From: Alexey Starikovskiy <aystarik@gmail.com>
To: "Moore, Robert" <robert.moore@intel.com>
Cc: "Lin, Ming M" <ming.m.lin@intel.com>,
Alexey Starikovskiy <astarikovskiy@suse.de>,
"Linux-acpi@vger.kernel.org" <Linux-acpi@vger.kernel.org>
Subject: Re: [PATCH 1/4] ACPICA: Remove use of caches in controlled way
Date: Tue, 12 May 2009 22:54:14 +0400 [thread overview]
Message-ID: <4A09C5D6.80409@gmail.com> (raw)
In-Reply-To: <4911F71203A09E4D9981D27F9D8308582281A5F5@orsmsx503.amr.corp.intel.com>
What bugs are you seeing? May I be of any help?
Regards,
Alex.
Moore, Robert wrote:
> Removing the cache has exposed a couple of bugs that we are currently working on. We need to make sure that we pass the ASLTS suite before the code is integrated/released.
>
> Bob
>
>
>
>> -----Original Message-----
>> From: Lin, Ming M
>> Sent: Monday, May 11, 2009 1:14 AM
>> To: Alexey Starikovskiy
>> Cc: Moore, Robert; Linux-acpi@vger.kernel.org
>> Subject: Re: [PATCH 1/4] ACPICA: Remove use of caches in controlled way
>>
>> Hi, Alexey
>>
>> The attachment are the linuxized version of your patches.
>> I added another patch (the 4th) to remove linux specific cache code.
>>
>> The patches are applied on top of linux-acpi-2.6/acpica branch.
>>
>> Do you have any suggestion what data I should collect to verify if these
>> patches have a performance gain?
>>
>> Lin Ming
>>
>> On Thu, 2009-05-07 at 23:52 +0800, Alexey Starikovskiy wrote:
>>
>>> Signed-off-by: Alexey Starikovskiy <astarikovskiy@suse.de>
>>> ---
>>>
>>> source/components/utilities/utcache.c | 99 +++++----------------------
>>>
>> ------
>>
>>> 1 files changed, 14 insertions(+), 85 deletions(-)
>>>
>>>
>>> diff --git a/source/components/utilities/utcache.c
>>>
>> b/source/components/utilities/utcache.c
>>
>>> index aabc0a2..d08d520 100644
>>> --- a/source/components/utilities/utcache.c
>>> +++ b/source/components/utilities/utcache.c
>>> @@ -287,49 +287,15 @@ AcpiOsReleaseObject (
>>> ACPI_MEMORY_LIST *Cache,
>>> void *Object)
>>> {
>>> - ACPI_STATUS Status;
>>> -
>>> -
>>> ACPI_FUNCTION_ENTRY ();
>>>
>>> -
>>> if (!Cache || !Object)
>>> {
>>> return (AE_BAD_PARAMETER);
>>> }
>>>
>>> - /* If cache is full, just free this object */
>>> -
>>> - if (Cache->CurrentDepth >= Cache->MaxDepth)
>>> - {
>>> - ACPI_FREE (Object);
>>> - ACPI_MEM_TRACKING (Cache->TotalFreed++);
>>> - }
>>> -
>>> - /* Otherwise put this object back into the cache */
>>> -
>>> - else
>>> - {
>>> - Status = AcpiUtAcquireMutex (ACPI_MTX_CACHES);
>>> - if (ACPI_FAILURE (Status))
>>> - {
>>> - return (Status);
>>> - }
>>> -
>>> - /* Mark the object as cached */
>>> -
>>> - ACPI_MEMSET (Object, 0xCA, Cache->ObjectSize);
>>> - ACPI_SET_DESCRIPTOR_TYPE (Object, ACPI_DESC_TYPE_CACHED);
>>> -
>>> - /* Put the object at the head of the cache list */
>>> -
>>> - * (ACPI_CAST_INDIRECT_PTR (char,
>>> - &(((char *) Object)[Cache->LinkOffset]))) = Cache->ListHead;
>>> - Cache->ListHead = Object;
>>> - Cache->CurrentDepth++;
>>> -
>>> - (void) AcpiUtReleaseMutex (ACPI_MTX_CACHES);
>>> - }
>>> + ACPI_FREE (Object);
>>> + ACPI_MEM_TRACKING (Cache->TotalFreed++);
>>>
>>> return (AE_OK);
>>> }
>>> @@ -353,8 +319,6 @@ AcpiOsAcquireObject (
>>> ACPI_MEMORY_LIST *Cache)
>>> {
>>> ACPI_STATUS Status;
>>> - void *Object;
>>> -
