1. Patchew
  2. EDK2
  3. View series

[edk2-devel] [edk2-staging][PATCH v2] BaseTools/Fmmt: Fix found the next FFS issue

Feng, YunhuaX posted 1 patch 3 years, 11 months ago
Failed in applying to current master (apply log)
BaseTools/Source/C/FMMT/FmmtLib.c | 7 +++++++
1 file changed, 7 insertions(+)
[edk2-devel] [edk2-staging][PATCH v2] BaseTools/Fmmt: Fix found the next FFS issue
Posted by Feng, YunhuaX 3 years, 11 months ago 
if the ffs size is invalid, break the iteration and return NOT_FOUND.

Cc: Ray Ni< ray.ni@intel.com>
Cc: Bob Feng <bob.c.feng@intel.com>
Cc: Liming Gao <liming.gao@intel.com>

Change-Id: I9e320d6176af350ff208901209f3f6c89e4e1924
Signed-off-by: Yunhua Feng <yunhuax.feng@intel.com>
---
 BaseTools/Source/C/FMMT/FmmtLib.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/BaseTools/Source/C/FMMT/FmmtLib.c b/BaseTools/Source/C/FMMT/FmmtLib.c
index 9ec511ef06..6858cfe778 100644
--- a/BaseTools/Source/C/FMMT/FmmtLib.c
+++ b/BaseTools/Source/C/FMMT/FmmtLib.c
@@ -1769,10 +1769,17 @@ FvBufFindNextFile (
       (*Key + sizeof (*fhdr)) < fvSize;
       *Key = (UINTN)ALIGN_POINTER (*Key, 8)
     ) {
     fhdr = (EFI_FFS_FILE_HEADER*) ((UINT8*)hdr + *Key);
     fsize = GetFfsFileLength (fhdr);
+    //
+    //if the ffs size is invalid, break the loop
+    //the size defined in EFI_FFS_FILE_HEADER is 3 bytes.
+    //
+    if (fsize == (UINTN)((FvbAttributes & EFI_FVB2_ERASE_POLARITY) ? 0xFFFFFF : 0x0)) {
+      break;
+    }
     if (!EFI_TEST_FFS_ATTRIBUTES_BIT(
           FvbAttributes,
           fhdr->State,
           EFI_FILE_HEADER_VALID
         ) ||
--
2.12.2.windows.2


-=-=-=-=-=-=-=-=-=-=-=-
Groups.io Links: You receive all messages sent to this group.

View/Reply Online (#58948): https://edk2.groups.io/g/devel/message/58948
Mute This Topic: https://groups.io/mt/74091408/1787277
Group Owner: devel+owner@edk2.groups.io
Unsubscribe: https://edk2.groups.io/g/devel/unsub  [importer@patchew.org]
-=-=-=-=-=-=-=-=-=-=-=-
Re: [edk2-devel] [edk2-staging][PATCH v2] BaseTools/Fmmt: Fix found the next FFS issue
Posted by Ni, Ray 3 years, 11 months ago 
Yunhua,
I understand that if the ffs size 0 should be specially handled because
otherwise the for-loop will be infinite.

But why do you need to check the size against 0xFFFFFF?

Why the break condition in for-loop " (*Key + sizeof (*fhdr)) < fvSize;" cannot
help when the fsize is too big such as 0xFFFFFF?

I suggest you explain what issue you want to fix and how this fixes the issue in
commit message in detail.

Thanks,
Ray

> -----Original Message-----
> From: Feng, YunhuaX <yunhuax.feng@intel.com>
> Sent: Saturday, May 9, 2020 3:03 PM
> To: devel@edk2.groups.io
> Cc: Ni, Ray <ray.ni@intel.com>; Gao, Liming <liming.gao@intel.com>; Feng, Bob C <bob.c.feng@intel.com>
> Subject: [edk2-staging][PATCH v2] BaseTools/Fmmt: Fix found the next FFS issue
> 
> if the ffs size is invalid, break the iteration and return NOT_FOUND.
> 
> Cc: Ray Ni< ray.ni@intel.com>
> Cc: Bob Feng <bob.c.feng@intel.com>
> Cc: Liming Gao <liming.gao@intel.com>
> 
> Change-Id: I9e320d6176af350ff208901209f3f6c89e4e1924
> Signed-off-by: Yunhua Feng <yunhuax.feng@intel.com>
> ---
>  BaseTools/Source/C/FMMT/FmmtLib.c | 7 +++++++
>  1 file changed, 7 insertions(+)
> 
> diff --git a/BaseTools/Source/C/FMMT/FmmtLib.c b/BaseTools/Source/C/FMMT/FmmtLib.c
> index 9ec511ef06..6858cfe778 100644
> --- a/BaseTools/Source/C/FMMT/FmmtLib.c
> +++ b/BaseTools/Source/C/FMMT/FmmtLib.c
> @@ -1769,10 +1769,17 @@ FvBufFindNextFile (
>        (*Key + sizeof (*fhdr)) < fvSize;
>        *Key = (UINTN)ALIGN_POINTER (*Key, 8)
>      ) {
>      fhdr = (EFI_FFS_FILE_HEADER*) ((UINT8*)hdr + *Key);
>      fsize = GetFfsFileLength (fhdr);
> +    //
> +    //if the ffs size is invalid, break the loop
> +    //the size defined in EFI_FFS_FILE_HEADER is 3 bytes.
> +    //
> +    if (fsize == (UINTN)((FvbAttributes & EFI_FVB2_ERASE_POLARITY) ? 0xFFFFFF : 0x0)) {
> +      break;
> +    }
>      if (!EFI_TEST_FFS_ATTRIBUTES_BIT(
>            FvbAttributes,
>            fhdr->State,
>            EFI_FILE_HEADER_VALID
>          ) ||
> --
> 2.12.2.windows.2
> 


-=-=-=-=-=-=-=-=-=-=-=-
Groups.io Links: You receive all messages sent to this group.

View/Reply Online (#58952): https://edk2.groups.io/g/devel/message/58952
Mute This Topic: https://groups.io/mt/74091408/1787277
Group Owner: devel+owner@edk2.groups.io
Unsubscribe: https://edk2.groups.io/g/devel/unsub  [importer@patchew.org]
-=-=-=-=-=-=-=-=-=-=-=-
Re: [edk2-devel] [edk2-staging][PATCH v2] BaseTools/Fmmt: Fix found the next FFS issue
Posted by Feng, YunhuaX 3 years, 11 months ago 
Hi Ray,
I encounter a FV image, behind valid FFS is filled with 0xff, and then filled with some 0x00.

In the for-loop, When the fsize == 0xFFFFFF, *key = *key + 1,  so not break.
when the fsize == 0x0, *key = *key + fsize, so it is dead loop.

I hope that if ffs size invalid, break the for-loop. 


-----Original Message-----
From: Ni, Ray <ray.ni@intel.com> 
Sent: Saturday, May 9, 2020 4:28 PM
To: Feng, YunhuaX <yunhuax.feng@intel.com>; devel@edk2.groups.io
Cc: Gao, Liming <liming.gao@intel.com>; Feng, Bob C <bob.c.feng@intel.com>
Subject: RE: [edk2-staging][PATCH v2] BaseTools/Fmmt: Fix found the next FFS issue

Yunhua,
I understand that if the ffs size 0 should be specially handled because otherwise the for-loop will be infinite.

But why do you need to check the size against 0xFFFFFF?

Why the break condition in for-loop " (*Key + sizeof (*fhdr)) < fvSize;" cannot help when the fsize is too big such as 0xFFFFFF?

I suggest you explain what issue you want to fix and how this fixes the issue in commit message in detail.

Thanks,
Ray

> -----Original Message-----
> From: Feng, YunhuaX <yunhuax.feng@intel.com>
> Sent: Saturday, May 9, 2020 3:03 PM
> To: devel@edk2.groups.io
> Cc: Ni, Ray <ray.ni@intel.com>; Gao, Liming <liming.gao@intel.com>; 
> Feng, Bob C <bob.c.feng@intel.com>
> Subject: [edk2-staging][PATCH v2] BaseTools/Fmmt: Fix found the next 
> FFS issue
> 
> if the ffs size is invalid, break the iteration and return NOT_FOUND.
> 
> Cc: Ray Ni< ray.ni@intel.com>
> Cc: Bob Feng <bob.c.feng@intel.com>
> Cc: Liming Gao <liming.gao@intel.com>
> 
> Change-Id: I9e320d6176af350ff208901209f3f6c89e4e1924
> Signed-off-by: Yunhua Feng <yunhuax.feng@intel.com>
> ---
>  BaseTools/Source/C/FMMT/FmmtLib.c | 7 +++++++
>  1 file changed, 7 insertions(+)
> 
> diff --git a/BaseTools/Source/C/FMMT/FmmtLib.c 
> b/BaseTools/Source/C/FMMT/FmmtLib.c
> index 9ec511ef06..6858cfe778 100644
> --- a/BaseTools/Source/C/FMMT/FmmtLib.c
> +++ b/BaseTools/Source/C/FMMT/FmmtLib.c
> @@ -1769,10 +1769,17 @@ FvBufFindNextFile (
>        (*Key + sizeof (*fhdr)) < fvSize;
>        *Key = (UINTN)ALIGN_POINTER (*Key, 8)
>      ) {
>      fhdr = (EFI_FFS_FILE_HEADER*) ((UINT8*)hdr + *Key);
>      fsize = GetFfsFileLength (fhdr);
> +    //
> +    //if the ffs size is invalid, break the loop
> +    //the size defined in EFI_FFS_FILE_HEADER is 3 bytes.
> +    //
> +    if (fsize == (UINTN)((FvbAttributes & EFI_FVB2_ERASE_POLARITY) ? 0xFFFFFF : 0x0)) {
> +      break;
> +    }
>      if (!EFI_TEST_FFS_ATTRIBUTES_BIT(
>            FvbAttributes,
>            fhdr->State,
>            EFI_FILE_HEADER_VALID
>          ) ||
> --
> 2.12.2.windows.2
> 



-=-=-=-=-=-=-=-=-=-=-=-
Groups.io Links: You receive all messages sent to this group.

View/Reply Online (#58953): https://edk2.groups.io/g/devel/message/58953
Mute This Topic: https://groups.io/mt/74091408/1787277
Group Owner: devel+owner@edk2.groups.io
Unsubscribe: https://edk2.groups.io/g/devel/unsub  [importer@patchew.org]
-=-=-=-=-=-=-=-=-=-=-=-

Patchew 2.1-devel-0870f84

© 2016 - 2024 Red Hat, Inc.