public inbox for devel@edk2.groups.io
 help / color / mirror / Atom feed
From: Laszlo Ersek <lersek@redhat.com>
To: nkvangup <narendra.k.vanguput@intel.com>, edk2-devel@lists.01.org
Cc: Yao Jiewen <jiewen.yao@intel.com>, Eric Dong <eric.dong@intel.com>
Subject: Re: [PATCH v4] UefiCpuPkg\CpuSmm: Save & restore CR2 on-demand paging in SMM
Date: Wed, 20 Mar 2019 17:31:19 +0100	[thread overview]
Message-ID: <a45d4389-133e-876e-5720-6ed97061846a@redhat.com> (raw)
In-Reply-To: <20190318143825.17352-1-narendra.k.vanguput@intel.com>

On 03/18/19 15:38, nkvangup wrote:
> BZ: https://bugzilla.tianocore.org/show_bug.cgi?id=1593
> 
> For every SMI occurrence, save and restore CR2 register only when SMM
> on-demand paging support is enabled in 64 bit operation mode.
> 
> Contributed-under: TianoCore Contribution Agreement 1.1
> Signed-off-by: Vanguput Narendra K <narendra.k.vanguput@intel.com>
> Cc: Eric Dong <eric.dong@intel.com>
> Cc: Ray Ni <ray.ni@intel.com>
> Cc: Laszlo Ersek <lersek@redhat.com>
> Cc: Yao Jiewen <jiewen.yao@intel.com>
> ---
>  UefiCpuPkg/PiSmmCpuDxeSmm/MpService.c   | 22 ++++++++++++++--------
>  UefiCpuPkg/PiSmmCpuDxeSmm/X64/PageTbl.c |  2 +-
>  2 files changed, 15 insertions(+), 9 deletions(-)
> 
> diff --git a/UefiCpuPkg/PiSmmCpuDxeSmm/MpService.c b/UefiCpuPkg/PiSmmCpuDxeSmm/MpService.c
> index 3b0b3b52ac..0c07b31c4f 100644
> --- a/UefiCpuPkg/PiSmmCpuDxeSmm/MpService.c
> +++ b/UefiCpuPkg/PiSmmCpuDxeSmm/MpService.c
> @@ -28,6 +28,7 @@ UINTN                                       mSemaphoreSize;
>  SPIN_LOCK                                   *mPFLock = NULL;
>  SMM_CPU_SYNC_MODE                           mCpuSmmSyncMode;
>  BOOLEAN                                     mMachineCheckSupported = FALSE;
> +BOOLEAN                                     mCpuSmmStaticPageTable = TRUE;

Hmmm. This change is a bit daring, but I think it could be valid.

- In the IA32 build, mCpuSmmStaticPageTable would never be modified, or
read, by *preexistent* code (because all that code is in X64/PageTbl.c).
And the new code, added by this patch, would (presumably) work fine,
with the initial TRUE value.

- In the X64 build, the preexistent code would never read the initial
value (which we now set to TRUE here), i.e. before overwriting the
variable from the PCD -- because that would mean a bug in the
preexistent code. (Well, unless that code relied on the zero initial
value of the variable).

(1) I think I'd like to defer on this to other UefiCpuPkg reviewers.
Honestly I find this style questionable. It makes me feel uncomfortable.
I'd prefer the new APIs with the separate IA32/X64 implementations that
I suggested in my v2 review. But if other reviewers like this one
better, I won't mind.

(After hearing their opinions, I'd attempt to find the time to
regression test the patch (or maybe v5), too.)

Assuming other reviewers prefer this approach over my suggestion, I have
some other comments:

>  
>  /**
>    Performs an atomic compare exchange operation to get semaphore.
> @@ -1111,10 +1112,13 @@ SmiRendezvous (
>  
>    ASSERT(CpuIndex < mMaxNumberOfCpus);
>  
> -  //
> -  // Save Cr2 because Page Fault exception in SMM may override its value
> -  //
> -  Cr2 = AsmReadCr2 ();
> +    if (!mCpuSmmStaticPageTable) {
> +    //
> +    // Save and restore Cr2 when using on-demand paging for above 4G memory because Page Fault
> +    // exception in SMM may override its value
> +    //
> +    Cr2 = AsmReadCr2 ();
> +  }

(2) The indentation of the "if" is broken.

(3) Given that we're already using two comment lines, I'd suggest not
exceeding 80 characters per line.

>  
>    //
>    // Perform CPU specific entry hooks
> @@ -1253,10 +1257,12 @@ SmiRendezvous (
>  
>  Exit:
>    SmmCpuFeaturesRendezvousExit (CpuIndex);
> -  //
> -  // Restore Cr2
> -  //
> -  AsmWriteCr2 (Cr2);
> +    if (!mCpuSmmStaticPageTable) {

(4) same as (2).

> +    //
> +    // Restore Cr2
> +    //
> +    AsmWriteCr2 (Cr2);
> +  }
>  }
>  
>  /**
> diff --git a/UefiCpuPkg/PiSmmCpuDxeSmm/X64/PageTbl.c b/UefiCpuPkg/PiSmmCpuDxeSmm/X64/PageTbl.c
> index 2c77cb47a4..e444b8a031 100644
> --- a/UefiCpuPkg/PiSmmCpuDxeSmm/X64/PageTbl.c
> +++ b/UefiCpuPkg/PiSmmCpuDxeSmm/X64/PageTbl.c
> @@ -21,7 +21,7 @@ WITHOUT WARRANTIES OR REPRESENTATIONS OF ANY KIND, EITHER EXPRESS OR IMPLIED.
>  
>  LIST_ENTRY                          mPagePool = INITIALIZE_LIST_HEAD_VARIABLE (mPagePool);
>  BOOLEAN                             m1GPageTableSupport = FALSE;
> -BOOLEAN                             mCpuSmmStaticPageTable;
> +extern BOOLEAN                      mCpuSmmStaticPageTable;

(5) This is generally not great style, and it conflicts with the
existent code of this driver. Namely, declarations of variables with
file scope, static storage duration, and external linkage, should go
into "PiSmmCpuDxeSmm.h"-- we already got a bunch of them there.

Thanks
Laszlo

>  
>  /**
>    Disable CET.
> 



  reply	other threads:[~2019-03-20 16:31 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-03-18 14:38 [PATCH v4] UefiCpuPkg\CpuSmm: Save & restore CR2 on-demand paging in SMM nkvangup
2019-03-20 16:31 ` Laszlo Ersek [this message]
2019-03-20 17:27   ` Vanguput, Narendra K
2019-03-21  0:56     ` Dong, Eric
2019-03-21  2:48       ` Vanguput, Narendra K

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=a45d4389-133e-876e-5720-6ed97061846a@redhat.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