>>>
>>> ACPI_FUNCTION_NAME (OsAcquireObject);
>>>
>>> @@ -372,61 +336,26 @@ AcpiOsAcquireObject (
>>>
>>> ACPI_MEM_TRACKING (Cache->Requests++);
>>>
>>> - /* Check the cache first */
>>> + /* The cache is empty, create a new object */
>>>
>>> - if (Cache->ListHead)
>>> - {
>>> - /* There is an object available, use it */
>>> -
>>> - Object = Cache->ListHead;
>>> - Cache->ListHead = *(ACPI_CAST_INDIRECT_PTR (char,
>>> - &(((char *) Object)[Cache-
>>> LinkOffset])));
>>> -
>>> - Cache->CurrentDepth--;
>>> -
>>> - ACPI_MEM_TRACKING (Cache->Hits++);
>>> - ACPI_DEBUG_PRINT ((ACPI_DB_EXEC,
>>> - "Object %p from %s cache\n", Object, Cache->ListName));
>>> -
>>> - Status = AcpiUtReleaseMutex (ACPI_MTX_CACHES);
>>> - if (ACPI_FAILURE (Status))
>>> - {
>>> - return (NULL);
>>> - }
>>> -
>>> - /* Clear (zero) the previously used Object */
>>> -
>>> - ACPI_MEMSET (Object, 0, Cache->ObjectSize);
>>> - }
>>> - else
>>> - {
>>> - /* The cache is empty, create a new object */
>>> -
>>> - ACPI_MEM_TRACKING (Cache->TotalAllocated++);
>>> + ACPI_MEM_TRACKING (Cache->TotalAllocated++);
>>>
>>> #ifdef ACPI_DBG_TRACK_ALLOCATIONS
>>> - if ((Cache->TotalAllocated - Cache->TotalFreed) > Cache-
>>> MaxOccupied)
>>> - {
>>> - Cache->MaxOccupied = Cache->TotalAllocated - Cache-
>>> TotalFreed;
>>> - }
>>> + if ((Cache->TotalAllocated - Cache->TotalFreed) > Cache-
>>> MaxOccupied)
>>> + {
>>> + Cache->MaxOccupied = Cache->TotalAllocated - Cache->TotalFreed;
>>> + }
>>> #endif
>>>
>>> - /* Avoid deadlock with ACPI_ALLOCATE_ZEROED */
>>> + /* Avoid deadlock with ACPI_ALLOCATE_ZEROED */
>>>
>>> - Status = AcpiUtReleaseMutex (ACPI_MTX_CACHES);
>>> - if (ACPI_FAILURE (Status))
>>> - {
>>> - return (NULL);
>>> - }
>>> -
>>> - Object = ACPI_ALLOCATE_ZEROED (Cache->ObjectSize);
>>> - if (!Object)
>>> - {
>>> - return (NULL);
>>> - }
>>> + Status = AcpiUtReleaseMutex (ACPI_MTX_CACHES);
>>> + if (ACPI_FAILURE (Status))
>>> + {
>>> + return (NULL);
>>> }
>>>
>>> - return (Object);
>>> + return ACPI_ALLOCATE_ZEROED (Cache->ObjectSize);
>>> }
>>> #endif /* ACPI_USE_LOCAL_CACHE */
>>>
>>>
>>> --
>>> To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
>>> the body of a message to majordomo@vger.kernel.org
>>> More majordomo info at http://vger.kernel.org/majordomo-info.html
>>>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
>
next prev parent reply other threads:[~2009-05-12 18:54 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-05-07 15:52 [PATCH 1/4] ACPICA: Remove use of caches in controlled way Alexey Starikovskiy
2009-05-07 15:52 ` [PATCH 2/4] ACPICA: Remove cache code Alexey Starikovskiy
2009-05-07 15:52 ` [PATCH 3/4] ACPICA: Drop Operand cache Alexey Starikovskiy
2009-05-07 15:52 ` [PATCH 4/4] ACPICA: Delete NextObject pointer Alexey Starikovskiy
2009-05-11 8:14 ` [PATCH 1/4] ACPICA: Remove use of caches in controlled way Lin Ming
2009-05-11 15:42 ` Moore, Robert
2009-05-12 18:54 ` Alexey Starikovskiy [this message]
2009-05-12 21:28 ` Moore, Robert
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=4A09C5D6.80409@gmail.com \
--to=aystarik@gmail.com \
--cc=Linux-acpi@vger.kernel.org \
--cc=astarikovskiy@suse.de \
--cc=ming.m.lin@intel.com \
--cc=robert.moore@intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox