From: "Nickle Wang via groups.io" <nicklew=nvidia.com@groups.io>
To: M M <mike.maslenkin@gmail.com>, "Chang, Abner" <Abner.Chang@amd.com>
Cc: "devel@edk2.groups.io" <devel@edk2.groups.io>,
Igor Kulchytskyy <igork@ami.com>
Subject: Re: [edk2-devel] [edk2-redfish-client][PATCH V2] RedfishClientPkg/FeatureDriver:\r ComputerSystem_1_13_0 driver
Date: Tue, 30 Jan 2024 07:58:27 +0000 [thread overview]
Message-ID: <PH8PR12MB70257E4CC1EDFB0E9C30E2C4D97D2@PH8PR12MB7025.namprd12.prod.outlook.com> (raw)
In-Reply-To: <4907CA4D-C19B-4603-9EBF-1A102EFC215F@gmail.com>
[-- Attachment #1: Type: text/plain, Size: 6624 bytes --]
Hi Mike,
How about we add code to release “PendingSettingUri” at the end of RedfishResourceConsumeResource(). I think this would help to address memory leak issue. And we keep “Private->Uri” to be just the reference to “Uri” from caller and we don’t release “Private->Uri”. Caller always has responsibility to release memory it allocated.
Then, I like to modify below code to match my logic above:
//
// Initialize collection path
//
CollectionUri = RedfishGetUri (ResourceUri);
if (CollectionUri == NULL) {
ASSERT (FALSE);
FreePool (ResourceUri);
return EFI_OUT_OF_RESOURCES;
}
Status = HandleResource (Private, CollectionUri);
if (EFI_ERROR (Status)) {
DEBUG ((DEBUG_ERROR, "%a, process external resource: %a failed: %r\n", __func__, CollectionUri, Status));
}
FreePool (CollectionUri);
In this way, we don’ have huge changes to fix memory leak issue as you mentioned. (removing Private->Uri) Please let me know if this makes sense to you or not.
Thanks,
Nickle
From: M M <mike.maslenkin@gmail.com>
Sent: Friday, January 26, 2024 11:41 PM
To: Chang, Abner <Abner.Chang@amd.com>
Cc: devel@edk2.groups.io; Nickle Wang <nicklew@nvidia.com>; Igor Kulchytskyy <igork@ami.com>
Subject: Re: [edk2-devel] [edk2-redfish-client][PATCH V2] RedfishClientPkg/FeatureDriver: ComputerSystem_1_13_0 driver
External email: Use caution opening links or attachments
BTW did you consider to remove this Private->Uri someday at all ?
I tried to remove it, but since it declared in a common header for all redfish client feature drivers
those changes was huge comparing to a small memory leak I tried to fix.
And now such pattern spreads into new feature drivers.
In short: there is no need to cache this value in Private->Uri, but pass it as an argument down to stack in EDKII_REDFISH_RESOURCE_CONFIG_PROTOCOL implementation.
I think it just small design flaw.
The code I'm talking about:
+ //
+ // Check and see if "@Redfish.Settings" exist or not.
+ //
+ ZeroMem (&PendingSettingResponse, sizeof (REDFISH_RESPONSE));
+ Status = GetPendingSettings (
+ Private->RedfishService,
+ Response.Payload,
+ &PendingSettingResponse,
+ &PendingSettingUri
+ );
+ if (!EFI_ERROR (Status)) {
+ DEBUG ((REDFISH_DEBUG_TRACE, "%a: @Redfish.Settings found: %s\n", __func__, PendingSettingUri));
+ Private->Uri = PendingSettingUri;
+ ExpectedResponse = &PendingSettingResponse;
+ } else {
+ DEBUG ((REDFISH_DEBUG_TRACE, "%a: No @Redfish.Settings is found\n", __func__));
+ Private->Uri = Uri;
+ ExpectedResponse = &Response;
+ }
In this pattern PendingSettingUri is leaked, since Private->Uri belongs to caller and never released.
But... in fact it is not true for Feature/Bios driver, and actually leaked resource is original Private->Uri string.
This is because these EDKII_REDFISH_RESOURCE_CONFIG_PROTOCOL function called by
HandleResource() function, which in turns called from here:
//
// Initialize collection path
//
Private->Uri = RedfishGetUri (ResourceUri);
if (Private->Uri == NULL) {
ASSERT (FALSE);
FreePool (ResourceUri);
return EFI_OUT_OF_RESOURCES;
}
Status = HandleResource (Private, Private->Uri);
if (EFI_ERROR (Status)) {
DEBUG ((DEBUG_ERROR, "%a, process external resource: %a failed: %r\n", __func__, Private->Uri, Status));
}
FreePool (Private->Uri);
Too many "Private->Uri" with a different value.
