the MMU.
*/
+#ifdef CONFIG_KASAN
+ bl kasan_early_init
+#endif
li r3,0
mr r4,r31
bl machine_init
diff --git a/arch/powerpc/kernel/head_8xx.S b/arch/powerpc/kernel/head_8xx.S
index 03c73b4c6435..d25adb6ef235 100644
--- a/arch/powerpc/kernel/head_8xx.S
+++ b/arch/powerpc/kernel/head_8xx.S
@@ -853,6 +853,9 @@ start_here:
/*
* Decide what sort of machine this is and initialize the MMU.
*/
+#ifdef CONFIG_KASAN
+ bl kasan_early_init
+#endif
li r3,0
mr r4,r31
bl machine_init
diff --git a/arch/powerpc/kernel/head_fsl_booke.S b/arch/powerpc/kernel/head_fsl_booke.S
index 1881127682e9..0fc38eb957b7 100644
--- a/arch/powerpc/kernel/head_fsl_booke.S
+++ b/arch/powerpc/kernel/head_fsl_booke.S
@@ -275,6 +275,9 @@ set_ivor:
/*
* Decide what sort of machine this is and initialize the MMU.
*/
+#ifdef CONFIG_KASAN
+ bl kasan_early_init
+#endif
mr r3,r30
mr r4,r31
bl machine_init
diff --git a/arch/powerpc/kernel/setup-common.c b/arch/powerpc/kernel/setup-common.c
index e7534f306c8e..3c6c5a43901e 100644
--- a/arch/powerpc/kernel/setup-common.c
+++ b/arch/powerpc/kernel/setup-common.c
@@ -67,6 +67,7 @@
#include <asm/livepatch.h>
#include <asm/mmu_context.h>
#include <asm/cpu_has_feature.h>
+#include <asm/kasan.h>
#include "setup.h"
@@ -865,6 +866,8 @@ static void smp_setup_pacas(void)
*/
void __init setup_arch(char **cmdline_p)
{
+ kasan_init();
+
*cmdline_p = boot_command_line;
/* Set a half-reasonable default so udelay does something sensible */
diff --git a/arch/powerpc/mm/Makefile b/arch/powerpc/mm/Makefile
index 240d73dce6bb..80382a2d169b 100644
--- a/arch/powerpc/mm/Makefile
+++ b/arch/powerpc/mm/Makefile
@@ -53,6 +53,7 @@ obj-$(CONFIG_PPC_COPRO_BASE) += copro_fault.o
obj-$(CONFIG_SPAPR_TCE_IOMMU) += mmu_context_iommu.o
obj-$(CONFIG_PPC_PTDUMP) += ptdump/
obj-$(CONFIG_PPC_MEM_KEYS) += pkeys.o
+obj-$(CONFIG_KASAN) += kasan/
# Disable kcov instrumentation on sensitive code
# This is necessary for booting with kcov enabled on book3e machines
diff --git a/arch/powerpc/mm/kasan/Makefile b/arch/powerpc/mm/kasan/Makefile
new file mode 100644
index 000000000000..6577897673dd
--- /dev/null
+++ b/arch/powerpc/mm/kasan/Makefile
@@ -0,0 +1,5 @@
+# SPDX-License-Identifier: GPL-2.0
+
+KASAN_SANITIZE := n
+
+obj-$(CONFIG_PPC32) += kasan_init_32.o
diff --git a/arch/powerpc/mm/kasan/kasan_init_32.c b/arch/powerpc/mm/kasan/kasan_init_32.c
new file mode 100644
index 000000000000..cc788917ce38
--- /dev/null
+++ b/arch/powerpc/mm/kasan/kasan_init_32.c
@@ -0,0 +1,152 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#define DISABLE_BRANCH_PROFILING
+
+#include <linux/kasan.h>
+#include <linux/printk.h>
+#include <linux/memblock.h>
+#include <linux/sched/task.h>
+#include <linux/vmalloc.h>
+#include <asm/pgalloc.h>
+#include <asm/code-patching.h>
+#include <mm/mmu_decl.h>
+
+static void kasan_populate_pte(pte_t *ptep, pgprot_t prot)
+{
+ unsigned long va = (unsigned long)kasan_early_shadow_page;
+ phys_addr_t pa = __pa(kasan_early_shadow_page);
+ int i;
+
+ for (i = 0; i < PTRS_PER_PTE; i++, ptep++)
+ __set_pte_at(&init_mm, va, ptep, pfn_pte(PHYS_PFN(pa), prot), 0);
+}
+
+static int kasan_init_shadow_page_tables(unsigned long k_start, unsigned long k_end)
+{
+ pmd_t *pmd;
+ unsigned long k_cur, k_next;
+
+ pmd = pmd_offset(pud_offset(pgd_offset_k(k_start), k_start), k_start);
+
+ for (k_cur = k_start; k_cur != k_end; k_cur = k_next, pmd++) {
+ pte_t *new;
+
+ k_next = pgd_addr_end(k_cur, k_end);
+ if ((void *)pmd_page_vaddr(*pmd) != kasan_early_shadow_pte)
+ continue;
+
+ new = pte_alloc_one_kernel(&init_mm);
+
+ if (!new)
+ return -ENOMEM;
+ kasan_populate_pte(new, PAGE_KERNEL_RO);
+ pmd_populate_kernel(&init_mm, pmd, new);
+ }
+ return 0;
+}
+
+static void __ref *kasan_get_one_page(void)
+{
+ if (slab_is_available())
+ return (void *)__get_free_page(GFP_KERNEL | __GFP_ZERO);
+
+ return memblock_alloc(PAGE_SIZE, PAGE_SIZE);
+}
+
+static int __ref kasan_init_region(void *start, size_t size)
+{
+ unsigned long k_start = (unsigned long)kasan_mem_to_shadow(start);
+ unsigned long k_end = (unsigned long)kasan_mem_to_shadow(start + size);
+ unsigned long k_cur;
+ pmd_t *pmd;
+ void *block = NULL;
+ int ret = kasan_init_shadow_page_tables(k_start, k_end);
+
+ if (ret)
+ return ret;
+
+ if (!slab_is_available())
+ block = memblock_alloc(k_end - k_start, PAGE_SIZE);
+
+ for (k_cur = k_start; k_cur < k_end; k_cur += PAGE_SIZE) {
+ void *va = block ? block + k_cur - k_start : kasan_get_one_page();
+ pte_t pte = pfn_pte(PHYS_PFN(__pa(va)), PAGE_KERNEL);
+
+ if (!va)
+ return -ENOMEM;
+
+ pmd = pmd_offset(pud_offset(pgd_offset_k(k_cur), k_cur), k_cur);
+ __set_pte_at(&init_mm, k_cur, pte_offset_kernel(pmd, k_cur), pte, 0);
+ }
+ flush_tlb_kernel_range(k_start, k_end);
+ return 0;
+}
+
+static void __init kasan_remap_early_shadow_ro(void)
+{
+ kasan_populate_pte(kasan_early_shadow_pte, PAGE_KERNEL_RO);
+ flush_tlb_kernel_range(KASAN_SHADOW_START, KASAN_SHADOW_END);
+}
+
+void __init kasan_init(void)
+{
+ int ret;
+ struct memblock_region *reg;
+
+ for_each_memblock(memory, reg) {
+ phys_addr_t base = reg->base;
+ phys_addr_t top = min(base + reg->size, total_lowmem);
+
+ if (base >= top)
+ continue;
+
+ ret = kasan_init_region(__va(base), top - base);
+ if (ret)
+ panic("kasan: kasan_init_region() failed");
+ }
+
+ kasan_remap_early_shadow_ro();
+
+ clear_page(kasan_early_shadow_page);
+
+ /* At this point kasan is fully initialized. Enable error messages */
+ init_task.kasan_depth = 0;
+ pr_info("KASAN init done\n");
+}
+
+#ifdef CONFIG_MODULES
+void *module_alloc(unsigned long size)
+{
+ void *base = vmalloc_exec(size);
+
+ if (!base)
+ return NULL;
+
+ if (!kasan_init_region(base, size))
+ return base;
+
+ vfree(base);
+
+ return NULL;
+}
+#endif
+
+void __init kasan_early_init(void)
+{
+ unsigned long addr = KASAN_SHADOW_START;
+ unsigned long end = KASAN_SHADOW_END;
+ unsigned long next;
+ pmd_t *pmd = pmd_offset(pud_offset(pgd_offset_k(addr), addr), addr);
+
+ BUILD_BUG_ON(KASAN_SHADOW_START & ~PGDIR_MASK);
+
+ kasan_populate_pte(kasan_early_shadow_pte, PAGE_KERNEL);
+
+ do {
+ next = pgd_addr_end(addr, end);
+ pmd_populate_kernel(&init_mm, pmd, kasan_early_shadow_pte);
+ } while (pmd++, addr = next, addr != end);
+
+ if (early_mmu_has_feature(MMU_FTR_HPTE_TABLE))
+ WARN(1, "KASAN not supported on hash 6xx");
+}
--
2.13.3
^ permalink raw reply related
* [PATCH v9 08/11] powerpc: disable KASAN instrumentation on early/critical files.
From: Christophe Leroy @ 2019-03-01 12:33 UTC (permalink / raw)
To: Benjamin Herrenschmidt, Paul Mackerras, Michael Ellerman,
Nicholas Piggin, Aneesh Kumar K.V, Andrey Ryabinin,
Alexander Potapenko, Dmitry Vyukov, Daniel Axtens
Cc: linux-mm, linuxppc-dev, linux-kernel, kasan-dev
In-Reply-To: <cover.1551443452.git.christophe.leroy@c-s.fr>
All files containing functions run before kasan_early_init() is called
must have KASAN instrumentation disabled.
For those file, branch profiling also have to be disabled otherwise
each if () generates a call to ftrace_likely_update().
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
---
arch/powerpc/kernel/Makefile | 12 ++++++++++++
arch/powerpc/lib/Makefile | 8 ++++++++
arch/powerpc/mm/Makefile | 6 ++++++
arch/powerpc/platforms/powermac/Makefile | 6 ++++++
arch/powerpc/purgatory/Makefile | 3 +++
arch/powerpc/xmon/Makefile | 1 +
6 files changed, 36 insertions(+)
diff --git a/arch/powerpc/kernel/Makefile b/arch/powerpc/kernel/Makefile
index 45e47752b692..0ea6c4aa3a20 100644
--- a/arch/powerpc/kernel/Makefile
+++ b/arch/powerpc/kernel/Makefile
@@ -31,6 +31,18 @@ CFLAGS_REMOVE_btext.o = $(CC_FLAGS_FTRACE)
CFLAGS_REMOVE_prom.o = $(CC_FLAGS_FTRACE)
endif
+KASAN_SANITIZE_early_32.o := n
+KASAN_SANITIZE_cputable.o := n
+KASAN_SANITIZE_prom_init.o := n
+KASAN_SANITIZE_btext.o := n
+
+ifdef CONFIG_KASAN
+CFLAGS_early_32.o += -DDISABLE_BRANCH_PROFILING
+CFLAGS_cputable.o += -DDISABLE_BRANCH_PROFILING
+CFLAGS_prom_init.o += -DDISABLE_BRANCH_PROFILING
+CFLAGS_btext.o += -DDISABLE_BRANCH_PROFILING
+endif
+
obj-y := cputable.o ptrace.o syscalls.o \
irq.o align.o signal_32.o pmc.o vdso.o \
process.o systbl.o idle.o \
diff --git a/arch/powerpc/lib/Makefile b/arch/powerpc/lib/Makefile
index 47a4de434c22..c55f9c27bf79 100644
--- a/arch/powerpc/lib/Makefile
+++ b/arch/powerpc/lib/Makefile
@@ -8,6 +8,14 @@ ccflags-$(CONFIG_PPC64) := $(NO_MINIMAL_TOC)
CFLAGS_REMOVE_code-patching.o = $(CC_FLAGS_FTRACE)
CFLAGS_REMOVE_feature-fixups.o = $(CC_FLAGS_FTRACE)
+KASAN_SANITIZE_code-patching.o := n
+KASAN_SANITIZE_feature-fixups.o := n
+
+ifdef CONFIG_KASAN
+CFLAGS_code-patching.o += -DDISABLE_BRANCH_PROFILING
+CFLAGS_feature-fixups.o += -DDISABLE_BRANCH_PROFILING
+endif
+
obj-y += alloc.o code-patching.o feature-fixups.o
ifndef CONFIG_KASAN
diff --git a/arch/powerpc/mm/Makefile b/arch/powerpc/mm/Makefile
index d52ec118e09d..240d73dce6bb 100644
--- a/arch/powerpc/mm/Makefile
+++ b/arch/powerpc/mm/Makefile
@@ -7,6 +7,12 @@ ccflags-$(CONFIG_PPC64) := $(NO_MINIMAL_TOC)
CFLAGS_REMOVE_slb.o = $(CC_FLAGS_FTRACE)
+KASAN_SANITIZE_ppc_mmu_32.o := n
+
+ifdef CONFIG_KASAN
+CFLAGS_ppc_mmu_32.o += -DDISABLE_BRANCH_PROFILING
+endif
+
obj-y := fault.o mem.o pgtable.o mmap.o \
init_$(BITS).o pgtable_$(BITS).o \
init-common.o mmu_context.o drmem.o
diff --git a/arch/powerpc/platforms/powermac/Makefile b/arch/powerpc/platforms/powermac/Makefile
index 923bfb340433..859efbacdf06 100644
--- a/arch/powerpc/platforms/powermac/Makefile
+++ b/arch/powerpc/platforms/powermac/Makefile
@@ -2,6 +2,12 @@
CFLAGS_bootx_init.o += -fPIC
CFLAGS_bootx_init.o += $(call cc-option, -fno-stack-protector)
+KASAN_SANITIZE_bootx_init.o := n
+
+ifdef CONFIG_KASAN
+CFLAGS_bootx_init.o += -DDISABLE_BRANCH_PROFILING
+endif
+
ifdef CONFIG_FUNCTION_TRACER
# Do not trace early boot code
CFLAGS_REMOVE_bootx_init.o = $(CC_FLAGS_FTRACE)
diff --git a/arch/powerpc/purgatory/Makefile b/arch/powerpc/purgatory/Makefile
index 4314ba5baf43..7c6d8b14f440 100644
--- a/arch/powerpc/purgatory/Makefile
+++ b/arch/powerpc/purgatory/Makefile
@@ -1,4 +1,7 @@
# SPDX-License-Identifier: GPL-2.0
+
+KASAN_SANITIZE := n
+
targets += trampoline.o purgatory.ro kexec-purgatory.c
LDFLAGS_purgatory.ro := -e purgatory_start -r --no-undefined
diff --git a/arch/powerpc/xmon/Makefile b/arch/powerpc/xmon/Makefile
index 3050f9323254..f142570ad860 100644
--- a/arch/powerpc/xmon/Makefile
+++ b/arch/powerpc/xmon/Makefile
@@ -7,6 +7,7 @@ subdir-ccflags-y := $(call cc-disable-warning, builtin-requires-header)
GCOV_PROFILE := n
KCOV_INSTRUMENT := n
UBSAN_SANITIZE := n
+KASAN_SANITIZE := n
# Disable ftrace for the entire directory
ORIG_CFLAGS := $(KBUILD_CFLAGS)
--
2.13.3
^ permalink raw reply related
* [PATCH v9 06/11] powerpc/32: make KVIRT_TOP dependent on FIXMAP_START
From: Christophe Leroy @ 2019-03-01 12:33 UTC (permalink / raw)
To: Benjamin Herrenschmidt, Paul Mackerras, Michael Ellerman,
Nicholas Piggin, Aneesh Kumar K.V, Andrey Ryabinin,
Alexander Potapenko, Dmitry Vyukov, Daniel Axtens
Cc: linux-mm, linuxppc-dev, linux-kernel, kasan-dev
In-Reply-To: <cover.1551443452.git.christophe.leroy@c-s.fr>
When we add KASAN shadow area, KVIRT_TOP can't be anymore fixed
at 0xfe000000.
This patch uses FIXADDR_START to define KVIRT_TOP.
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
---
arch/powerpc/include/asm/book3s/32/pgtable.h | 13 ++++++++++---
arch/powerpc/include/asm/nohash/32/pgtable.h | 13 ++++++++++---
2 files changed, 20 insertions(+), 6 deletions(-)
diff --git a/arch/powerpc/include/asm/book3s/32/pgtable.h b/arch/powerpc/include/asm/book3s/32/pgtable.h
index aa8406b8f7ba..838de59f6754 100644
--- a/arch/powerpc/include/asm/book3s/32/pgtable.h
+++ b/arch/powerpc/include/asm/book3s/32/pgtable.h
@@ -134,15 +134,24 @@ static inline bool pte_user(pte_t pte)
#define PGDIR_MASK (~(PGDIR_SIZE-1))
#define USER_PTRS_PER_PGD (TASK_SIZE / PGDIR_SIZE)
+
+#ifndef __ASSEMBLY__
+
+int map_kernel_page(unsigned long va, phys_addr_t pa, pgprot_t prot);
+
+#endif /* !__ASSEMBLY__ */
+
/*
* This is the bottom of the PKMAP area with HIGHMEM or an arbitrary
* value (for now) on others, from where we can start layout kernel
* virtual space that goes below PKMAP and FIXMAP
*/
+#include <asm/fixmap.h>
+
#ifdef CONFIG_HIGHMEM
#define KVIRT_TOP PKMAP_BASE
#else
-#define KVIRT_TOP (0xfe000000UL) /* for now, could be FIXMAP_BASE ? */
+#define KVIRT_TOP FIXADDR_START
#endif
/*
@@ -373,8 +382,6 @@ static inline void __ptep_set_access_flags(struct vm_area_struct *vma,
#define __pte_to_swp_entry(pte) ((swp_entry_t) { pte_val(pte) >> 3 })
#define __swp_entry_to_pte(x) ((pte_t) { (x).val << 3 })
-int map_kernel_page(unsigned long va, phys_addr_t pa, pgprot_t prot);
-
/* Generic accessors to PTE bits */
static inline int pte_write(pte_t pte) { return !!(pte_val(pte) & _PAGE_RW);}
static inline int pte_read(pte_t pte) { return 1; }
diff --git a/arch/powerpc/include/asm/nohash/32/pgtable.h b/arch/powerpc/include/asm/nohash/32/pgtable.h
index bed433358260..0284f8f5305f 100644
--- a/arch/powerpc/include/asm/nohash/32/pgtable.h
+++ b/arch/powerpc/include/asm/nohash/32/pgtable.h
@@ -64,15 +64,24 @@ extern int icache_44x_need_flush;
#define pgd_ERROR(e) \
pr_err("%s:%d: bad pgd %08lx.\n", __FILE__, __LINE__, pgd_val(e))
+#ifndef __ASSEMBLY__
+
+int map_kernel_page(unsigned long va, phys_addr_t pa, pgprot_t prot);
+
+#endif /* !__ASSEMBLY__ */
+
+
/*
* This is the bottom of the PKMAP area with HIGHMEM or an arbitrary
* value (for now) on others, from where we can start layout kernel
* virtual space that goes below PKMAP and FIXMAP
*/
+#include <asm/fixmap.h>
+
#ifdef CONFIG_HIGHMEM
#define KVIRT_TOP PKMAP_BASE
#else
-#define KVIRT_TOP (0xfe000000UL) /* for now, could be FIXMAP_BASE ? */
+#define KVIRT_TOP FIXADDR_START
#endif
/*
@@ -379,8 +388,6 @@ static inline int pte_young(pte_t pte)
#define __pte_to_swp_entry(pte) ((swp_entry_t) { pte_val(pte) >> 3 })
#define __swp_entry_to_pte(x) ((pte_t) { (x).val << 3 })
-int map_kernel_page(unsigned long va, phys_addr_t pa, pgprot_t prot);
-
#endif /* !__ASSEMBLY__ */
#endif /* __ASM_POWERPC_NOHASH_32_PGTABLE_H */
--
2.13.3
^ permalink raw reply related
* [PATCH v9 05/11] powerpc/32: use memset() instead of memset_io() to zero BSS
From: Christophe Leroy @ 2019-03-01 12:33 UTC (permalink / raw)
To: Benjamin Herrenschmidt, Paul Mackerras, Michael Ellerman,
Nicholas Piggin, Aneesh Kumar K.V, Andrey Ryabinin,
Alexander Potapenko, Dmitry Vyukov, Daniel Axtens
Cc: linux-mm, linuxppc-dev, linux-kernel, kasan-dev
In-Reply-To: <cover.1551443452.git.christophe.leroy@c-s.fr>
Since commit 400c47d81ca38 ("powerpc32: memset: only use dcbz once cache is
enabled"), memset() can be used before activation of the cache,
so no need to use memset_io() for zeroing the BSS.
Acked-by: Dmitry Vyukov <dvyukov@google.com>
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
---
arch/powerpc/kernel/early_32.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/arch/powerpc/kernel/early_32.c b/arch/powerpc/kernel/early_32.c
index cf3cdd81dc47..3482118ffe76 100644
--- a/arch/powerpc/kernel/early_32.c
+++ b/arch/powerpc/kernel/early_32.c
@@ -21,8 +21,8 @@ notrace unsigned long __init early_init(unsigned long dt_ptr)
{
unsigned long offset = reloc_offset();
- /* First zero the BSS -- use memset_io, some platforms don't have caches on yet */
- memset_io((void __iomem *)PTRRELOC(&__bss_start), 0, __bss_stop - __bss_start);
+ /* First zero the BSS */
+ memset(PTRRELOC(&__bss_start), 0, __bss_stop - __bss_start);
/*
* Identify the CPU type and fix up code sections
--
2.13.3
^ permalink raw reply related
* [PATCH v9 01/11] powerpc/32: Move early_init() in a separate file
From: Christophe Leroy @ 2019-03-01 12:33 UTC (permalink / raw)
To: Benjamin Herrenschmidt, Paul Mackerras, Michael Ellerman,
Nicholas Piggin, Aneesh Kumar K.V, Andrey Ryabinin,
Alexander Potapenko, Dmitry Vyukov, Daniel Axtens
Cc: linux-mm, linuxppc-dev, linux-kernel, kasan-dev
In-Reply-To: <cover.1551443452.git.christophe.leroy@c-s.fr>
In preparation of KASAN, move early_init() into a separate
file in order to allow deactivation of KASAN for that function.
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
---
arch/powerpc/kernel/Makefile | 2 +-
arch/powerpc/kernel/early_32.c | 36 ++++++++++++++++++++++++++++++++++++
arch/powerpc/kernel/setup_32.c | 28 ----------------------------
3 files changed, 37 insertions(+), 29 deletions(-)
create mode 100644 arch/powerpc/kernel/early_32.c
diff --git a/arch/powerpc/kernel/Makefile b/arch/powerpc/kernel/Makefile
index cddadccf551d..45e47752b692 100644
--- a/arch/powerpc/kernel/Makefile
+++ b/arch/powerpc/kernel/Makefile
@@ -93,7 +93,7 @@ extra-y += vmlinux.lds
obj-$(CONFIG_RELOCATABLE) += reloc_$(BITS).o
-obj-$(CONFIG_PPC32) += entry_32.o setup_32.o
+obj-$(CONFIG_PPC32) += entry_32.o setup_32.o early_32.o
obj-$(CONFIG_PPC64) += dma-iommu.o iommu.o
obj-$(CONFIG_KGDB) += kgdb.o
obj-$(CONFIG_BOOTX_TEXT) += btext.o
diff --git a/arch/powerpc/kernel/early_32.c b/arch/powerpc/kernel/early_32.c
new file mode 100644
index 000000000000..cf3cdd81dc47
--- /dev/null
+++ b/arch/powerpc/kernel/early_32.c
@@ -0,0 +1,36 @@
+// SPDX-License-Identifier: GPL-2.0
+
+/*
+ * Early init before relocation
+ */
+
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <asm/setup.h>
+#include <asm/sections.h>
+#include <asm/asm-prototypes.h>
+
+/*
+ * We're called here very early in the boot.
+ *
+ * Note that the kernel may be running at an address which is different
+ * from the address that it was linked at, so we must use RELOC/PTRRELOC
+ * to access static data (including strings). -- paulus
+ */
+notrace unsigned long __init early_init(unsigned long dt_ptr)
+{
+ unsigned long offset = reloc_offset();
+
+ /* First zero the BSS -- use memset_io, some platforms don't have caches on yet */
+ memset_io((void __iomem *)PTRRELOC(&__bss_start), 0, __bss_stop - __bss_start);
+
+ /*
+ * Identify the CPU type and fix up code sections
+ * that depend on which cpu we have.
+ */
+ identify_cpu(offset, mfspr(SPRN_PVR));
+
+ apply_feature_fixups();
+
+ return KERNELBASE + offset;
+}
diff --git a/arch/powerpc/kernel/setup_32.c b/arch/powerpc/kernel/setup_32.c
index 1f0b7629c1a6..3f0d51b4b2f5 100644
--- a/arch/powerpc/kernel/setup_32.c
+++ b/arch/powerpc/kernel/setup_32.c
@@ -63,34 +63,6 @@ EXPORT_SYMBOL(DMA_MODE_READ);
EXPORT_SYMBOL(DMA_MODE_WRITE);
/*
- * We're called here very early in the boot.
- *
- * Note that the kernel may be running at an address which is different
- * from the address that it was linked at, so we must use RELOC/PTRRELOC
- * to access static data (including strings). -- paulus
- */
-notrace unsigned long __init early_init(unsigned long dt_ptr)
-{
- unsigned long offset = reloc_offset();
-
- /* First zero the BSS -- use memset_io, some platforms don't have
- * caches on yet */
- memset_io((void __iomem *)PTRRELOC(&__bss_start), 0,
- __bss_stop - __bss_start);
-
- /*
- * Identify the CPU type and fix up code sections
- * that depend on which cpu we have.
- */
- identify_cpu(offset, mfspr(SPRN_PVR));
-
- apply_feature_fixups();
-
- return KERNELBASE + offset;
-}
-
-
-/*
* This is run before start_kernel(), the kernel has been relocated
* and we are running with enough of the MMU enabled to have our
* proper kernel virtual addresses
--
2.13.3
^ permalink raw reply related
* [PATCH v9 00/11] KASAN for powerpc/32
From: Christophe Leroy @ 2019-03-01 12:33 UTC (permalink / raw)
To: Benjamin Herrenschmidt, Paul Mackerras, Michael Ellerman,
Nicholas Piggin, Aneesh Kumar K.V, Andrey Ryabinin,
Alexander Potapenko, Dmitry Vyukov, Daniel Axtens
Cc: linux-mm, linuxppc-dev, linux-kernel, kasan-dev
This series adds KASAN support to powerpc/32
Tested on nohash/32 (8xx) and book3s/32 (mpc832x ie 603).
Boot tested on qemu mac99
Changes in v9:
- Fixed fixmap IMMR alignment issue on 8xx with KASAN enabled.
- Set up final shadow page tables before switching to the final hash table on hash32
- Using PAGE_READONLY instead of PAGE_KERNEL_RO on hash32
- Use flash_tlb_kernel_range() instead of flash_tlb_mm() which doesn't work for kernel on some subarches.
- use __set_pte_at() instead of pte_update() to install final page tables
Changes in v8:
- Fixed circular issue between pgtable.h and fixmap.h
- Added missing includes in ppc64 string files
- Fixed kasan string related macro names for ppc64.
- Fixed most checkpatch messages
- build tested on kisskb (http://kisskb.ellerman.id.au/kisskb/head/6e65827de2fe71d21682dafd9084ed2cc6e06d4f/)
- moved CONFIG_KASAN_SHADOW_OFFSET in Kconfig.debug
Changes in v7:
- split in several smaller patches
- prom_init now has its own string functions
- full deactivation of powerpc-optimised string functions when KASAN is active
- shadow area now at a fixed place on very top of kernel virtual space.
- Early static hash table for hash book3s/32.
- Full support of both inline and outline instrumentation for both hash and nohash ppc32
- Earlier full activation of kasan.
Changes in v6:
- Fixed oops on module loading (due to access to RO shadow zero area).
- Added support for hash book3s/32, thanks to Daniel's patch to differ KASAN activation.
- Reworked handling of optimised string functions (dedicated patch for it)
- Reordered some files to ease adding of book3e/64 support.
Changes in v5:
- Added KASAN_SHADOW_OFFSET in Makefile, otherwise we fallback to KASAN_MINIMAL
and some stuff like stack instrumentation is not performed
- Moved calls to kasan_early_init() in head.S because stack instrumentation
in machine_init was performed before the call to kasan_early_init()
- Mapping kasan_early_shadow_page RW in kasan_early_init() and
remaping RO later in kasan_init()
- Allocating a big memblock() for shadow area, falling back to PAGE_SIZE blocks in case of failure.
Changes in v4:
- Comments from Andrey (DISABLE_BRANCH_PROFILING, Activation of reports)
- Proper initialisation of shadow area in kasan_init()
- Panic in case Hash table is required.
- Added comments in patch one to explain why *t = *s becomes memcpy(t, s, ...)
- Call of kasan_init_tags()
Changes in v3:
- Removed the printk() in kasan_early_init() to avoid build failure (see https://github.com/linuxppc/issues/issues/218)
- Added necessary changes in asm/book3s/32/pgtable.h to get it work on powerpc 603 family
- Added a few KASAN_SANITIZE_xxx.o := n to successfully boot on powerpc 603 family
Changes in v2:
- Rebased.
- Using __set_pte_at() to build the early table.
- Worked around and got rid of the patch adding asm/page.h in asm/pgtable-types.h
==> might be fixed independently but not needed for this serie.
Christophe Leroy (11):
powerpc/32: Move early_init() in a separate file
powerpc: prepare string/mem functions for KASAN
powerpc/prom_init: don't use string functions from lib/
powerpc/mm: don't use direct assignation during early boot.
powerpc/32: use memset() instead of memset_io() to zero BSS
powerpc/32: make KVIRT_TOP dependent on FIXMAP_START
powerpc/32: prepare shadow area for KASAN
powerpc: disable KASAN instrumentation on early/critical files.
powerpc/32: Add KASAN support
powerpc/32s: move hash code patching out of MMU_init_hw()
powerpc/32s: set up an early static hash table for KASAN.
arch/powerpc/Kconfig | 1 +
arch/powerpc/Kconfig.debug | 5 +
arch/powerpc/include/asm/book3s/32/pgtable.h | 13 +-
arch/powerpc/include/asm/fixmap.h | 5 +
arch/powerpc/include/asm/kasan.h | 39 +++++
arch/powerpc/include/asm/nohash/32/pgtable.h | 13 +-
arch/powerpc/include/asm/string.h | 32 +++-
arch/powerpc/kernel/Makefile | 14 +-
arch/powerpc/kernel/cputable.c | 13 +-
arch/powerpc/kernel/early_32.c | 36 +++++
arch/powerpc/kernel/head_32.S | 49 ++++--
arch/powerpc/kernel/head_40x.S | 3 +
arch/powerpc/kernel/head_44x.S | 3 +
arch/powerpc/kernel/head_8xx.S | 3 +
arch/powerpc/kernel/head_fsl_booke.S | 3 +
arch/powerpc/kernel/prom_init.c | 213 +++++++++++++++++++++------
arch/powerpc/kernel/prom_init_check.sh | 12 +-
arch/powerpc/kernel/setup-common.c | 3 +
arch/powerpc/kernel/setup_32.c | 28 ----
arch/powerpc/lib/Makefile | 19 ++-
arch/powerpc/lib/copy_32.S | 15 +-
arch/powerpc/lib/mem_64.S | 11 +-
arch/powerpc/lib/memcpy_64.S | 5 +-
arch/powerpc/mm/Makefile | 7 +
arch/powerpc/mm/kasan/Makefile | 5 +
arch/powerpc/mm/kasan/kasan_init_32.c | 189 ++++++++++++++++++++++++
arch/powerpc/mm/mem.c | 4 +
arch/powerpc/mm/mmu_decl.h | 2 +
arch/powerpc/mm/ppc_mmu_32.c | 36 +++--
arch/powerpc/mm/ptdump/ptdump.c | 8 +
arch/powerpc/platforms/powermac/Makefile | 6 +
arch/powerpc/purgatory/Makefile | 3 +
arch/powerpc/xmon/Makefile | 1 +
33 files changed, 676 insertions(+), 123 deletions(-)
create mode 100644 arch/powerpc/include/asm/kasan.h
create mode 100644 arch/powerpc/kernel/early_32.c
create mode 100644 arch/powerpc/mm/kasan/Makefile
create mode 100644 arch/powerpc/mm/kasan/kasan_init_32.c
--
2.13.3
^ permalink raw reply
* [PATCH v9 03/11] powerpc/prom_init: don't use string functions from lib/
From: Christophe Leroy @ 2019-03-01 12:33 UTC (permalink / raw)
To: Benjamin Herrenschmidt, Paul Mackerras, Michael Ellerman,
Nicholas Piggin, Aneesh Kumar K.V, Andrey Ryabinin,
Alexander Potapenko, Dmitry Vyukov, Daniel Axtens
Cc: linux-mm, linuxppc-dev, linux-kernel, kasan-dev
In-Reply-To: <cover.1551443452.git.christophe.leroy@c-s.fr>
When KASAN is active, the string functions in lib/ are doing the
KASAN checks. This is too early for prom_init.
This patch implements dedicated string functions for prom_init,
which will be compiled in with KASAN disabled.
Size of prom_init before the patch:
text data bss dec hex filename
12060 488 6960 19508 4c34 arch/powerpc/kernel/prom_init.o
Size of prom_init after the patch:
text data bss dec hex filename
12460 488 6960 19908 4dc4 arch/powerpc/kernel/prom_init.o
This increases the size of prom_init a bit, but as prom_init is
in __init section, it is freed after boot anyway.
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
---
arch/powerpc/kernel/prom_init.c | 213 ++++++++++++++++++++++++++-------
arch/powerpc/kernel/prom_init_check.sh | 2 +-
2 files changed, 173 insertions(+), 42 deletions(-)
diff --git a/arch/powerpc/kernel/prom_init.c b/arch/powerpc/kernel/prom_init.c
index f33ff4163a51..56d5639ab082 100644
--- a/arch/powerpc/kernel/prom_init.c
+++ b/arch/powerpc/kernel/prom_init.c
@@ -224,6 +224,137 @@ static bool __prombss rtas_has_query_cpu_stopped;
#define PHANDLE_VALID(p) ((p) != 0 && (p) != PROM_ERROR)
#define IHANDLE_VALID(i) ((i) != 0 && (i) != PROM_ERROR)
+/* Copied from lib/string.c and lib/kstrtox.c */
+
+static int __init prom_strcmp(const char *cs, const char *ct)
+{
+ unsigned char c1, c2;
+
+ while (1) {
+ c1 = *cs++;
+ c2 = *ct++;
+ if (c1 != c2)
+ return c1 < c2 ? -1 : 1;
+ if (!c1)
+ break;
+ }
+ return 0;
+}
+
+static char __init *prom_strcpy(char *dest, const char *src)
+{
+ char *tmp = dest;
+
+ while ((*dest++ = *src++) != '\0')
+ /* nothing */;
+ return tmp;
+}
+
+static int __init prom_strncmp(const char *cs, const char *ct, size_t count)
+{
+ unsigned char c1, c2;
+
+ while (count) {
+ c1 = *cs++;
+ c2 = *ct++;
+ if (c1 != c2)
+ return c1 < c2 ? -1 : 1;
+ if (!c1)
+ break;
+ count--;
+ }
+ return 0;
+}
+
+static size_t __init prom_strlen(const char *s)
+{
+ const char *sc;
+
+ for (sc = s; *sc != '\0'; ++sc)
+ /* nothing */;
+ return sc - s;
+}
+
+static int __init prom_memcmp(const void *cs, const void *ct, size_t count)
+{
+ const unsigned char *su1, *su2;
+ int res = 0;
+
+ for (su1 = cs, su2 = ct; 0 < count; ++su1, ++su2, count--)
+ if ((res = *su1 - *su2) != 0)
+ break;
+ return res;
+}
+
+static char __init *prom_strstr(const char *s1, const char *s2)
+{
+ size_t l1, l2;
+
+ l2 = prom_strlen(s2);
+ if (!l2)
+ return (char *)s1;
+ l1 = prom_strlen(s1);
+ while (l1 >= l2) {
+ l1--;
+ if (!prom_memcmp(s1, s2, l2))
+ return (char *)s1;
+ s1++;
+ }
+ return NULL;
+}
+
+#ifdef CONFIG_CMDLINE
+static size_t __init prom_strlcpy(char *dest, const char *src, size_t size)
+{
+ size_t ret = prom_strlen(src);
+
+ if (size) {
+ size_t len = (ret >= size) ? size - 1 : ret;
+ memcpy(dest, src, len);
+ dest[len] = '\0';
+ }
+ return ret;
+}
+#endif
+
+#ifdef CONFIG_PPC_PSERIES
+static int __init prom_strtobool(const char *s, bool *res)
+{
+ if (!s)
+ return -EINVAL;
+
+ switch (s[0]) {
+ case 'y':
+ case 'Y':
+ case '1':
+ *res = true;
+ return 0;
+ case 'n':
+ case 'N':
+ case '0':
+ *res = false;
+ return 0;
+ case 'o':
+ case 'O':
+ switch (s[1]) {
+ case 'n':
+ case 'N':
+ *res = true;
+ return 0;
+ case 'f':
+ case 'F':
+ *res = false;
+ return 0;
+ default:
+ break;
+ }
+ default:
+ break;
+ }
+
+ return -EINVAL;
+}
+#endif
/* This is the one and *ONLY* place where we actually call open
* firmware.
@@ -555,7 +686,7 @@ static int __init prom_setprop(phandle node, const char *nodename,
add_string(&p, tohex((u32)(unsigned long) value));
add_string(&p, tohex(valuelen));
add_string(&p, tohex(ADDR(pname)));
- add_string(&p, tohex(strlen(pname)));
+ add_string(&p, tohex(prom_strlen(pname)));
add_string(&p, "property");
*p = 0;
return call_prom("interpret", 1, 1, (u32)(unsigned long) cmd);
@@ -639,25 +770,25 @@ static void __init early_cmdline_parse(void)
l = prom_getprop(prom.chosen, "bootargs", p, COMMAND_LINE_SIZE-1);
#ifdef CONFIG_CMDLINE
if (l <= 0 || p[0] == '\0') /* dbl check */
- strlcpy(prom_cmd_line,
+ prom_strlcpy(prom_cmd_line,
CONFIG_CMDLINE, sizeof(prom_cmd_line));
#endif /* CONFIG_CMDLINE */
prom_printf("command line: %s\n", prom_cmd_line);
#ifdef CONFIG_PPC64
- opt = strstr(prom_cmd_line, "iommu=");
+ opt = prom_strstr(prom_cmd_line, "iommu=");
if (opt) {
prom_printf("iommu opt is: %s\n", opt);
opt += 6;
while (*opt && *opt == ' ')
opt++;
- if (!strncmp(opt, "off", 3))
+ if (!prom_strncmp(opt, "off", 3))
prom_iommu_off = 1;
- else if (!strncmp(opt, "force", 5))
+ else if (!prom_strncmp(opt, "force", 5))
prom_iommu_force_on = 1;
}
#endif
- opt = strstr(prom_cmd_line, "mem=");
+ opt = prom_strstr(prom_cmd_line, "mem=");
if (opt) {
opt += 4;
prom_memory_limit = prom_memparse(opt, (const char **)&opt);
@@ -669,13 +800,13 @@ static void __init early_cmdline_parse(void)
#ifdef CONFIG_PPC_PSERIES
prom_radix_disable = !IS_ENABLED(CONFIG_PPC_RADIX_MMU_DEFAULT);
- opt = strstr(prom_cmd_line, "disable_radix");
+ opt = prom_strstr(prom_cmd_line, "disable_radix");
if (opt) {
opt += 13;
if (*opt && *opt == '=') {
bool val;
- if (kstrtobool(++opt, &val))
+ if (prom_strtobool(++opt, &val))
prom_radix_disable = false;
else
prom_radix_disable = val;
@@ -1028,7 +1159,7 @@ static int __init prom_count_smt_threads(void)
type[0] = 0;
prom_getprop(node, "device_type", type, sizeof(type));
- if (strcmp(type, "cpu"))
+ if (prom_strcmp(type, "cpu"))
continue;
/*
* There is an entry for each smt thread, each entry being
@@ -1475,7 +1606,7 @@ static void __init prom_init_mem(void)
*/
prom_getprop(node, "name", type, sizeof(type));
}
- if (strcmp(type, "memory"))
+ if (prom_strcmp(type, "memory"))
continue;
plen = prom_getprop(node, "reg", regbuf, sizeof(regbuf));
@@ -1756,19 +1887,19 @@ static void __init prom_initialize_tce_table(void)
prom_getprop(node, "device_type", type, sizeof(type));
prom_getprop(node, "model", model, sizeof(model));
- if ((type[0] == 0) || (strstr(type, "pci") == NULL))
+ if ((type[0] == 0) || (prom_strstr(type, "pci") == NULL))
continue;
/* Keep the old logic intact to avoid regression. */
if (compatible[0] != 0) {
- if ((strstr(compatible, "python") == NULL) &&
- (strstr(compatible, "Speedwagon") == NULL) &&
- (strstr(compatible, "Winnipeg") == NULL))
+ if ((prom_strstr(compatible, "python") == NULL) &&
+ (prom_strstr(compatible, "Speedwagon") == NULL) &&
+ (prom_strstr(compatible, "Winnipeg") == NULL))
continue;
} else if (model[0] != 0) {
- if ((strstr(model, "ython") == NULL) &&
- (strstr(model, "peedwagon") == NULL) &&
- (strstr(model, "innipeg") == NULL))
+ if ((prom_strstr(model, "ython") == NULL) &&
+ (prom_strstr(model, "peedwagon") == NULL) &&
+ (prom_strstr(model, "innipeg") == NULL))
continue;
}
@@ -1917,12 +2048,12 @@ static void __init prom_hold_cpus(void)
type[0] = 0;
prom_getprop(node, "device_type", type, sizeof(type));
- if (strcmp(type, "cpu") != 0)
+ if (prom_strcmp(type, "cpu") != 0)
continue;
/* Skip non-configured cpus. */
if (prom_getprop(node, "status", type, sizeof(type)) > 0)
- if (strcmp(type, "okay") != 0)
+ if (prom_strcmp(type, "okay") != 0)
continue;
reg = cpu_to_be32(-1); /* make sparse happy */
@@ -1998,9 +2129,9 @@ static void __init prom_find_mmu(void)
return;
version[sizeof(version) - 1] = 0;
/* XXX might need to add other versions here */
- if (strcmp(version, "Open Firmware, 1.0.5") == 0)
+ if (prom_strcmp(version, "Open Firmware, 1.0.5") == 0)
of_workarounds = OF_WA_CLAIM;
- else if (strncmp(version, "FirmWorks,3.", 12) == 0) {
+ else if (prom_strncmp(version, "FirmWorks,3.", 12) == 0) {
of_workarounds = OF_WA_CLAIM | OF_WA_LONGTRAIL;
call_prom("interpret", 1, 1, "dev /memory 0 to allow-reclaim");
} else
@@ -2033,7 +2164,7 @@ static void __init prom_init_stdout(void)
call_prom("instance-to-path", 3, 1, prom.stdout, path, 255);
prom_printf("OF stdout device is: %s\n", of_stdout_device);
prom_setprop(prom.chosen, "/chosen", "linux,stdout-path",
- path, strlen(path) + 1);
+ path, prom_strlen(path) + 1);
/* instance-to-package fails on PA-Semi */
stdout_node = call_prom("instance-to-package", 1, 1, prom.stdout);
@@ -2043,7 +2174,7 @@ static void __init prom_init_stdout(void)
/* If it's a display, note it */
memset(type, 0, sizeof(type));
prom_getprop(stdout_node, "device_type", type, sizeof(type));
- if (strcmp(type, "display") == 0)
+ if (prom_strcmp(type, "display") == 0)
prom_setprop(stdout_node, path, "linux,boot-display", NULL, 0);
}
}
@@ -2064,19 +2195,19 @@ static int __init prom_find_machine_type(void)
compat[len] = 0;
while (i < len) {
char *p = &compat[i];
- int sl = strlen(p);
+ int sl = prom_strlen(p);
if (sl == 0)
break;
- if (strstr(p, "Power Macintosh") ||
- strstr(p, "MacRISC"))
+ if (prom_strstr(p, "Power Macintosh") ||
+ prom_strstr(p, "MacRISC"))
return PLATFORM_POWERMAC;
#ifdef CONFIG_PPC64
/* We must make sure we don't detect the IBM Cell
* blades as pSeries due to some firmware issues,
* so we do it here.