PS: Funny, just pay attention to the log trace above - %a used for Private->Uri. Probably it was one of the factors
induced me to write previous message :)
Regards,
Mike.
On 26. 1. 2024., at 17:13, Chang, Abner <Abner.Chang@amd.com<mailto:Abner.Chang@amd.com>> wrote:
[AMD Official Use Only - General]
It is no problem. 😊 Thank you Mike for looking into this patch.
Abner
-----Original Message-----
From: M M <mike.maslenkin@gmail.com<mailto:mike.maslenkin@gmail.com>>
Sent: Friday, January 26, 2024 10:11 PM
To: Chang, Abner <Abner.Chang@amd.com<mailto:Abner.Chang@amd.com>>
Cc: devel@edk2.groups.io<mailto:devel@edk2.groups.io>; Nickle Wang <nicklew@nvidia.com<mailto:nicklew@nvidia.com>>; Igor
Kulchytskyy <igork@ami.com<mailto:igork@ami.com>>
Subject: Re: [edk2-devel] [edk2-redfish-client][PATCH V2]
RedfishClientPkg/FeatureDriver: ComputerSystem_1_13_0 driver
Caution: This message originated from an External Source. Use proper caution
when opening attachments, clicking links, or responding.
You are right.
Sorry for the noise!
No problems here.
I mixed up with tags in my editor while looked into patch.
Regards,
Mike
On 26. 1. 2024., at 17:03, Chang, Abner <Abner.Chang@amd.com<mailto:Abner.Chang@amd.com>> wrote:
[AMD Official Use Only - General]
Hi Mike,
I can't identify the issue on %s as Private->Uri is defined as EFI_STRING, or I
miss something?
Thanks
Abner
-----Original Message-----
From: M M <mike.maslenkin@gmail.com<mailto:mike.maslenkin@gmail.com>>
Sent: Friday, January 26, 2024 9:55 PM
To: devel@edk2.groups.io<mailto:devel@edk2.groups.io>; Nickle Wang <nicklew@nvidia.com<mailto:nicklew@nvidia.com>>
Cc: Chang, Abner <Abner.Chang@amd.com<mailto:Abner.Chang@amd.com>>; Igor Kulchytskyy
<igork@ami.com<mailto:igork@ami.com>>
Subject: Re: [edk2-devel] [edk2-redfish-client][PATCH V2]
RedfishClientPkg/FeatureDriver: ComputerSystem_1_13_0 driver
Caution: This message originated from an External Source. Use proper
caution
when opening attachments, clicking links, or responding.
Hi Abner,
On 26. 1. 2024., at 05:20, Nickle Wang via groups.io<http://groups.io/>
<nicklew=nvidia.com@groups.io<mailto:nicklew=nvidia.com@groups.io>> wrote:
Hi Abner,
Same minor issue as 1_5_0. Please add "%a:" to below DEBUG call.
+ DEBUG ((DEBUG_MANAGEABILITY, " No platform Redfish
ConfigureLang
found for %s\n", __func__, Private->Uri));
Regards,
Nickle
There are two issues in this string. The second one is incorrect %s format
used
for Private->Uri.
This issue exists in V3 not only in the line mentioned above.
Regards,
Mike.
-=-=-=-=-=-=-=-=-=-=-=-
Groups.io Links: You receive all messages sent to this group.
View/Reply Online (#114765): https://edk2.groups.io/g/devel/message/114765
Mute This Topic: https://groups.io/mt/103967446/7686176
Group Owner: devel+owner@edk2.groups.io
Unsubscribe: https://edk2.groups.io/g/devel/unsub [rebecca@openfw.io]
-=-=-=-=-=-=-=-=-=-=-=-
[-- Attachment #2: Type: text/html, Size: 23610 bytes --]
prev parent reply other threads:[~2024-01-30 7:58 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-01-26 1:00 [edk2-devel] [edk2-redfish-client][PATCH V2] RedfishClientPkg/FeatureDriver: ComputerSystem_1_13_0 driver Chang, Abner via groups.io
2024-01-26 2:20 ` Nickle Wang via groups.io
2024-01-26 2:32 ` Chang, Abner via groups.io
2024-01-26 13:55 ` Mike Maslenkin
2024-01-26 14:03 ` Chang, Abner via groups.io
2024-01-26 14:11 ` Mike Maslenkin
2024-01-26 14:13 ` Chang, Abner via groups.io
2024-01-26 15:41 ` Mike Maslenkin
2024-01-30 7:58 ` Nickle Wang via groups.io [this message]
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-list 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=PH8PR12MB70257E4CC1EDFB0E9C30E2C4D97D2@PH8PR12MB7025.namprd12.prod.outlook.com \
--to=devel@edk2.groups.io \
/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