*/
- if (strstr(p, "IBM,CBEA") ||
- strstr(p, "IBM,CPBW-1.0"))
+ if (prom_strstr(p, "IBM,CBEA") ||
+ prom_strstr(p, "IBM,CPBW-1.0"))
return PLATFORM_GENERIC;
#endif /* CONFIG_PPC64 */
i += sl + 1;
@@ -2093,7 +2224,7 @@ static int __init prom_find_machine_type(void)
compat, sizeof(compat)-1);
if (len <= 0)
return PLATFORM_GENERIC;
- if (strcmp(compat, "chrp"))
+ if (prom_strcmp(compat, "chrp"))
return PLATFORM_GENERIC;
/* Default to pSeries. We need to know if we are running LPAR */
@@ -2155,7 +2286,7 @@ static void __init prom_check_displays(void)
for (node = 0; prom_next_node(&node); ) {
memset(type, 0, sizeof(type));
prom_getprop(node, "device_type", type, sizeof(type));
- if (strcmp(type, "display") != 0)
+ if (prom_strcmp(type, "display") != 0)
continue;
/* It seems OF doesn't null-terminate the path :-( */
@@ -2259,9 +2390,9 @@ static unsigned long __init dt_find_string(char *str)
s = os = (char *)dt_string_start;
s += 4;
while (s < (char *)dt_string_end) {
- if (strcmp(s, str) == 0)
+ if (prom_strcmp(s, str) == 0)
return s - os;
- s += strlen(s) + 1;
+ s += prom_strlen(s) + 1;
}
return 0;
}
@@ -2294,7 +2425,7 @@ static void __init scan_dt_build_strings(phandle node,
}
/* skip "name" */
- if (strcmp(namep, "name") == 0) {
+ if (prom_strcmp(namep, "name") == 0) {
*mem_start = (unsigned long)namep;
prev_name = "name";
continue;
@@ -2306,7 +2437,7 @@ static void __init scan_dt_build_strings(phandle node,
namep = sstart + soff;
} else {
/* Trim off some if we can */
- *mem_start = (unsigned long)namep + strlen(namep) + 1;
+ *mem_start = (unsigned long)namep + prom_strlen(namep) + 1;
dt_string_end = *mem_start;
}
prev_name = namep;
@@ -2375,7 +2506,7 @@ static void __init scan_dt_build_struct(phandle node, unsigned long *mem_start,
break;
/* skip "name" */
- if (strcmp(pname, "name") == 0) {
+ if (prom_strcmp(pname, "name") == 0) {
prev_name = "name";
continue;
}
@@ -2406,7 +2537,7 @@ static void __init scan_dt_build_struct(phandle node, unsigned long *mem_start,
call_prom("getprop", 4, 1, node, pname, valp, l);
*mem_start = _ALIGN(*mem_start, 4);
- if (!strcmp(pname, "phandle"))
+ if (!prom_strcmp(pname, "phandle"))
has_phandle = 1;
}
@@ -2476,8 +2607,8 @@ static void __init flatten_device_tree(void)
/* Add "phandle" in there, we'll need it */
namep = make_room(&mem_start, &mem_end, 16, 1);
- strcpy(namep, "phandle");
- mem_start = (unsigned long)namep + strlen(namep) + 1;
+ prom_strcpy(namep, "phandle");
+ mem_start = (unsigned long)namep + prom_strlen(namep) + 1;
/* Build string array */
prom_printf("Building dt strings...\n");
@@ -2799,7 +2930,7 @@ static void __init fixup_device_tree_efika(void)
rv = prom_getprop(node, "model", prop, sizeof(prop));
if (rv == PROM_ERROR)
return;
- if (strcmp(prop, "EFIKA5K2"))
+ if (prom_strcmp(prop, "EFIKA5K2"))
return;
prom_printf("Applying EFIKA device tree fixups\n");
@@ -2807,13 +2938,13 @@ static void __init fixup_device_tree_efika(void)
/* Claiming to be 'chrp' is death */
node = call_prom("finddevice", 1, 1, ADDR("/"));
rv = prom_getprop(node, "device_type", prop, sizeof(prop));
- if (rv != PROM_ERROR && (strcmp(prop, "chrp") == 0))
+ if (rv != PROM_ERROR && (prom_strcmp(prop, "chrp") == 0))
prom_setprop(node, "/", "device_type", "efika", sizeof("efika"));
/* CODEGEN,description is exposed in /proc/cpuinfo so
fix that too */
rv = prom_getprop(node, "CODEGEN,description", prop, sizeof(prop));
- if (rv != PROM_ERROR && (strstr(prop, "CHRP")))
+ if (rv != PROM_ERROR && (prom_strstr(prop, "CHRP")))
prom_setprop(node, "/", "CODEGEN,description",
"Efika 5200B PowerPC System",
sizeof("Efika 5200B PowerPC System"));
diff --git a/arch/powerpc/kernel/prom_init_check.sh b/arch/powerpc/kernel/prom_init_check.sh
index 181fd10008ef..4cac45cb5de5 100644
--- a/arch/powerpc/kernel/prom_init_check.sh
+++ b/arch/powerpc/kernel/prom_init_check.sh
@@ -27,7 +27,7 @@ fi
WHITELIST="add_reloc_offset __bss_start __bss_stop copy_and_flush
_end enter_prom $MEM_FUNCS reloc_offset __secondary_hold
__secondary_hold_acknowledge __secondary_hold_spinloop __start
-strcmp strcpy strlcpy strlen strncmp strstr kstrtobool logo_linux_clut224
+logo_linux_clut224
reloc_got2 kernstart_addr memstart_addr linux_banner _stext
__prom_init_toc_start __prom_init_toc_end btext_setup_display TOC."
--
2.13.3
^ permalink raw reply related
* [PATCH v9 02/11] powerpc: prepare string/mem functions for KASAN
From: Christophe Leroy @ 2019-03-01 12:33 UTC (permalink / raw)
To: Benjamin Herrenschmidt, Paul Mackerras, Michael Ellerman,
Nicholas Piggin, Aneesh Kumar K.V, Andrey Ryabinin,
Alexander Potapenko, Dmitry Vyukov, Daniel Axtens
Cc: linux-mm, linuxppc-dev, linux-kernel, kasan-dev
In-Reply-To: <cover.1551443452.git.christophe.leroy@c-s.fr>
CONFIG_KASAN implements wrappers for memcpy() memmove() and memset()
Those wrappers are doing the verification then call respectively
__memcpy() __memmove() and __memset(). The arches are therefore
expected to rename their optimised functions that way.
For files on which KASAN is inhibited, #defines are used to allow
them to directly call optimised versions of the functions without
going through the KASAN wrappers.
See commit 393f203f5fd5 ("x86_64: kasan: add interceptors for
memset/memmove/memcpy functions") for details.
Other string / mem functions do not (yet) have kasan wrappers,
we therefore have to fallback to the generic versions when
KASAN is active, otherwise KASAN checks will be skipped.
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
---
arch/powerpc/include/asm/kasan.h | 15 +++++++++++++++
arch/powerpc/include/asm/string.h | 32 +++++++++++++++++++++++++++++---
arch/powerpc/kernel/prom_init_check.sh | 10 +++++++++-
arch/powerpc/lib/Makefile | 11 ++++++++---
arch/powerpc/lib/copy_32.S | 15 +++++++++------
arch/powerpc/lib/mem_64.S | 11 +++++++----
arch/powerpc/lib/memcpy_64.S | 5 +++--
7 files changed, 80 insertions(+), 19 deletions(-)
create mode 100644 arch/powerpc/include/asm/kasan.h
diff --git a/arch/powerpc/include/asm/kasan.h b/arch/powerpc/include/asm/kasan.h
new file mode 100644
index 000000000000..c3161b8fc017
--- /dev/null
+++ b/arch/powerpc/include/asm/kasan.h
@@ -0,0 +1,15 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef __ASM_KASAN_H
+#define __ASM_KASAN_H
+
+#ifdef CONFIG_KASAN
+#define _GLOBAL_KASAN(fn) .weak fn ; _GLOBAL(__##fn) ; _GLOBAL(fn)
+#define _GLOBAL_TOC_KASAN(fn) .weak fn ; _GLOBAL_TOC(__##fn) ; _GLOBAL_TOC(fn)
+#define EXPORT_SYMBOL_KASAN(fn) EXPORT_SYMBOL(__##fn) ; EXPORT_SYMBOL(fn)
+#else
+#define _GLOBAL_KASAN(fn) _GLOBAL(fn)
+#define _GLOBAL_TOC_KASAN(fn) _GLOBAL_TOC(fn)
+#define EXPORT_SYMBOL_KASAN(fn) EXPORT_SYMBOL(fn)
+#endif
+
+#endif
diff --git a/arch/powerpc/include/asm/string.h b/arch/powerpc/include/asm/string.h
index 1647de15a31e..9bf6dffb4090 100644
--- a/arch/powerpc/include/asm/string.h
+++ b/arch/powerpc/include/asm/string.h
@@ -4,14 +4,17 @@
#ifdef __KERNEL__
+#ifndef CONFIG_KASAN
#define __HAVE_ARCH_STRNCPY
#define __HAVE_ARCH_STRNCMP
+#define __HAVE_ARCH_MEMCHR
+#define __HAVE_ARCH_MEMCMP
+#define __HAVE_ARCH_MEMSET16
+#endif
+
#define __HAVE_ARCH_MEMSET
#define __HAVE_ARCH_MEMCPY
#define __HAVE_ARCH_MEMMOVE
-#define __HAVE_ARCH_MEMCMP
-#define __HAVE_ARCH_MEMCHR
-#define __HAVE_ARCH_MEMSET16
#define __HAVE_ARCH_MEMCPY_FLUSHCACHE
extern char * strcpy(char *,const char *);
@@ -27,7 +30,27 @@ extern int memcmp(const void *,const void *,__kernel_size_t);
extern void * memchr(const void *,int,__kernel_size_t);
extern void * memcpy_flushcache(void *,const void *,__kernel_size_t);
+void *__memset(void *s, int c, __kernel_size_t count);
+void *__memcpy(void *to, const void *from, __kernel_size_t n);
+void *__memmove(void *to, const void *from, __kernel_size_t n);
+
+#if defined(CONFIG_KASAN) && !defined(__SANITIZE_ADDRESS__)
+/*
+ * For files that are not instrumented (e.g. mm/slub.c) we
+ * should use not instrumented version of mem* functions.
+ */
+#define memcpy(dst, src, len) __memcpy(dst, src, len)
+#define memmove(dst, src, len) __memmove(dst, src, len)
+#define memset(s, c, n) __memset(s, c, n)
+
+#ifndef __NO_FORTIFY
+#define __NO_FORTIFY /* FORTIFY_SOURCE uses __builtin_memcpy, etc. */
+#endif
+
+#endif
+
#ifdef CONFIG_PPC64
+#ifndef CONFIG_KASAN
#define __HAVE_ARCH_MEMSET32
#define __HAVE_ARCH_MEMSET64
@@ -49,8 +72,11 @@ static inline void *memset64(uint64_t *p, uint64_t v, __kernel_size_t n)
{
return __memset64(p, v, n * 8);
}
+#endif
#else
+#ifndef CONFIG_KASAN
#define __HAVE_ARCH_STRLEN
+#endif
extern void *memset16(uint16_t *, uint16_t, __kernel_size_t);
#endif
diff --git a/arch/powerpc/kernel/prom_init_check.sh b/arch/powerpc/kernel/prom_init_check.sh
index 667df97d2595..181fd10008ef 100644
--- a/arch/powerpc/kernel/prom_init_check.sh
+++ b/arch/powerpc/kernel/prom_init_check.sh
@@ -16,8 +16,16 @@
# If you really need to reference something from prom_init.o add
# it to the list below:
+grep "^CONFIG_KASAN=y$" .config >/dev/null
+if [ $? -eq 0 ]
+then
+ MEM_FUNCS="__memcpy __memset"
+else
+ MEM_FUNCS="memcpy memset"
+fi
+
WHITELIST="add_reloc_offset __bss_start __bss_stop copy_and_flush
-_end enter_prom memcpy memset reloc_offset __secondary_hold
+_end enter_prom $MEM_FUNCS reloc_offset __secondary_hold
__secondary_hold_acknowledge __secondary_hold_spinloop __start
strcmp strcpy strlcpy strlen strncmp strstr kstrtobool logo_linux_clut224
reloc_got2 kernstart_addr memstart_addr linux_banner _stext
diff --git a/arch/powerpc/lib/Makefile b/arch/powerpc/lib/Makefile
index 79396e184bca..47a4de434c22 100644
--- a/arch/powerpc/lib/Makefile
+++ b/arch/powerpc/lib/Makefile
@@ -8,9 +8,14 @@ ccflags-$(CONFIG_PPC64) := $(NO_MINIMAL_TOC)
CFLAGS_REMOVE_code-patching.o = $(CC_FLAGS_FTRACE)
CFLAGS_REMOVE_feature-fixups.o = $(CC_FLAGS_FTRACE)
-obj-y += string.o alloc.o code-patching.o feature-fixups.o
+obj-y += alloc.o code-patching.o feature-fixups.o
-obj-$(CONFIG_PPC32) += div64.o copy_32.o crtsavres.o strlen_32.o
+ifndef CONFIG_KASAN
+obj-y += string.o memcmp_$(BITS).o
+obj-$(CONFIG_PPC32) += strlen_32.o
+endif
+
+obj-$(CONFIG_PPC32) += div64.o copy_32.o crtsavres.o
obj-$(CONFIG_FUNCTION_ERROR_INJECTION) += error-inject.o
@@ -34,7 +39,7 @@ obj64-$(CONFIG_KPROBES_SANITY_TEST) += test_emulate_step.o \
test_emulate_step_exec_instr.o
obj-y += checksum_$(BITS).o checksum_wrappers.o \
- string_$(BITS).o memcmp_$(BITS).o
+ string_$(BITS).o
obj-y += sstep.o ldstfp.o quad.o
obj64-y += quad.o
diff --git a/arch/powerpc/lib/copy_32.S b/arch/powerpc/lib/copy_32.S
index ba66846fe973..fc4fa7246200 100644
--- a/arch/powerpc/lib/copy_32.S
+++ b/arch/powerpc/lib/copy_32.S
@@ -14,6 +14,7 @@
#include <asm/ppc_asm.h>
#include <asm/export.h>
#include <asm/code-patching-asm.h>
+#include <asm/kasan.h>
#define COPY_16_BYTES \
lwz r7,4(r4); \
@@ -68,6 +69,7 @@ CACHELINE_BYTES = L1_CACHE_BYTES
LG_CACHELINE_BYTES = L1_CACHE_SHIFT
CACHELINE_MASK = (L1_CACHE_BYTES-1)
+#ifndef CONFIG_KASAN
_GLOBAL(memset16)
rlwinm. r0 ,r5, 31, 1, 31
addi r6, r3, -4
@@ -81,6 +83,7 @@ _GLOBAL(memset16)
sth r4, 4(r6)
blr
EXPORT_SYMBOL(memset16)
+#endif
/*
* Use dcbz on the complete cache lines in the destination
@@ -91,7 +94,7 @@ EXPORT_SYMBOL(memset16)
* We therefore skip the optimised bloc that uses dcbz. This jump is
* replaced by a nop once cache is active. This is done in machine_init()
*/
-_GLOBAL(memset)
+_GLOBAL_KASAN(memset)
cmplwi 0,r5,4
blt 7f
@@ -150,7 +153,7 @@ _GLOBAL(memset)
9: stbu r4,1(r6)
bdnz 9b
blr
-EXPORT_SYMBOL(memset)
+EXPORT_SYMBOL_KASAN(memset)
/*
* This version uses dcbz on the complete cache lines in the
@@ -163,12 +166,12 @@ EXPORT_SYMBOL(memset)
* We therefore jump to generic_memcpy which doesn't use dcbz. This jump is
* replaced by a nop once cache is active. This is done in machine_init()
*/
-_GLOBAL(memmove)
+_GLOBAL_KASAN(memmove)
cmplw 0,r3,r4
bgt backwards_memcpy
/* fall through */
-_GLOBAL(memcpy)
+_GLOBAL_KASAN(memcpy)
1: b generic_memcpy
patch_site 1b, patch__memcpy_nocache
@@ -242,8 +245,8 @@ _GLOBAL(memcpy)
stbu r0,1(r6)
bdnz 40b
65: blr
-EXPORT_SYMBOL(memcpy)
-EXPORT_SYMBOL(memmove)
+EXPORT_SYMBOL_KASAN(memcpy)
+EXPORT_SYMBOL_KASAN(memmove)
generic_memcpy:
srwi. r7,r5,3
diff --git a/arch/powerpc/lib/mem_64.S b/arch/powerpc/lib/mem_64.S
index 3c3be02f33b7..7cd6cf6822a2 100644
--- a/arch/powerpc/lib/mem_64.S
+++ b/arch/powerpc/lib/mem_64.S
@@ -12,7 +12,9 @@
#include <asm/errno.h>
#include <asm/ppc_asm.h>
#include <asm/export.h>
+#include <asm/kasan.h>
+#ifndef CONFIG_KASAN
_GLOBAL(__memset16)
rlwimi r4,r4,16,0,15
/* fall through */
@@ -29,8 +31,9 @@ _GLOBAL(__memset64)
EXPORT_SYMBOL(__memset16)
EXPORT_SYMBOL(__memset32)
EXPORT_SYMBOL(__memset64)
+#endif
-_GLOBAL(memset)
+_GLOBAL_KASAN(memset)
neg r0,r3
rlwimi r4,r4,8,16,23
andi. r0,r0,7 /* # bytes to be 8-byte aligned */
@@ -95,9 +98,9 @@ _GLOBAL(memset)
10: bflr 31
stb r4,0(r6)
blr
-EXPORT_SYMBOL(memset)
+EXPORT_SYMBOL_KASAN(memset)
-_GLOBAL_TOC(memmove)
+_GLOBAL_TOC_KASAN(memmove)
cmplw 0,r3,r4
bgt backwards_memcpy
b memcpy
@@ -138,4 +141,4 @@ _GLOBAL(backwards_memcpy)
beq 2b
mtctr r7
b 1b
-EXPORT_SYMBOL(memmove)
+EXPORT_SYMBOL_KASAN(memmove)
diff --git a/arch/powerpc/lib/memcpy_64.S b/arch/powerpc/lib/memcpy_64.S
index 273ea67e60a1..862b515b8868 100644
--- a/arch/powerpc/lib/memcpy_64.S
+++ b/arch/powerpc/lib/memcpy_64.S
@@ -11,6 +11,7 @@
#include <asm/export.h>
#include <asm/asm-compat.h>
#include <asm/feature-fixups.h>
+#include <asm/kasan.h>
#ifndef SELFTEST_CASE
/* For big-endian, 0 == most CPUs, 1 == POWER6, 2 == Cell */
@@ -18,7 +19,7 @@
#endif
.align 7
-_GLOBAL_TOC(memcpy)
+_GLOBAL_TOC_KASAN(memcpy)
BEGIN_FTR_SECTION
#ifdef __LITTLE_ENDIAN__
cmpdi cr7,r5,0
@@ -229,4 +230,4 @@ END_FTR_SECTION_IFCLR(CPU_FTR_UNALIGNED_LD_STD)
4: ld r3,-STACKFRAMESIZE+STK_REG(R31)(r1) /* return dest pointer */
blr
#endif
-EXPORT_SYMBOL(memcpy)
+EXPORT_SYMBOL_KASAN(memcpy)
--
2.13.3
^ permalink raw reply related
* [PATCH v9 04/11] powerpc/mm: don't use direct assignation during early boot.
From: Christophe Leroy @ 2019-03-01 12:33 UTC (permalink / raw)
To: Benjamin Herrenschmidt, Paul Mackerras, Michael Ellerman,
Nicholas Piggin, Aneesh Kumar K.V, Andrey Ryabinin,
Alexander Potapenko, Dmitry Vyukov, Daniel Axtens
Cc: linux-mm, linuxppc-dev, linux-kernel, kasan-dev
In-Reply-To: <cover.1551443452.git.christophe.leroy@c-s.fr>
In kernel/cputable.c, explicitly use memcpy() instead of *y = *x;
This will allow GCC to replace it with __memcpy() when KASAN is
selected.
Acked-by: Dmitry Vyukov <dvyukov@google.com>
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
---
arch/powerpc/kernel/cputable.c | 13 ++++++++++---
1 file changed, 10 insertions(+), 3 deletions(-)
diff --git a/arch/powerpc/kernel/cputable.c b/arch/powerpc/kernel/cputable.c
index 1eab54bc6ee9..cd12f362b61f 100644
--- a/arch/powerpc/kernel/cputable.c
+++ b/arch/powerpc/kernel/cputable.c
@@ -2147,7 +2147,11 @@ void __init set_cur_cpu_spec(struct cpu_spec *s)
struct cpu_spec *t = &the_cpu_spec;
t = PTRRELOC(t);
- *t = *s;
+ /*
+ * use memcpy() instead of *t = *s so that GCC replaces it
+ * by __memcpy() when KASAN is active
+ */
+ memcpy(t, s, sizeof(*t));
*PTRRELOC(&cur_cpu_spec) = &the_cpu_spec;
}
@@ -2161,8 +2165,11 @@ static struct cpu_spec * __init setup_cpu_spec(unsigned long offset,
t = PTRRELOC(t);
old = *t;
- /* Copy everything, then do fixups */
- *t = *s;
+ /*
+ * Copy everything, then do fixups. Use memcpy() instead of *t = *s
+ * so that GCC replaces it by __memcpy() when KASAN is active
+ */
+ memcpy(t, s, sizeof(*t));
/*
* If we are overriding a previous value derived from the real
--
2.13.3
^ permalink raw reply related
* Re: DMA & POWER8
From: Илья Кузнецов @ 2019-03-01 10:32 UTC (permalink / raw)
To: linuxppc-dev, Benjamin Herrenschmidt; +Cc: a.fomichev
In-Reply-To: <20190227155702.ngm3l5bjsde2vvrw@yadro.com>
I'm sorry for previous email content without greeting.
A bit more description… We are working on enabling RDMA over PCIe NTB on
OpenPOWER using https://github.com/ntrdma project. During initialization
dma_request_channel() founds no available channels. We also found that there no
channels at all in dma_device_list. This is probably powerpc specific problem.
Could someone explain if that is hardware limitation or arch specific part
should be changed?
Ben, a while ago you answered regarding DMA master on POWER8, may be you have
idea here also?
Thank you in advance!
Илья Кузнецов <ilya@yadro.com> wrote on Ср [2019-фев-27 18:57:02 +0300]:
>We're trying to use Slave-DMA API of the DMA Engine in the linux kernel on
>POWER8 OpenPOWER platform. And we discovered that dma_request_channel() always fails.
>Does it mean that Turismo hardware does not support Slave-DMA? Or DMA
>subsystem in the linux kernel needs to be fixed, doesn't it?
^ permalink raw reply
* [PATCH V2] ASoC: fsl_asrc: add constraint for the asrc of older version
From: S.j. Wang @ 2019-03-01 8:37 UTC (permalink / raw)
To: timur@kernel.org, nicoleotsuka@gmail.com, Xiubo.Lee@gmail.com,
festevam@gmail.com, broonie@kernel.org,
alsa-devel@alsa-project.org
Cc: linuxppc-dev@lists.ozlabs.org
There is constraint for the channel number setting on the
asrc of older version (e.g. imx35), the channel number should
be even, odd number isn't valid.
So add this constraint when the asrc of older version is used.
Signed-off-by: Shengjiu Wang <shengjiu.wang@nxp.com>
---
changes in v2
- switch to add constraint in startup function
sound/soc/fsl/fsl_asrc.c | 14 ++++++++++++++
1 file changed, 14 insertions(+)
diff --git a/sound/soc/fsl/fsl_asrc.c b/sound/soc/fsl/fsl_asrc.c
index 528e8b108422..31494f225037 100644
--- a/sound/soc/fsl/fsl_asrc.c
+++ b/sound/soc/fsl/fsl_asrc.c
@@ -445,6 +445,19 @@ struct dma_chan *fsl_asrc_get_dma_channel(struct fsl_asrc_pair *pair, bool dir)
}
EXPORT_SYMBOL_GPL(fsl_asrc_get_dma_channel);
+static int fsl_asrc_dai_startup(struct snd_pcm_substream *substream,
+ struct snd_soc_dai *dai)
+{
+ struct fsl_asrc *asrc_priv = snd_soc_dai_get_drvdata(dai);
+
+ /* channel_bits = 3 means older version on imx35*/
+ if (asrc_priv->channel_bits == 3)
+ snd_pcm_hw_constraint_step(substream->runtime, 0,
+ SNDRV_PCM_HW_PARAM_CHANNELS, 2);
+
+ return 0;
+}
+
static int fsl_asrc_dai_hw_params(struct snd_pcm_substream *substream,
struct snd_pcm_hw_params *params,
struct snd_soc_dai *dai)
@@ -539,6 +552,7 @@ static int fsl_asrc_dai_trigger(struct snd_pcm_substream *substream, int cmd,
}
static const struct snd_soc_dai_ops fsl_asrc_dai_ops = {
+ .startup = fsl_asrc_dai_startup,
.hw_params = fsl_asrc_dai_hw_params,
.hw_free = fsl_asrc_dai_hw_free,
.trigger = fsl_asrc_dai_trigger,
--
1.9.1
^ permalink raw reply related
* RE: [PATCH] ASoC: fsl_asrc: add protection for the asrc of older version
From: S.j. Wang @ 2019-03-01 8:32 UTC (permalink / raw)
To: Nicolin Chen
Cc: alsa-devel@alsa-project.org, timur@kernel.org,
Xiubo.Lee@gmail.com, festevam@gmail.com, broonie@kernel.org,
linuxppc-dev@lists.ozlabs.org
In-Reply-To: <20190301080812.GA5244@Asurada>
Hi
>
> On Fri, Mar 01, 2019 at 06:55:25AM +0000, S.j. Wang wrote:
>
> > > Alternatively, I feel instead of error-out at here, should we add a
> > > HW constraint or at least fence it off at the beginning of the
> > > hw_params()? This is actually nothing specific to the pair-request
> > > function but a hardware constraint.
> >
> > How about add constraint in startup?
> > static int fsl_asrc_dai_startup(struct snd_pcm_substream *substream,
> > struct snd_soc_dai *dai) {
> > struct fsl_asrc *asrc_priv = snd_soc_dai_get_drvdata(dai);
> >
> > if (asrc_priv->channel_bits == 3) {
> > snd_pcm_hw_constraint_step(substream->runtime, 0,
> > SNDRV_PCM_HW_PARAM_CHANNELS, 2);
> > }
> >
> > return 0;
> > }
>
> Yea, that looks good to me. Better to have a line of comments to tell that
> "bits==3" means older version -- maybe we should have something much
> more clear than using channel_bits but it is fine for now since they only
> differ here.
Ok , will send v2.
^ permalink raw reply
* Re: [PATCH] ASoC: fsl_asrc: add protection for the asrc of older version
From: Nicolin Chen @ 2019-03-01 8:08 UTC (permalink / raw)
To: S.j. Wang
Cc: alsa-devel@alsa-project.org, timur@kernel.org,
Xiubo.Lee@gmail.com, festevam@gmail.com, broonie@kernel.org,
linuxppc-dev@lists.ozlabs.org
In-Reply-To: <VI1PR0402MB339254C31997242633511D12E3760@VI1PR0402MB3392.eurprd04.prod.outlook.com>
On Fri, Mar 01, 2019 at 06:55:25AM +0000, S.j. Wang wrote:
> > Alternatively, I feel instead of error-out at here, should we add a HW
> > constraint or at least fence it off at the beginning of the hw_params()? This
> > is actually nothing specific to the pair-request function but a hardware
> > constraint.
>
> How about add constraint in startup?
> static int fsl_asrc_dai_startup(struct snd_pcm_substream *substream,
> struct snd_soc_dai *dai)
> {
> struct fsl_asrc *asrc_priv = snd_soc_dai_get_drvdata(dai);
>
> if (asrc_priv->channel_bits == 3) {
> snd_pcm_hw_constraint_step(substream->runtime, 0,
> SNDRV_PCM_HW_PARAM_CHANNELS, 2);
> }
>
> return 0;
> }
Yea, that looks good to me. Better to have a line of comments
to tell that "bits==3" means older version -- maybe we should
have something much more clear than using channel_bits but it
is fine for now since they only differ here.
^ permalink raw reply
* Re: [PATCH] KVM: PPC: powerpc: Add count cache flush parameters to kvmppc_get_cpu_char()
From: Paul Mackerras @ 2019-03-01 8:01 UTC (permalink / raw)
To: Michael Ellerman; +Cc: linuxppc-dev, kvm-ppc, Suraj Jitindar Singh, kvm
In-Reply-To: <87bm2vjdbw.fsf@concordia.ellerman.id.au>
On Fri, Mar 01, 2019 at 05:52:51PM +1100, Michael Ellerman wrote:
> Suraj Jitindar Singh <sjitindarsingh@gmail.com> writes:
>
> > Add KVM_PPC_CPU_CHAR_BCCTR_FLUSH_ASSIST &
> > KVM_PPC_CPU_BEHAV_FLUSH_COUNT_CACHE to the characteristics returned from
> > the H_GET_CPU_CHARACTERISTICS H-CALL, as queried from either the
> > hypervisor or the device tree.
> >
> > Signed-off-by: Suraj Jitindar Singh <sjitindarsingh@gmail.com>
> > ---
> > arch/powerpc/include/uapi/asm/kvm.h | 2 ++
> > arch/powerpc/kvm/powerpc.c | 18 ++++++++++++++----
> > 2 files changed, 16 insertions(+), 4 deletions(-)
> >
> > diff --git a/arch/powerpc/kvm/powerpc.c b/arch/powerpc/kvm/powerpc.c
> > index b90a7d154180..a99dcac91e50 100644
> > --- a/arch/powerpc/kvm/powerpc.c
> > +++ b/arch/powerpc/kvm/powerpc.c
> > @@ -2251,12 +2253,16 @@ static int kvmppc_get_cpu_char(struct kvm_ppc_cpu_char *cp)
> > if (have_fw_feat(fw_features, "enabled",
> > "fw-count-cache-disabled"))
> > cp->character |= KVM_PPC_CPU_CHAR_COUNT_CACHE_DIS;
> > + if (have_fw_feat(fw_features, "enabled",
> > + "fw-count-cache-flush-bcctr2,0,0"))
>
> I don't think there's any reason KVM needs to be querying the device
> tree directly, is there?
>
> Can't it just use the security flags (security_features.h), that are
> initialised by the powernv platform code based on the device tree.
My recollection is that the security flags didn't have all the
information we need. It's possible that's no longer true.
I merged the patch and sent a pull request to Paolo; using the same
pattern as the existing code made it a low-risk patch and I wanted to
get it in for 5.1. For 5.2 we can look at changing it over if you
like.
Paul.
^ permalink raw reply
* RE: [PATCH] ASoC: fsl_asrc: add protection for the asrc of older version
From: S.j. Wang @ 2019-03-01 6:55 UTC (permalink / raw)
To: Nicolin Chen
Cc: alsa-devel@alsa-project.org, timur@kernel.org,
Xiubo.Lee@gmail.com, festevam@gmail.com, broonie@kernel.org,
linuxppc-dev@lists.ozlabs.org
In-Reply-To: <20190301025356.GA1371@Asurada-Nvidia.nvidia.com>
Hi
>
> Hi Shengjiu,
>
> On Fri, Mar 01, 2019 at 02:32:38AM +0000, S.j. Wang wrote:
> > There is a constraint for the channel number setting on the asrc of
> > older version (e.g. imx35), the channel number should be even, odd
> > number isn't valid.
> >
> > So add protection when the asrc of older version is used.
> >
> > Signed-off-by: Shengjiu Wang <shengjiu.wang@nxp.com>
> > ---
> > sound/soc/fsl/fsl_asrc.c | 3 ++-
> > 1 file changed, 2 insertions(+), 1 deletion(-)
> >
> > diff --git a/sound/soc/fsl/fsl_asrc.c b/sound/soc/fsl/fsl_asrc.c index
> > 528e8b108422..b3b3c5e15ef1 100644
> > --- a/sound/soc/fsl/fsl_asrc.c
> > +++ b/sound/soc/fsl/fsl_asrc.c
> > @@ -109,7 +109,8 @@ static int fsl_asrc_request_pair(int channels, struct
> fsl_asrc_pair *pair)
> > if (index == ASRC_INVALID_PAIR) {
> > dev_err(dev, "all pairs are busy now\n");
> > ret = -EBUSY;
> > - } else if (asrc_priv->channel_avail < channels) {
> > + } else if (asrc_priv->channel_avail < channels ||
> > + (asrc_priv->channel_bits < 4 && channels % 2 != 0)) {
> > dev_err(dev, "can't afford required channels: %d\n",
> channels);
>
> I feel it'd be better to have another else-if, since the existing error message
> is against something else.
>
> + } else if (asrc_priv->channel_bits < 4 && channels & 1) {
> + /* old version of ASRC has channel_bits = 3 */
> + dev_err(dev, "does not support odd channel number\n");
> + ret = -EINVAL;
>
> Alternatively, I feel instead of error-out at here, should we add a HW
> constraint or at least fence it off at the beginning of the hw_params()? This
> is actually nothing specific to the pair-request function but a hardware
> constraint.
How about add constraint in startup?
static int fsl_asrc_dai_startup(struct snd_pcm_substream *substream,
struct snd_soc_dai *dai)
{
struct fsl_asrc *asrc_priv = snd_soc_dai_get_drvdata(dai);
if (asrc_priv->channel_bits == 3) {
snd_pcm_hw_constraint_step(substream->runtime, 0,
SNDRV_PCM_HW_PARAM_CHANNELS, 2);
}
return 0;
}
Best regards
Wang shengjiu
^ permalink raw reply
* Re: [PATCH] KVM: PPC: powerpc: Add count cache flush parameters to kvmppc_get_cpu_char()
From: Michael Ellerman @ 2019-03-01 6:52 UTC (permalink / raw)
To: Suraj Jitindar Singh, kvm-ppc
Cc: linuxppc-dev, paulus, Suraj Jitindar Singh, kvm
In-Reply-To: <20190301032516.29919-1-sjitindarsingh@gmail.com>
Suraj Jitindar Singh <sjitindarsingh@gmail.com> writes:
> Add KVM_PPC_CPU_CHAR_BCCTR_FLUSH_ASSIST &
> KVM_PPC_CPU_BEHAV_FLUSH_COUNT_CACHE to the characteristics returned from
> the H_GET_CPU_CHARACTERISTICS H-CALL, as queried from either the
> hypervisor or the device tree.
>
> Signed-off-by: Suraj Jitindar Singh <sjitindarsingh@gmail.com>
> ---
> arch/powerpc/include/uapi/asm/kvm.h | 2 ++
> arch/powerpc/kvm/powerpc.c | 18 ++++++++++++++----
> 2 files changed, 16 insertions(+), 4 deletions(-)
>
> diff --git a/arch/powerpc/kvm/powerpc.c b/arch/powerpc/kvm/powerpc.c
> index b90a7d154180..a99dcac91e50 100644
> --- a/arch/powerpc/kvm/powerpc.c
> +++ b/arch/powerpc/kvm/powerpc.c
> @@ -2251,12 +2253,16 @@ static int kvmppc_get_cpu_char(struct kvm_ppc_cpu_char *cp)
> if (have_fw_feat(fw_features, "enabled",
> "fw-count-cache-disabled"))
> cp->character |= KVM_PPC_CPU_CHAR_COUNT_CACHE_DIS;
> + if (have_fw_feat(fw_features, "enabled",
> + "fw-count-cache-flush-bcctr2,0,0"))
I don't think there's any reason KVM needs to be querying the device
tree directly, is there?
Can't it just use the security flags (security_features.h), that are
initialised by the powernv platform code based on the device tree.
I seem to remember we didn't do it that way originally just because we
didn't have time to get everything lined up.
cheers
^ permalink raw reply
* RE: [alsa-devel] [PATCH] ASoC: cs42xx8: Remove S32_LE in format list
From: S.j. Wang @ 2019-03-01 6:32 UTC (permalink / raw)
To: Matt Flax, Nicolin Chen
Cc: alsa-devel@alsa-project.org, timur@kernel.org,
Xiubo.Lee@gmail.com, festevam@gmail.com, broonie@kernel.org,
linuxppc-dev@lists.ozlabs.org
In-Reply-To: <5c60b8c7-cb85-dec8-0358-6773251d3f2c@flatmax.org>
>
> It would be a mistake to remove S32_LE from the format list.
>
> If you removed that, then we wouldn't be able to do 32 bit native playback
> - it would force users to go through an ALSA plugin.
>
> If you look at the cs42448 data sheet for example, the I2S and TDM bus
> uses a 32 bit word size. For that reason, native S32_LE format is handled by
> the codec. While the codec only uses the 24 MSBs, the I2S bus is 32 bit.
This case is covered by S24_LE I think. The S32_LE means the data is 32bit and slot width
Is 32bit, this is not in data sheet.
Best regards
Wang shengjiu
>
> Matt
>
> On 1/3/19 3:50 pm, Nicolin Chen wrote:
> > Hi Shengjiu,
> >
> > On Thu, Feb 28, 2019 at 05:56:31AM +0000, S.j. Wang wrote:
> >> cs42xx8 is a 24-bit A/D and 24-bit D/A device, so the S32_LE should
> >> not be in the supported format list.
> >>
> >> Signed-off-by: Shengjiu Wang <shengjiu.wang@nxp.com>
> >> ---
> >> sound/soc/codecs/cs42xx8.c | 3 +--
> > This is for cs42xx8 codec driver, but its maintainers aren't seemly in
> > To/Cc list. Please run ./scripts/get_maintainer.pl for it and resend.
> > _______________________________________________
> > Alsa-devel mailing list
> > Alsa-devel@alsa-project.org
> >
> https://emea01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fma
> i
> > lman.alsa-project.org%2Fmailman%2Flistinfo%2Falsa-
> devel&data=02%7C
> >
> 01%7Cshengjiu.wang%40nxp.com%7Ce5686d72bd7843c5962508d69e0370a4
> %7C686e
> >
> a1d3bc2b4c6fa92cd99c5c301635%7C0%7C1%7C636870134975623123&
> sdata=Rr
> >
> YwfC4Y3xssED%2FzRum%2FXABGJTHvRrYwnRNgUL7u2lU%3D&reserved
> =0
^ permalink raw reply
* [PATCH v2] powernv: powercap: Add hard minimum powercap
From: Shilpasri G Bhat @ 2019-03-01 6:25 UTC (permalink / raw)
To: mpe; +Cc: ego, linuxppc-dev, linux-kernel, Shilpasri G Bhat, dja
In POWER9, OCC(On-Chip-Controller) provides for hard and soft system
powercapping range. The hard powercap range is guaranteed while soft
powercap may or may not be enforced by OCC due to various power-thermal
reasons based on system configuration and workloads. This patch adds
a sysfs file to export the hard minimum powercap limit to allow the
user to set the appropriate powercap value that can be managed by the
system.
Signed-off-by: Shilpasri G Bhat <shilpa.bhat@linux.vnet.ibm.com>
---
Changes from V1:
- s/asserted/enforced by OCC
.../ABI/testing/sysfs-firmware-opal-powercap | 10 ++++
arch/powerpc/platforms/powernv/opal-powercap.c | 66 +++++++++-------------
2 files changed, 37 insertions(+), 39 deletions(-)
diff --git a/Documentation/ABI/testing/sysfs-firmware-opal-powercap b/Documentation/ABI/testing/sysfs-firmware-opal-powercap
index c9b66ec..e37d43e 100644
--- a/Documentation/ABI/testing/sysfs-firmware-opal-powercap
+++ b/Documentation/ABI/testing/sysfs-firmware-opal-powercap
@@ -29,3 +29,13 @@ Description: System powercap directory and attributes applicable for
creates a request for setting a new-powercap. The
powercap requested must be between powercap-min
and powercap-max.
+
+What: /sys/firmware/opal/powercap/system-powercap/powercap-hard-min
+Date: March 2019
+Contact: Linux for PowerPC mailing list <linuxppc-dev@ozlabs.org>
+Description: Hard minimum powercap
+
+ This file provides the hard minimum powercap limit in Watts.
+ The powercap value above hard minimum limit is guaranteed to be
+ enforced by OCC and the powercap value below this limit may or
+ may not be guaranteed.
diff --git a/arch/powerpc/platforms/powernv/opal-powercap.c b/arch/powerpc/platforms/powernv/opal-powercap.c
index d90ee4f..38408e7 100644
--- a/arch/powerpc/platforms/powernv/opal-powercap.c
+++ b/arch/powerpc/platforms/powernv/opal-powercap.c
@@ -139,10 +139,24 @@ static void powercap_add_attr(int handle, const char *name,
attr->handle = handle;
sysfs_attr_init(&attr->attr.attr);
attr->attr.attr.name = name;
- attr->attr.attr.mode = 0444;
+
+ if (!strncmp(name, "powercap-current", strlen(name))) {
+ attr->attr.attr.mode = 0664;
+ attr->attr.store = powercap_store;
+ } else {
+ attr->attr.attr.mode = 0444;
+ }
+
attr->attr.show = powercap_show;
}
+static const char * const powercap_strs[] = {
+ "powercap-max",
+ "powercap-min",
+ "powercap-current",
+ "powercap-hard-min",
+};
+
void __init opal_powercap_init(void)
{
struct device_node *powercap, *node;
@@ -167,60 +181,34 @@ void __init opal_powercap_init(void)
i = 0;
for_each_child_of_node(powercap, node) {
- u32 cur, min, max;
- int j = 0;
- bool has_cur = false, has_min = false, has_max = false;
+ u32 id;
+ int j, count = 0;
- if (!of_property_read_u32(node, "powercap-min", &min)) {
- j++;
- has_min = true;
- }
-
- if (!of_property_read_u32(node, "powercap-max", &max)) {
- j++;
- has_max = true;
- }
+ for (j = 0; j < ARRAY_SIZE(powercap_strs); j++)
+ if (!of_property_read_u32(node, powercap_strs[j], &id))
+ count++;
- if (!of_property_read_u32(node, "powercap-current", &cur)) {
- j++;
- has_cur = true;
- }
-
- pcaps[i].pattrs = kcalloc(j, sizeof(struct powercap_attr),
+ pcaps[i].pattrs = kcalloc(count, sizeof(struct powercap_attr),
GFP_KERNEL);
if (!pcaps[i].pattrs)
goto out_pcaps_pattrs;
- pcaps[i].pg.attrs = kcalloc(j + 1, sizeof(struct attribute *),
+ pcaps[i].pg.attrs = kcalloc(count + 1,
+ sizeof(struct attribute *),
GFP_KERNEL);
if (!pcaps[i].pg.attrs) {
kfree(pcaps[i].pattrs);
goto out_pcaps_pattrs;
}
- j = 0;
pcaps[i].pg.name = kasprintf(GFP_KERNEL, "%pOFn", node);
- if (has_min) {
- powercap_add_attr(min, "powercap-min",
- &pcaps[i].pattrs[j]);
- pcaps[i].pg.attrs[j] = &pcaps[i].pattrs[j].attr.attr;
- j++;
- }
-
- if (has_max) {
- powercap_add_attr(max, "powercap-max",
- &pcaps[i].pattrs[j]);
- pcaps[i].pg.attrs[j] = &pcaps[i].pattrs[j].attr.attr;
- j++;
- }
+ for (j = 0; j < ARRAY_SIZE(powercap_strs); j++) {
+ if (of_property_read_u32(node, powercap_strs[j], &id))
+ continue;
- if (has_cur) {
- powercap_add_attr(cur, "powercap-current",
+ powercap_add_attr(id, powercap_strs[j],
&pcaps[i].pattrs[j]);
- pcaps[i].pattrs[j].attr.attr.mode |= 0220;
- pcaps[i].pattrs[j].attr.store = powercap_store;
pcaps[i].pg.attrs[j] = &pcaps[i].pattrs[j].attr.attr;
- j++;
}
if (sysfs_create_group(powercap_kobj, &pcaps[i].pg)) {
--
1.8.3.1
^ permalink raw reply related
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox